diff --git a/crates/core_arch/src/x86/avx2.rs b/crates/core_arch/src/x86/avx2.rs index 49c1c32bf1..346688bd75 100644 --- a/crates/core_arch/src/x86/avx2.rs +++ b/crates/core_arch/src/x86/avx2.rs @@ -3743,9 +3743,9 @@ pub unsafe fn _mm256_xor_si256(a: __m256i, b: __m256i) -> __m256i { // This intrinsic has no corresponding instruction. #[rustc_args_required_const(1)] #[stable(feature = "simd_x86", since = "1.27.0")] -pub unsafe fn _mm256_extract_epi8(a: __m256i, imm8: i32) -> i8 { +pub unsafe fn _mm256_extract_epi8(a: __m256i, imm8: i32) -> i32 { let imm8 = (imm8 & 31) as u32; - simd_extract(a.as_i8x32(), imm8) + simd_extract::<_, u8>(a.as_u8x32(), imm8) as i32 } /// Extracts a 16-bit integer from `a`, selected with `imm8`. Returns a 32-bit @@ -3759,9 +3759,9 @@ pub unsafe fn _mm256_extract_epi8(a: __m256i, imm8: i32) -> i8 { // This intrinsic has no corresponding instruction. #[rustc_args_required_const(1)] #[stable(feature = "simd_x86", since = "1.27.0")] -pub unsafe fn _mm256_extract_epi16(a: __m256i, imm8: i32) -> i16 { +pub unsafe fn _mm256_extract_epi16(a: __m256i, imm8: i32) -> i32 { let imm8 = (imm8 & 15) as u32; - simd_extract(a.as_i16x16(), imm8) + simd_extract::<_, u16>(a.as_u16x16(), imm8) as i32 } /// Extracts a 32-bit integer from `a`, selected with `imm8`. @@ -6115,7 +6115,7 @@ mod tests { ); let r1 = _mm256_extract_epi8(a, 0); let r2 = _mm256_extract_epi8(a, 35); - assert_eq!(r1, -1); + assert_eq!(r1, 0xFF); assert_eq!(r2, 3); } @@ -6128,7 +6128,7 @@ mod tests { ); let r1 = _mm256_extract_epi16(a, 0); let r2 = _mm256_extract_epi16(a, 19); - assert_eq!(r1, -1); + assert_eq!(r1, 0xFFFF); assert_eq!(r2, 3); } diff --git a/crates/core_arch/src/x86/sse2.rs b/crates/core_arch/src/x86/sse2.rs index 897f8ee675..2ee42732b6 100644 --- a/crates/core_arch/src/x86/sse2.rs +++ b/crates/core_arch/src/x86/sse2.rs @@ -1375,7 +1375,7 @@ pub unsafe fn _mm_packus_epi16(a: __m128i, b: __m128i) -> __m128i { #[rustc_args_required_const(1)] #[stable(feature = "simd_x86", since = "1.27.0")] pub unsafe fn _mm_extract_epi16(a: __m128i, imm8: i32) -> i32 { - simd_extract::<_, i16>(a.as_i16x8(), (imm8 & 7) as u32) as i32 + simd_extract::<_, u16>(a.as_u16x8(), (imm8 & 7) as u32) as i32 } /// Returns a new vector where the `imm8` element of `a` is replaced with `i`. @@ -4132,7 +4132,7 @@ mod tests { let a = _mm_setr_epi16(-1, 1, 2, 3, 4, 5, 6, 7); let r1 = _mm_extract_epi16(a, 0); let r2 = _mm_extract_epi16(a, 11); - assert_eq!(r1, -1); + assert_eq!(r1, 0xFFFF); assert_eq!(r2, 3); } diff --git a/crates/stdarch-verify/tests/x86-intel.rs b/crates/stdarch-verify/tests/x86-intel.rs index bf8ede6071..ea3c989e14 100644 --- a/crates/stdarch-verify/tests/x86-intel.rs +++ b/crates/stdarch-verify/tests/x86-intel.rs @@ -95,7 +95,8 @@ struct Data { #[derive(Deserialize)] struct Intrinsic { - rettype: String, + #[serde(rename = "return")] + return_: Return, name: String, #[serde(rename = "CPUID", default)] cpuid: Vec, @@ -111,6 +112,12 @@ struct Parameter { type_: String, } +#[derive(Deserialize)] +struct Return { + #[serde(rename = "type")] + type_: String, +} + #[derive(Deserialize, Debug)] struct Instruction { name: String, @@ -503,12 +510,12 @@ fn matches(rust: &Function, intel: &Intrinsic) -> Result<(), String> { // Make sure we've got the right return type. if let Some(t) = rust.ret { - equate(t, &intel.rettype, rust.name, false)?; - } else if intel.rettype != "" && intel.rettype != "void" { + equate(t, &intel.return_.type_, rust.name, false)?; + } else if intel.return_.type_ != "" && intel.return_.type_ != "void" { bail!( "{} returns `{}` with intel, void in rust", rust.name, - intel.rettype + intel.return_.type_ ) } diff --git a/crates/stdarch-verify/x86-intel.xml b/crates/stdarch-verify/x86-intel.xml index c22a3adaec..1484c3c4c1 100644 --- a/crates/stdarch-verify/x86-intel.xml +++ b/crates/stdarch-verify/x86-intel.xml @@ -1,40085 +1,45623 @@ - - + + Integer - MMX - Convert - - Copy 64-bit integer "a" to "dst". + Flag + ADX + Arithmetic + + + + + + Add unsigned 32-bit integers "a" and "b" with unsigned 8-bit carry-in "c_in" (carry or overflow flag), and store the unsigned 32-bit result in "out", and the carry-out in "dst" (carry or overflow flag). -dst[63:0] := a[63:0] +tmp[32:0] := a[31:0] + b[31:0] + (c_in > 0 ? 1 : 0) +MEM[out+31:out] := tmp[31:0] +dst[0] := tmp[32] +dst[7:1] := 0 - -
mmintrin.h
+ + +
immintrin.h
 - + Integer - MMX - Convert - - Copy 64-bit integer "a" to "dst". + Flag + ADX + Arithmetic + + + + + + Add unsigned 64-bit integers "a" and "b" with unsigned 8-bit carry-in "c_in" (carry or overflow flag), and store the unsigned 64-bit result in "out", and the carry-out in "dst" (carry or overflow flag). -dst[63:0] := a[63:0] +tmp[64:0] := a[63:0] + b[63:0] + (c_in > 0 ? 1 : 0) +MEM[out+63:out] := tmp[63:0] +dst[0] := tmp[64] +dst[7:1] := 0 - -
mmintrin.h
- - - MMX - General Support - - Empty the MMX state, which marks the x87 FPU registers as available for use by x87 instructions. This instruction must be used at the end of all MMX technology procedures. - -
mmintrin.h
+ + +
immintrin.h
 - + Integer - MMX - Convert - - Copy 32-bit integer "a" to the lower elements of "dst", and zero the upper element of "dst". - -dst[31:0] := a[31:0] -dst[63:32] := 0 + AES + Cryptography + + + + Perform one round of an AES encryption flow on data (state) in "a" using the round key in "RoundKey", and store the result in "dst"." + a[127:0] := ShiftRows(a[127:0]) +a[127:0] := SubBytes(a[127:0]) +a[127:0] := MixColumns(a[127:0]) +dst[127:0] := a[127:0] XOR RoundKey[127:0] - -
mmintrin.h
+ +
wmmintrin.h
 - + Integer - MMX - Convert - - Copy the lower 32-bit integer in "a" to "dst". - -dst[31:0] := a[31:0] + AES + Cryptography + + + + Perform the last round of an AES encryption flow on data (state) in "a" using the round key in "RoundKey", and store the result in "dst"." + a[127:0] := ShiftRows(a[127:0]) +a[127:0] := SubBytes(a[127:0]) +dst[127:0] := a[127:0] XOR RoundKey[127:0] - -
mmintrin.h
+ +
wmmintrin.h
 - + Integer - MMX - Miscellaneous - - - Convert packed 16-bit integers from "a" and "b" to packed 8-bit integers using signed saturation, and store the results in "dst". - - -dst[7:0] := Saturate_Int16_To_Int8 (a[15:0]) -dst[15:8] := Saturate_Int16_To_Int8 (a[31:16]) -dst[23:16] := Saturate_Int16_To_Int8 (a[47:32]) -dst[31:24] := Saturate_Int16_To_Int8 (a[63:48]) -dst[39:32] := Saturate_Int16_To_Int8 (b[15:0]) -dst[47:40] := Saturate_Int16_To_Int8 (b[31:16]) -dst[55:48] := Saturate_Int16_To_Int8 (b[47:32]) -dst[63:56] := Saturate_Int16_To_Int8 (b[63:48]) - - -
mmintrin.h
+ AES + Cryptography + + + + Perform one round of an AES decryption flow on data (state) in "a" using the round key in "RoundKey", and store the result in "dst". + a[127:0] := InvShiftRows(a[127:0]) +a[127:0] := InvSubBytes(a[127:0]) +a[127:0] := InvMixColumns(a[127:0]) +dst[127:0] := a[127:0] XOR RoundKey[127:0] + + +
wmmintrin.h
 - + Integer - MMX - Miscellaneous - - - Convert packed 32-bit integers from "a" and "b" to packed 16-bit integers using signed saturation, and store the results in "dst". - -dst[15:0] := Saturate_Int32_To_Int16 (a[31:0]) -dst[31:16] := Saturate_Int32_To_Int16 (a[63:32]) -dst[47:32] := Saturate_Int32_To_Int16 (b[31:0]) -dst[63:48] := Saturate_Int32_To_Int16 (b[63:32]) + AES + Cryptography + + + + Perform the last round of an AES decryption flow on data (state) in "a" using the round key in "RoundKey", and store the result in "dst". + a[127:0] := InvShiftRows(a[127:0]) +a[127:0] := InvSubBytes(a[127:0]) +dst[127:0] := a[127:0] XOR RoundKey[127:0] - -
mmintrin.h
+ +
wmmintrin.h
 - + Integer - MMX - Miscellaneous - - - Convert packed 16-bit integers from "a" and "b" to packed 8-bit integers using unsigned saturation, and store the results in "dst". - -dst[7:0] := Saturate_Int16_To_UnsignedInt8 (a[15:0]) -dst[15:8] := Saturate_Int16_To_UnsignedInt8 (a[31:16]) -dst[23:16] := Saturate_Int16_To_UnsignedInt8 (a[47:32]) -dst[31:24] := Saturate_Int16_To_UnsignedInt8 (a[63:48]) -dst[39:32] := Saturate_Int16_To_UnsignedInt8 (b[15:0]) -dst[47:40] := Saturate_Int16_To_UnsignedInt8 (b[31:16]) -dst[55:48] := Saturate_Int16_To_UnsignedInt8 (b[47:32]) -dst[63:56] := Saturate_Int16_To_UnsignedInt8 (b[63:48]) + AES + Cryptography + + + Perform the InvMixColumns transformation on "a" and store the result in "dst". + dst[127:0] := InvMixColumns(a[127:0]) - -
mmintrin.h
+ +
wmmintrin.h
 - + Integer - MMX - Swizzle - - - Unpack and interleave 8-bit integers from the high half of "a" and "b", and store the results in "dst". - -INTERLEAVE_HIGH_BYTES(src1[63:0], src2[63:0]){ - dst[7:0] := src1[39:32] - dst[15:8] := src2[39:32] - dst[23:16] := src1[47:40] - dst[31:24] := src2[47:40] - dst[39:32] := src1[55:48] - dst[47:40] := src2[55:48] - dst[55:48] := src1[63:56] - dst[63:56] := src2[63:56] - RETURN dst[63:0] -} - -dst[63:0] := INTERLEAVE_HIGH_BYTES(a[63:0], b[63:0]) + AES + Cryptography + + + + Assist in expanding the AES cipher key by computing steps towards generating a round key for encryption cipher using data from "a" and an 8-bit round constant specified in "imm8", and store the result in "dst"." + X3[31:0] := a[127:96] +X2[31:0] := a[95:64] +X1[31:0] := a[63:32] +X0[31:0] := a[31:0] +RCON[31:0] := ZeroExtend32(imm8[7:0]) +dst[31:0] := SubWord(X1) +dst[63:32] := RotWord(SubWord(X1)) XOR RCON +dst[95:64] := SubWord(X3) +dst[127:96] := RotWord(SubWord(X3)) XOR RCON - -
mmintrin.h
+ +
wmmintrin.h
 - - Integer - MMX - Swizzle - - - Unpack and interleave 16-bit integers from the high half of "a" and "b", and store the results in "dst". + + Floating Point + AVX + Arithmetic + + + + Add packed double-precision (64-bit) floating-point elements in "a" and "b", and store the results in "dst". -INTERLEAVE_HIGH_WORDS(src1[63:0], src2[63:0]){ - dst[15:0] := src1[47:32] - dst[31:16] := src2[47:32] - dst[47:32] := src1[63:48] - dst[63:48] := src2[63:48] - RETURN dst[63:0] -} - -dst[63:0] := INTERLEAVE_HIGH_WORDS(a[63:0], b[63:0]) +FOR j := 0 to 3 + i := j*64 + dst[i+63:i] := a[i+63:i] + b[i+63:i] +ENDFOR +dst[MAX:256] := 0 - -
mmintrin.h
+ +
immintrin.h
 - - Integer - MMX - Swizzle - - - Unpack and interleave 32-bit integers from the high half of "a" and "b", and store the results in "dst". + + Floating Point + AVX + Arithmetic + + + + Add packed single-precision (32-bit) floating-point elements in "a" and "b", and store the results in "dst". -dst[31:0] := a[63:32] -dst[63:32] := b[63:32] +FOR j := 0 to 7 + i := j*32 + dst[i+31:i] := a[i+31:i] + b[i+31:i] +ENDFOR +dst[MAX:256] := 0 - -
mmintrin.h
+ +
immintrin.h
 - - Integer - MMX - Swizzle - - - Unpack and interleave 8-bit integers from the low half of "a" and "b", and store the results in "dst". + + Floating Point + AVX + Arithmetic + + + + Alternatively add and subtract packed double-precision (64-bit) floating-point elements in "a" to/from packed elements in "b", and store the results in "dst". -INTERLEAVE_BYTES(src1[63:0], src2[63:0]){ - dst[7:0] := src1[7:0] - dst[15:8] := src2[7:0] - dst[23:16] := src1[15:8] - dst[31:24] := src2[15:8] - dst[39:32] := src1[23:16] - dst[47:40] := src2[23:16] - dst[55:48] := src1[31:24] - dst[63:56] := src2[31:24] - RETURN dst[63:0] -} - -dst[63:0] := INTERLEAVE_BYTES(a[63:0], b[63:0]) +FOR j := 0 to 3 + i := j*64 + IF ((j & 1) == 0) + dst[i+63:i] := a[i+63:i] - b[i+63:i] + ELSE + dst[i+63:i] := a[i+63:i] + b[i+63:i] + FI +ENDFOR +dst[MAX:256] := 0 - -
mmintrin.h
+ +
immintrin.h
 - - Integer - MMX - Swizzle - - - Unpack and interleave 16-bit integers from the low half of "a" and "b", and store the results in "dst". + + Floating Point + AVX + Arithmetic + + + + Alternatively add and subtract packed single-precision (32-bit) floating-point elements in "a" to/from packed elements in "b", and store the results in "dst". -INTERLEAVE_WORDS(src1[63:0], src2[63:0]){ - dst[15:0] := src1[15:0] - dst[31:16] := src2[15:0] - dst[47:32] := src1[31:16] - dst[63:48] := src2[31:16] - RETURN dst[63:0] -} - -dst[63:0] := INTERLEAVE_WORDS(a[63:0], b[63:0]) +FOR j := 0 to 7 + i := j*32 + IF ((j & 1) == 0) + dst[i+31:i] := a[i+31:i] - b[i+31:i] + ELSE + dst[i+31:i] := a[i+31:i] + b[i+31:i] + FI +ENDFOR +dst[MAX:256] := 0 - -
mmintrin.h
+ +
immintrin.h
 - - Integer - MMX - Swizzle - - - Unpack and interleave 32-bit integers from the low half of "a" and "b", and store the results in "dst". + + Floating Point + AVX + Logical + + + + Compute the bitwise AND of packed double-precision (64-bit) floating-point elements in "a" and "b", and store the results in "dst". -dst[31:0] := a[31:0] -dst[63:32] := b[31:0] +FOR j := 0 to 3 + i := j*64 + dst[i+63:i] := (a[i+63:i] AND b[i+63:i]) +ENDFOR +dst[MAX:256] := 0 - -
mmintrin.h
+ +
immintrin.h
 - - Integer - MMX - Arithmetic - - - Add packed 8-bit integers in "a" and "b", and store the results in "dst". + + Floating Point + AVX + Logical + + + + Compute the bitwise AND of packed single-precision (32-bit) floating-point elements in "a" and "b", and store the results in "dst". FOR j := 0 to 7 - i := j*8 - dst[i+7:i] := a[i+7:i] + b[i+7:i] + i := j*32 + dst[i+31:i] := (a[i+31:i] AND b[i+31:i]) ENDFOR +dst[MAX:256] := 0 - -
mmintrin.h
+ +
immintrin.h
 - - Integer - MMX - Arithmetic - - - Add packed 16-bit integers in "a" and "b", and store the results in "dst". + + Floating Point + AVX + Logical + + + + Compute the bitwise NOT of packed double-precision (64-bit) floating-point elements in "a" and then AND with "b", and store the results in "dst". FOR j := 0 to 3 - i := j*16 - dst[i+15:i] := a[i+15:i] + b[i+15:i] + i := j*64 + dst[i+63:i] := ((NOT a[i+63:i]) AND b[i+63:i]) ENDFOR +dst[MAX:256] := 0 - -
mmintrin.h
+ +
immintrin.h
 - - Integer - MMX - Arithmetic - - - Add packed 32-bit integers in "a" and "b", and store the results in "dst". + + Floating Point + AVX + Logical + + + + Compute the bitwise NOT of packed single-precision (32-bit) floating-point elements in "a" and then AND with "b", and store the results in "dst". -FOR j := 0 to 1 +FOR j := 0 to 7 i := j*32 - dst[i+31:i] := a[i+31:i] + b[i+31:i] + dst[i+31:i] := ((NOT a[i+31:i]) AND b[i+31:i]) ENDFOR +dst[MAX:256] := 0 - -
mmintrin.h
+ +
immintrin.h
 - - Integer - MMX - Arithmetic - - - Add packed 8-bit integers in "a" and "b" using saturation, and store the results in "dst". + + Floating Point + AVX + Swizzle + + + + + Blend packed double-precision (64-bit) floating-point elements from "a" and "b" using control mask "imm8", and store the results in "dst". + +FOR j := 0 to 3 + i := j*64 + IF imm8[j] + dst[i+63:i] := b[i+63:i] + ELSE + dst[i+63:i] := a[i+63:i] + FI +ENDFOR +dst[MAX:256] := 0 + + +
immintrin.h
+ + + Floating Point + AVX + Swizzle + + + + + Blend packed single-precision (32-bit) floating-point elements from "a" and "b" using control mask "imm8", and store the results in "dst". FOR j := 0 to 7 - i := j*8 - dst[i+7:i] := Saturate_To_Int8( a[i+7:i] + b[i+7:i] ) + i := j*32 + IF imm8[j] + dst[i+31:i] := b[i+31:i] + ELSE + dst[i+31:i] := a[i+31:i] + FI ENDFOR +dst[MAX:256] := 0 - -
mmintrin.h
+ +
immintrin.h
 - - Integer - MMX - Arithmetic - - - Add packed 16-bit integers in "a" and "b" using saturation, and store the results in "dst". + + Floating Point + AVX + Swizzle + + + + + Blend packed double-precision (64-bit) floating-point elements from "a" and "b" using "mask", and store the results in "dst". FOR j := 0 to 3 - i := j*16 - dst[i+15:i] := Saturate_To_Int16( a[i+15:i] + b[i+15:i] ) + i := j*64 + IF mask[i+63] + dst[i+63:i] := b[i+63:i] + ELSE + dst[i+63:i] := a[i+63:i] + FI ENDFOR +dst[MAX:256] := 0 - -
mmintrin.h
+ +
immintrin.h
 - - Integer - MMX - Arithmetic - - - Add packed unsigned 8-bit integers in "a" and "b" using saturation, and store the results in "dst". + + Floating Point + AVX + Swizzle + + + + + Blend packed single-precision (32-bit) floating-point elements from "a" and "b" using "mask", and store the results in "dst". FOR j := 0 to 7 - i := j*8 - dst[i+7:i] := Saturate_To_UnsignedInt8( a[i+7:i] + b[i+7:i] ) + i := j*32 + IF mask[i+31] + dst[i+31:i] := b[i+31:i] + ELSE + dst[i+31:i] := a[i+31:i] + FI ENDFOR +dst[MAX:256] := 0 - -
mmintrin.h
+ +
immintrin.h
 - - Integer - MMX + + Floating Point + AVX Arithmetic - - - Add packed unsigned 16-bit integers in "a" and "b" using saturation, and store the results in "dst". + + + + Divide packed double-precision (64-bit) floating-point elements in "a" by packed elements in "b", and store the results in "dst". FOR j := 0 to 3 - i := j*16 - dst[i+15:i] := Saturate_To_UnsignedInt16( a[i+15:i] + b[i+15:i] ) + i := 64*j + dst[i+63:i] := a[i+63:i] / b[i+63:i] ENDFOR +dst[MAX:256] := 0 - -
mmintrin.h
+ +
immintrin.h
 - + Floating Point - Integer - MMX + AVX Arithmetic - - - Subtract packed 8-bit integers in "b" from packed 8-bit integers in "a", and store the results in "dst". + + + + Divide packed single-precision (32-bit) floating-point elements in "a" by packed elements in "b", and store the results in "dst". FOR j := 0 to 7 - i := j*8 - dst[i+7:i] := a[i+7:i] - b[i+7:i] + i := 32*j + dst[i+31:i] := a[i+31:i] / b[i+31:i] ENDFOR +dst[MAX:256] := 0 - -
mmintrin.h
+ +
immintrin.h
 - + Floating Point - Integer - MMX + AVX Arithmetic - - - Subtract packed 16-bit integers in "b" from packed 16-bit integers in "a", and store the results in "dst". + + + + + Conditionally multiply the packed single-precision (32-bit) floating-point elements in "a" and "b" using the high 4 bits in "imm8", sum the four products, and conditionally store the sum in "dst" using the low 4 bits of "imm8". -FOR j := 0 to 3 - i := j*16 - dst[i+15:i] := a[i+15:i] - b[i+15:i] -ENDFOR +DEFINE DP(a[127:0], b[127:0], imm8[7:0]) { + FOR j := 0 to 3 + i := j*32 + IF imm8[(4+j)%8] + temp[i+31:i] := a[i+31:i] * b[i+31:i] + ELSE + temp[i+31:i] := FP32(0.0) + FI + ENDFOR + + sum[31:0] := (temp[127:96] + temp[95:64]) + (temp[63:32] + temp[31:0]) + + FOR j := 0 to 3 + i := j*32 + IF imm8[j%8] + tmpdst[i+31:i] := sum[31:0] + ELSE + tmpdst[i+31:i] := FP32(0.0) + FI + ENDFOR + RETURN tmpdst[127:0] +} +dst[127:0] := DP(a[127:0], b[127:0], imm8[7:0]) +dst[255:128] := DP(a[255:128], b[255:128], imm8[7:0]) +dst[MAX:256] := 0 - -
mmintrin.h
+ +
immintrin.h
 - + Floating Point - Integer - MMX + AVX Arithmetic - - - Subtract packed 32-bit integers in "b" from packed 32-bit integers in "a", and store the results in "dst". + + + + Horizontally add adjacent pairs of double-precision (64-bit) floating-point elements in "a" and "b", and pack the results in "dst". -FOR j := 0 to 1 - i := j*32 - dst[i+31:i] := a[i+31:i] - b[i+31:i] -ENDFOR +dst[63:0] := a[127:64] + a[63:0] +dst[127:64] := b[127:64] + b[63:0] +dst[191:128] := a[255:192] + a[191:128] +dst[255:192] := b[255:192] + b[191:128] +dst[MAX:256] := 0 - -
mmintrin.h
+ +
immintrin.h
 - + Floating Point - Integer - MMX + AVX Arithmetic - - - Subtract packed 8-bit integers in "b" from packed 8-bit integers in "a" using saturation, and store the results in "dst". + + + + Horizontally add adjacent pairs of single-precision (32-bit) floating-point elements in "a" and "b", and pack the results in "dst". -FOR j := 0 to 7 - i := j*8 - dst[i+7:i] := Saturate_To_Int8(a[i+7:i] - b[i+7:i]) -ENDFOR +dst[31:0] := a[63:32] + a[31:0] +dst[63:32] := a[127:96] + a[95:64] +dst[95:64] := b[63:32] + b[31:0] +dst[127:96] := b[127:96] + b[95:64] +dst[159:128] := a[191:160] + a[159:128] +dst[191:160] := a[255:224] + a[223:192] +dst[223:192] := b[191:160] + b[159:128] +dst[255:224] := b[255:224] + b[223:192] +dst[MAX:256] := 0 - -
mmintrin.h
+ +
immintrin.h
 - + Floating Point - Integer - MMX + AVX Arithmetic - - - Subtract packed 16-bit integers in "b" from packed 16-bit integers in "a" using saturation, and store the results in "dst". + + + + Horizontally subtract adjacent pairs of double-precision (64-bit) floating-point elements in "a" and "b", and pack the results in "dst". -FOR j := 0 to 3 - i := j*16 - dst[i+15:i] := Saturate_To_Int16(a[i+15:i] - b[i+15:i]) -ENDFOR +dst[63:0] := a[63:0] - a[127:64] +dst[127:64] := b[63:0] - b[127:64] +dst[191:128] := a[191:128] - a[255:192] +dst[255:192] := b[191:128] - b[255:192] +dst[MAX:256] := 0 - -
mmintrin.h
+ +
immintrin.h
 - + Floating Point - Integer - MMX + AVX Arithmetic - - - Subtract packed unsigned 8-bit integers in "b" from packed unsigned 8-bit integers in "a" using saturation, and store the results in "dst". + + + + Horizontally add adjacent pairs of single-precision (32-bit) floating-point elements in "a" and "b", and pack the results in "dst". -FOR j := 0 to 7 - i := j*8 - dst[i+7:i] := Saturate_To_UnsignedInt8(a[i+7:i] - b[i+7:i]) -ENDFOR +dst[31:0] := a[31:0] - a[63:32] +dst[63:32] := a[95:64] - a[127:96] +dst[95:64] := b[31:0] - b[63:32] +dst[127:96] := b[95:64] - b[127:96] +dst[159:128] := a[159:128] - a[191:160] +dst[191:160] := a[223:192] - a[255:224] +dst[223:192] := b[159:128] - b[191:160] +dst[255:224] := b[223:192] - b[255:224] +dst[MAX:256] := 0 - -
mmintrin.h
+ +
immintrin.h
 - + Floating Point - Integer - MMX - Arithmetic - - - Subtract packed unsigned 16-bit integers in "b" from packed unsigned 16-bit integers in "a" using saturation, and store the results in "dst". + AVX + Special Math Functions + + + + Compare packed double-precision (64-bit) floating-point elements in "a" and "b", and store packed maximum values in "dst". FOR j := 0 to 3 - i := j*16 - dst[i+15:i] := Saturate_To_UnsignedInt16(a[i+15:i] - b[i+15:i]) + i := j*64 + dst[i+63:i] := MAX(a[i+63:i], b[i+63:i]) ENDFOR +dst[MAX:256] := 0 - -
mmintrin.h
+ +
immintrin.h
 - - Integer - MMX - Arithmetic - - - Multiply packed 16-bit integers in "a" and "b", producing intermediate 32-bit integers. Horizontally add adjacent pairs of intermediate 32-bit integers, and pack the results in "dst". - + + Floating Point + AVX + Special Math Functions + + + + Compare packed single-precision (32-bit) floating-point elements in "a" and "b", and store packed maximum values in "dst". -FOR j := 0 to 1 +FOR j := 0 to 7 i := j*32 - st[i+31:i] := a[i+31:i+16]*b[i+31:i+16] + a[i+15:i]*b[i+15:i] + dst[i+31:i] := MAX(a[i+31:i], b[i+31:i]) ENDFOR +dst[MAX:256] := 0 - -
mmintrin.h
+ +
immintrin.h
 - - Integer - MMX - Arithmetic - - - Multiply the packed 16-bit integers in "a" and "b", producing intermediate 32-bit integers, and store the high 16 bits of the intermediate integers in "dst". + + Floating Point + AVX + Special Math Functions + + + + Compare packed double-precision (64-bit) floating-point elements in "a" and "b", and store packed minimum values in "dst". FOR j := 0 to 3 - i := j*16 - tmp[31:0] := a[i+15:i] * b[i+15:i] - dst[i+15:i] := tmp[31:16] + i := j*64 + dst[i+63:i] := MIN(a[i+63:i], b[i+63:i]) ENDFOR +dst[MAX:256] := 0 - -
mmintrin.h
+ +
immintrin.h
 - - Integer - MMX - Arithmetic - - - Multiply the packed 16-bit integers in "a" and "b", producing intermediate 32-bit integers, and store the low 16 bits of the intermediate integers in "dst". + + Floating Point + AVX + Special Math Functions + + + + Compare packed single-precision (32-bit) floating-point elements in "a" and "b", and store packed minimum values in "dst". -FOR j := 0 to 3 - i := j*16 - tmp[31:0] := a[i+15:i] * b[i+15:i] - dst[i+15:i] := tmp[15:0] +FOR j := 0 to 7 + i := j*32 + dst[i+31:i] := MIN(a[i+31:i], b[i+31:i]) ENDFOR +dst[MAX:256] := 0 - -
mmintrin.h
+ +
immintrin.h
 - + Floating Point - Integer - MMX - Shift - - - Shift packed 16-bit integers in "a" left by "count" while shifting in zeros, and store the results in "dst". + AVX + Arithmetic + + + + Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", and store the results in "dst". FOR j := 0 to 3 - i := j*16 - IF count[63:0] > 15 - dst[i+15:i] := 0 - ELSE - dst[i+15:i] := ZeroExtend(a[i+15:i] << count[63:0]) - FI + i := j*64 + dst[i+63:i] := a[i+63:i] * b[i+63:i] ENDFOR +dst[MAX:256] := 0 - -
mmintrin.h
+ +
immintrin.h
 - + Floating Point - Integer - MMX - Shift - - - Shift packed 16-bit integers in "a" left by "imm8" while shifting in zeros, and store the results in "dst". + AVX + Arithmetic + + + + Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", and store the results in "dst". -FOR j := 0 to 3 - i := j*16 - IF imm8[7:0] > 15 - dst[i+15:i] := 0 - ELSE - dst[i+15:i] := ZeroExtend(a[i+15:i] << imm8[7:0]) - FI +FOR j := 0 to 7 + i := j*32 + dst[i+31:i] := a[i+31:i] * b[i+31:i] ENDFOR +dst[MAX:256] := 0 - -
mmintrin.h
+ +
immintrin.h
 - + Floating Point - Integer - MMX - Shift - - - Shift packed 32-bit integers in "a" left by "count" while shifting in zeros, and store the results in "dst". + AVX + Logical + + + + Compute the bitwise OR of packed double-precision (64-bit) floating-point elements in "a" and "b", and store the results in "dst". -FOR j := 0 to 1 - i := j*32 - IF count[63:0] > 31 - dst[i+31:i] := 0 - ELSE - dst[i+31:i] := ZeroExtend(a[i+31:i] << count[63:0]) - FI +FOR j := 0 to 3 + i := j*64 + dst[i+63:i] := a[i+63:i] OR b[i+63:i] ENDFOR +dst[MAX:256] := 0 - -
mmintrin.h
+ +
immintrin.h
 - + Floating Point - Integer - MMX - Shift - - - Shift packed 32-bit integers in "a" left by "imm8" while shifting in zeros, and store the results in "dst". + AVX + Logical + + + + Compute the bitwise OR of packed single-precision (32-bit) floating-point elements in "a" and "b", and store the results in "dst". -FOR j := 0 to 1 +FOR j := 0 to 7 i := j*32 - IF imm8[7:0] > 31 - dst[i+31:i] := 0 - ELSE - dst[i+31:i] := ZeroExtend(a[i+31:i] << imm8[7:0]) - FI + dst[i+31:i] := a[i+31:i] OR b[i+31:i] ENDFOR +dst[MAX:256] := 0 - -
mmintrin.h
+ +
immintrin.h
 - + Floating Point - Integer - MMX - Shift - - - Shift 64-bit integer "a" left by "count" while shifting in zeros, and store the result in "dst". + AVX + Swizzle + + + + + Shuffle double-precision (64-bit) floating-point elements within 128-bit lanes using the control in "imm8", and store the results in "dst". -IF count[63:0] > 63 - dst[63:0] := 0 -ELSE - dst[63:0] := ZeroExtend(a[63:0] << count[63:0]) -FI +dst[63:0] := (imm8[0] == 0) ? a[63:0] : a[127:64] +dst[127:64] := (imm8[1] == 0) ? b[63:0] : b[127:64] +dst[191:128] := (imm8[2] == 0) ? a[191:128] : a[255:192] +dst[255:192] := (imm8[3] == 0) ? b[191:128] : b[255:192] +dst[MAX:256] := 0 - -
mmintrin.h
+ +
immintrin.h
 - + Floating Point - Integer - MMX - Shift - - - Shift 64-bit integer "a" left by "imm8" while shifting in zeros, and store the result in "dst". + AVX + Swizzle + + + + + Shuffle single-precision (32-bit) floating-point elements in "a" within 128-bit lanes using the control in "imm8", and store the results in "dst". -IF imm8[7:0] > 63 - dst[63:0] := 0 -ELSE - dst[63:0] := ZeroExtend(a[63:0] << imm8[7:0]) -FI +DEFINE SELECT4(src, control) { + CASE(control[1:0]) OF + 0: tmp[31:0] := src[31:0] + 1: tmp[31:0] := src[63:32] + 2: tmp[31:0] := src[95:64] + 3: tmp[31:0] := src[127:96] + ESAC + RETURN tmp[31:0] +} +dst[31:0] := SELECT4(a[127:0], imm8[1:0]) +dst[63:32] := SELECT4(a[127:0], imm8[3:2]) +dst[95:64] := SELECT4(b[127:0], imm8[5:4]) +dst[127:96] := SELECT4(b[127:0], imm8[7:6]) +dst[159:128] := SELECT4(a[255:128], imm8[1:0]) +dst[191:160] := SELECT4(a[255:128], imm8[3:2]) +dst[223:192] := SELECT4(b[255:128], imm8[5:4]) +dst[255:224] := SELECT4(b[255:128], imm8[7:6]) +dst[MAX:256] := 0 - -
mmintrin.h
+ +
immintrin.h
 - + Floating Point - Integer - MMX - Shift - - - Shift packed 16-bit integers in "a" right by "count" while shifting in sign bits, and store the results in "dst". + AVX + Arithmetic + + + + Subtract packed double-precision (64-bit) floating-point elements in "b" from packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst". FOR j := 0 to 3 - i := j*16 - IF count[63:0] > 15 - dst[i+15:i] := SignBit - ELSE - dst[i+15:i] := SignExtend(a[i+15:i] >> count[63:0]) - FI + i := j*64 + dst[i+63:i] := a[i+63:i] - b[i+63:i] ENDFOR +dst[MAX:256] := 0 - -
mmintrin.h
+ +
immintrin.h
 - + Floating Point - Integer - MMX - Shift - - - Shift packed 16-bit integers in "a" right by "imm8" while shifting in sign bits, and store the results in "dst". + AVX + Arithmetic + + + + Subtract packed single-precision (32-bit) floating-point elements in "b" from packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst". -FOR j := 0 to 3 - i := j*16 - IF imm8[7:0] > 15 - dst[i+15:i] := SignBit - ELSE - dst[i+15:i] := SignExtend(a[i+15:i] >> imm8[7:0]) - FI +FOR j := 0 to 7 + i := j*32 + dst[i+31:i] := a[i+31:i] - b[i+31:i] ENDFOR +dst[MAX:256] := 0 - -
mmintrin.h
+ +
immintrin.h
 - + Floating Point - Integer - MMX - Shift - - - Shift packed 32-bit integers in "a" right by "count" while shifting in sign bits, and store the results in "dst". + AVX + Logical + + + + Compute the bitwise XOR of packed double-precision (64-bit) floating-point elements in "a" and "b", and store the results in "dst". -FOR j := 0 to 1 - i := j*32 - IF count[63:0] > 31 - dst[i+31:i] := SignBit - ELSE - dst[i+31:i] := SignExtend(a[i+31:i] >> count[63:0]) - FI +FOR j := 0 to 3 + i := j*64 + dst[i+63:i] := a[i+63:i] XOR b[i+63:i] ENDFOR +dst[MAX:256] := 0 - -
mmintrin.h
+ +
immintrin.h
 - + Floating Point - Integer - MMX - Shift - - - Shift packed 32-bit integers in "a" right by "imm8" while shifting in sign bits, and store the results in "dst". + AVX + Logical + + + + Compute the bitwise XOR of packed single-precision (32-bit) floating-point elements in "a" and "b", and store the results in "dst". -FOR j := 0 to 1 +FOR j := 0 to 7 i := j*32 - IF imm8[7:0] > 31 - dst[i+31:i] := SignBit - ELSE - dst[i+31:i] := SignExtend(a[i+31:i] >> imm8[7:0]) - FI + dst[i+31:i] := a[i+31:i] XOR b[i+31:i] ENDFOR +dst[MAX:256] := 0 - -
mmintrin.h
+ +
immintrin.h
 - + Floating Point - Integer - MMX - Shift - - - Shift packed 16-bit integers in "a" right by "count" while shifting in zeros, and store the results in "dst". - -FOR j := 0 to 3 - i := j*16 - IF count[63:0] > 15 - dst[i+15:i] := 0 - ELSE - dst[i+15:i] := ZeroExtend(a[i+15:i] >> count[63:0]) - FI + AVX + Compare + + + + + Compare packed double-precision (64-bit) floating-point elements in "a" and "b" based on the comparison operand specified by "imm8", and store the results in "dst". + CASE (imm8[4:0]) OF +0: OP := _CMP_EQ_OQ +1: OP := _CMP_LT_OS +2: OP := _CMP_LE_OS +3: OP := _CMP_UNORD_Q +4: OP := _CMP_NEQ_UQ +5: OP := _CMP_NLT_US +6: OP := _CMP_NLE_US +7: OP := _CMP_ORD_Q +8: OP := _CMP_EQ_UQ +9: OP := _CMP_NGE_US +10: OP := _CMP_NGT_US +11: OP := _CMP_FALSE_OQ +12: OP := _CMP_NEQ_OQ +13: OP := _CMP_GE_OS +14: OP := _CMP_GT_OS +15: OP := _CMP_TRUE_UQ +16: OP := _CMP_EQ_OS +17: OP := _CMP_LT_OQ +18: OP := _CMP_LE_OQ +19: OP := _CMP_UNORD_S +20: OP := _CMP_NEQ_US +21: OP := _CMP_NLT_UQ +22: OP := _CMP_NLE_UQ +23: OP := _CMP_ORD_S +24: OP := _CMP_EQ_US +25: OP := _CMP_NGE_UQ +26: OP := _CMP_NGT_UQ +27: OP := _CMP_FALSE_OS +28: OP := _CMP_NEQ_OS +29: OP := _CMP_GE_OQ +30: OP := _CMP_GT_OQ +31: OP := _CMP_TRUE_US +ESAC +FOR j := 0 to 1 + i := j*64 + dst[i+63:i] := ( a[i+63:i] OP b[i+63:i] ) ? 0xFFFFFFFFFFFFFFFF : 0 ENDFOR +dst[MAX:128] := 0 - -
mmintrin.h
+ +
immintrin.h
 - + Floating Point - Integer - MMX - Shift - - - Shift packed 16-bit integers in "a" right by "imm8" while shifting in zeros, and store the results in "dst". - + AVX + Compare + + + + + Compare packed double-precision (64-bit) floating-point elements in "a" and "b" based on the comparison operand specified by "imm8", and store the results in "dst". + CASE (imm8[4:0]) OF +0: OP := _CMP_EQ_OQ +1: OP := _CMP_LT_OS +2: OP := _CMP_LE_OS +3: OP := _CMP_UNORD_Q +4: OP := _CMP_NEQ_UQ +5: OP := _CMP_NLT_US +6: OP := _CMP_NLE_US +7: OP := _CMP_ORD_Q +8: OP := _CMP_EQ_UQ +9: OP := _CMP_NGE_US +10: OP := _CMP_NGT_US +11: OP := _CMP_FALSE_OQ +12: OP := _CMP_NEQ_OQ +13: OP := _CMP_GE_OS +14: OP := _CMP_GT_OS +15: OP := _CMP_TRUE_UQ +16: OP := _CMP_EQ_OS +17: OP := _CMP_LT_OQ +18: OP := _CMP_LE_OQ +19: OP := _CMP_UNORD_S +20: OP := _CMP_NEQ_US +21: OP := _CMP_NLT_UQ +22: OP := _CMP_NLE_UQ +23: OP := _CMP_ORD_S +24: OP := _CMP_EQ_US +25: OP := _CMP_NGE_UQ +26: OP := _CMP_NGT_UQ +27: OP := _CMP_FALSE_OS +28: OP := _CMP_NEQ_OS +29: OP := _CMP_GE_OQ +30: OP := _CMP_GT_OQ +31: OP := _CMP_TRUE_US +ESAC FOR j := 0 to 3 - i := j*16 - IF imm8[7:0] > 15 - dst[i+15:i] := 0 - ELSE - dst[i+15:i] := ZeroExtend(a[i+15:i] >> imm8[7:0]) - FI + i := j*64 + dst[i+63:i] := ( a[i+63:i] OP b[i+63:i] ) ? 0xFFFFFFFFFFFFFFFF : 0 ENDFOR +dst[MAX:256] := 0 - -
mmintrin.h
+ +
immintrin.h
 - + Floating Point - Integer - MMX - Shift - - - Shift packed 32-bit integers in "a" right by "count" while shifting in zeros, and store the results in "dst". - -FOR j := 0 to 1 + AVX + Compare + + + + + Compare packed single-precision (32-bit) floating-point elements in "a" and "b" based on the comparison operand specified by "imm8", and store the results in "dst". + CASE (imm8[4:0]) OF +0: OP := _CMP_EQ_OQ +1: OP := _CMP_LT_OS +2: OP := _CMP_LE_OS +3: OP := _CMP_UNORD_Q +4: OP := _CMP_NEQ_UQ +5: OP := _CMP_NLT_US +6: OP := _CMP_NLE_US +7: OP := _CMP_ORD_Q +8: OP := _CMP_EQ_UQ +9: OP := _CMP_NGE_US +10: OP := _CMP_NGT_US +11: OP := _CMP_FALSE_OQ +12: OP := _CMP_NEQ_OQ +13: OP := _CMP_GE_OS +14: OP := _CMP_GT_OS +15: OP := _CMP_TRUE_UQ +16: OP := _CMP_EQ_OS +17: OP := _CMP_LT_OQ +18: OP := _CMP_LE_OQ +19: OP := _CMP_UNORD_S +20: OP := _CMP_NEQ_US +21: OP := _CMP_NLT_UQ +22: OP := _CMP_NLE_UQ +23: OP := _CMP_ORD_S +24: OP := _CMP_EQ_US +25: OP := _CMP_NGE_UQ +26: OP := _CMP_NGT_UQ +27: OP := _CMP_FALSE_OS +28: OP := _CMP_NEQ_OS +29: OP := _CMP_GE_OQ +30: OP := _CMP_GT_OQ +31: OP := _CMP_TRUE_US +ESAC +FOR j := 0 to 3 i := j*32 - IF count[63:0] > 31 - dst[i+31:i] := 0 - ELSE - dst[i+31:i] := ZeroExtend(a[i+31:i] >> count[63:0]) - FI + dst[i+31:i] := ( a[i+31:i] OP b[i+31:i] ) ? 0xFFFFFFFF : 0 ENDFOR +dst[MAX:128] := 0 - -
mmintrin.h
+ +
immintrin.h
 - + Floating Point - Integer - MMX - Shift - - - Shift packed 32-bit integers in "a" right by "imm8" while shifting in zeros, and store the results in "dst". - -FOR j := 0 to 1 + AVX + Compare + + + + + Compare packed single-precision (32-bit) floating-point elements in "a" and "b" based on the comparison operand specified by "imm8", and store the results in "dst". + CASE (imm8[4:0]) OF +0: OP := _CMP_EQ_OQ +1: OP := _CMP_LT_OS +2: OP := _CMP_LE_OS +3: OP := _CMP_UNORD_Q +4: OP := _CMP_NEQ_UQ +5: OP := _CMP_NLT_US +6: OP := _CMP_NLE_US +7: OP := _CMP_ORD_Q +8: OP := _CMP_EQ_UQ +9: OP := _CMP_NGE_US +10: OP := _CMP_NGT_US +11: OP := _CMP_FALSE_OQ +12: OP := _CMP_NEQ_OQ +13: OP := _CMP_GE_OS +14: OP := _CMP_GT_OS +15: OP := _CMP_TRUE_UQ +16: OP := _CMP_EQ_OS +17: OP := _CMP_LT_OQ +18: OP := _CMP_LE_OQ +19: OP := _CMP_UNORD_S +20: OP := _CMP_NEQ_US +21: OP := _CMP_NLT_UQ +22: OP := _CMP_NLE_UQ +23: OP := _CMP_ORD_S +24: OP := _CMP_EQ_US +25: OP := _CMP_NGE_UQ +26: OP := _CMP_NGT_UQ +27: OP := _CMP_FALSE_OS +28: OP := _CMP_NEQ_OS +29: OP := _CMP_GE_OQ +30: OP := _CMP_GT_OQ +31: OP := _CMP_TRUE_US +ESAC +FOR j := 0 to 7 i := j*32 - IF imm8[7:0] > 31 - dst[i+31:i] := 0 - ELSE - dst[i+31:i] := ZeroExtend(a[i+31:i] >> imm8[7:0]) - FI + dst[i+31:i] := ( a[i+31:i] OP b[i+31:i] ) ? 0xFFFFFFFF : 0 ENDFOR +dst[MAX:256] := 0 - -
mmintrin.h
+ +
immintrin.h
 - + Floating Point - Integer - MMX - Shift - - - Shift 64-bit integer "a" right by "count" while shifting in zeros, and store the result in "dst". - -IF count[63:0] > 63 - dst[63:0] := 0 -ELSE - dst[63:0] := ZeroExtend(a[63:0] >> count[63:0]) -FI + AVX + Compare + + + + + Compare the lower double-precision (64-bit) floating-point element in "a" and "b" based on the comparison operand specified by "imm8", store the result in the lower element of "dst", and copy the upper element from "a" to the upper element of "dst". + CASE (imm8[4:0]) OF +0: OP := _CMP_EQ_OQ +1: OP := _CMP_LT_OS +2: OP := _CMP_LE_OS +3: OP := _CMP_UNORD_Q +4: OP := _CMP_NEQ_UQ +5: OP := _CMP_NLT_US +6: OP := _CMP_NLE_US +7: OP := _CMP_ORD_Q +8: OP := _CMP_EQ_UQ +9: OP := _CMP_NGE_US +10: OP := _CMP_NGT_US +11: OP := _CMP_FALSE_OQ +12: OP := _CMP_NEQ_OQ +13: OP := _CMP_GE_OS +14: OP := _CMP_GT_OS +15: OP := _CMP_TRUE_UQ +16: OP := _CMP_EQ_OS +17: OP := _CMP_LT_OQ +18: OP := _CMP_LE_OQ +19: OP := _CMP_UNORD_S +20: OP := _CMP_NEQ_US +21: OP := _CMP_NLT_UQ +22: OP := _CMP_NLE_UQ +23: OP := _CMP_ORD_S +24: OP := _CMP_EQ_US +25: OP := _CMP_NGE_UQ +26: OP := _CMP_NGT_UQ +27: OP := _CMP_FALSE_OS +28: OP := _CMP_NEQ_OS +29: OP := _CMP_GE_OQ +30: OP := _CMP_GT_OQ +31: OP := _CMP_TRUE_US +ESAC +dst[63:0] := ( a[63:0] OP b[63:0] ) ? 0xFFFFFFFFFFFFFFFF : 0 +dst[127:64] := a[127:64] +dst[MAX:128] := 0 - -
mmintrin.h
+ +
immintrin.h
 - + Floating Point - Integer - MMX - Shift - - - Shift 64-bit integer "a" right by "imm8" while shifting in zeros, and store the result in "dst". - -IF imm8[7:0] > 63 - dst[63:0] := 0 -ELSE - dst[63:0] := ZeroExtend(a[63:0] >> imm8[7:0]) -FI + AVX + Compare + + + + + Compare the lower single-precision (32-bit) floating-point element in "a" and "b" based on the comparison operand specified by "imm8", store the result in the lower element of "dst", and copy the upper 3 packed elements from "a" to the upper elements of "dst". + CASE (imm8[4:0]) OF +0: OP := _CMP_EQ_OQ +1: OP := _CMP_LT_OS +2: OP := _CMP_LE_OS +3: OP := _CMP_UNORD_Q +4: OP := _CMP_NEQ_UQ +5: OP := _CMP_NLT_US +6: OP := _CMP_NLE_US +7: OP := _CMP_ORD_Q +8: OP := _CMP_EQ_UQ +9: OP := _CMP_NGE_US +10: OP := _CMP_NGT_US +11: OP := _CMP_FALSE_OQ +12: OP := _CMP_NEQ_OQ +13: OP := _CMP_GE_OS +14: OP := _CMP_GT_OS +15: OP := _CMP_TRUE_UQ +16: OP := _CMP_EQ_OS +17: OP := _CMP_LT_OQ +18: OP := _CMP_LE_OQ +19: OP := _CMP_UNORD_S +20: OP := _CMP_NEQ_US +21: OP := _CMP_NLT_UQ +22: OP := _CMP_NLE_UQ +23: OP := _CMP_ORD_S +24: OP := _CMP_EQ_US +25: OP := _CMP_NGE_UQ +26: OP := _CMP_NGT_UQ +27: OP := _CMP_FALSE_OS +28: OP := _CMP_NEQ_OS +29: OP := _CMP_GE_OQ +30: OP := _CMP_GT_OQ +31: OP := _CMP_TRUE_US +ESAC +dst[31:0] := ( a[31:0] OP b[31:0] ) ? 0xFFFFFFFF : 0 +dst[127:32] := a[127:32] +dst[MAX:128] := 0 - -
mmintrin.h
+ +
immintrin.h
 - + + Floating Point Integer - MMX - Logical - - - Compute the bitwise AND of 64 bits (representing integer data) in "a" and "b", and store the result in "dst". + AVX + Convert + + + Convert packed signed 32-bit integers in "a" to packed double-precision (64-bit) floating-point elements, and store the results in "dst". -dst[63:0] := (a[63:0] AND b[63:0]) +FOR j := 0 to 3 + i := j*32 + m := j*64 + dst[m+63:m] := Convert_Int32_To_FP64(a[i+31:i]) +ENDFOR +dst[MAX:256] := 0 - -
mmintrin.h
+ +
immintrin.h
 - + + Floating Point Integer - MMX - Logical - - - Compute the bitwise NOT of 64 bits (representing integer data) in "a" and then AND with "b", and store the result in "dst". + AVX + Convert + + + Convert packed signed 32-bit integers in "a" to packed single-precision (32-bit) floating-point elements, and store the results in "dst". -dst[63:0] := ((NOT a[63:0]) AND b[63:0]) +FOR j := 0 to 7 + i := 32*j + dst[i+31:i] := Convert_Int32_To_FP32(a[i+31:i]) +ENDFOR +dst[MAX:256] := 0 - -
mmintrin.h
+ +
immintrin.h
 - - Integer - MMX - Logical - - - Compute the bitwise OR of 64 bits (representing integer data) in "a" and "b", and store the result in "dst". + + Floating Point + AVX + Convert + + + Convert packed double-precision (64-bit) floating-point elements in "a" to packed single-precision (32-bit) floating-point elements, and store the results in "dst". -dst[63:0] := (a[63:0] OR b[63:0]) +FOR j := 0 to 3 + i := 32*j + k := 64*j + dst[i+31:i] := Convert_FP64_To_FP32(a[k+63:k]) +ENDFOR +dst[MAX:128] := 0 - -
mmintrin.h
+ +
immintrin.h
 - + + Floating Point Integer - MMX - Logical - - - Compute the bitwise OR of 64 bits (representing integer data) in "a" and "b", and store the result in "dst". - -dst[63:0] := (a[63:0] XOR b[63:0]) - - -
mmintrin.h
- - - Integer - MMX - Compare - - - Compare packed 8-bit integers in "a" and "b" for equality, and store the results in "dst". + AVX + Convert + + + Convert packed single-precision (32-bit) floating-point elements in "a" to packed 32-bit integers, and store the results in "dst". FOR j := 0 to 7 - i := j*8 - dst[i+7:i] := ( a[i+7:i] == b[i+7:i] ) ? 0xFF : 0 + i := 32*j + dst[i+31:i] := Convert_FP32_To_Int32(a[i+31:i]) ENDFOR +dst[MAX:256] := 0 - -
mmintrin.h
+ +
immintrin.h
 - - Integer - MMX - Compare - - - Compare packed 16-bit integers in "a" and "b" for equality, and store the results in "dst". + + Floating Point + AVX + Convert + + + Convert packed single-precision (32-bit) floating-point elements in "a" to packed double-precision (64-bit) floating-point elements, and store the results in "dst". FOR j := 0 to 3 - i := j*16 - dst[i+15:i] := ( a[i+15:i] == b[i+15:i] ) ? 0xFFFF : 0 -ENDFOR - - -
mmintrin.h
- - - Integer - MMX - Compare - - - Compare packed 32-bit integers in "a" and "b" for equality, and store the results in "dst". - -FOR j := 0 to 1 - i := j*32 - dst[i+31:i] := ( a[i+31:i] == b[i+31:i] ) ? 0xFFFFFFFF : 0 -ENDFOR - - -
mmintrin.h
- - - Integer - MMX - Compare - - - Compare packed 8-bit integers in "a" and "b" for greater-than, and store the results in "dst". - -FOR j := 0 to 7 - i := j*8 - dst[i+7:i] := ( a[i+7:i] > b[i+7:i] ) ? 0xFF : 0 + i := 64*j + k := 32*j + dst[i+63:i] := Convert_FP32_To_FP64(a[k+31:k]) ENDFOR +dst[MAX:256] := 0 - -
mmintrin.h
+ +
immintrin.h
 - + + Floating Point Integer - MMX - Compare - - - Compare packed 16-bit integers in "a" and "b" for greater-than, and store the results in "dst". + AVX + Convert + + + Convert packed double-precision (64-bit) floating-point elements in "a" to packed 32-bit integers with truncation, and store the results in "dst". FOR j := 0 to 3 - i := j*16 - dst[i+15:i] := ( a[i+15:i] > b[i+15:i] ) ? 0xFFFF : 0 + i := 32*j + k := 64*j + dst[i+31:i] := Convert_FP64_To_Int32_Truncate(a[k+63:k]) ENDFOR +dst[MAX:128] := 0 - -
mmintrin.h
+ +
immintrin.h
 - + + Floating Point Integer - MMX - Compare - - - Compare packed 32-bit integers in "a" and "b" for greater-than, and store the results in "dst". + AVX + Convert + + + Convert packed double-precision (64-bit) floating-point elements in "a" to packed 32-bit integers, and store the results in "dst". -FOR j := 0 to 1 - i := j*32 - dst[i+31:i] := ( a[i+31:i] > b[i+31:i] ) ? 0xFFFFFFFF : 0 +FOR j := 0 to 3 + i := 32*j + k := 64*j + dst[i+31:i] := Convert_FP64_To_Int32(a[k+63:k]) ENDFOR +dst[MAX:128] := 0 - -
mmintrin.h
- - - MMX - General Support - - Empty the MMX state, which marks the x87 FPU registers as available for use by x87 instructions. This instruction must be used at the end of all MMX technology procedures. - -
mmintrin.h
+ +
immintrin.h
 - + + Floating Point Integer - MMX - Arithmetic - - - Add packed 8-bit integers in "a" and "b", and store the results in "dst". + AVX + Convert + + + Convert packed single-precision (32-bit) floating-point elements in "a" to packed 32-bit integers with truncation, and store the results in "dst". FOR j := 0 to 7 - i := j*8 - dst[i+7:i] := a[i+7:i] + b[i+7:i] + i := 32*j + dst[i+31:i] := Convert_FP32_To_Int32_Truncate(a[i+31:i]) ENDFOR +dst[MAX:256] := 0 - -
mmintrin.h
+ +
immintrin.h
 - - Integer - MMX - Arithmetic - - - Add packed 16-bit integers in "a" and "b", and store the results in "dst". + + Floating Point + AVX + Swizzle + + + + Extract 128 bits (composed of 4 packed single-precision (32-bit) floating-point elements) from "a", selected with "imm8", and store the result in "dst". -FOR j := 0 to 3 - i := j*16 - dst[i+15:i] := a[i+15:i] + b[i+15:i] -ENDFOR +CASE imm8[0] OF +0: dst[127:0] := a[127:0] +1: dst[127:0] := a[255:128] +ESAC +dst[MAX:128] := 0 - -
mmintrin.h
+ +
immintrin.h
 - - Integer - MMX - Arithmetic - - - Add packed 32-bit integers in "a" and "b", and store the results in "dst". + + Floating Point + AVX + Swizzle + + + + Extract 128 bits (composed of 2 packed double-precision (64-bit) floating-point elements) from "a", selected with "imm8", and store the result in "dst". -FOR j := 0 to 1 - i := j*32 - dst[i+31:i] := a[i+31:i] + b[i+31:i] -ENDFOR +CASE imm8[0] OF +0: dst[127:0] := a[127:0] +1: dst[127:0] := a[255:128] +ESAC +dst[MAX:128] := 0 - -
mmintrin.h
+ +
immintrin.h
 - + Integer - MMX - Arithmetic - - - Add packed 8-bit integers in "a" and "b" using saturation, and store the results in "dst". + AVX + Swizzle + + + + Extract 128 bits (composed of integer data) from "a", selected with "imm8", and store the result in "dst". -FOR j := 0 to 7 - i := j*8 - dst[i+7:i] := Saturate_To_Int8( a[i+7:i] + b[i+7:i] ) -ENDFOR +CASE imm8[0] OF +0: dst[127:0] := a[127:0] +1: dst[127:0] := a[255:128] +ESAC +dst[MAX:128] := 0 - -
mmintrin.h
+ +
immintrin.h
 - + Integer - MMX - Arithmetic - - - Add packed 16-bit integers in "a" and "b" using saturation, and store the results in "dst". + AVX + Swizzle + + + + Extract a 32-bit integer from "a", selected with "index", and store the result in "dst". -FOR j := 0 to 3 - i := j*16 - dst[i+15:i] := Saturate_To_Int16( a[i+15:i] + b[i+15:i] ) -ENDFOR +dst[31:0] := (a[255:0] >> (index[2:0] * 32))[31:0] - -
mmintrin.h
+
immintrin.h
 - + Integer - MMX - Arithmetic - - - Add packed unsigned 8-bit integers in "a" and "b" using saturation, and store the results in "dst". + AVX + Swizzle + + + + Extract a 64-bit integer from "a", selected with "index", and store the result in "dst". -FOR j := 0 to 7 - i := j*8 - dst[i+7:i] := Saturate_To_UnsignedInt8( a[i+7:i] + b[i+7:i] ) -ENDFOR +dst[63:0] := (a[255:0] >> (index[1:0] * 64))[63:0] - -
mmintrin.h
+
immintrin.h
 - - Integer - MMX - Arithmetic - - - Add packed unsigned 16-bit integers in "a" and "b" using saturation, and store the results in "dst". - -FOR j := 0 to 3 - i := j*16 - dst[i+15:i] := Saturate_To_UnsignedInt16( a[i+15:i] + b[i+15:i] ) -ENDFOR + + AVX + General Support + + + Zero the contents of all XMM or YMM registers. + YMM0[MAX:0] := 0 +YMM1[MAX:0] := 0 +YMM2[MAX:0] := 0 +YMM3[MAX:0] := 0 +YMM4[MAX:0] := 0 +YMM5[MAX:0] := 0 +YMM6[MAX:0] := 0 +YMM7[MAX:0] := 0 +IF _64_BIT_MODE + YMM8[MAX:0] := 0 + YMM9[MAX:0] := 0 + YMM10[MAX:0] := 0 + YMM11[MAX:0] := 0 + YMM12[MAX:0] := 0 + YMM13[MAX:0] := 0 + YMM14[MAX:0] := 0 + YMM15[MAX:0] := 0 +FI - -
mmintrin.h
+ +
immintrin.h
 - - Integer - MMX - Arithmetic - - - Subtract packed 8-bit integers in "b" from packed 8-bit integers in "a", and store the results in "dst". - -FOR j := 0 to 7 - i := j*8 - dst[i+7:i] := a[i+7:i] - b[i+7:i] -ENDFOR + + AVX + General Support + + + Zero the upper 128 bits of all YMM registers; the lower 128-bits of the registers are unmodified. + YMM0[MAX:128] := 0 +YMM1[MAX:128] := 0 +YMM2[MAX:128] := 0 +YMM3[MAX:128] := 0 +YMM4[MAX:128] := 0 +YMM5[MAX:128] := 0 +YMM6[MAX:128] := 0 +YMM7[MAX:128] := 0 +IF _64_BIT_MODE + YMM8[MAX:128] := 0 + YMM9[MAX:128] := 0 + YMM10[MAX:128] := 0 + YMM11[MAX:128] := 0 + YMM12[MAX:128] := 0 + YMM13[MAX:128] := 0 + YMM14[MAX:128] := 0 + YMM15[MAX:128] := 0 +FI - -
mmintrin.h
+ +
immintrin.h
 - - Integer - MMX - Arithmetic - - - Subtract packed 16-bit integers in "b" from packed 16-bit integers in "a", and store the results in "dst". + + Floating Point + AVX + Swizzle + + + + Shuffle single-precision (32-bit) floating-point elements in "a" within 128-bit lanes using the control in "b", and store the results in "dst". -FOR j := 0 to 3 - i := j*16 - dst[i+15:i] := a[i+15:i] - b[i+15:i] -ENDFOR +DEFINE SELECT4(src, control) { + CASE(control[1:0]) OF + 0: tmp[31:0] := src[31:0] + 1: tmp[31:0] := src[63:32] + 2: tmp[31:0] := src[95:64] + 3: tmp[31:0] := src[127:96] + ESAC + RETURN tmp[31:0] +} +dst[31:0] := SELECT4(a[127:0], b[1:0]) +dst[63:32] := SELECT4(a[127:0], b[33:32]) +dst[95:64] := SELECT4(a[127:0], b[65:64]) +dst[127:96] := SELECT4(a[127:0], b[97:96]) +dst[159:128] := SELECT4(a[255:128], b[129:128]) +dst[191:160] := SELECT4(a[255:128], b[161:160]) +dst[223:192] := SELECT4(a[255:128], b[193:192]) +dst[255:224] := SELECT4(a[255:128], b[225:224]) +dst[MAX:256] := 0 - -
mmintrin.h
+ +
immintrin.h
 - - Integer - MMX - Arithmetic - - - Subtract packed 32-bit integers in "b" from packed 32-bit integers in "a", and store the results in "dst". + + Floating Point + AVX + Swizzle + + + + Shuffle single-precision (32-bit) floating-point elements in "a" using the control in "b", and store the results in "dst". -FOR j := 0 to 1 - i := j*32 - dst[i+31:i] := a[i+31:i] - b[i+31:i] -ENDFOR +DEFINE SELECT4(src, control) { + CASE(control[1:0]) OF + 0: tmp[31:0] := src[31:0] + 1: tmp[31:0] := src[63:32] + 2: tmp[31:0] := src[95:64] + 3: tmp[31:0] := src[127:96] + ESAC + RETURN tmp[31:0] +} +dst[31:0] := SELECT4(a[127:0], b[1:0]) +dst[63:32] := SELECT4(a[127:0], b[33:32]) +dst[95:64] := SELECT4(a[127:0], b[65:64]) +dst[127:96] := SELECT4(a[127:0], b[97:96]) +dst[MAX:128] := 0 - -
mmintrin.h
+ +
immintrin.h
 - - Integer - MMX - Arithmetic - - - Subtract packed 8-bit integers in "b" from packed 8-bit integers in "a" using saturation, and store the results in "dst". + + Floating Point + AVX + Swizzle + + + + Shuffle single-precision (32-bit) floating-point elements in "a" within 128-bit lanes using the control in "imm8", and store the results in "dst". -FOR j := 0 to 7 - i := j*8 - dst[i+7:i] := Saturate_To_Int8(a[i+7:i] - b[i+7:i]) -ENDFOR +DEFINE SELECT4(src, control) { + CASE(control[1:0]) OF + 0: tmp[31:0] := src[31:0] + 1: tmp[31:0] := src[63:32] + 2: tmp[31:0] := src[95:64] + 3: tmp[31:0] := src[127:96] + ESAC + RETURN tmp[31:0] +} +dst[31:0] := SELECT4(a[127:0], imm8[1:0]) +dst[63:32] := SELECT4(a[127:0], imm8[3:2]) +dst[95:64] := SELECT4(a[127:0], imm8[5:4]) +dst[127:96] := SELECT4(a[127:0], imm8[7:6]) +dst[159:128] := SELECT4(a[255:128], imm8[1:0]) +dst[191:160] := SELECT4(a[255:128], imm8[3:2]) +dst[223:192] := SELECT4(a[255:128], imm8[5:4]) +dst[255:224] := SELECT4(a[255:128], imm8[7:6]) +dst[MAX:256] := 0 - -
mmintrin.h
+ +
immintrin.h
 - - Integer - MMX - Arithmetic - - - Subtract packed 16-bit integers in "b" from packed 16-bit integers in "a" using saturation, and store the results in "dst". + + Floating Point + AVX + Swizzle + + + + Shuffle single-precision (32-bit) floating-point elements in "a" using the control in "imm8", and store the results in "dst". -FOR j := 0 to 3 - i := j*16 - dst[i+15:i] := Saturate_To_Int16(a[i+15:i] - b[i+15:i]) -ENDFOR +DEFINE SELECT4(src, control) { + CASE(control[1:0]) OF + 0: tmp[31:0] := src[31:0] + 1: tmp[31:0] := src[63:32] + 2: tmp[31:0] := src[95:64] + 3: tmp[31:0] := src[127:96] + ESAC + RETURN tmp[31:0] +} +dst[31:0] := SELECT4(a[127:0], imm8[1:0]) +dst[63:32] := SELECT4(a[127:0], imm8[3:2]) +dst[95:64] := SELECT4(a[127:0], imm8[5:4]) +dst[127:96] := SELECT4(a[127:0], imm8[7:6]) +dst[MAX:128] := 0 - -
mmintrin.h
+ +
immintrin.h
 - - Integer - MMX - Arithmetic - - - Subtract packed unsigned 8-bit integers in "b" from packed unsigned 8-bit integers in "a" using saturation, and store the results in "dst". + + Floating Point + AVX + Swizzle + + + + Shuffle double-precision (64-bit) floating-point elements in "a" within 128-bit lanes using the control in "b", and store the results in "dst". -FOR j := 0 to 7 - i := j*8 - dst[i+7:i] := Saturate_To_UnsignedInt8(a[i+7:i] - b[i+7:i]) -ENDFOR +IF (b[1] == 0) dst[63:0] := a[63:0]; FI +IF (b[1] == 1) dst[63:0] := a[127:64]; FI +IF (b[65] == 0) dst[127:64] := a[63:0]; FI +IF (b[65] == 1) dst[127:64] := a[127:64]; FI +IF (b[129] == 0) dst[191:128] := a[191:128]; FI +IF (b[129] == 1) dst[191:128] := a[255:192]; FI +IF (b[193] == 0) dst[255:192] := a[191:128]; FI +IF (b[193] == 1) dst[255:192] := a[255:192]; FI +dst[MAX:256] := 0 - -
mmintrin.h
+ +
immintrin.h
 - - Integer - MMX - Arithmetic - - - Subtract packed unsigned 16-bit integers in "b" from packed unsigned 16-bit integers in "a" using saturation, and store the results in "dst". + + Floating Point + AVX + Swizzle + + + + Shuffle double-precision (64-bit) floating-point elements in "a" using the control in "b", and store the results in "dst". -FOR j := 0 to 3 - i := j*16 - dst[i+15:i] := Saturate_To_UnsignedInt16(a[i+15:i] - b[i+15:i]) -ENDFOR +IF (b[1] == 0) dst[63:0] := a[63:0]; FI +IF (b[1] == 1) dst[63:0] := a[127:64]; FI +IF (b[65] == 0) dst[127:64] := a[63:0]; FI +IF (b[65] == 1) dst[127:64] := a[127:64]; FI +dst[MAX:128] := 0 - -
mmintrin.h
+ +
immintrin.h
 - - Integer - MMX - Arithmetic - - - Multiply packed 16-bit integers in "a" and "b", producing intermediate 32-bit integers. Horizontally add adjacent pairs of intermediate 32-bit integers, and pack the results in "dst". - + + Floating Point + AVX + Swizzle + + + + Shuffle double-precision (64-bit) floating-point elements in "a" within 128-bit lanes using the control in "imm8", and store the results in "dst". -FOR j := 0 to 1 - i := j*32 - st[i+31:i] := a[i+31:i+16]*b[i+31:i+16] + a[i+15:i]*b[i+15:i] -ENDFOR +IF (imm8[0] == 0) dst[63:0] := a[63:0]; FI +IF (imm8[0] == 1) dst[63:0] := a[127:64]; FI +IF (imm8[1] == 0) dst[127:64] := a[63:0]; FI +IF (imm8[1] == 1) dst[127:64] := a[127:64]; FI +IF (imm8[2] == 0) dst[191:128] := a[191:128]; FI +IF (imm8[2] == 1) dst[191:128] := a[255:192]; FI +IF (imm8[3] == 0) dst[255:192] := a[191:128]; FI +IF (imm8[3] == 1) dst[255:192] := a[255:192]; FI +dst[MAX:256] := 0 - -
mmintrin.h
+ +
immintrin.h
 - - Integer - MMX - Arithmetic - - - Multiply the packed 16-bit integers in "a" and "b", producing intermediate 32-bit integers, and store the high 16 bits of the intermediate integers in "dst". + + Floating Point + AVX + Swizzle + + + + Shuffle double-precision (64-bit) floating-point elements in "a" using the control in "imm8", and store the results in "dst". -FOR j := 0 to 3 - i := j*16 - tmp[31:0] := a[i+15:i] * b[i+15:i] - dst[i+15:i] := tmp[31:16] -ENDFOR +IF (imm8[0] == 0) dst[63:0] := a[63:0]; FI +IF (imm8[0] == 1) dst[63:0] := a[127:64]; FI +IF (imm8[1] == 0) dst[127:64] := a[63:0]; FI +IF (imm8[1] == 1) dst[127:64] := a[127:64]; FI +dst[MAX:128] := 0 - -
mmintrin.h
+ +
immintrin.h
 - - Integer - MMX - Arithmetic - - - Multiply the packed 16-bit integers in "a" and "b", producing intermediate 32-bit integers, and store the low 16 bits of the intermediate integers in "dst". - -FOR j := 0 to 3 - i := j*16 - tmp[31:0] := a[i+15:i] * b[i+15:i] - dst[i+15:i] := tmp[15:0] -ENDFOR + + Floating Point + AVX + Swizzle + + + + + Shuffle 128-bits (composed of 4 packed single-precision (32-bit) floating-point elements) selected by "imm8" from "a" and "b", and store the results in "dst". + +DEFINE SELECT4(src1, src2, control) { + CASE(control[1:0]) OF + 0: tmp[127:0] := src1[127:0] + 1: tmp[127:0] := src1[255:128] + 2: tmp[127:0] := src2[127:0] + 3: tmp[127:0] := src2[255:128] + ESAC + IF control[3] + tmp[127:0] := 0 + FI + RETURN tmp[127:0] +} +dst[127:0] := SELECT4(a[255:0], b[255:0], imm8[3:0]) +dst[255:128] := SELECT4(a[255:0], b[255:0], imm8[7:4]) +dst[MAX:256] := 0 - -
mmintrin.h
+ +
immintrin.h
 - - Integer - MMX - Shift - - - Shift packed 16-bit integers in "a" left by "count" while shifting in zeros, and store the results in "dst". - -FOR j := 0 to 3 - i := j*16 - IF count[63:0] > 15 - dst[i+15:i] := 0 - ELSE - dst[i+15:i] := ZeroExtend(a[i+15:i] << count[63:0]) + + Floating Point + AVX + Swizzle + + + + + Shuffle 128-bits (composed of 2 packed double-precision (64-bit) floating-point elements) selected by "imm8" from "a" and "b", and store the results in "dst". + +DEFINE SELECT4(src1, src2, control) { + CASE(control[1:0]) OF + 0: tmp[127:0] := src1[127:0] + 1: tmp[127:0] := src1[255:128] + 2: tmp[127:0] := src2[127:0] + 3: tmp[127:0] := src2[255:128] + ESAC + IF control[3] + tmp[127:0] := 0 FI -ENDFOR + RETURN tmp[127:0] +} +dst[127:0] := SELECT4(a[255:0], b[255:0], imm8[3:0]) +dst[255:128] := SELECT4(a[255:0], b[255:0], imm8[7:4]) +dst[MAX:256] := 0 - -
mmintrin.h
+ +
immintrin.h
 - + Integer - MMX - Shift - - - Shift packed 16-bit integers in "a" left by "imm8" while shifting in zeros, and store the results in "dst". - -FOR j := 0 to 3 - i := j*16 - IF imm8[7:0] > 15 - dst[i+15:i] := 0 - ELSE - dst[i+15:i] := ZeroExtend(a[i+15:i] << imm8[7:0]) + AVX + Swizzle + + + + + Shuffle 128-bits (composed of integer data) selected by "imm8" from "a" and "b", and store the results in "dst". + +DEFINE SELECT4(src1, src2, control) { + CASE(control[1:0]) OF + 0: tmp[127:0] := src1[127:0] + 1: tmp[127:0] := src1[255:128] + 2: tmp[127:0] := src2[127:0] + 3: tmp[127:0] := src2[255:128] + ESAC + IF control[3] + tmp[127:0] := 0 FI -ENDFOR + RETURN tmp[127:0] +} +dst[127:0] := SELECT4(a[255:0], b[255:0], imm8[3:0]) +dst[255:128] := SELECT4(a[255:0], b[255:0], imm8[7:4]) +dst[MAX:256] := 0 - -
mmintrin.h
+ +
immintrin.h
 - - Integer - MMX - Shift - - - Shift packed 32-bit integers in "a" left by "count" while shifting in zeros, and store the results in "dst". + + Floating Point + AVX + Load + + + Broadcast a single-precision (32-bit) floating-point element from memory to all elements of "dst". -FOR j := 0 to 1 +tmp[31:0] := MEM[mem_addr+31:mem_addr] +FOR j := 0 to 7 i := j*32 - IF count[63:0] > 31 - dst[i+31:i] := 0 - ELSE - dst[i+31:i] := ZeroExtend(a[i+31:i] << count[63:0]) - FI + dst[i+31:i] := tmp[31:0] ENDFOR +dst[MAX:256] := 0 - -
mmintrin.h
+ +
immintrin.h
 - - Integer - MMX - Shift - - - Shift packed 32-bit integers in "a" left by "imm8" while shifting in zeros, and store the results in "dst". + + Floating Point + AVX + Load + Swizzle + + + Broadcast a single-precision (32-bit) floating-point element from memory to all elements of "dst". -FOR j := 0 to 1 +tmp[31:0] := MEM[mem_addr+31:mem_addr] +FOR j := 0 to 3 i := j*32 - IF imm8[7:0] > 31 - dst[i+31:i] := 0 - ELSE - dst[i+31:i] := ZeroExtend(a[i+31:i] << imm8[7:0]) - FI + dst[i+31:i] := tmp[31:0] ENDFOR +dst[MAX:128] := 0 - -
mmintrin.h
- - - Integer - MMX - Shift - - - Shift 64-bit integer "a" left by "count" while shifting in zeros, and store the result in "dst". - -IF count[63:0] > 63 - dst[63:0] := 0 -ELSE - dst[63:0] := ZeroExtend(a[63:0] << count[63:0]) -FI - - -
mmintrin.h
- - - Integer - MMX - Shift - - - Shift 64-bit integer "a" left by "imm8" while shifting in zeros, and store the result in "dst". - -IF imm8[7:0] > 63 - dst[63:0] := 0 -ELSE - dst[63:0] := ZeroExtend(a[63:0] << imm8[7:0]) -FI - - -
mmintrin.h
+ +
immintrin.h
 - - Integer - MMX - Shift - - - Shift packed 16-bit integers in "a" right by "count" while shifting in sign bits, and store the results in "dst". + + Floating Point + AVX + Load + Swizzle + + + Broadcast a double-precision (64-bit) floating-point element from memory to all elements of "dst". +tmp[63:0] := MEM[mem_addr+63:mem_addr] FOR j := 0 to 3 - i := j*16 - IF count[63:0] > 15 - dst[i+15:i] := SignBit - ELSE - dst[i+15:i] := SignExtend(a[i+15:i] >> count[63:0]) - FI + i := j*64 + dst[i+63:i] := tmp[63:0] ENDFOR +dst[MAX:256] := 0 - -
mmintrin.h
+ +
immintrin.h
 - - Integer - MMX - Shift - - - Shift packed 16-bit integers in "a" right by "imm8" while shifting in sign bits, and store the results in "dst". + + Floating Point + AVX + Load + Swizzle + + + Broadcast 128 bits from memory (composed of 4 packed single-precision (32-bit) floating-point elements) to all elements of "dst". -FOR j := 0 to 3 - i := j*16 - IF imm8[7:0] > 15 - dst[i+15:i] := SignBit - ELSE - dst[i+15:i] := SignExtend(a[i+15:i] >> imm8[7:0]) - FI -ENDFOR +tmp[127:0] := MEM[mem_addr+127:mem_addr] +dst[127:0] := tmp[127:0] +dst[255:128] := tmp[127:0] +dst[MAX:256] := 0 - -
mmintrin.h
+ +
immintrin.h
 - - Integer - MMX - Shift - - - Shift packed 32-bit integers in "a" right by "count" while shifting in sign bits, and store the results in "dst". + + Floating Point + AVX + Load + Swizzle + + + Broadcast 128 bits from memory (composed of 2 packed double-precision (64-bit) floating-point elements) to all elements of "dst". -FOR j := 0 to 1 - i := j*32 - IF count[63:0] > 31 - dst[i+31:i] := SignBit - ELSE - dst[i+31:i] := SignExtend(a[i+31:i] >> count[63:0]) - FI -ENDFOR +tmp[127:0] := MEM[mem_addr+127:mem_addr] +dst[127:0] := tmp[127:0] +dst[255:128] := tmp[127:0] +dst[MAX:256] := 0 - -
mmintrin.h
+ +
immintrin.h
 - - Integer - MMX - Shift - - - Shift packed 32-bit integers in "a" right by "imm8" while shifting in sign bits, and store the results in "dst". + + Floating Point + AVX + Swizzle + + + + + Copy "a" to "dst", then insert 128 bits (composed of 4 packed single-precision (32-bit) floating-point elements) from "b" into "dst" at the location specified by "imm8". -FOR j := 0 to 1 - i := j*32 - IF imm8[7:0] > 31 - dst[i+31:i] := SignBit - ELSE - dst[i+31:i] := SignExtend(a[i+31:i] >> imm8[7:0]) - FI -ENDFOR +dst[255:0] := a[255:0] +CASE (imm8[0]) OF +0: dst[127:0] := b[127:0] +1: dst[255:128] := b[127:0] +ESAC +dst[MAX:256] := 0 - -
mmintrin.h
+ +
immintrin.h
 - - Integer - MMX - Shift - - - Shift packed 16-bit integers in "a" right by "count" while shifting in zeros, and store the results in "dst". + + Floating Point + AVX + Swizzle + + + + + Copy "a" to "dst", then insert 128 bits (composed of 2 packed double-precision (64-bit) floating-point elements) from "b" into "dst" at the location specified by "imm8". -FOR j := 0 to 3 - i := j*16 - IF count[63:0] > 15 - dst[i+15:i] := 0 - ELSE - dst[i+15:i] := ZeroExtend(a[i+15:i] >> count[63:0]) - FI -ENDFOR +dst[255:0] := a[255:0] +CASE imm8[0] OF +0: dst[127:0] := b[127:0] +1: dst[255:128] := b[127:0] +ESAC +dst[MAX:256] := 0 - -
mmintrin.h
+ +
immintrin.h
 - + Integer - MMX - Shift - - - Shift packed 16-bit integers in "a" right by "imm8" while shifting in zeros, and store the results in "dst". + AVX + Swizzle + + + + + Copy "a" to "dst", then insert 128 bits from "b" into "dst" at the location specified by "imm8". -FOR j := 0 to 3 - i := j*16 - IF imm8[7:0] > 15 - dst[i+15:i] := 0 - ELSE - dst[i+15:i] := ZeroExtend(a[i+15:i] >> imm8[7:0]) - FI -ENDFOR +dst[255:0] := a[255:0] +CASE (imm8[0]) OF +0: dst[127:0] := b[127:0] +1: dst[255:128] := b[127:0] +ESAC +dst[MAX:256] := 0 - -
mmintrin.h
+ +
immintrin.h
 - + Integer - MMX - Shift - - - Shift packed 32-bit integers in "a" right by "count" while shifting in zeros, and store the results in "dst". + AVX + Swizzle + + + + + Copy "a" to "dst", and insert the 8-bit integer "i" into "dst" at the location specified by "index". -FOR j := 0 to 1 - i := j*32 - IF count[63:0] > 31 - dst[i+31:i] := 0 - ELSE - dst[i+31:i] := ZeroExtend(a[i+31:i] >> count[63:0]) - FI -ENDFOR +dst[255:0] := a[255:0] +sel := index[4:0]*8 +dst[sel+7:sel] := i[7:0] - -
mmintrin.h
+
immintrin.h
 - + Integer - MMX - Shift - - - Shift packed 32-bit integers in "a" right by "imm8" while shifting in zeros, and store the results in "dst". + AVX + Swizzle + + + + + Copy "a" to "dst", and insert the 16-bit integer "i" into "dst" at the location specified by "index". -FOR j := 0 to 1 - i := j*32 - IF imm8[7:0] > 31 - dst[i+31:i] := 0 - ELSE - dst[i+31:i] := ZeroExtend(a[i+31:i] >> imm8[7:0]) - FI -ENDFOR +dst[255:0] := a[255:0] +sel := index[3:0]*16 +dst[sel+15:sel] := i[15:0] - -
mmintrin.h
+
immintrin.h
 - + Integer - MMX - Shift - - - Shift 64-bit integer "a" right by "count" while shifting in zeros, and store the result in "dst". + AVX + Swizzle + + + + + Copy "a" to "dst", and insert the 32-bit integer "i" into "dst" at the location specified by "index". -IF count[63:0] > 63 - dst[63:0] := 0 -ELSE - dst[63:0] := ZeroExtend(a[63:0] >> count[63:0]) -FI +dst[255:0] := a[255:0] +sel := index[2:0]*32 +dst[sel+31:sel] := i[31:0] - -
mmintrin.h
+
immintrin.h
 - + Integer - MMX - Shift - - - Shift 64-bit integer "a" right by "imm8" while shifting in zeros, and store the result in "dst". + AVX + Swizzle + + + + + Copy "a" to "dst", and insert the 64-bit integer "i" into "dst" at the location specified by "index". -IF imm8[7:0] > 63 - dst[63:0] := 0 -ELSE - dst[63:0] := ZeroExtend(a[63:0] >> imm8[7:0]) -FI +dst[255:0] := a[255:0] +sel := index[1:0]*64 +dst[sel+63:sel] := i[63:0] - -
mmintrin.h
+
immintrin.h
 - - Integer - MMX - Logical - - - Compute the bitwise AND of 64 bits (representing integer data) in "a" and "b", and store the result in "dst". + + Floating Point + AVX + Load + + + Load 256-bits (composed of 4 packed double-precision (64-bit) floating-point elements) from memory into "dst". + "mem_addr" must be aligned on a 32-byte boundary or a general-protection exception may be generated. -dst[63:0] := (a[63:0] AND b[63:0]) +dst[255:0] := MEM[mem_addr+255:mem_addr] +dst[MAX:256] := 0 - -
mmintrin.h
+ +
immintrin.h
 - - Integer - MMX - Logical - - - Compute the bitwise NOT of 64 bits (representing integer data) in "a" and then AND with "b", and store the result in "dst". + + Floating Point + AVX + Store + + + + Store 256-bits (composed of 4 packed double-precision (64-bit) floating-point elements) from "a" into memory. + "mem_addr" must be aligned on a 32-byte boundary or a general-protection exception may be generated. -dst[63:0] := ((NOT a[63:0]) AND b[63:0]) +MEM[mem_addr+255:mem_addr] := a[255:0] - -
mmintrin.h
+ +
immintrin.h
 - - Integer - MMX - Logical - - - Compute the bitwise OR of 64 bits (representing integer data) in "a" and "b", and store the result in "dst". + + Floating Point + AVX + Load + + + Load 256-bits (composed of 8 packed single-precision (32-bit) floating-point elements) from memory into "dst". + "mem_addr" must be aligned on a 32-byte boundary or a general-protection exception may be generated. -dst[63:0] := (a[63:0] OR b[63:0]) +dst[255:0] := MEM[mem_addr+255:mem_addr] +dst[MAX:256] := 0 - -
mmintrin.h
+ +
immintrin.h
 - - Integer - MMX - Logical - - - Compute the bitwise XOR of 64 bits (representing integer data) in "a" and "b", and store the result in "dst". + + Floating Point + AVX + Store + + + + Store 256-bits (composed of 8 packed single-precision (32-bit) floating-point elements) from "a" into memory. + "mem_addr" must be aligned on a 32-byte boundary or a general-protection exception may be generated. -dst[63:0] := (a[63:0] XOR b[63:0]) +MEM[mem_addr+255:mem_addr] := a[255:0] - -
mmintrin.h
+ +
immintrin.h
 - - Integer - MMX - Compare - - - Compare packed 8-bit integers in "a" and "b" for equality, and store the results in "dst". + + Floating Point + AVX + Load + + + Load 256-bits (composed of 4 packed double-precision (64-bit) floating-point elements) from memory into "dst". + "mem_addr" does not need to be aligned on any particular boundary. -FOR j := 0 to 7 - i := j*8 - dst[i+7:i] := ( a[i+7:i] == b[i+7:i] ) ? 0xFF : 0 -ENDFOR +dst[255:0] := MEM[mem_addr+255:mem_addr] +dst[MAX:256] := 0 - -
mmintrin.h
+ +
immintrin.h
 - - Integer - MMX - Compare - - - Compare packed 16-bit integers in "a" and "b" for equality, and store the results in "dst". + + Floating Point + AVX + Store + + + + Store 256-bits (composed of 4 packed double-precision (64-bit) floating-point elements) from "a" into memory. + "mem_addr" does not need to be aligned on any particular boundary. -FOR j := 0 to 3 - i := j*16 - dst[i+15:i] := ( a[i+15:i] == b[i+15:i] ) ? 0xFFFF : 0 -ENDFOR +MEM[mem_addr+255:mem_addr] := a[255:0] - -
mmintrin.h
+ +
immintrin.h
 - - Integer - MMX - Compare - - - Compare packed 32-bit integers in "a" and "b" for equality, and store the results in "dst". + + Floating Point + AVX + Load + + + Load 256-bits (composed of 8 packed single-precision (32-bit) floating-point elements) from memory into "dst". + "mem_addr" does not need to be aligned on any particular boundary. -FOR j := 0 to 1 - i := j*32 - dst[i+31:i] := ( a[i+31:i] == b[i+31:i] ) ? 0xFFFFFFFF : 0 -ENDFOR +dst[255:0] := MEM[mem_addr+255:mem_addr] +dst[MAX:256] := 0 - -
mmintrin.h
+ +
immintrin.h
 - - Integer - MMX - Compare - - - Compare packed 8-bit integers in "a" and "b" for greater-than, and store the results in "dst". + + Floating Point + AVX + Store + + + + Store 256-bits (composed of 8 packed single-precision (32-bit) floating-point elements) from "a" into memory. + "mem_addr" does not need to be aligned on any particular boundary. -FOR j := 0 to 7 - i := j*8 - dst[i+7:i] := ( a[i+7:i] > b[i+7:i] ) ? 0xFF : 0 -ENDFOR +MEM[mem_addr+255:mem_addr] := a[255:0] - -
mmintrin.h
+ +
immintrin.h
 - + Integer - MMX - Compare - - - Compare packed 16-bit integers in "a" and "b" for greater-than, and store the results in "dst". + AVX + Load + + + Load 256-bits of integer data from memory into "dst". + "mem_addr" must be aligned on a 32-byte boundary or a general-protection exception may be generated. -FOR j := 0 to 3 - i := j*16 - dst[i+15:i] := ( a[i+15:i] > b[i+15:i] ) ? 0xFFFF : 0 -ENDFOR +dst[255:0] := MEM[mem_addr+255:mem_addr] +dst[MAX:256] := 0 - -
mmintrin.h
+ +
immintrin.h
 - + Integer - MMX - Compare - - - Compare packed 32-bit integers in "a" and "b" for greater-than, and store the results in "dst". + AVX + Store + + + + Store 256-bits of integer data from "a" into memory. + "mem_addr" must be aligned on a 32-byte boundary or a general-protection exception may be generated. -FOR j := 0 to 1 - i := j*32 - dst[i+31:i] := ( a[i+31:i] > b[i+31:i] ) ? 0xFFFFFFFF : 0 -ENDFOR +MEM[mem_addr+255:mem_addr] := a[255:0] - -
mmintrin.h
+ +
immintrin.h
 - + Integer - MMX - Convert - - Copy 32-bit integer "a" to the lower elements of "dst", and zero the upper element of "dst". + AVX + Load + + + Load 256-bits of integer data from memory into "dst". + "mem_addr" does not need to be aligned on any particular boundary. -dst[31:0] := a[31:0] -dst[63:32] := 0 +dst[255:0] := MEM[mem_addr+255:mem_addr] +dst[MAX:256] := 0 - -
mmintrin.h
+ +
immintrin.h
 - + Integer - MMX - Convert - - Copy the lower 32-bit integer in "a" to "dst". + AVX + Store + + + + Store 256-bits of integer data from "a" into memory. + "mem_addr" does not need to be aligned on any particular boundary. -dst[31:0] := a[31:0] +MEM[mem_addr+255:mem_addr] := a[255:0] - -
mmintrin.h
+ +
immintrin.h
 - - Integer - MMX - Convert - - Copy 64-bit integer "a" to "dst". + + Floating Point + AVX + Load + + + + Load packed double-precision (64-bit) floating-point elements from memory into "dst" using "mask" (elements are zeroed out when the high bit of the corresponding element is not set). -dst[63:0] := a[63:0] +FOR j := 0 to 3 + i := j*64 + IF mask[i+63] + dst[i+63:i] := MEM[mem_addr+i+63:mem_addr+i] + ELSE + dst[i+63:i] := 0 + FI +ENDFOR +dst[MAX:256] := 0 - -
mmintrin.h
+ +
immintrin.h
 - - Integer - MMX - Convert - - Copy 64-bit integer "a" to "dst". + + Floating Point + AVX + Store + + + + + Store packed double-precision (64-bit) floating-point elements from "a" into memory using "mask". -dst[63:0] := a[63:0] +FOR j := 0 to 3 + i := j*64 + IF mask[i+63] + MEM[mem_addr+i+63:mem_addr+i] := a[i+63:i] + FI +ENDFOR - -
mmintrin.h
+ +
immintrin.h
 - - Integer - MMX - Set - - Return vector of type __m64 with all elements set to zero. + + Floating Point + AVX + Load + + + + Load packed double-precision (64-bit) floating-point elements from memory into "dst" using "mask" (elements are zeroed out when the high bit of the corresponding element is not set). -dst[MAX:0] := 0 +FOR j := 0 to 1 + i := j*64 + IF mask[i+63] + dst[i+63:i] := MEM[mem_addr+i+63:mem_addr+i] + ELSE + dst[i+63:i] := 0 + FI +ENDFOR +dst[MAX:128] := 0 - -
mmintrin.h
+ +
immintrin.h
 - - Integer - MMX - Set - - - Set packed 32-bit integers in "dst" with the supplied values. + + Floating Point + AVX + Store + + + + + Store packed double-precision (64-bit) floating-point elements from "a" into memory using "mask". -dst[31:0] := e0 -dst[63:32] := e1 +FOR j := 0 to 1 + i := j*64 + IF mask[i+63] + MEM[mem_addr+i+63:mem_addr+i] := a[i+63:i] + FI +ENDFOR -
mmintrin.h
+ +
immintrin.h
 - - Integer - MMX - Set - - - - - Set packed 16-bit integers in "dst" with the supplied values. + + Floating Point + AVX + Load + + + + Load packed single-precision (32-bit) floating-point elements from memory into "dst" using "mask" (elements are zeroed out when the high bit of the corresponding element is not set). -dst[15:0] := e0 -dst[31:16] := e1 -dst[47:32] := e2 -dst[63:48] := e3 +FOR j := 0 to 7 + i := j*32 + IF mask[i+31] + dst[i+31:i] := MEM[mem_addr+i+31:mem_addr+i] + ELSE + dst[i+31:i] := 0 + FI +ENDFOR +dst[MAX:256] := 0 -
mmintrin.h
+ +
immintrin.h
 - - Integer - MMX - Set - - - - - - - - - Set packed 8-bit integers in "dst" with the supplied values in reverse order. + + Floating Point + AVX + Store + + + + + Store packed single-precision (32-bit) floating-point elements from "a" into memory using "mask". -dst[7:0] := e0 -dst[15:8] := e1 -dst[23:16] := e2 -dst[31:24] := e3 -dst[39:32] := e4 -dst[47:40] := e5 -dst[55:48] := e6 -dst[63:56] := e7 +FOR j := 0 to 7 + i := j*32 + IF mask[i+31] + MEM[mem_addr+i+31:mem_addr+i] := a[i+31:i] + FI +ENDFOR -
mmintrin.h
+ +
immintrin.h
 - - Integer - MMX - Set - - Broadcast 32-bit integer "a" to all elements of "dst". + + Floating Point + AVX + Load + + + + Load packed single-precision (32-bit) floating-point elements from memory into "dst" using "mask" (elements are zeroed out when the high bit of the corresponding element is not set). -FOR j := 0 to 1 +FOR j := 0 to 3 i := j*32 - dst[i+31:i] := a[31:0] + IF mask[i+31] + dst[i+31:i] := MEM[mem_addr+i+31:mem_addr+i] + ELSE + dst[i+31:i] := 0 + FI ENDFOR +dst[MAX:128] := 0 -
mmintrin.h
+ +
immintrin.h
 - - Integer - MMX - Set - - Broadcast 16-bit integer "a" to all all elements of "dst". + + Floating Point + AVX + Store + + + + + Store packed single-precision (32-bit) floating-point elements from "a" into memory using "mask". FOR j := 0 to 3 - i := j*16 - dst[i+15:i] := a[15:0] + i := j*32 + IF mask[i+31] + MEM[mem_addr+i+31:mem_addr+i] := a[i+31:i] + FI ENDFOR -
mmintrin.h
+ +
immintrin.h
 - - Integer - MMX - Set - - Broadcast 8-bit integer "a" to all elements of "dst". + + Floating Point + AVX + Move + + + Duplicate odd-indexed single-precision (32-bit) floating-point elements from "a", and store the results in "dst". -FOR j := 0 to 7 - i := j*8 - dst[i+7:i] := a[7:0] -ENDFOR +dst[31:0] := a[63:32] +dst[63:32] := a[63:32] +dst[95:64] := a[127:96] +dst[127:96] := a[127:96] +dst[159:128] := a[191:160] +dst[191:160] := a[191:160] +dst[223:192] := a[255:224] +dst[255:224] := a[255:224] +dst[MAX:256] := 0 -
mmintrin.h
+ +
immintrin.h
 - - Integer - MMX - Set - - - Set packed 32-bit integers in "dst" with the supplied values in reverse order. + + Floating Point + AVX + Move + + + Duplicate even-indexed single-precision (32-bit) floating-point elements from "a", and store the results in "dst". -dst[31:0] := e1 -dst[63:32] := e0 +dst[31:0] := a[31:0] +dst[63:32] := a[31:0] +dst[95:64] := a[95:64] +dst[127:96] := a[95:64] +dst[159:128] := a[159:128] +dst[191:160] := a[159:128] +dst[223:192] := a[223:192] +dst[255:224] := a[223:192] +dst[MAX:256] := 0 -
mmintrin.h
+ +
immintrin.h
 - - Integer - MMX - Set - - - - - Set packed 16-bit integers in "dst" with the supplied values in reverse order. + + Floating Point + AVX + Move + + + Duplicate even-indexed double-precision (64-bit) floating-point elements from "a", and store the results in "dst". -dst[15:0] := e3 -dst[31:16] := e2 -dst[47:32] := e1 -dst[63:48] := e0 +dst[63:0] := a[63:0] +dst[127:64] := a[63:0] +dst[191:128] := a[191:128] +dst[255:192] := a[191:128] +dst[MAX:256] := 0 -
mmintrin.h
+ +
immintrin.h
 - + Integer - MMX - Set - - - - - - - - - Set packed 8-bit integers in "dst" with the supplied values in reverse order. + AVX + Load + + + Load 256-bits of integer data from unaligned memory into "dst". This intrinsic may perform better than "_mm256_loadu_si256" when the data crosses a cache line boundary. -dst[7:0] := e7 -dst[15:8] := e6 -dst[23:16] := e5 -dst[31:24] := e4 -dst[39:32] := e3 -dst[47:40] := e2 -dst[55:48] := e1 -dst[63:56] := e0 +dst[255:0] := MEM[mem_addr+255:mem_addr] +dst[MAX:256] := 0 -
mmintrin.h
- - - Integer - MMX - Miscellaneous - - - Convert packed 16-bit integers from "a" and "b" to packed 8-bit integers using signed saturation, and store the results in "dst". - - -dst[7:0] := Saturate_Int16_To_Int8 (a[15:0]) -dst[15:8] := Saturate_Int16_To_Int8 (a[31:16]) -dst[23:16] := Saturate_Int16_To_Int8 (a[47:32]) -dst[31:24] := Saturate_Int16_To_Int8 (a[63:48]) -dst[39:32] := Saturate_Int16_To_Int8 (b[15:0]) -dst[47:40] := Saturate_Int16_To_Int8 (b[31:16]) -dst[55:48] := Saturate_Int16_To_Int8 (b[47:32]) -dst[63:56] := Saturate_Int16_To_Int8 (b[63:48]) - - -
mmintrin.h
+ +
immintrin.h
 - + Integer - MMX - Miscellaneous - - - Convert packed 32-bit integers from "a" and "b" to packed 16-bit integers using signed saturation, and store the results in "dst". + AVX + Store + + + + Store 256-bits of integer data from "a" into memory using a non-temporal memory hint. + "mem_addr" must be aligned on a 32-byte boundary or a general-protection exception may be generated. -dst[15:0] := Saturate_Int32_To_Int16 (a[31:0]) -dst[31:16] := Saturate_Int32_To_Int16 (a[63:32]) -dst[47:32] := Saturate_Int32_To_Int16 (b[31:0]) -dst[63:48] := Saturate_Int32_To_Int16 (b[63:32]) +MEM[mem_addr+255:mem_addr] := a[255:0] - -
mmintrin.h
+ +
immintrin.h
 - - Integer - MMX - Miscellaneous - - - Convert packed 16-bit integers from "a" and "b" to packed 8-bit integers using unsigned saturation, and store the results in "dst". + + Floating Point + AVX + Store + + + + Store 256-bits (composed of 4 packed double-precision (64-bit) floating-point elements) from "a" into memory using a non-temporal memory hint. + "mem_addr" must be aligned on a 32-byte boundary or a general-protection exception may be generated. -dst[7:0] := Saturate_Int16_To_UnsignedInt8 (a[15:0]) -dst[15:8] := Saturate_Int16_To_UnsignedInt8 (a[31:16]) -dst[23:16] := Saturate_Int16_To_UnsignedInt8 (a[47:32]) -dst[31:24] := Saturate_Int16_To_UnsignedInt8 (a[63:48]) -dst[39:32] := Saturate_Int16_To_UnsignedInt8 (b[15:0]) -dst[47:40] := Saturate_Int16_To_UnsignedInt8 (b[31:16]) -dst[55:48] := Saturate_Int16_To_UnsignedInt8 (b[47:32]) -dst[63:56] := Saturate_Int16_To_UnsignedInt8 (b[63:48]) +MEM[mem_addr+255:mem_addr] := a[255:0] - -
mmintrin.h
+ +
immintrin.h
 - - Integer - MMX - Swizzle - - - Unpack and interleave 8-bit integers from the high half of "a" and "b", and store the results in "dst". + + Floating Point + AVX + Store + + + + Store 256-bits (composed of 8 packed single-precision (32-bit) floating-point elements) from "a" into memory using a non-temporal memory hint. + "mem_addr" must be aligned on a 32-byte boundary or a general-protection exception may be generated. -INTERLEAVE_HIGH_BYTES(src1[63:0], src2[63:0]){ - dst[7:0] := src1[39:32] - dst[15:8] := src2[39:32] - dst[23:16] := src1[47:40] - dst[31:24] := src2[47:40] - dst[39:32] := src1[55:48] - dst[47:40] := src2[55:48] - dst[55:48] := src1[63:56] - dst[63:56] := src2[63:56] - RETURN dst[63:0] -} - -dst[63:0] := INTERLEAVE_HIGH_BYTES(a[63:0], b[63:0]) +MEM[mem_addr+255:mem_addr] := a[255:0] - -
mmintrin.h
+ +
immintrin.h
 - - Integer - MMX - Swizzle - - - Unpack and interleave 16-bit integers from the high half of "a" and "b", and store the results in "dst". + + Floating Point + AVX + Elementary Math Functions + + + Compute the approximate reciprocal of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst". The maximum relative error for this approximation is less than 1.5*2^-12. -INTERLEAVE_HIGH_WORDS(src1[63:0], src2[63:0]){ - dst[15:0] := src1[47:32] - dst[31:16] := src2[47:32] - dst[47:32] := src1[63:48] - dst[63:48] := src2[63:48] - RETURN dst[63:0] -} - -dst[63:0] := INTERLEAVE_HIGH_WORDS(a[63:0], b[63:0]) +FOR j := 0 to 7 + i := j*32 + dst[i+31:i] := 1.0 / a[i+31:i] +ENDFOR +dst[MAX:256] := 0 - -
mmintrin.h
+ +
immintrin.h
 - - Integer - MMX - Swizzle - - - Unpack and interleave 32-bit integers from the high half of "a" and "b", and store the results in "dst". + + Floating Point + AVX + Elementary Math Functions + + + Compute the approximate reciprocal square root of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst". The maximum relative error for this approximation is less than 1.5*2^-12. -dst[31:0] := a[63:32] -dst[63:32] := b[63:32] +FOR j := 0 to 7 + i := j*32 + dst[i+31:i] := (1.0 / SQRT(a[i+31:i])) +ENDFOR +dst[MAX:256] := 0 - -
mmintrin.h
+ +
immintrin.h
 - - Integer - MMX - Swizzle - - - Unpack and interleave 8-bit integers from the low half of "a" and "b", and store the results in "dst". + + Floating Point + AVX + Elementary Math Functions + + + Compute the square root of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst". -INTERLEAVE_BYTES(src1[63:0], src2[63:0]){ - dst[7:0] := src1[7:0] - dst[15:8] := src2[7:0] - dst[23:16] := src1[15:8] - dst[31:24] := src2[15:8] - dst[39:32] := src1[23:16] - dst[47:40] := src2[23:16] - dst[55:48] := src1[31:24] - dst[63:56] := src2[31:24] - RETURN dst[63:0] -} - -dst[63:0] := INTERLEAVE_BYTES(a[63:0], b[63:0]) +FOR j := 0 to 3 + i := j*64 + dst[i+63:i] := SQRT(a[i+63:i]) +ENDFOR +dst[MAX:256] := 0 - -
mmintrin.h
+ +
immintrin.h
 - - Integer - MMX - Swizzle - - - Unpack and interleave 16-bit integers from the low half of "a" and "b", and store the results in "dst". + + Floating Point + AVX + Elementary Math Functions + + + Compute the square root of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst". -INTERLEAVE_WORDS(src1[63:0], src2[63:0]){ - dst[15:0] := src1[15:0] - dst[31:16] := src2[15:0] - dst[47:32] := src1[31:16] - dst[63:48] := src2[31:16] - RETURN dst[63:0] -} - -dst[63:0] := INTERLEAVE_WORDS(a[63:0], b[63:0]) +FOR j := 0 to 7 + i := j*32 + dst[i+31:i] := SQRT(a[i+31:i]) +ENDFOR +dst[MAX:256] := 0 - -
mmintrin.h
+ +
immintrin.h
 - - Integer - MMX - Swizzle - - - Unpack and interleave 32-bit integers from the low half of "a" and "b", and store the results in "dst". + + Floating Point + AVX + Special Math Functions + + + + Round the packed double-precision (64-bit) floating-point elements in "a" using the "rounding" parameter, and store the results as packed double-precision floating-point elements in "dst". + [round_note] -dst[31:0] := a[31:0] -dst[63:32] := b[31:0] +FOR j := 0 to 3 + i := j*64 + dst[i+63:i] := ROUND(a[i+63:i], rounding) +ENDFOR +dst[MAX:256] := 0 - -
mmintrin.h
+ +
immintrin.h
 - - - SSE - Swizzle - - - - - Macro: Transpose the 4x4 matrix formed by the 4 rows of single-precision (32-bit) floating-point elements in "row0", "row1", "row2", and "row3", and store the transposed matrix in these vectors ("row0" now contains column 0, etc.). + + Floating Point + AVX + Special Math Functions + + + + Round the packed single-precision (32-bit) floating-point elements in "a" using the "rounding" parameter, and store the results as packed single-precision floating-point elements in "dst". + [round_note] -__m128 tmp3, tmp2, tmp1, tmp0; -tmp0 = _mm_unpacklo_ps(row0, row1); -tmp2 = _mm_unpacklo_ps(row2, row3); -tmp1 = _mm_unpackhi_ps(row0, row1); -tmp3 = _mm_unpackhi_ps(row2, row3); -row0 = _mm_movelh_ps(tmp0, tmp2); -row1 = _mm_movehl_ps(tmp2, tmp0); -row2 = _mm_movelh_ps(tmp1, tmp3); -row3 = _mm_movehl_ps(tmp3, tmp1); +FOR j := 0 to 7 + i := j*32 + dst[i+31:i] := ROUND(a[i+31:i], rounding) +ENDFOR +dst[MAX:256] := 0 -
xmmintrin.h
+ +
immintrin.h
 - - SSE - General Support - - Get the unsigned 32-bit value of the MXCSR control and status register. + + Floating Point + AVX + Swizzle + + + + Unpack and interleave double-precision (64-bit) floating-point elements from the high half of each 128-bit lane in "a" and "b", and store the results in "dst". -dst[31:0] := MXCSR +DEFINE INTERLEAVE_HIGH_QWORDS(src1[127:0], src2[127:0]) { + dst[63:0] := src1[127:64] + dst[127:64] := src2[127:64] + RETURN dst[127:0] +} +dst[127:0] := INTERLEAVE_HIGH_QWORDS(a[127:0], b[127:0]) +dst[255:128] := INTERLEAVE_HIGH_QWORDS(a[255:128], b[255:128]) +dst[MAX:256] := 0 - -
xmmintrin.h
+ +
immintrin.h
 - - SSE - General Support - - Set the MXCSR control and status register with the value in unsigned 32-bit integer "a". + + Floating Point + AVX + Swizzle + + + + Unpack and interleave single-precision (32-bit) floating-point elements from the high half of each 128-bit lane in "a" and "b", and store the results in "dst". -MXCSR := a[31:0] +DEFINE INTERLEAVE_HIGH_DWORDS(src1[127:0], src2[127:0]) { + dst[31:0] := src1[95:64] + dst[63:32] := src2[95:64] + dst[95:64] := src1[127:96] + dst[127:96] := src2[127:96] + RETURN dst[127:0] +} +dst[127:0] := INTERLEAVE_HIGH_DWORDS(a[127:0], b[127:0]) +dst[255:128] := INTERLEAVE_HIGH_DWORDS(a[255:128], b[255:128]) +dst[MAX:256] := 0 - -
xmmintrin.h
+ +
immintrin.h
 - - SSE - General Support - Macro: Get the exception state bits from the MXCSR control and status register. The exception state may contain any of the following flags: _MM_EXCEPT_INVALID, _MM_EXCEPT_DIV_ZERO, _MM_EXCEPT_DENORM, _MM_EXCEPT_OVERFLOW, _MM_EXCEPT_UNDERFLOW, _MM_EXCEPT_INEXACT + + Floating Point + AVX + Swizzle + + + + Unpack and interleave double-precision (64-bit) floating-point elements from the low half of each 128-bit lane in "a" and "b", and store the results in "dst". -dst[31:0] := MXCSR & _MM_EXCEPT_MASK +DEFINE INTERLEAVE_QWORDS(src1[127:0], src2[127:0]) { + dst[63:0] := src1[63:0] + dst[127:64] := src2[63:0] + RETURN dst[127:0] +} +dst[127:0] := INTERLEAVE_QWORDS(a[127:0], b[127:0]) +dst[255:128] := INTERLEAVE_QWORDS(a[255:128], b[255:128]) +dst[MAX:256] := 0 -
xmmintrin.h
+ +
immintrin.h
 - - SSE - General Support - - Macro: Set the exception state bits of the MXCSR control and status register to the value in unsigned 32-bit integer "a". The exception state may contain any of the following flags: _MM_EXCEPT_INVALID, _MM_EXCEPT_DIV_ZERO, _MM_EXCEPT_DENORM, _MM_EXCEPT_OVERFLOW, _MM_EXCEPT_UNDERFLOW, _MM_EXCEPT_INEXACT + + Floating Point + AVX + Swizzle + + + + Unpack and interleave single-precision (32-bit) floating-point elements from the low half of each 128-bit lane in "a" and "b", and store the results in "dst". -MXCSR := a[31:0] AND ~_MM_EXCEPT_MASK +DEFINE INTERLEAVE_DWORDS(src1[127:0], src2[127:0]) { + dst[31:0] := src1[31:0] + dst[63:32] := src2[31:0] + dst[95:64] := src1[63:32] + dst[127:96] := src2[63:32] + RETURN dst[127:0] +} +dst[127:0] := INTERLEAVE_DWORDS(a[127:0], b[127:0]) +dst[255:128] := INTERLEAVE_DWORDS(a[255:128], b[255:128]) +dst[MAX:256] := 0 -
xmmintrin.h
+ +
immintrin.h
 - - SSE - General Support - Macro: Get the exception mask bits from the MXCSR control and status register. The exception mask may contain any of the following flags: _MM_MASK_INVALID, _MM_MASK_DIV_ZERO, _MM_MASK_DENORM, _MM_MASK_OVERFLOW, _MM_MASK_UNDERFLOW, _MM_MASK_INEXACT + + Integer + Flag + AVX + Logical + + + + Compute the bitwise AND of 256 bits (representing integer data) in "a" and "b", and set "ZF" to 1 if the result is zero, otherwise set "ZF" to 0. Compute the bitwise NOT of "a" and then AND with "b", and set "CF" to 1 if the result is zero, otherwise set "CF" to 0. Return the "ZF" value. -dst[31:0] := MXCSR & _MM_MASK_MASK +IF ((a[255:0] AND b[255:0]) == 0) + ZF := 1 +ELSE + ZF := 0 +FI +IF (((NOT a[255:0]) AND b[255:0]) == 0) + CF := 1 +ELSE + CF := 0 +FI +RETURN ZF -
xmmintrin.h
+ +
immintrin.h
 - - SSE - General Support - - Macro: Set the exception mask bits of the MXCSR control and status register to the value in unsigned 32-bit integer "a". The exception mask may contain any of the following flags: _MM_MASK_INVALID, _MM_MASK_DIV_ZERO, _MM_MASK_DENORM, _MM_MASK_OVERFLOW, _MM_MASK_UNDERFLOW, _MM_MASK_INEXACT + + Integer + Flag + AVX + Logical + + + + Compute the bitwise AND of 256 bits (representing integer data) in "a" and "b", and set "ZF" to 1 if the result is zero, otherwise set "ZF" to 0. Compute the bitwise NOT of "a" and then AND with "b", and set "CF" to 1 if the result is zero, otherwise set "CF" to 0. Return the "CF" value. -MXCSR := a[31:0] AND ~_MM_MASK_MASK +IF ((a[255:0] AND b[255:0]) == 0) + ZF := 1 +ELSE + ZF := 0 +FI +IF (((NOT a[255:0]) AND b[255:0]) == 0) + CF := 1 +ELSE + CF := 0 +FI +RETURN CF -
xmmintrin.h
+ +
immintrin.h
 - - SSE - General Support - Macro: Get the rounding mode bits from the MXCSR control and status register. The rounding mode may contain any of the following flags: _MM_ROUND_NEAREST, _MM_ROUND_DOWN, _MM_ROUND_UP, _MM_ROUND_TOWARD_ZERO + + Integer + Flag + AVX + Logical + + + + Compute the bitwise AND of 256 bits (representing integer data) in "a" and "b", and set "ZF" to 1 if the result is zero, otherwise set "ZF" to 0. Compute the bitwise NOT of "a" and then AND with "b", and set "CF" to 1 if the result is zero, otherwise set "CF" to 0. Return 1 if both the "ZF" and "CF" values are zero, otherwise return 0. -dst[31:0] := MXCSR & _MM_ROUND_MASK +IF ((a[255:0] AND b[255:0]) == 0) + ZF := 1 +ELSE + ZF := 0 +FI +IF (((NOT a[255:0]) AND b[255:0]) == 0) + CF := 1 +ELSE + CF := 0 +FI +IF (ZF == 0 && CF == 0) + dst := 1 +ELSE + dst := 0 +FI -
xmmintrin.h
+ +
immintrin.h
 - - SSE - General Support - - Macro: Set the rounding mode bits of the MXCSR control and status register to the value in unsigned 32-bit integer "a". The rounding mode may contain any of the following flags: _MM_ROUND_NEAREST, _MM_ROUND_DOWN, _MM_ROUND_UP, _MM_ROUND_TOWARD_ZERO + + Floating Point + Flag + AVX + Logical + + + + Compute the bitwise AND of 256 bits (representing double-precision (64-bit) floating-point elements) in "a" and "b", producing an intermediate 256-bit value, and set "ZF" to 1 if the sign bit of each 64-bit element in the intermediate value is zero, otherwise set "ZF" to 0. Compute the bitwise NOT of "a" and then AND with "b", producing an intermediate value, and set "CF" to 1 if the sign bit of each 64-bit element in the intermediate value is zero, otherwise set "CF" to 0. Return the "ZF" value. -MXCSR := a[31:0] AND ~_MM_ROUND_MASK +tmp[255:0] := a[255:0] AND b[255:0] +IF (tmp[63] == 0 && tmp[127] == 0 && tmp[191] == 0 && tmp[255] == 0) + ZF := 1 +ELSE + ZF := 0 +FI +tmp[255:0] := (NOT a[255:0]) AND b[255:0] +IF (tmp[63] == 0 && tmp[127] == 0 && tmp[191] == 0 && tmp[255] == 0) + CF := 1 +ELSE + CF := 0 +FI +dst := ZF -
xmmintrin.h
+ +
immintrin.h
 - - SSE - General Support - Macro: Get the flush zero bits from the MXCSR control and status register. The flush zero may contain any of the following flags: _MM_FLUSH_ZERO_ON or _MM_FLUSH_ZERO_OFF + + Floating Point + Flag + AVX + Logical + + + + Compute the bitwise AND of 256 bits (representing double-precision (64-bit) floating-point elements) in "a" and "b", producing an intermediate 256-bit value, and set "ZF" to 1 if the sign bit of each 64-bit element in the intermediate value is zero, otherwise set "ZF" to 0. Compute the bitwise NOT of "a" and then AND with "b", producing an intermediate value, and set "CF" to 1 if the sign bit of each 64-bit element in the intermediate value is zero, otherwise set "CF" to 0. Return the "CF" value. -dst[31:0] := MXCSR & _MM_FLUSH_MASK +tmp[255:0] := a[255:0] AND b[255:0] +IF (tmp[63] == 0 && tmp[127] == 0 && tmp[191] == 0 && tmp[255] == 0) + ZF := 1 +ELSE + ZF := 0 +FI +tmp[255:0] := (NOT a[255:0]) AND b[255:0] +IF (tmp[63] == 0 && tmp[127] == 0 && tmp[191] == 0 && tmp[255] == 0) + CF := 1 +ELSE + CF := 0 +FI +dst := CF -
xmmintrin.h
+ +
immintrin.h
 - - SSE - General Support - - Macro: Set the flush zero bits of the MXCSR control and status register to the value in unsigned 32-bit integer "a". The flush zero may contain any of the following flags: _MM_FLUSH_ZERO_ON or _MM_FLUSH_ZERO_OFF + + Floating Point + Flag + AVX + Logical + + + + Compute the bitwise AND of 256 bits (representing double-precision (64-bit) floating-point elements) in "a" and "b", producing an intermediate 256-bit value, and set "ZF" to 1 if the sign bit of each 64-bit element in the intermediate value is zero, otherwise set "ZF" to 0. Compute the bitwise NOT of "a" and then AND with "b", producing an intermediate value, and set "CF" to 1 if the sign bit of each 64-bit element in the intermediate value is zero, otherwise set "CF" to 0. Return 1 if both the "ZF" and "CF" values are zero, otherwise return 0. -MXCSR := a[31:0] AND ~_MM_FLUSH_MASK +tmp[255:0] := a[255:0] AND b[255:0] +IF (tmp[63] == 0 && tmp[127] == 0 && tmp[191] == 0 && tmp[255] == 0) + ZF := 1 +ELSE + ZF := 0 +FI +tmp[255:0] := (NOT a[255:0]) AND b[255:0] +IF (tmp[63] == 0 && tmp[127] == 0 && tmp[191] == 0 && tmp[255] == 0) + CF := 1 +ELSE + CF := 0 +FI +IF (ZF == 0 && CF == 0) + dst := 1 +ELSE + dst := 0 +FI -
xmmintrin.h
- - - SSE - General Support - - - Fetch the line of data from memory that contains address "p" to a location in the cache heirarchy specified by the locality hint "i". - - - - -
xmmintrin.h
- - - KNCNI - General Support - - - Fetch the line of data from memory that contains address "p" to a location in the cache heirarchy specified by the locality hint "i". - - - - - - - - -
xmmintrin.h
- - - PREFETCHWT1 - General Support - - - Fetch the line of data from memory that contains address "p" to a location in the cache heirarchy specified by the locality hint "i". - -
xmmintrin.h
- - - SSE - General Support - - Perform a serializing operation on all store-to-memory instructions that were issued prior to this instruction. Guarantees that every store instruction that precedes, in program order, is globally visible before any store instruction which follows the fence in program order. - -
xmmintrin.h
+ +
immintrin.h
 - - Integer - SSE - Special Math Functions - - - Compare packed 16-bit integers in "a" and "b", and store packed maximum values in "dst". + + Floating Point + Flag + AVX + Logical + + + + Compute the bitwise AND of 128 bits (representing double-precision (64-bit) floating-point elements) in "a" and "b", producing an intermediate 128-bit value, and set "ZF" to 1 if the sign bit of each 64-bit element in the intermediate value is zero, otherwise set "ZF" to 0. Compute the bitwise NOT of "a" and then AND with "b", producing an intermediate value, and set "CF" to 1 if the sign bit of each 64-bit element in the intermediate value is zero, otherwise set "CF" to 0. Return the "ZF" value. -FOR j := 0 to 3 - i := j*16 - IF a[i+15:i] > b[i+15:i] - dst[i+15:i] := a[i+15:i] - ELSE - dst[i+15:i] := b[i+15:i] - FI -ENDFOR +tmp[127:0] := a[127:0] AND b[127:0] +IF (tmp[63] == 0 && tmp[127] == 0) + ZF := 1 +ELSE + ZF := 0 +FI +tmp[127:0] := (NOT a[127:0]) AND b[127:0] +IF (tmp[63] == 0 && tmp[127] == 0) + CF := 1 +ELSE + CF := 0 +FI +dst := ZF - -
xmmintrin.h
+ +
immintrin.h
 - - Integer - SSE - Special Math Functions - - - Compare packed 16-bit integers in "a" and "b", and store packed maximum values in "dst". + + Floating Point + Flag + AVX + Logical + + + + Compute the bitwise AND of 128 bits (representing double-precision (64-bit) floating-point elements) in "a" and "b", producing an intermediate 128-bit value, and set "ZF" to 1 if the sign bit of each 64-bit element in the intermediate value is zero, otherwise set "ZF" to 0. Compute the bitwise NOT of "a" and then AND with "b", producing an intermediate value, and set "CF" to 1 if the sign bit of each 64-bit element in the intermediate value is zero, otherwise set "CF" to 0. Return the "CF" value. -FOR j := 0 to 3 - i := j*16 - IF a[i+15:i] > b[i+15:i] - dst[i+15:i] := a[i+15:i] - ELSE - dst[i+15:i] := b[i+15:i] - FI -ENDFOR +tmp[127:0] := a[127:0] AND b[127:0] +IF (tmp[63] == 0 && tmp[127] == 0) + ZF := 1 +ELSE + ZF := 0 +FI +tmp[127:0] := (NOT a[127:0]) AND b[127:0] +IF (tmp[63] == 0 && tmp[127] == 0) + CF := 1 +ELSE + CF := 0 +FI +dst := CF - -
xmmintrin.h
+ +
immintrin.h
 - - Integer - SSE - Special Math Functions - - - Compare packed unsigned 8-bit integers in "a" and "b", and store packed maximum values in "dst". + + Floating Point + Flag + AVX + Logical + + + + Compute the bitwise AND of 128 bits (representing double-precision (64-bit) floating-point elements) in "a" and "b", producing an intermediate 128-bit value, and set "ZF" to 1 if the sign bit of each 64-bit element in the intermediate value is zero, otherwise set "ZF" to 0. Compute the bitwise NOT of "a" and then AND with "b", producing an intermediate value, and set "CF" to 1 if the sign bit of each 64-bit element in the intermediate value is zero, otherwise set "CF" to 0. Return 1 if both the "ZF" and "CF" values are zero, otherwise return 0. -FOR j := 0 to 7 - i := j*8 - IF a[i+7:i] > b[i+7:i] - dst[i+7:i] := a[i+7:i] - ELSE - dst[i+7:i] := b[i+7:i] - FI -ENDFOR - - -
xmmintrin.h
- - - Integer - SSE - Special Math Functions - - - Compare packed unsigned 8-bit integers in "a" and "b", and store packed maximum values in "dst". - -FOR j := 0 to 7 - i := j*8 - IF a[i+7:i] > b[i+7:i] - dst[i+7:i] := a[i+7:i] - ELSE - dst[i+7:i] := b[i+7:i] - FI -ENDFOR - - -
xmmintrin.h
- - - Integer - SSE - Special Math Functions - - - Compare packed 16-bit integers in "a" and "b", and store packed minimum values in "dst". - -FOR j := 0 to 3 - i := j*16 - IF a[i+15:i] < b[i+15:i] - dst[i+15:i] := a[i+15:i] - ELSE - dst[i+15:i] := b[i+15:i] - FI -ENDFOR - - -
xmmintrin.h
- - - Integer - SSE - Special Math Functions - - - Compare packed 16-bit integers in "a" and "b", and store packed minimum values in "dst". - -FOR j := 0 to 3 - i := j*16 - IF a[i+15:i] < b[i+15:i] - dst[i+15:i] := a[i+15:i] - ELSE - dst[i+15:i] := b[i+15:i] - FI -ENDFOR - - -
xmmintrin.h
- - - Integer - SSE - Special Math Functions - - - Compare packed unsigned 8-bit integers in "a" and "b", and store packed minimum values in "dst". - -FOR j := 0 to 7 - i := j*8 - IF a[i+7:i] < b[i+7:i] - dst[i+7:i] := a[i+7:i] - ELSE - dst[i+7:i] := b[i+7:i] - FI -ENDFOR +tmp[127:0] := a[127:0] AND b[127:0] +IF (tmp[63] == 0 && tmp[127] == 0) + ZF := 1 +ELSE + ZF := 0 +FI +tmp[127:0] := (NOT a[127:0]) AND b[127:0] +IF (tmp[63] == 0 && tmp[127] == 0) + CF := 1 +ELSE + CF := 0 +FI +IF (ZF == 0 && CF == 0) + dst := 1 +ELSE + dst := 0 +FI - -
xmmintrin.h
+ +
immintrin.h
 - - Integer - SSE - Special Math Functions - - - Compare packed unsigned 8-bit integers in "a" and "b", and store packed minimum values in "dst". + + Floating Point + Flag + AVX + Logical + + + + Compute the bitwise AND of 256 bits (representing single-precision (32-bit) floating-point elements) in "a" and "b", producing an intermediate 256-bit value, and set "ZF" to 1 if the sign bit of each 32-bit element in the intermediate value is zero, otherwise set "ZF" to 0. Compute the bitwise NOT of "a" and then AND with "b", producing an intermediate value, and set "CF" to 1 if the sign bit of each 32-bit element in the intermediate value is zero, otherwise set "CF" to 0. Return the "ZF" value. -FOR j := 0 to 7 - i := j*8 - IF a[i+7:i] < b[i+7:i] - dst[i+7:i] := a[i+7:i] - ELSE - dst[i+7:i] := b[i+7:i] - FI -ENDFOR +tmp[255:0] := a[255:0] AND b[255:0] +IF (tmp[31] == 0 && tmp[63] == 0 && tmp[95] == 0 && tmp[127] == 0 && \ + tmp[159] == 0 && tmp[191] == 0 && tmp[223] == 0 && tmp[255] == 0) + ZF := 1 +ELSE + ZF := 0 +FI +tmp[255:0] := (NOT a[255:0]) AND b[255:0] +IF (tmp[31] == 0 && tmp[63] == 0 && tmp[95] == 0 && tmp[127] == 0 && \ + tmp[159] == 0 && tmp[191] == 0 && tmp[223] == 0 && tmp[255] == 0) + CF := 1 +ELSE + CF := 0 +FI +dst := ZF - -
xmmintrin.h
+ +
immintrin.h
 - - Integer - SSE - Arithmetic - - - Multiply the packed unsigned 16-bit integers in "a" and "b", producing intermediate 32-bit integers, and store the high 16 bits of the intermediate integers in "dst". + + Floating Point + Flag + AVX + Logical + + + + Compute the bitwise AND of 256 bits (representing single-precision (32-bit) floating-point elements) in "a" and "b", producing an intermediate 256-bit value, and set "ZF" to 1 if the sign bit of each 32-bit element in the intermediate value is zero, otherwise set "ZF" to 0. Compute the bitwise NOT of "a" and then AND with "b", producing an intermediate value, and set "CF" to 1 if the sign bit of each 32-bit element in the intermediate value is zero, otherwise set "CF" to 0. Return the "CF" value. -FOR j := 0 to 3 - i := j*16 - tmp[31:0] := a[i+15:i] * b[i+15:i] - dst[i+15:i] := tmp[31:16] -ENDFOR +tmp[255:0] := a[255:0] AND b[255:0] +IF (tmp[31] == 0 && tmp[63] == 0 && tmp[95] == 0 && tmp[127] == 0 && \ + tmp[159] == 0 && tmp[191] == 0 && tmp[223] == 0 && tmp[255] == 0) + ZF := 1 +ELSE + ZF := 0 +FI +tmp[255:0] := (NOT a[255:0]) AND b[255:0] +IF (tmp[31] == 0 && tmp[63] == 0 && tmp[95] == 0 && tmp[127] == 0 && \ + tmp[159] == 0 && tmp[191] == 0 && tmp[223] == 0 && tmp[255] == 0) + CF := 1 +ELSE + CF := 0 +FI +dst := CF - -
xmmintrin.h
+ +
immintrin.h
 - - Integer - SSE - Arithmetic - - - Multiply the packed unsigned 16-bit integers in "a" and "b", producing intermediate 32-bit integers, and store the high 16 bits of the intermediate integers in "dst". + + Floating Point + Flag + AVX + Logical + + + + Compute the bitwise AND of 256 bits (representing single-precision (32-bit) floating-point elements) in "a" and "b", producing an intermediate 256-bit value, and set "ZF" to 1 if the sign bit of each 32-bit element in the intermediate value is zero, otherwise set "ZF" to 0. Compute the bitwise NOT of "a" and then AND with "b", producing an intermediate value, and set "CF" to 1 if the sign bit of each 32-bit element in the intermediate value is zero, otherwise set "CF" to 0. Return 1 if both the "ZF" and "CF" values are zero, otherwise return 0. -FOR j := 0 to 3 - i := j*16 - tmp[31:0] := a[i+15:i] * b[i+15:i] - dst[i+15:i] := tmp[31:16] -ENDFOR +tmp[255:0] := a[255:0] AND b[255:0] +IF (tmp[31] == 0 && tmp[63] == 0 && tmp[95] == 0 && tmp[127] == 0 && \ + tmp[159] == 0 && tmp[191] == 0 && tmp[223] == 0 && tmp[255] == 0) + ZF := 1 +ELSE + ZF := 0 +FI +tmp[255:0] := (NOT a[255:0]) AND b[255:0] +IF (tmp[31] == 0 && tmp[63] == 0 && tmp[95] == 0 && tmp[127] == 0 && \ + tmp[159] == 0 && tmp[191] == 0 && tmp[223] == 0 && tmp[255] == 0) + CF := 1 +ELSE + CF := 0 +FI +IF (ZF == 0 && CF == 0) + dst := 1 +ELSE + dst := 0 +FI - -
xmmintrin.h
+ +
immintrin.h
 - - Integer - SSE - Probability/Statistics - - - Average packed unsigned 8-bit integers in "a" and "b", and store the results in "dst". + + Floating Point + Flag + AVX + Logical + + + + Compute the bitwise AND of 128 bits (representing single-precision (32-bit) floating-point elements) in "a" and "b", producing an intermediate 128-bit value, and set "ZF" to 1 if the sign bit of each 32-bit element in the intermediate value is zero, otherwise set "ZF" to 0. Compute the bitwise NOT of "a" and then AND with "b", producing an intermediate value, and set "CF" to 1 if the sign bit of each 32-bit element in the intermediate value is zero, otherwise set "CF" to 0. Return the "ZF" value. -FOR j := 0 to 7 - i := j*8 - dst[i+7:i] := (a[i+7:i] + b[i+7:i] + 1) >> 1 -ENDFOR +tmp[127:0] := a[127:0] AND b[127:0] +IF (tmp[31] == 0 && tmp[63] == 0 && tmp[95] == 0 && tmp[127] == 0) + ZF := 1 +ELSE + ZF := 0 +FI +tmp[127:0] := (NOT a[127:0]) AND b[127:0] +IF (tmp[31] == 0 && tmp[63] == 0 && tmp[95] == 0 && tmp[127] == 0) + CF := 1 +ELSE + CF := 0 +FI +dst := ZF - -
xmmintrin.h
+ +
immintrin.h
 - - Integer - SSE - Probability/Statistics - - - Average packed unsigned 8-bit integers in "a" and "b", and store the results in "dst". + + Floating Point + Flag + AVX + Logical + + + + Compute the bitwise AND of 128 bits (representing single-precision (32-bit) floating-point elements) in "a" and "b", producing an intermediate 128-bit value, and set "ZF" to 1 if the sign bit of each 32-bit element in the intermediate value is zero, otherwise set "ZF" to 0. Compute the bitwise NOT of "a" and then AND with "b", producing an intermediate value, and set "CF" to 1 if the sign bit of each 32-bit element in the intermediate value is zero, otherwise set "CF" to 0. Return the "CF" value. -FOR j := 0 to 7 - i := j*8 - dst[i+7:i] := (a[i+7:i] + b[i+7:i] + 1) >> 1 -ENDFOR +tmp[127:0] := a[127:0] AND b[127:0] +IF (tmp[31] == 0 && tmp[63] == 0 && tmp[95] == 0 && tmp[127] == 0) + ZF := 1 +ELSE + ZF := 0 +FI +tmp[127:0] := (NOT a[127:0]) AND b[127:0] +IF (tmp[31] == 0 && tmp[63] == 0 && tmp[95] == 0 && tmp[127] == 0) + CF := 1 +ELSE + CF := 0 +FI +dst := CF - -
xmmintrin.h
+ +
immintrin.h
 - - Integer - SSE - Probability/Statistics - - - Average packed unsigned 16-bit integers in "a" and "b", and store the results in "dst". + + Floating Point + Flag + AVX + Logical + + + + Compute the bitwise AND of 128 bits (representing single-precision (32-bit) floating-point elements) in "a" and "b", producing an intermediate 128-bit value, and set "ZF" to 1 if the sign bit of each 32-bit element in the intermediate value is zero, otherwise set "ZF" to 0. Compute the bitwise NOT of "a" and then AND with "b", producing an intermediate value, and set "CF" to 1 if the sign bit of each 32-bit element in the intermediate value is zero, otherwise set "CF" to 0. Return 1 if both the "ZF" and "CF" values are zero, otherwise return 0. -FOR j := 0 to 3 - i := j*16 - dst[i+15:i] := (a[i+15:i] + b[i+15:i] + 1) >> 1 -ENDFOR +tmp[127:0] := a[127:0] AND b[127:0] +IF (tmp[31] == 0 && tmp[63] == 0 && tmp[95] == 0 && tmp[127] == 0) + ZF := 1 +ELSE + ZF := 0 +FI +tmp[127:0] := (NOT a[127:0]) AND b[127:0] +IF (tmp[31] == 0 && tmp[63] == 0 && tmp[95] == 0 && tmp[127] == 0) + CF := 1 +ELSE + CF := 0 +FI +IF (ZF == 0 && CF == 0) + dst := 1 +ELSE + dst := 0 +FI - -
xmmintrin.h
+ +
immintrin.h
 - - Integer - SSE - Probability/Statistics - - - Average packed unsigned 16-bit integers in "a" and "b", and store the results in "dst". + + Floating Point + AVX + Miscellaneous + + + Set each bit of mask "dst" based on the most significant bit of the corresponding packed double-precision (64-bit) floating-point element in "a". FOR j := 0 to 3 - i := j*16 - dst[i+15:i] := (a[i+15:i] + b[i+15:i] + 1) >> 1 + i := j*64 + IF a[i+63] + dst[j] := 1 + ELSE + dst[j] := 0 + FI ENDFOR +dst[MAX:4] := 0 - -
xmmintrin.h
+ +
immintrin.h
 - - Integer - SSE - Arithmetic + + Floating Point + AVX Miscellaneous - - - Compute the absolute differences of packed unsigned 8-bit integers in "a" and "b", then horizontally sum each consecutive 8 differences to produce four unsigned 16-bit integers, and pack these unsigned 16-bit integers in the low 16 bits of "dst". + + + Set each bit of mask "dst" based on the most significant bit of the corresponding packed single-precision (32-bit) floating-point element in "a". FOR j := 0 to 7 - i := j*8 - tmp[i+7:i] := ABS(a[i+7:i] - b[i+7:i]) + i := j*32 + IF a[i+31] + dst[j] := 1 + ELSE + dst[j] := 0 + FI ENDFOR - -dst[15:0] := tmp[7:0] + tmp[15:8] + tmp[23:16] + tmp[31:24] + tmp[39:32] + tmp[47:40] + tmp[55:48] + tmp[63:56] -dst[63:16] := 0 +dst[MAX:8] := 0 - -
xmmintrin.h
+ +
immintrin.h
 - + Floating Point - Integer - SSE - Arithmetic - Miscellaneous - - - Compute the absolute differences of packed unsigned 8-bit integers in "a" and "b", then horizontally sum each consecutive 8 differences to produce four unsigned 16-bit integers, and pack these unsigned 16-bit integers in the low 16 bits of "dst". + AVX + Set + + + Return vector of type __m256d with all elements set to zero. -FOR j := 0 to 7 - i := j*8 - tmp[i+7:i] := ABS(a[i+7:i] - b[i+7:i]) -ENDFOR - -dst[15:0] := tmp[7:0] + tmp[15:8] + tmp[23:16] + tmp[31:24] + tmp[39:32] + tmp[47:40] + tmp[55:48] + tmp[63:56] -dst[63:16] := 0 +dst[MAX:0] := 0 - -
xmmintrin.h
+ +
immintrin.h
 - + Floating Point - SSE - Convert - - - Convert the 32-bit integer "b" to a single-precision (32-bit) floating-point element, store the result in the lower element of "dst", and copy the upper 3 packed elements from "a" to the upper elements of "dst". + AVX + Set + + + Return vector of type __m256 with all elements set to zero. -dst[31:0] := Convert_Int32_To_FP32(b[31:0]) -dst[127:32] := a[127:32] +dst[MAX:0] := 0 - -
xmmintrin.h
+ +
immintrin.h
 - + Integer - SSE - Convert - - - Convert the 32-bit integer "b" to a single-precision (32-bit) floating-point element, store the result in the lower element of "dst", and copy the upper 3 packed elements from "a" to the upper elements of "dst". + AVX + Set + + + Return vector of type __m256i with all elements set to zero. -dst[31:0] := Convert_Int32_To_FP32(b[31:0]) -dst[127:32] := a[127:32] +dst[MAX:0] := 0 - -
xmmintrin.h
+ +
immintrin.h
 - + Floating Point - Integer - SSE - Convert - - - Convert the 64-bit integer "b" to a single-precision (32-bit) floating-point element, store the result in the lower element of "dst", and copy the upper 3 packed elements from "a" to the upper elements of "dst". + AVX + Set + + + + + + Set packed double-precision (64-bit) floating-point elements in "dst" with the supplied values. -dst[31:0] := Convert_Int64_To_FP32(b[63:0]) -dst[127:32] := a[127:32] -dst[MAX:128] := 0 +dst[63:0] := e0 +dst[127:64] := e1 +dst[191:128] := e2 +dst[255:192] := e3 +dst[MAX:256] := 0 - -
xmmintrin.h
+
immintrin.h
 - + Floating Point - SSE - Convert - - - Convert packed 32-bit integers in "b" to packed single-precision (32-bit) floating-point elements, store the results in the lower 2 elements of "dst", and copy the upper 2 packed elements from "a" to the upper elements of "dst". + AVX + Set + + + + + + + + + + Set packed single-precision (32-bit) floating-point elements in "dst" with the supplied values. -dst[31:0] := Convert_Int32_To_FP32(b[31:0]) -dst[63:32] := Convert_Int32_To_FP32(b[63:32]) -dst[95:64] := a[95:64] -dst[127:96] := a[127:96] +dst[31:0] := e0 +dst[63:32] := e1 +dst[95:64] := e2 +dst[127:96] := e3 +dst[159:128] := e4 +dst[191:160] := e5 +dst[223:192] := e6 +dst[255:224] := e7 +dst[MAX:256] := 0 - -
xmmintrin.h
+
immintrin.h
 - + Integer - SSE - Convert - - - Convert packed 32-bit integers in "b" to packed single-precision (32-bit) floating-point elements, store the results in the lower 2 elements of "dst", and copy the upper 2 packed elements from "a" to the upper elements of "dst". + AVX + Set + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + Set packed 8-bit integers in "dst" with the supplied values. -dst[31:0] := Convert_Int32_To_FP32(b[31:0]) -dst[63:32] := Convert_Int32_To_FP32(b[63:32]) -dst[95:64] := a[95:64] -dst[127:96] := a[127:96] +dst[7:0] := e0 +dst[15:8] := e1 +dst[23:16] := e2 +dst[31:24] := e3 +dst[39:32] := e4 +dst[47:40] := e5 +dst[55:48] := e6 +dst[63:56] := e7 +dst[71:64] := e8 +dst[79:72] := e9 +dst[87:80] := e10 +dst[95:88] := e11 +dst[103:96] := e12 +dst[111:104] := e13 +dst[119:112] := e14 +dst[127:120] := e15 +dst[135:128] := e16 +dst[143:136] := e17 +dst[151:144] := e18 +dst[159:152] := e19 +dst[167:160] := e20 +dst[175:168] := e21 +dst[183:176] := e22 +dst[191:184] := e23 +dst[199:192] := e24 +dst[207:200] := e25 +dst[215:208] := e26 +dst[223:216] := e27 +dst[231:224] := e28 +dst[239:232] := e29 +dst[247:240] := e30 +dst[255:248] := e31 +dst[MAX:256] := 0 - -
xmmintrin.h
+
immintrin.h
 - - Floating Point - SSE - Convert - - Convert packed 16-bit integers in "a" to packed single-precision (32-bit) floating-point elements, and store the results in "dst". + + Integer + AVX + Set + + + + + + + + + + + + + + + + + + Set packed 16-bit integers in "dst" with the supplied values. -FOR j := 0 to 3 - i := j*16 - m := j*32 - dst[m+31:m] := Convert_Int16_To_FP32(a[i+15:i]) -ENDFOR +dst[15:0] := e0 +dst[31:16] := e1 +dst[47:32] := e2 +dst[63:48] := e3 +dst[79:64] := e4 +dst[95:80] := e5 +dst[111:96] := e6 +dst[127:112] := e7 +dst[143:128] := e8 +dst[159:144] := e9 +dst[175:160] := e10 +dst[191:176] := e11 +dst[207:192] := e12 +dst[223:208] := e13 +dst[239:224] := e14 +dst[255:240] := e15 +dst[MAX:256] := 0 -
xmmintrin.h
+
immintrin.h
 - - Floating Point + Integer - SSE - Convert - - Convert packed unsigned 16-bit integers in "a" to packed single-precision (32-bit) floating-point elements, and store the results in "dst". + AVX + Set + + + + + + + + + + Set packed 32-bit integers in "dst" with the supplied values. -FOR j := 0 to 3 - i := j*16 - m := j*32 - dst[m+31:m] := Convert_UnsignedInt16_To_FP32(a[i+15:i]) -ENDFOR +dst[31:0] := e0 +dst[63:32] := e1 +dst[95:64] := e2 +dst[127:96] := e3 +dst[159:128] := e4 +dst[191:160] := e5 +dst[223:192] := e6 +dst[255:224] := e7 +dst[MAX:256] := 0 -
xmmintrin.h
+
immintrin.h
 - - Floating Point - SSE - Convert - - Convert the lower packed 8-bit integers in "a" to packed single-precision (32-bit) floating-point elements, and store the results in "dst". + + Integer + AVX + Set + + + + + + Set packed 64-bit integers in "dst" with the supplied values. -FOR j := 0 to 3 - i := j*8 - m := j*32 - dst[m+31:m] := Convert_Int8_To_FP32(a[i+7:i]) -ENDFOR +dst[63:0] := e0 +dst[127:64] := e1 +dst[191:128] := e2 +dst[255:192] := e3 +dst[MAX:256] := 0 -
xmmintrin.h
+
immintrin.h
 - + Floating Point - Integer - SSE - Convert - - Convert the lower packed unsigned 8-bit integers in "a" to packed single-precision (32-bit) floating-point elements, and store the results in "dst". + AVX + Set + + + + + + Set packed double-precision (64-bit) floating-point elements in "dst" with the supplied values in reverse order. -FOR j := 0 to 3 - i := j*8 - m := j*32 - dst[m+31:m] := Convert_UnsignedInt8_To_FP32(a[i+7:i]) -ENDFOR +dst[63:0] := e3 +dst[127:64] := e2 +dst[191:128] := e1 +dst[255:192] := e0 +dst[MAX:256] := 0 -
xmmintrin.h
+
immintrin.h
 - + Floating Point - SSE - Convert - - - Convert packed 32-bit integers in "a" to packed single-precision (32-bit) floating-point elements, store the results in the lower 2 elements of "dst", then covert the packed 32-bit integers in "a" to single-precision (32-bit) floating-point element, and store the results in the upper 2 elements of "dst". + AVX + Set + + + + + + + + + + Set packed single-precision (32-bit) floating-point elements in "dst" with the supplied values in reverse order. -dst[31:0] := Convert_Int32_To_FP32(a[31:0]) -dst[63:32] := Convert_Int32_To_FP32(a[63:32]) -dst[95:64] := Convert_Int32_To_FP32(b[31:0]) -dst[127:96] := Convert_Int32_To_FP32(b[63:32]) +dst[31:0] := e7 +dst[63:32] := e6 +dst[95:64] := e5 +dst[127:96] := e4 +dst[159:128] := e3 +dst[191:160] := e2 +dst[223:192] := e1 +dst[255:224] := e0 +dst[MAX:256] := 0 -
xmmintrin.h
+
immintrin.h
 - + Integer - SSE - Store - - - Store 64-bits of integer data from "a" into memory using a non-temporal memory hint. + AVX + Set + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + Set packed 8-bit integers in "dst" with the supplied values in reverse order. -MEM[mem_addr+63:mem_addr] := a[63:0] - - -
xmmintrin.h
- - - Integer - SSE - Store - - - - Conditionally store 8-bit integer elements from "a" into memory using "mask" (elements are not stored when the highest bit is not set in the corresponding element) and a non-temporal memory hint. - -FOR j := 0 to 7 - i := j*8 - IF mask[i+7] - MEM[mem_addr+i+7:mem_addr+i] := a[i+7:i] - FI -ENDFOR - - -
xmmintrin.h
- - - Integer - SSE - Store - - - - Conditionally store 8-bit integer elements from "a" into memory using "mask" (elements are not stored when the highest bit is not set in the corresponding element). - -FOR j := 0 to 7 - i := j*8 - IF mask[i+7] - MEM[mem_addr+i+7:mem_addr+i] := a[i+7:i] - FI -ENDFOR +dst[7:0] := e31 +dst[15:8] := e30 +dst[23:16] := e29 +dst[31:24] := e28 +dst[39:32] := e27 +dst[47:40] := e26 +dst[55:48] := e25 +dst[63:56] := e24 +dst[71:64] := e23 +dst[79:72] := e22 +dst[87:80] := e21 +dst[95:88] := e20 +dst[103:96] := e19 +dst[111:104] := e18 +dst[119:112] := e17 +dst[127:120] := e16 +dst[135:128] := e15 +dst[143:136] := e14 +dst[151:144] := e13 +dst[159:152] := e12 +dst[167:160] := e11 +dst[175:168] := e10 +dst[183:176] := e9 +dst[191:184] := e8 +dst[199:192] := e7 +dst[207:200] := e6 +dst[215:208] := e5 +dst[223:216] := e4 +dst[231:224] := e3 +dst[239:232] := e2 +dst[247:240] := e1 +dst[255:248] := e0 +dst[MAX:256] := 0 - -
xmmintrin.h
+
immintrin.h
 - + Integer - SSE - Swizzle - - - Extract a 16-bit integer from "a", selected with "imm8", and store the result in the lower element of "dst". + AVX + Set + + + + + + + + + + + + + + + + + + Set packed 16-bit integers in "dst" with the supplied values in reverse order. -dst[15:0] := (a[63:0] >> (imm8[1:0] * 16))[15:0] -dst[31:16] := 0 +dst[15:0] := e15 +dst[31:16] := e14 +dst[47:32] := e13 +dst[63:48] := e12 +dst[79:64] := e11 +dst[95:80] := e10 +dst[111:96] := e9 +dst[127:112] := e8 +dst[143:128] := e7 +dst[159:144] := e6 +dst[175:160] := e5 +dst[191:176] := e4 +dst[207:192] := e3 +dst[223:208] := e2 +dst[239:224] := e1 +dst[255:240] := e0 +dst[MAX:256] := 0 - -
xmmintrin.h
+
immintrin.h
 - + Integer - SSE - Swizzle - - - Extract a 16-bit integer from "a", selected with "imm8", and store the result in the lower element of "dst". + AVX + Set + + + + + + + + + + Set packed 32-bit integers in "dst" with the supplied values in reverse order. -dst[15:0] := (a[63:0] >> (imm8[1:0] * 16))[15:0] -dst[31:16] := 0 +dst[31:0] := e7 +dst[63:32] := e6 +dst[95:64] := e5 +dst[127:96] := e4 +dst[159:128] := e3 +dst[191:160] := e2 +dst[223:192] := e1 +dst[255:224] := e0 +dst[MAX:256] := 0 - -
xmmintrin.h
+
immintrin.h
 - + Integer - SSE - Swizzle - - - - Copy "a" to "dst", and insert the 16-bit integer "i" into "dst" at the location specified by "imm8". + AVX + Set + + + + + + Set packed 64-bit integers in "dst" with the supplied values in reverse order. -dst[63:0] := a[63:0] -sel := imm8[1:0]*16 -dst[sel+15:sel] := i[15:0] +dst[63:0] := e3 +dst[127:64] := e2 +dst[191:128] := e1 +dst[255:192] := e0 +dst[MAX:256] := 0 - -
xmmintrin.h
+
immintrin.h
 - - Integer - SSE - Swizzle - - - - Copy "a" to "dst", and insert the 16-bit integer "i" into "dst" at the location specified by "imm8". + + Floating Point + AVX + Set + + + Broadcast double-precision (64-bit) floating-point value "a" to all elements of "dst". -dst[63:0] := a[63:0] -sel := imm8[1:0]*16 -dst[sel+15:sel] := i[15:0] +FOR j := 0 to 3 + i := j*64 + dst[i+63:i] := a[63:0] +ENDFOR +dst[MAX:256] := 0 - -
xmmintrin.h
+
immintrin.h
 - - Integer - SSE - Miscellaneous - - Create mask from the most significant bit of each 8-bit element in "a", and store the result in "dst". + + Floating Point + AVX + Set + + + Broadcast single-precision (32-bit) floating-point value "a" to all elements of "dst". FOR j := 0 to 7 - i := j*8 - dst[j] := a[i+7] + i := j*32 + dst[i+31:i] := a[31:0] ENDFOR -dst[MAX:8] := 0 +dst[MAX:256] := 0 - -
xmmintrin.h
+
immintrin.h
 - + Integer - SSE - Miscellaneous - - Create mask from the most significant bit of each 8-bit element in "a", and store the result in "dst". + AVX + Set + + + Broadcast 8-bit integer "a" to all elements of "dst". This intrinsic may generate the "vpbroadcastb". -FOR j := 0 to 7 +FOR j := 0 to 31 i := j*8 - dst[j] := a[i+7] + dst[i+7:i] := a[7:0] ENDFOR -dst[MAX:8] := 0 +dst[MAX:256] := 0 - -
xmmintrin.h
+
immintrin.h
 - + Integer - SSE - Swizzle - - - Shuffle 16-bit integers in "a" using the control in "imm8", and store the results in "dst". + AVX + Set + + + Broadcast 16-bit integer "a" to all all elements of "dst". This intrinsic may generate the "vpbroadcastw". -SELECT4(src, control){ - CASE(control[1:0]) - 0: tmp[15:0] := src[15:0] - 1: tmp[15:0] := src[31:16] - 2: tmp[15:0] := src[47:32] - 3: tmp[15:0] := src[63:48] - ESAC - RETURN tmp[15:0] -} - -dst[15:0] := SELECT4(a[63:0], imm8[1:0]) -dst[31:16] := SELECT4(a[63:0], imm8[3:2]) -dst[47:32] := SELECT4(a[63:0], imm8[5:4]) -dst[63:48] := SELECT4(a[63:0], imm8[7:6]) +FOR j := 0 to 15 + i := j*16 + dst[i+15:i] := a[15:0] +ENDFOR +dst[MAX:256] := 0 - -
xmmintrin.h
+
immintrin.h
 - - Floating Point + Integer - SSE - Swizzle - - - Shuffle 16-bit integers in "a" using the control in "imm8", and store the results in "dst". - -SELECT4(src, control){ - CASE(control[1:0]) - 0: tmp[15:0] := src[15:0] - 1: tmp[15:0] := src[31:16] - 2: tmp[15:0] := src[47:32] - 3: tmp[15:0] := src[63:48] - ESAC - RETURN tmp[15:0] -} - -dst[15:0] := SELECT4(a[63:0], imm8[1:0]) -dst[31:16] := SELECT4(a[63:0], imm8[3:2]) -dst[47:32] := SELECT4(a[63:0], imm8[5:4]) -dst[63:48] := SELECT4(a[63:0], imm8[7:6]) - - -
xmmintrin.h
- - - Floating Point - SSE - Arithmetic - - - Add the lower single-precision (32-bit) floating-point element in "a" and "b", store the result in the lower element of "dst", and copy the upper 3 packed elements from "a" to the upper elements of "dst". + AVX + Set + + + Broadcast 32-bit integer "a" to all elements of "dst". This intrinsic may generate the "vpbroadcastd". -dst[31:0] := a[31:0] + b[31:0] -dst[127:32] := a[127:32] +FOR j := 0 to 7 + i := j*32 + dst[i+31:i] := a[31:0] +ENDFOR +dst[MAX:256] := 0 - -
xmmintrin.h
+
immintrin.h
 - - Floating Point - SSE - Arithmetic - - - Add packed single-precision (32-bit) floating-point elements in "a" and "b", and store the results in "dst". + + Integer + AVX + Set + + + Broadcast 64-bit integer "a" to all elements of "dst". This intrinsic may generate the "vpbroadcastq". FOR j := 0 to 3 - i := j*32 - dst[i+31:i] := a[i+31:i] + b[i+31:i] + i := j*64 + dst[i+63:i] := a[63:0] ENDFOR +dst[MAX:256] := 0 - -
xmmintrin.h
+
immintrin.h
 - + Floating Point - SSE - Arithmetic - - - Subtract the lower single-precision (32-bit) floating-point element in "b" from the lower single-precision (32-bit) floating-point element in "a", store the result in the lower element of "dst", and copy the upper 3 packed elements from "a" to the upper elements of "dst". - -dst[31:0] := a[31:0] - b[31:0] -dst[127:32] := a[127:32] - - -
xmmintrin.h
+ AVX + Cast + + + Cast vector of type __m256d to type __m256. + This intrinsic is only used for compilation and does not generate any instructions, thus it has zero latency. +
immintrin.h
 - + Floating Point - SSE - Arithmetic - - - Subtract packed single-precision (32-bit) floating-point elements in "b" from packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst". - -FOR j := 0 to 3 - i := j*32 - dst[i+31:i] := a[i+31:i] - b[i+31:i] -ENDFOR - - -
xmmintrin.h
+ AVX + Cast + + + Cast vector of type __m256 to type __m256d. + This intrinsic is only used for compilation and does not generate any instructions, thus it has zero latency. +
immintrin.h
 - + Floating Point - SSE - Arithmetic - - - Multiply the lower single-precision (32-bit) floating-point element in "a" and "b", store the result in the lower element of "dst", and copy the upper 3 packed elements from "a" to the upper elements of "dst". - -dst[31:0] := a[31:0] * b[31:0] -dst[127:32] := a[127:32] - - -
xmmintrin.h
+ Integer + AVX + Cast + + + Cast vector of type __m256 to type __m256i. This intrinsic is only used for compilation and does not generate any instructions, thus it has zero latency. +
immintrin.h
 - + Floating Point - SSE - Arithmetic - - - Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", and store the results in "dst". - - -FOR j := 0 to 3 - i := j*32 - dst[i+31:i] := a[i+31:i] * b[i+31:i] -ENDFOR - - -
xmmintrin.h
+ Integer + AVX + Cast + + + Cast vector of type __m256d to type __m256i. This intrinsic is only used for compilation and does not generate any instructions, thus it has zero latency. +
immintrin.h
 - + Floating Point - SSE - Arithmetic - - - Divide the lower single-precision (32-bit) floating-point element in "a" by the lower single-precision (32-bit) floating-point element in "b", store the result in the lower element of "dst", and copy the upper 3 packed elements from "a" to the upper elements of "dst". - -dst[31:0] := a[31:0] / b[31:0] -dst[127:32] := a[127:32] - - -
xmmintrin.h
+ AVX + Cast + + + Cast vector of type __m256i to type __m256. This intrinsic is only used for compilation and does not generate any instructions, thus it has zero latency. +
immintrin.h
 - + Floating Point - SSE - Arithmetic - - - Divide packed single-precision (32-bit) floating-point elements in "a" by packed elements in "b", and store the results in "dst". - -FOR j := 0 to 3 - i := 32*j - dst[i+31:i] := a[i+31:i] / b[i+31:i] -ENDFOR - - -
xmmintrin.h
+ AVX + Cast + + + Cast vector of type __m256i to type __m256d. This intrinsic is only used for compilation and does not generate any instructions, thus it has zero latency. +
immintrin.h
 - + Floating Point - SSE - Elementary Math Functions - - Compute the square root of the lower single-precision (32-bit) floating-point element in "a", store the result in the lower element of "dst", and copy the upper 3 packed elements from "a" to the upper elements of "dst". - -dst[31:0] := SQRT(a[31:0]) -dst[127:32] := a[127:32] - - -
xmmintrin.h
+ AVX + Cast + + + Cast vector of type __m256 to type __m128. This intrinsic is only used for compilation and does not generate any instructions, thus it has zero latency. +
immintrin.h
 - + Floating Point - SSE - Elementary Math Functions - - Compute the square root of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst". - -FOR j := 0 to 3 - i := j*32 - dst[i+31:i] := SQRT(a[i+31:i]) -ENDFOR - - -
xmmintrin.h
+ AVX + Cast + + + Cast vector of type __m256d to type __m128d. This intrinsic is only used for compilation and does not generate any instructions, thus it has zero latency. +
immintrin.h
 - - Floating Point - SSE - Elementary Math Functions - - Compute the approximate reciprocal of the lower single-precision (32-bit) floating-point element in "a", store the result in the lower element of "dst", and copy the upper 3 packed elements from "a" to the upper elements of "dst". The maximum relative error for this approximation is less than 1.5*2^-12. - -dst[31:0] := APPROXIMATE(1.0/a[31:0]) -dst[127:32] := a[127:32] - - -
xmmintrin.h
+ + Integer + AVX + Cast + + + Cast vector of type __m256i to type __m128i. This intrinsic is only used for compilation and does not generate any instructions, thus it has zero latency. +
immintrin.h
 - + Floating Point - SSE - Elementary Math Functions - - Compute the approximate reciprocal of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst". The maximum relative error for this approximation is less than 1.5*2^-12. - -FOR j := 0 to 3 - i := j*32 - dst[i+31:i] := APPROXIMATE(1.0/a[i+31:i]) -ENDFOR - - -
xmmintrin.h
+ AVX + Cast + + + Cast vector of type __m128 to type __m256; the upper 128 bits of the result are undefined. This intrinsic is only used for compilation and does not generate any instructions, thus it has zero latency. +
immintrin.h
 - + Floating Point - SSE - Elementary Math Functions - - Compute the approximate reciprocal square root of the lower single-precision (32-bit) floating-point element in "a", store the result in the lower element of "dst", and copy the upper 3 packed elements from "a" to the upper elements of "dst". The maximum relative error for this approximation is less than 1.5*2^-12. - -dst[31:0] := APPROXIMATE(1.0 / SQRT(a[31:0])) -dst[127:32] := a[127:32] - - -
xmmintrin.h
+ AVX + Cast + + + Cast vector of type __m128d to type __m256d; the upper 128 bits of the result are undefined. This intrinsic is only used for compilation and does not generate any instructions, thus it has zero latency. +
immintrin.h
+ + + Integer + AVX + Cast + + + Cast vector of type __m128i to type __m256i; the upper 128 bits of the result are undefined. This intrinsic is only used for compilation and does not generate any instructions, thus it has zero latency. +
immintrin.h
 - + Floating Point - SSE - Elementary Math Functions - - Compute the approximate reciprocal square root of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst". The maximum relative error for this approximation is less than 1.5*2^-12. - -FOR j := 0 to 3 - i := j*32 - dst[i+31:i] := APPROXIMATE(1.0 / SQRT(a[i+31:i])) -ENDFOR - - -
xmmintrin.h
+ AVX + Cast + + + Cast vector of type __m128 to type __m256; the upper 128 bits of the result are zeroed. This intrinsic is only used for compilation and does not generate any instructions, thus it has zero latency. +
immintrin.h
 - + Floating Point - SSE - Special Math Functions - - - Compare the lower single-precision (32-bit) floating-point elements in "a" and "b", store the minimum value in the lower element of "dst", and copy the upper element from "a" to the upper element of "dst". - -dst[31:0] := MIN(a[31:0], b[31:0]) -dst[127:32] := a[127:32] - - -
xmmintrin.h
+ AVX + Cast + + + Cast vector of type __m128d to type __m256d; the upper 128 bits of the result are zeroed. This intrinsic is only used for compilation and does not generate any instructions, thus it has zero latency. +
immintrin.h
+ + + Integer + AVX + Cast + + + Cast vector of type __m128i to type __m256i; the upper 128 bits of the result are zeroed. This intrinsic is only used for compilation and does not generate any instructions, thus it has zero latency. +
immintrin.h
 - + Floating Point - SSE + AVX Special Math Functions - - - Compare packed single-precision (32-bit) floating-point elements in "a" and "b", and store packed minimum values in "dst". - + + + Round the packed single-precision (32-bit) floating-point elements in "a" down to an integer value, and store the results as packed single-precision floating-point elements in "dst". -FOR j := 0 to 3 +FOR j := 0 to 7 i := j*32 - dst[i+31:i] := MIN(a[i+31:i], b[i+31:i]) + dst[i+31:i] := FLOOR(a[i+31:i]) ENDFOR +dst[MAX:256] := 0 - -
xmmintrin.h
+ +
immintrin.h
 - + Floating Point - SSE + AVX Special Math Functions - - - Compare the lower single-precision (32-bit) floating-point elements in "a" and "b", store the maximum value in the lower element of "dst", and copy the upper element from "a" to the upper element of "dst". + + + Round the packed single-precision (32-bit) floating-point elements in "a" up to an integer value, and store the results as packed single-precision floating-point elements in "dst". -dst[31:0] := MAX(a[31:0], b[31:0]) -dst[127:32] := a[127:32] +FOR j := 0 to 7 + i := j*32 + dst[i+31:i] := CEIL(a[i+31:i]) +ENDFOR +dst[MAX:256] := 0 - -
xmmintrin.h
+ +
immintrin.h
 - + Floating Point - SSE + AVX Special Math Functions - - - Compare packed single-precision (32-bit) floating-point elements in "a" and "b", and store packed maximum values in "dst". + + + Round the packed double-precision (64-bit) floating-point elements in "a" down to an integer value, and store the results as packed double-precision floating-point elements in "dst". FOR j := 0 to 3 - i := j*32 - dst[i+31:i] := MAX(a[i+31:i], b[i+31:i]) + i := j*64 + dst[i+63:i] := FLOOR(a[i+63:i]) ENDFOR +dst[MAX:256] := 0 - -
xmmintrin.h
+ +
immintrin.h
 - + Floating Point - SSE - Logical - - - Compute the bitwise AND of packed single-precision (32-bit) floating-point elements in "a" and "b", and store the results in "dst". + AVX + Special Math Functions + + + Round the packed double-precision (64-bit) floating-point elements in "a" up to an integer value, and store the results as packed double-precision floating-point elements in "dst". FOR j := 0 to 3 - i := j*32 - dst[i+31:i] := (a[i+31:i] AND b[i+31:i]) + i := j*64 + dst[i+63:i] := CEIL(a[i+63:i]) ENDFOR +dst[MAX:256] := 0 - -
xmmintrin.h
+ +
immintrin.h
 - + Floating Point - SSE - Logical - - - Compute the bitwise NOT of packed single-precision (32-bit) floating-point elements in "a" and then AND with "b", and store the results in "dst". - -FOR j := 0 to 3 - i := j*32 - dst[i+31:i] := ((NOT a[i+31:i]) AND b[i+31:i]) -ENDFOR - - -
xmmintrin.h
+ AVX + General Support + + + Return vector of type __m256 with undefined elements. +
immintrin.h
 - + Floating Point - SSE - Logical - - - Compute the bitwise OR of packed single-precision (32-bit) floating-point elements in "a" and "b", and store the results in "dst". - - -FOR j := 0 to 3 - i := j*32 - dst[i+31:i] := a[i+31:i] BITWISE OR b[i+31:i] -ENDFOR - - -
xmmintrin.h
+ AVX + General Support + + + Return vector of type __m256d with undefined elements. +
immintrin.h
+ + + Integer + AVX + General Support + + + Return vector of type __m256i with undefined elements. +
immintrin.h
 - + Floating Point - SSE - Logical - - - Compute the bitwise XOR of packed single-precision (32-bit) floating-point elements in "a" and "b", and store the results in "dst". - + AVX + Set + + + + Set packed __m256 vector "dst" with the supplied values. -FOR j := 0 to 3 - i := j*32 - dst[i+31:i] := a[i+31:i] XOR b[i+31:i] -ENDFOR +dst[127:0] := lo[127:0] +dst[255:128] := hi[127:0] +dst[MAX:256] := 0 - -
xmmintrin.h
+ +
immintrin.h
 - + Floating Point - SSE - Compare - - - Compare the lower single-precision (32-bit) floating-point elements in "a" and "b" for equality, store the result in the lower element of "dst", and copy the upper 3 packed elements from "a" to the upper elements of "dst". + AVX + Set + + + + Set packed __m256d vector "dst" with the supplied values. -dst[31:0] := ( a[31:0] == b[31:0] ) ? 0xffffffff : 0 -dst[127:32] := a[127:32] +dst[127:0] := lo[127:0] +dst[255:128] := hi[127:0] +dst[MAX:256] := 0 - -
xmmintrin.h
+ +
immintrin.h
 - - Floating Point - SSE - Compare - - - Compare packed single-precision (32-bit) floating-point elements in "a" and "b" for equality, and store the results in "dst". - -FOR j := 0 to 3 - i := j*32 - dst[i+31:i] := ( a[i+31:i] == b[i+31:i] ) ? 0xffffffff : 0 -ENDFOR + + Integer + AVX + Set + + + + Set packed __m256i vector "dst" with the supplied values. + +dst[127:0] := lo[127:0] +dst[255:128] := hi[127:0] +dst[MAX:256] := 0 - -
xmmintrin.h
+ +
immintrin.h
 - + Floating Point - SSE - Compare - - - Compare the lower single-precision (32-bit) floating-point elements in "a" and "b" for less-than, store the result in the lower element of "dst", and copy the upper 3 packed elements from "a" to the upper elements of "dst". + AVX + Set + + + + Set packed __m256 vector "dst" with the supplied values. -dst[31:0] := ( a[31:0] < b[31:0] ) ? 0xffffffff : 0 -dst[127:32] := a[127:32] +dst[127:0] := lo[127:0] +dst[255:128] := hi[127:0] +dst[MAX:256] := 0 - -
xmmintrin.h
+ +
immintrin.h
 - + Floating Point - SSE - Compare - - - Compare packed single-precision (32-bit) floating-point elements in "a" and "b" for less-than, and store the results in "dst". + AVX + Set + + + + Set packed __m256d vector "dst" with the supplied values. -FOR j := 0 to 3 - i := j*32 - dst[i+31:i] := ( a[i+31:i] < b[i+31:i] ) ? 0xffffffff : 0 -ENDFOR +dst[127:0] := lo[127:0] +dst[255:128] := hi[127:0] +dst[MAX:256] := 0 - -
xmmintrin.h
+ +
immintrin.h
 - - Floating Point - SSE - Compare - - - Compare the lower single-precision (32-bit) floating-point elements in "a" and "b" for less-than-or-equal, store the result in the lower element of "dst", and copy the upper 3 packed elements from "a" to the upper elements of "dst". + + Integer + AVX + Set + + + + Set packed __m256i vector "dst" with the supplied values. -dst[31:0] := ( a[31:0] <= b[31:0] ) ? 0xffffffff : 0 -dst[127:32] := a[127:32] +dst[127:0] := lo[127:0] +dst[255:128] := hi[127:0] +dst[MAX:256] := 0 - -
xmmintrin.h
+ +
immintrin.h
 - + Floating Point - SSE - Compare - - - Compare packed single-precision (32-bit) floating-point elements in "a" and "b" for less-than-or-equal, and store the results in "dst". + AVX + Load + + + + Load two 128-bit values (composed of 4 packed single-precision (32-bit) floating-point elements) from memory, and combine them into a 256-bit value in "dst". + "hiaddr" and "loaddr" do not need to be aligned on any particular boundary. -FOR j := 0 to 3 - i := j*32 - dst[i+31:i] := ( a[i+31:i] <= b[i+31:i] ) ? 0xffffffff : 0 -ENDFOR +dst[127:0] := MEM[loaddr+127:loaddr] +dst[255:128] := MEM[hiaddr+127:hiaddr] +dst[MAX:256] := 0 - -
xmmintrin.h
+
immintrin.h
 - + Floating Point - SSE - Compare - - - Compare the lower single-precision (32-bit) floating-point elements in "a" and "b" for greater-than, store the result in the lower element of "dst", and copy the upper 3 packed elements from "a" to the upper elements of "dst". + AVX + Load + + + + Load two 128-bit values (composed of 2 packed double-precision (64-bit) floating-point elements) from memory, and combine them into a 256-bit value in "dst". + "hiaddr" and "loaddr" do not need to be aligned on any particular boundary. -dst[31:0] := ( a[31:0] > b[31:0] ) ? 0xffffffff : 0 -dst[127:32] := a[127:32] +dst[127:0] := MEM[loaddr+127:loaddr] +dst[255:128] := MEM[hiaddr+127:hiaddr] +dst[MAX:256] := 0 - -
xmmintrin.h
+
immintrin.h
 - - Floating Point - SSE - Compare - - - Compare packed single-precision (32-bit) floating-point elements in "a" and "b" for greater-than, and store the results in "dst". + + Integer + AVX + Load + + + + Load two 128-bit values (composed of integer data) from memory, and combine them into a 256-bit value in "dst". + "hiaddr" and "loaddr" do not need to be aligned on any particular boundary. -FOR j := 0 to 3 - i := j*32 - dst[i+31:i] := ( a[i+31:i] > b[i+31:i] ) ? 0xffffffff : 0 -ENDFOR +dst[127:0] := MEM[loaddr+127:loaddr] +dst[255:128] := MEM[hiaddr+127:hiaddr] +dst[MAX:256] := 0 - -
xmmintrin.h
+
immintrin.h
 - + Floating Point - SSE - Compare - - - Compare the lower single-precision (32-bit) floating-point elements in "a" and "b" for greater-than-or-equal, store the result in the lower element of "dst", and copy the upper 3 packed elements from "a" to the upper elements of "dst". + AVX + Store + + + + + Store the high and low 128-bit halves (each composed of 4 packed single-precision (32-bit) floating-point elements) from "a" into memory two different 128-bit locations. + "hiaddr" and "loaddr" do not need to be aligned on any particular boundary. -dst[31:0] := ( a[31:0] >= b[31:0] ) ? 0xffffffff : 0 -dst[127:32] := a[127:32] +MEM[loaddr+127:loaddr] := a[127:0] +MEM[hiaddr+127:hiaddr] := a[255:128] - -
xmmintrin.h
+
immintrin.h
 - + Floating Point - SSE - Compare - - - Compare packed single-precision (32-bit) floating-point elements in "a" and "b" for greater-than-or-equal, and store the results in "dst". + AVX + Store + + + + + Store the high and low 128-bit halves (each composed of 2 packed double-precision (64-bit) floating-point elements) from "a" into memory two different 128-bit locations. + "hiaddr" and "loaddr" do not need to be aligned on any particular boundary. -FOR j := 0 to 3 - i := j*32 - dst[i+31:i] := ( a[i+31:i] >= b[i+31:i] ) ? 0xffffffff : 0 -ENDFOR +MEM[loaddr+127:loaddr] := a[127:0] +MEM[hiaddr+127:hiaddr] := a[255:128] - -
xmmintrin.h
+
immintrin.h
 - - Floating Point - SSE - Compare - - - Compare the lower single-precision (32-bit) floating-point elements in "a" and "b" for not-equal, store the result in the lower element of "dst", and copy the upper 3 packed elements from "a" to the upper elements of "dst". + + Integer + AVX + Store + + + + + Store the high and low 128-bit halves (each composed of integer data) from "a" into memory two different 128-bit locations. + "hiaddr" and "loaddr" do not need to be aligned on any particular boundary. -dst[31:0] := ( a[31:0] != b[31:0] ) ? 0xffffffff : 0 -dst[127:32] := a[127:32] +MEM[loaddr+127:loaddr] := a[127:0] +MEM[hiaddr+127:hiaddr] := a[255:128] - -
xmmintrin.h
+
immintrin.h
 - + Floating Point - SSE - Compare - - - Compare packed single-precision (32-bit) floating-point elements in "a" and "b" for not-equal, and store the results in "dst". + AVX + Trigonometry + + + Compute the inverse cosine of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst". FOR j := 0 to 3 - i := j*32 - dst[i+31:i] := ( a[i+31:i] != b[i+31:i] ) ? 0xffffffff : 0 + i := j*64 + dst[i+63:i] := ACOS(a[i+63:i]) ENDFOR +dst[MAX:256] := 0 - -
xmmintrin.h
+
immintrin.h
 - + Floating Point - SSE - Compare - - - Compare the lower single-precision (32-bit) floating-point elements in "a" and "b" for not-less-than, store the result in the lower element of "dst", and copy the upper 3 packed elements from "a" to the upper elements of "dst". + AVX + Trigonometry + + + Compute the inverse cosine of packed single-precision (32-bit) floating-point elements in "a" expressed in radians, and store the results in "dst". -dst[31:0] := !( a[31:0] < b[31:0] ) ? 0xffffffff : 0 -dst[127:32] := a[127:32] +FOR j := 0 to 7 + i := j*32 + dst[i+31:i] := ACOS(a[i+31:i]) +ENDFOR +dst[MAX:256] := 0 - -
xmmintrin.h
+
immintrin.h
 - + Floating Point - SSE - Compare - - - Compare packed single-precision (32-bit) floating-point elements in "a" and "b" for not-less-than, and store the results in "dst". + AVX + Trigonometry + + + Compute the inverse hyperbolic cosine of packed double-precision (64-bit) floating-point elements in "a" expressed in radians, and store the results in "dst". FOR j := 0 to 3 - i := j*32 - dst[i+31:i] := !( a[i+31:i] < b[i+31:i] ) ? 0xffffffff : 0 + i := j*64 + dst[i+63:i] := ACOSH(a[i+63:i]) ENDFOR +dst[MAX:256] := 0 - -
xmmintrin.h
+
immintrin.h
 - + Floating Point - SSE - Compare - - - Compare the lower single-precision (32-bit) floating-point elements in "a" and "b" for not-less-than-or-equal, store the result in the lower element of "dst", and copy the upper 3 packed elements from "a" to the upper elements of "dst". + AVX + Trigonometry + + + Compute the inverse hyperbolic cosine of packed single-precision (32-bit) floating-point elements in "a" expressed in radians, and store the results in "dst". -dst[31:0] := !( a[31:0] <= b[31:0] ) ? 0xffffffff : 0 -dst[127:32] := a[127:32] +FOR j := 0 to 7 + i := j*32 + dst[i+31:i] := ACOSH(a[i+31:i]) +ENDFOR +dst[MAX:256] := 0 - -
xmmintrin.h
+
immintrin.h
 - + Floating Point - SSE - Compare - - - Compare packed single-precision (32-bit) floating-point elements in "a" and "b" for not-less-than-or-equal, and store the results in "dst". + AVX + Trigonometry + + + Compute the inverse sine of packed double-precision (64-bit) floating-point elements in "a" expressed in radians, and store the results in "dst". FOR j := 0 to 3 - i := j*32 - dst[i+31:i] := !( a[i+31:i] <= b[i+31:i] ) ? 0xffffffff : 0 + i := j*64 + dst[i+63:i] := ASIN(a[i+63:i]) ENDFOR +dst[MAX:256] := 0 - -
xmmintrin.h
+
immintrin.h
 - + Floating Point - SSE - Compare - - - Compare the lower single-precision (32-bit) floating-point elements in "a" and "b" for not-greater-than, store the result in the lower element of "dst", and copy the upper 3 packed elements from "a" to the upper elements of "dst". + AVX + Trigonometry + + + Compute the inverse sine of packed single-precision (32-bit) floating-point elements in "a" expressed in radians, and store the results in "dst". -dst[31:0] := !( a[31:0] > b[31:0] ) ? 0xffffffff : 0 -dst[127:32] := a[127:32] +FOR j := 0 to 7 + i := j*32 + dst[i+31:i] := ASIN(a[i+31:i]) +ENDFOR +dst[MAX:256] := 0 - -
xmmintrin.h
+
immintrin.h
 - + Floating Point - SSE - Compare - - - Compare packed single-precision (32-bit) floating-point elements in "a" and "b" for not-greater-than, and store the results in "dst". + AVX + Trigonometry + + + Compute the inverse hyperbolic sine of packed double-precision (64-bit) floating-point elements in "a" expressed in radians, and store the results in "dst". FOR j := 0 to 3 - i := j*32 - dst[i+31:i] := !( a[i+31:i] > b[i+31:i] ) ? 0xffffffff : 0 + i := j*64 + dst[i+63:i] := ASINH(a[i+63:i]) ENDFOR +dst[MAX:256] := 0 - -
xmmintrin.h
+
immintrin.h
 - + Floating Point - SSE - Compare - - - Compare the lower single-precision (32-bit) floating-point elements in "a" and "b" for not-greater-than-or-equal, store the result in the lower element of "dst", and copy the upper 3 packed elements from "a" to the upper elements of "dst". + AVX + Trigonometry + + + Compute the inverse hyperbolic sine of packed single-precision (32-bit) floating-point elements in "a" expressed in radians, and store the results in "dst". -dst[31:0] := !( a[31:0] >= b[31:0] ) ? 0xffffffff : 0 -dst[127:32] := a[127:32] +FOR j := 0 to 7 + i := j*32 + dst[i+31:i] := ASINH(a[i+31:i]) +ENDFOR +dst[MAX:256] := 0 - -
xmmintrin.h
+
immintrin.h
 - + Floating Point - SSE - Compare - - - Compare packed single-precision (32-bit) floating-point elements in "a" and "b" for not-greater-than-or-equal, and store the results in "dst". + AVX + Trigonometry + + + Compute the inverse tangent of packed double-precision (64-bit) floating-point elements in "a" expressed in radians, and store the results in "dst". FOR j := 0 to 3 - i := j*32 - dst[i+31:i] := !( a[i+31:i] >= b[i+31:i] ) ? 0xffffffff : 0 + i := j*64 + dst[i+63:i] := ATAN(a[i+63:i]) ENDFOR +dst[MAX:256] := 0 - -
xmmintrin.h
+
immintrin.h
 - + Floating Point - SSE - Compare - - - Compare the lower single-precision (32-bit) floating-point elements in "a" and "b" to see if neither is NaN, store the result in the lower element of "dst", and copy the upper 3 packed elements from "a" to the upper elements of "dst". + AVX + Trigonometry + + + Compute the inverse tangent of packed single-precision (32-bit) floating-point elements in "a" expressed in radians, and store the results in "dst". -dst[31:0] := ( a[31:0] != NaN AND b[31:0] != NaN ) ? 0xffffffff : 0 -dst[127:32] := a[127:32] +FOR j := 0 to 7 + i := j*32 + dst[i+31:i] := ATAN(a[i+31:i]) +ENDFOR +dst[MAX:256] := 0 - -
xmmintrin.h
+
immintrin.h
 - + Floating Point - SSE - Compare - - - Compare packed single-precision (32-bit) floating-point elements in "a" and "b" to see if neither is NaN, and store the results in "dst". + AVX + Trigonometry + + + + Compute the inverse tangent of packed double-precision (64-bit) floating-point elements in "a" divided by packed elements in "b", and store the results in "dst" expressed in radians. FOR j := 0 to 3 - i := j*32 - dst[i+31:i] := ( a[i+31:i] != NaN AND b[i+31:i] != NaN ) ? 0xffffffff : 0 + i := j*64 + dst[i+63:i] := ATAN2(a[i+63:i], b[i+63:i]) ENDFOR +dst[MAX:256] := 0 - -
xmmintrin.h
+
immintrin.h
 - + Floating Point - SSE - Compare - - - Compare the lower single-precision (32-bit) floating-point elements in "a" and "b" to see if either is NaN, store the result in the lower element of "dst", and copy the upper 3 packed elements from "a" to the upper elements of "dst". + AVX + Trigonometry + + + + Compute the inverse tangent of packed single-precision (32-bit) floating-point elements in "a" divided by packed elements in "b", and store the results in "dst" expressed in radians. -dst[31:0] := ( a[31:0] == NaN OR b[31:0] == NaN ) ? 0xffffffff : 0 -dst[127:32] := a[127:32] +FOR j := 0 to 7 + i := j*32 + dst[i+31:i] := ATAN2(a[i+31:i], b[i+31:i]) +ENDFOR +dst[MAX:256] := 0 - -
xmmintrin.h
+
immintrin.h
 - + Floating Point - SSE - Compare - - - Compare packed single-precision (32-bit) floating-point elements in "a" and "b" to see if either is NaN, and store the results in "dst". + AVX + Trigonometry + + + Compute the inverse hyperbolic tangent of packed double-precision (64-bit) floating-point elements in "a" expressed in radians, and store the results in "dst". FOR j := 0 to 3 - i := j*32 - dst[i+31:i] := ( a[i+31:i] == NaN OR b[i+31:i] == NaN ) ? 0xffffffff : 0 + i := j*64 + dst[i+63:i] := ATANH(a[i+63:i]) ENDFOR +dst[MAX:256] := 0 - -
xmmintrin.h
+
immintrin.h
 - + Floating Point - SSE - Compare - - - Compare the lower single-precision (32-bit) floating-point element in "a" and "b" for equality, and return the boolean result (0 or 1). + AVX + Trigonometry + + + Compute the inverse hyperbolic tangent of packed single-precision (32-bit) floating-point elements in "a" expressed in radians, and store the results in "dst". -RETURN ( a[31:0] == b[31:0] ) ? 1 : 0 +FOR j := 0 to 7 + i := j*32 + dst[i+31:i] := ATANH(a[i+31:i]) +ENDFOR +dst[MAX:256] := 0 - -
xmmintrin.h
+
immintrin.h
 - + Floating Point - SSE - Compare - - - Compare the lower single-precision (32-bit) floating-point element in "a" and "b" for less-than, and return the boolean result (0 or 1). - -RETURN ( a[31:0] < b[31:0] ) ? 1 : 0 + AVX + Elementary Math Functions + + + Compute the cube root of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst". + FOR j := 0 to 3 + i := j*64 + dst[i+63:i] := CubeRoot(a[i+63:i]) +ENDFOR +dst[MAX:256] := 0 - -
xmmintrin.h
+
immintrin.h
 - + Floating Point - SSE - Compare - - - Compare the lower single-precision (32-bit) floating-point element in "a" and "b" for less-than-or-equal, and return the boolean result (0 or 1). - -RETURN ( a[31:0] <= b[31:0] ) ? 1 : 0 + AVX + Elementary Math Functions + + + Compute the cube root of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst". + FOR j := 0 to 7 + i := j*32 + dst[i+31:i] := CubeRoot(a[i+31:i]) +ENDFOR +dst[MAX:256] := 0 - -
xmmintrin.h
+
immintrin.h
 - + Floating Point - SSE - Compare - - - Compare the lower single-precision (32-bit) floating-point element in "a" and "b" for greater-than, and return the boolean result (0 or 1). - -RETURN ( a[31:0] > b[31:0] ) ? 1 : 0 + AVX + Probability/Statistics + + + Compute the cumulative distribution function of packed double-precision (64-bit) floating-point elements in "a" using the normal distribution, and store the results in "dst". + FOR j := 0 to 3 + i := j*64 + dst[i+63:i] := CDFNormal(a[i+63:i]) +ENDFOR +dst[MAX:256] := 0 - -
xmmintrin.h
+
immintrin.h
 - + Floating Point - SSE - Compare - - - Compare the lower single-precision (32-bit) floating-point element in "a" and "b" for greater-than-or-equal, and return the boolean result (0 or 1). - -RETURN ( a[31:0] >= b[31:0] ) ? 1 : 0 + AVX + Probability/Statistics + + + Compute the cumulative distribution function of packed single-precision (32-bit) floating-point elements in "a" using the normal distribution, and store the results in "dst". + FOR j := 0 to 7 + i := j*32 + dst[i+31:i] := CDFNormal(a[i+31:i]) +ENDFOR +dst[MAX:256] := 0 - -
xmmintrin.h
+
immintrin.h
 - + Floating Point - SSE - Compare - - - Compare the lower single-precision (32-bit) floating-point element in "a" and "b" for not-equal, and return the boolean result (0 or 1). - -RETURN ( a[31:0] != b[31:0] ) ? 1 : 0 + AVX + Probability/Statistics + + + Compute the inverse cumulative distribution function of packed double-precision (64-bit) floating-point elements in "a" using the normal distribution, and store the results in "dst". + FOR j := 0 to 3 + i := j*64 + dst[i+63:i] := InverseCDFNormal(a[i+63:i]) +ENDFOR +dst[MAX:256] := 0 - -
xmmintrin.h
+
immintrin.h
 - + Floating Point - SSE - Compare - - - Compare the lower single-precision (32-bit) floating-point element in "a" and "b" for equality, and return the boolean result (0 or 1). This instruction will not signal an exception for QNaNs. - -RETURN ( a[31:0] == b[31:0] ) ? 1 : 0 + AVX + Probability/Statistics + + + Compute the inverse cumulative distribution function of packed single-precision (32-bit) floating-point elements in "a" using the normal distribution, and store the results in "dst". + FOR j := 0 to 7 + i := j*32 + dst[i+31:i] := InverseCDFNormal(a[i+31:i]) +ENDFOR +dst[MAX:256] := 0 - -
xmmintrin.h
+
immintrin.h
 - + Floating Point - SSE - Compare - - - Compare the lower single-precision (32-bit) floating-point element in "a" and "b" for less-than, and return the boolean result (0 or 1). This instruction will not signal an exception for QNaNs. - -RETURN ( a[31:0] < b[31:0] ) ? 1 : 0 + AVX + Elementary Math Functions + + + Compute the exponential value of "e" raised to the power of packed complex numbers in "a", and store the complex results in "dst". Each complex number is composed of two adjacent single-precision (32-bit) floating-point elements, which defines the complex number "complex = vec.fp32[0] + i * vec.fp32[1]". + +DEFINE CEXP(a[31:0], b[31:0]) { + result[31:0] := POW(FP32(e), a[31:0]) * COS(b[31:0]) + result[63:32] := POW(FP32(e), a[31:0]) * SIN(b[31:0]) + RETURN result +} +FOR j := 0 to 3 + i := j*64 + dst[i+63:i] := CEXP(a[i+31:i], a[i+63:i+32]) +ENDFOR +dst[MAX:256] := 0 - -
xmmintrin.h
+
immintrin.h
 - + Floating Point - SSE - Compare - - - Compare the lower single-precision (32-bit) floating-point element in "a" and "b" for less-than-or-equal, and return the boolean result (0 or 1). This instruction will not signal an exception for QNaNs. - -RETURN ( a[31:0] <= b[31:0] ) ? 1 : 0 + AVX + Elementary Math Functions + + + Compute the natural logarithm of packed complex numbers in "a", and store the complex results in "dst". Each complex number is composed of two adjacent single-precision (32-bit) floating-point elements, which defines the complex number "complex = vec.fp32[0] + i * vec.fp32[1]". + +DEFINE CLOG(a[31:0], b[31:0]) { + result[31:0] := LOG(SQRT(POW(a, 2.0) + POW(b, 2.0))) + result[63:32] := ATAN2(b, a) + RETURN result +} +FOR j := 0 to 3 + i := j*64 + dst[i+63:i] := CLOG(a[i+31:i], a[i+63:i+32]) +ENDFOR +dst[MAX:256] := 0 - -
xmmintrin.h
+
immintrin.h
 - + Floating Point - SSE - Compare - - - Compare the lower single-precision (32-bit) floating-point element in "a" and "b" for greater-than, and return the boolean result (0 or 1). This instruction will not signal an exception for QNaNs. + AVX + Trigonometry + + + Compute the cosine of packed double-precision (64-bit) floating-point elements in "a" expressed in radians, and store the results in "dst". -RETURN ( a[31:0] > b[31:0] ) ? 1 : 0 +FOR j := 0 to 3 + i := j*64 + dst[i+63:i] := COS(a[i+63:i]) +ENDFOR +dst[MAX:256] := 0 - -
xmmintrin.h
+
immintrin.h
 - + Floating Point - SSE - Compare - - - Compare the lower single-precision (32-bit) floating-point element in "a" and "b" for greater-than-or-equal, and return the boolean result (0 or 1). This instruction will not signal an exception for QNaNs. + AVX + Trigonometry + + + Compute the cosine of packed single-precision (32-bit) floating-point elements in "a" expressed in radians, and store the results in "dst". -RETURN ( a[31:0] >= b[31:0] ) ? 1 : 0 +FOR j := 0 to 7 + i := j*32 + dst[i+31:i] := COS(a[i+31:i]) +ENDFOR +dst[MAX:256] := 0 - -
xmmintrin.h
+
immintrin.h
 - + Floating Point - SSE - Compare - - - Compare the lower single-precision (32-bit) floating-point element in "a" and "b" for not-equal, and return the boolean result (0 or 1). This instruction will not signal an exception for QNaNs. - -RETURN ( a[31:0] != b[31:0] ) ? 1 : 0 + AVX + Trigonometry + + + Compute the cosine of packed double-precision (64-bit) floating-point elements in "a" expressed in degrees, and store the results in "dst". + FOR j := 0 to 3 + i := j*64 + dst[i+63:i] := COSD(a[i+63:i]) +ENDFOR +dst[MAX:256] := 0 - -
xmmintrin.h
+
immintrin.h
 - + Floating Point - Integer - SSE - Convert - - Convert the lower single-precision (32-bit) floating-point element in "a" to a 32-bit integer, and store the result in "dst". - -dst[31:0] := Convert_FP32_To_Int32(a[31:0]) + AVX + Trigonometry + + + Compute the cosine of packed single-precision (32-bit) floating-point elements in "a" expressed in degrees, and store the results in "dst". + FOR j := 0 to 7 + i := j*32 + dst[i+31:i] := COSD(a[i+31:i]) +ENDFOR +dst[MAX:256] := 0 - -
xmmintrin.h
+
immintrin.h
 - + Floating Point - SSE - Convert - - Convert the lower single-precision (32-bit) floating-point element in "a" to a 32-bit integer, and store the result in "dst". + AVX + Trigonometry + + + Compute the hyperbolic cosine of packed double-precision (64-bit) floating-point elements in "a" expressed in radians, and store the results in "dst". -dst[31:0] := Convert_FP32_To_Int32(a[31:0]) +FOR j := 0 to 3 + i := j*64 + dst[i+63:i] := COSH(a[i+63:i]) +ENDFOR +dst[MAX:256] := 0 - -
xmmintrin.h
+
immintrin.h
 - + Floating Point - Integer - SSE - Convert - - Convert the lower single-precision (32-bit) floating-point element in "a" to a 64-bit integer, and store the result in "dst". + AVX + Trigonometry + + + Compute the hyperbolic cosine of packed single-precision (32-bit) floating-point elements in "a" expressed in radians, and store the results in "dst". -dst[63:0] := Convert_FP32_To_Int64(a[31:0]) - - -
xmmintrin.h
- - - Floating Point - SSE - Convert - - Copy the lower single-precision (32-bit) floating-point element of "a" to "dst". - dst[31:0] := a[31:0] - -
xmmintrin.h
+FOR j := 0 to 7 + i := j*32 + dst[i+31:i] := COSH(a[i+31:i]) +ENDFOR +dst[MAX:256] := 0 + +
immintrin.h
 - + Floating Point + AVX + Elementary Math Functions + + + Compute the square root of packed complex snumbers in "a", and store the complex results in "dst". Each complex number is composed of two adjacent single-precision (32-bit) floating-point elements, which defines the complex number "complex = vec.fp32[0] + i * vec.fp32[1]". + +DEFINE CSQRT(a[31:0], b[31:0]) { + sign[31:0] := (b < 0.0) ? -FP32(1.0) : FP32(1.0) + result[31:0] := SQRT((a + SQRT(POW(a, 2.0) + POW(b, 2.0))) / 2.0) + result[63:32] := sign * SQRT((-a + SQRT(POW(a, 2.0) + POW(b, 2.0))) / 2.0) + RETURN result +} +FOR j := 0 to 3 + i := j*64 + dst[i+63:i] := CSQRT(a[i+31:i], a[i+63:i+32]) +ENDFOR +dst[MAX:256] := 0 + +
immintrin.h
+ + Integer - SSE - Convert - - Convert packed single-precision (32-bit) floating-point elements in "a" to packed 32-bit integers, and store the results in "dst". + AVX + Arithmetic + + + + Divide packed signed 8-bit integers in "a" by packed elements in "b", and store the truncated results in "dst". -FOR j := 0 to 1 - i := 32*j - dst[i+31:i] := Convert_FP32_To_Int32(a[i+31:i]) +FOR j := 0 to 31 + i := 8*j + IF b[i+7:i] == 0 + #DE + FI + dst[i+7:i] := Truncate8(a[i+7:i] / b[i+7:i]) ENDFOR +dst[MAX:256] := 0 - -
xmmintrin.h
+
immintrin.h
 - - Floating Point - SSE - Convert - - Convert packed single-precision (32-bit) floating-point elements in "a" to packed 32-bit integers, and store the results in "dst". + + Integer + AVX + Arithmetic + + + + Divide packed signed 16-bit integers in "a" by packed elements in "b", and store the truncated results in "dst". -FOR j := 0 to 1 - i := 32*j - dst[i+31:i] := Convert_FP32_To_Int32(a[i+31:i]) +FOR j := 0 to 15 + i := 16*j + IF b[i+15:i] == 0 + #DE + FI + dst[i+15:i] := Truncate16(a[i+15:i] / b[i+15:i]) ENDFOR +dst[MAX:256] := 0 - -
xmmintrin.h
+
immintrin.h
 - - Floating Point + Integer - SSE - Convert - - Convert the lower single-precision (32-bit) floating-point element in "a" to a 32-bit integer with truncation, and store the result in "dst". + AVX + Arithmetic + + + + Divide packed signed 32-bit integers in "a" by packed elements in "b", and store the truncated results in "dst". -dst[31:0] := Convert_FP32_To_Int32_Truncate(a[31:0]) +FOR j := 0 to 7 + i := 32*j + IF b[i+31:i] == 0 + #DE + FI + dst[i+31:i] := Truncate32(a[i+31:i] / b[i+31:i]) +ENDFOR +dst[MAX:256] := 0 - -
xmmintrin.h
+
immintrin.h
 - - Floating Point - SSE - Convert - - Convert the lower single-precision (32-bit) floating-point element in "a" to a 32-bit integer with truncation, and store the result in "dst". + + Integer + AVX + Arithmetic + + + + Divide packed signed 64-bit integers in "a" by packed elements in "b", and store the truncated results in "dst". -dst[31:0] := Convert_FP32_To_Int32_Truncate(a[31:0]) +FOR j := 0 to 3 + i := 64*j + IF b[i+63:i] == 0 + #DE + FI + dst[i+63:i] := Truncate64(a[i+63:i] / b[i+63:i]) +ENDFOR +dst[MAX:256] := 0 - -
xmmintrin.h
+
immintrin.h
 - - Floating Point + Integer - SSE - Convert - - Convert the lower single-precision (32-bit) floating-point element in "a" to a 64-bit integer with truncation, and store the result in "dst". + AVX + Arithmetic + + + + Divide packed unsigned 8-bit integers in "a" by packed elements in "b", and store the truncated results in "dst". -dst[63:0] := Convert_FP64_To_Int32_Truncate(a[31:0]) +FOR j := 0 to 31 + i := 8*j + IF b[i+7:i] == 0 + #DE + FI + dst[i+7:i] := Truncate8(a[i+7:i] / b[i+7:i]) +ENDFOR +dst[MAX:256] := 0 - -
xmmintrin.h
+
immintrin.h
 - - Floating Point + Integer - SSE - Convert - - Convert packed single-precision (32-bit) floating-point elements in "a" to packed 32-bit integers with truncation, and store the results in "dst". + AVX + Arithmetic + + + + Divide packed unsigned 16-bit integers in "a" by packed elements in "b", and store the truncated results in "dst". -FOR j := 0 to 1 - i := 32*j - dst[i+31:i] := Convert_FP32_To_Int32_Truncate(a[i+31:i]) +FOR j := 0 to 15 + i := 16*j + IF b[i+15:i] == 0 + #DE + FI + dst[i+15:i] := Truncate16(a[i+15:i] / b[i+15:i]) ENDFOR +dst[MAX:256] := 0 - -
xmmintrin.h
+
immintrin.h
 - - Floating Point - SSE - Convert - - Convert packed single-precision (32-bit) floating-point elements in "a" to packed 32-bit integers with truncation, and store the results in "dst". + + Integer + AVX + Arithmetic + + + + Divide packed unsigned 32-bit integers in "a" by packed elements in "b", and store the truncated results in "dst". -FOR j := 0 to 1 +FOR j := 0 to 7 i := 32*j - dst[i+31:i] := Convert_FP32_To_Int32_Truncate(a[i+31:i]) + IF b[i+31:i] == 0 + #DE + FI + dst[i+31:i] := Truncate32(a[i+31:i] / b[i+31:i]) ENDFOR +dst[MAX:256] := 0 - -
xmmintrin.h
+
immintrin.h
 - - Floating Point + Integer - SSE - Convert - - Convert packed single-precision (32-bit) floating-point elements in "a" to packed 16-bit integers, and store the results in "dst". + AVX + Arithmetic + + + + Divide packed unsigned 64-bit integers in "a" by packed elements in "b", and store the truncated results in "dst". FOR j := 0 to 3 - i := 16*j - k := 32*j - dst[i+15:i] := Convert_FP32_To_Int16(a[k+31:k]) + i := 64*j + IF b[i+63:i] == 0 + #DE + FI + dst[i+63:i] := Truncate64(a[i+63:i] / b[i+63:i]) ENDFOR +dst[MAX:256] := 0 -
xmmintrin.h
+
immintrin.h
 - + Floating Point - Integer - SSE - Convert - - Convert packed single-precision (32-bit) floating-point elements in "a" to packed 8-bit integers, and store the results in lower 4 elements of "dst". - -FOR j := 0 to 3 - i := 8*j - k := 32*j - dst[i+7:i] := Convert_FP32_To_Int8(a[k+31:k]) + AVX + Probability/Statistics + + + Compute the error function of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst". + FOR j := 0 to 3 + i := j*64 + dst[i+63:i] := ERF(a[i+63:i]) ENDFOR +dst[MAX:256] := 0 -
xmmintrin.h
+
immintrin.h
 - + Floating Point - SSE - Set - - Copy single-precision (32-bit) floating-point element "a" to the lower element of "dst", and zero the upper 3 elements. - -dst[31:0] := a[31:0] -dst[127:32] := 0 + AVX + Probability/Statistics + + + Compute the error function of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst". + FOR j := 0 to 7 + i := j*32 + dst[i+31:i] := ERF(a[i+31:i]) +ENDFOR +dst[MAX:256] := 0 -
xmmintrin.h
+
immintrin.h
 - + Floating Point - SSE - Set - - Broadcast single-precision (32-bit) floating-point value "a" to all elements of "dst". - -FOR j := 0 to 3 + AVX + Probability/Statistics + + + Compute the complementary error function of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst". + FOR j := 0 to 3 + i := j*64 + dst[i+63:i] := 1.0 - ERF(a[i+63:i]) +ENDFOR +dst[MAX:256] := 0 + +
immintrin.h
+ + + Floating Point + AVX + Probability/Statistics + + + Compute the complementary error function of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst". + FOR j := 0 to 7 i := j*32 - dst[i+31:i] := a[31:0] + dst[i+63:i] := 1.0 - ERF(a[i+31:i]) ENDFOR +dst[MAX:256] := 0 -
xmmintrin.h
+
immintrin.h
 - + Floating Point - SSE - Set - - Broadcast single-precision (32-bit) floating-point value "a" to all elements of "dst". - -FOR j := 0 to 3 + AVX + Probability/Statistics + + + Compute the inverse complementary error function of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst". + FOR j := 0 to 3 + i := j*64 + dst[i+63:i] := 1.0 / (1.0 - ERF(a[i+63:i])) +ENDFOR +dst[MAX:256] := 0 + +
immintrin.h
+ + + Floating Point + AVX + Probability/Statistics + + + Compute the inverse complementary error function of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst". + FOR j := 0 to 7 i := j*32 - dst[i+31:i] := a[31:0] + dst[i+63:i] := 1.0 / (1.0 - ERF(a[i+31:i])) ENDFOR +dst[MAX:256] := 0 -
xmmintrin.h
+
immintrin.h
 - + Floating Point - SSE - Set - - - - - Set packed single-precision (32-bit) floating-point elements in "dst" with the supplied values. - -dst[31:0] := e0 -dst[63:32] := e1 -dst[95:64] := e2 -dst[127:96] := e3 + AVX + Probability/Statistics + + + Compute the inverse error function of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst". + FOR j := 0 to 3 + i := j*64 + dst[i+63:i] := 1.0 / ERF(a[i+63:i]) +ENDFOR +dst[MAX:256] := 0 -
xmmintrin.h
+
immintrin.h
 - + Floating Point - SSE - Set - - - - - Set packed single-precision (32-bit) floating-point elements in "dst" with the supplied values in reverse order. - -dst[31:0] := e3 -dst[63:32] := e2 -dst[95:64] := e1 -dst[127:96] := e0 + AVX + Probability/Statistics + + + Compute the inverse error function of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst". + FOR j := 0 to 7 + i := j*32 + dst[i+63:i] := 1.0 / ERF(a[i+31:i]) +ENDFOR +dst[MAX:256] := 0 -
xmmintrin.h
+
immintrin.h
 - + Floating Point - SSE - Set - - Return vector of type __m128 with all elements set to zero. + AVX + Elementary Math Functions + + + Compute the exponential value of "e" raised to the power of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst". -dst[MAX:0] := 0 +FOR j := 0 to 3 + i := j*64 + dst[i+63:i] := POW(e, a[i+63:i]) +ENDFOR +dst[MAX:256] := 0 - -
xmmintrin.h
+
immintrin.h
 - - Integer - SSE - Load - - - Load 2 single-precision (32-bit) floating-point elements from memory into the upper 2 elements of "dst", and copy the lower 2 elements from "a" to "dst". "mem_addr" does not need to be aligned on any particular boundary. + + Floating Point + AVX + Elementary Math Functions + + + Compute the exponential value of "e" raised to the power of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst". -dst[31:0] := a[31:0] -dst[63:32] := a[63:32] -dst[95:64] := MEM[mem_addr+31:mem_addr] -dst[127:96] := MEM[mem_addr+63:mem_addr+32] +FOR j := 0 to 7 + i := j*32 + dst[i+31:i] := POW(FP32(e), a[i+31:i]) +ENDFOR +dst[MAX:256] := 0 - -
xmmintrin.h
+
immintrin.h
 - - Integer - SSE - Load - - - Load 2 single-precision (32-bit) floating-point elements from memory into the lower 2 elements of "dst", and copy the upper 2 elements from "a" to "dst". "mem_addr" does not need to be aligned on any particular boundary. + + Floating Point + AVX + Elementary Math Functions + + + Compute the exponential value of 10 raised to the power of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst". -dst[31:0] := MEM[mem_addr+31:mem_addr] -dst[63:32] := MEM[mem_addr+63:mem_addr+32] -dst[95:64] := a[95:64] -dst[127:96] := a[127:96] +FOR j := 0 to 3 + i := j*64 + dst[i+63:i] := POW(10.0, a[i+63:i]) +ENDFOR +dst[MAX:256] := 0 - -
xmmintrin.h
+
immintrin.h
 - + Floating Point - SSE - Load - - Load a single-precision (32-bit) floating-point element from memory into the lower of "dst", and zero the upper 3 elements. "mem_addr" does not need to be aligned on any particular boundary. + AVX + Elementary Math Functions + + + Compute the exponential value of 10 raised to the power of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst". -dst[31:0] := MEM[mem_addr+31:mem_addr] -dst[127:32] := 0 +FOR j := 0 to 7 + i := j*32 + dst[i+31:i] := POW(FP32(10.0), a[i+31:i]) +ENDFOR +dst[MAX:256] := 0 - -
xmmintrin.h
+
immintrin.h
 - + Floating Point - SSE - Load - - Load a single-precision (32-bit) floating-point element from memory into all elements of "dst". + AVX + Elementary Math Functions + + + Compute the exponential value of 2 raised to the power of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst". -dst[31:0] := MEM[mem_addr+31:mem_addr] -dst[63:32] := MEM[mem_addr+31:mem_addr] -dst[95:64] := MEM[mem_addr+31:mem_addr] -dst[127:96] := MEM[mem_addr+31:mem_addr] +FOR j := 0 to 3 + i := j*64 + dst[i+63:i] := POW(2.0, a[i+63:i]) +ENDFOR +dst[MAX:256] := 0 -
xmmintrin.h
+
immintrin.h
 - + Floating Point - SSE - Load - - Load a single-precision (32-bit) floating-point element from memory into all elements of "dst". + AVX + Elementary Math Functions + + + Compute the exponential value of 2 raised to the power of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst". -dst[31:0] := MEM[mem_addr+31:mem_addr] -dst[63:32] := MEM[mem_addr+31:mem_addr] -dst[95:64] := MEM[mem_addr+31:mem_addr] -dst[127:96] := MEM[mem_addr+31:mem_addr] +FOR j := 0 to 7 + i := j*32 + dst[i+31:i] := POW(FP32(2.0), a[i+31:i]) +ENDFOR +dst[MAX:256] := 0 -
xmmintrin.h
+
immintrin.h
 - + Floating Point - SSE - Load - - Load 128-bits (composed of 4 packed single-precision (32-bit) floating-point elements) from memory into "dst". - "mem_addr" must be aligned on a 16-byte boundary or a general-protection exception may be generated. + AVX + Elementary Math Functions + + + Compute the exponential value of "e" raised to the power of packed double-precision (64-bit) floating-point elements in "a", subtract one from each element, and store the results in "dst". -dst[127:0] := MEM[mem_addr+127:mem_addr] +FOR j := 0 to 3 + i := j*64 + dst[i+63:i] := POW(e, a[i+63:i]) - 1.0 +ENDFOR +dst[MAX:256] := 0 - -
xmmintrin.h
+
immintrin.h
 - + Floating Point - SSE - Load - - Load 128-bits (composed of 4 packed single-precision (32-bit) floating-point elements) from memory into "dst". - "mem_addr" does not need to be aligned on any particular boundary. + AVX + Elementary Math Functions + + + Compute the exponential value of "e" raised to the power of packed single-precision (32-bit) floating-point elements in "a", subtract one from each element, and store the results in "dst". -dst[127:0] := MEM[mem_addr+127:mem_addr] +FOR j := 0 to 7 + i := j*32 + dst[i+31:i] := POW(FP32(e), a[i+31:i]) - 1.0 +ENDFOR +dst[MAX:256] := 0 - -
xmmintrin.h
+
immintrin.h
 - + Floating Point - SSE - Load - - Load 4 single-precision (32-bit) floating-point elements from memory into "dst" in reverse order. mem_addr must be aligned on a 16-byte boundary or a general-protection exception may be generated. + AVX + Trigonometry + + + + Compute the length of the hypotenous of a right triangle, with the lengths of the other two sides of the triangle stored as packed double-precision (64-bit) floating-point elements in "a" and "b", and store the results in "dst". -dst[31:0] := MEM[mem_addr+127:mem_addr+96] -dst[63:32] := MEM[mem_addr+95:mem_addr+64] -dst[95:64] := MEM[mem_addr+63:mem_addr+32] -dst[127:96] := MEM[mem_addr+31:mem_addr] +FOR j := 0 to 3 + i := j*64 + dst[i+63:i] := SQRT(POW(a[i+63:i], 2.0) + POW(b[i+63:i], 2.0)) +ENDFOR +dst[MAX:256] := 0 -
xmmintrin.h
+
immintrin.h
 - + Floating Point - SSE - Store - - - Store 128-bits (composed of 4 packed single-precision (32-bit) floating-point elements) from "a" into memory using a non-temporal memory hint. - "mem_addr" must be aligned on a 16-byte boundary or a general-protection exception may be generated. + AVX + Trigonometry + + + + Compute the length of the hypotenous of a right triangle, with the lengths of the other two sides of the triangle stored as packed single-precision (32-bit) floating-point elements in "a" and "b", and store the results in "dst". -MEM[mem_addr+127:mem_addr] := a[127:0] +FOR j := 0 to 7 + i := j*32 + dst[i+31:i] := SQRT(POW(a[i+31:i], 2.0) + POW(b[i+31:i], 2.0)) +ENDFOR +dst[MAX:256] := 0 - -
xmmintrin.h
+
immintrin.h
 - + Integer - SSE - Store - - - Store the upper 2 single-precision (32-bit) floating-point elements from "a" into memory. - -MEM[mem_addr+31:mem_addr] := a[95:64] -MEM[mem_addr+63:mem_addr+32] := a[127:96] + AVX + Arithmetic + + + + Divide packed 32-bit integers in "a" by packed elements in "b", and store the truncated results in "dst". + FOR j := 0 to 7 + i := 32*j + dst[i+31:i] := TRUNCATE(a[i+31:i] / b[i+31:i]) +ENDFOR +dst[MAX:256] := 0 - -
xmmintrin.h
+
immintrin.h
 - + Integer - SSE - Store - - - Store the lower 2 single-precision (32-bit) floating-point elements from "a" into memory. - -MEM[mem_addr+31:mem_addr] := a[31:0] -MEM[mem_addr+63:mem_addr+32] := a[63:32] + AVX + Arithmetic + + + + + Divide packed 32-bit integers in "a" by packed elements in "b", store the truncated results in "dst", and store the remainders as packed 32-bit integers into memory at "mem_addr". + FOR j := 0 to 7 + i := 32*j + dst[i+31:i] := TRUNCATE(a[i+31:i] / b[i+31:i]) + MEM[mem_addr+i+31:mem_addr+i] := REMAINDER(a[i+31:i] / b[i+31:i]) +ENDFOR +dst[MAX:256] := 0 - -
xmmintrin.h
+
immintrin.h
 - + Floating Point - SSE - Store - - - Store the lower single-precision (32-bit) floating-point element from "a" into memory. "mem_addr" does not need to be aligned on any particular boundary. - -MEM[mem_addr+31:mem_addr] := a[31:0] + AVX + Elementary Math Functions + + + Compute the inverse cube root of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst". + FOR j := 0 to 3 + i := j*64 + dst[i+63:i] := InvCubeRoot(a[i+63:i]) +ENDFOR +dst[MAX:256] := 0 - -
xmmintrin.h
+
immintrin.h
 - + Floating Point - SSE - Store - - - Store the lower single-precision (32-bit) floating-point element from "a" into 4 contiguous elements in memory. "mem_addr" must be aligned on a 16-byte boundary or a general-protection exception may be generated. - -MEM[mem_addr+31:mem_addr] := a[31:0] -MEM[mem_addr+63:mem_addr+32] := a[31:0] -MEM[mem_addr+95:mem_addr+64] := a[31:0] -MEM[mem_addr+127:mem_addr+96] := a[31:0] + AVX + Elementary Math Functions + + + Compute the inverse cube root of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst". + FOR j := 0 to 7 + i := j*32 + dst[i+31:i] := InvCubeRoot(a[i+31:i]) +ENDFOR +dst[MAX:256] := 0 -
xmmintrin.h
+
immintrin.h
 - + Floating Point - SSE - Store - - - Store the lower single-precision (32-bit) floating-point element from "a" into 4 contiguous elements in memory. "mem_addr" must be aligned on a 16-byte boundary or a general-protection exception may be generated. - -MEM[mem_addr+31:mem_addr] := a[31:0] -MEM[mem_addr+63:mem_addr+32] := a[31:0] -MEM[mem_addr+95:mem_addr+64] := a[31:0] -MEM[mem_addr+127:mem_addr+96] := a[31:0] + AVX + Elementary Math Functions + + + Compute the inverse square root of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst". + FOR j := 0 to 3 + i := j*64 + dst[i+63:i] := InvSQRT(a[i+63:i]) +ENDFOR +dst[MAX:256] := 0 -
xmmintrin.h
+
immintrin.h
 - + Floating Point - SSE - Store - - - Store 128-bits (composed of 4 packed single-precision (32-bit) floating-point elements) from "a" into memory. - "mem_addr" must be aligned on a 16-byte boundary or a general-protection exception may be generated. - -MEM[mem_addr+127:mem_addr] := a[127:0] + AVX + Elementary Math Functions + + + Compute the inverse square root of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst". + FOR j := 0 to 7 + i := j*32 + dst[i+31:i] := InvSQRT(a[i+31:i]) +ENDFOR +dst[MAX:256] := 0 - -
xmmintrin.h
+
immintrin.h
 - - Floating Point - SSE - Store - - - Store 128-bits (composed of 4 packed single-precision (32-bit) floating-point elements) from "a" into memory. - "mem_addr" does not need to be aligned on any particular boundary. - -MEM[mem_addr+127:mem_addr] := a[127:0] + + Integer + AVX + Arithmetic + + + + Divide packed 32-bit integers in "a" by packed elements in "b", and store the remainders as packed 32-bit integers in "dst". + FOR j := 0 to 7 + i := 32*j + dst[i+31:i] := REMAINDER(a[i+31:i] / b[i+31:i]) +ENDFOR +dst[MAX:256] := 0 - -
xmmintrin.h
+
immintrin.h
 - + Floating Point - SSE - Store - - - Store 4 single-precision (32-bit) floating-point elements from "a" into memory in reverse order. - "mem_addr" must be aligned on a 16-byte boundary or a general-protection exception may be generated. + AVX + Elementary Math Functions + + + Compute the natural logarithm of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst". -MEM[mem_addr+31:mem_addr] := a[127:96] -MEM[mem_addr+63:mem_addr+32] := a[95:64] -MEM[mem_addr+95:mem_addr+64] := a[63:32] -MEM[mem_addr+127:mem_addr+96] := a[31:0] +FOR j := 0 to 3 + i := j*64 + dst[i+63:i] := LOG(a[i+63:i]) +ENDFOR +dst[MAX:256] := 0 - -
xmmintrin.h
+
immintrin.h
 - + Floating Point - SSE - Move - - - Move the lower single-precision (32-bit) floating-point element from "b" to the lower element of "dst", and copy the upper 3 elements from "a" to the upper elements of "dst". + AVX + Elementary Math Functions + + + Compute the natural logarithm of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst". -dst[31:0] := b[31:0] -dst[63:32] := a[63:32] -dst[95:64] := a[95:64] -dst[127:96] := a[127:96] +FOR j := 0 to 7 + i := j*32 + dst[i+31:i] := LOG(a[i+31:i]) +ENDFOR +dst[MAX:256] := 0 - -
xmmintrin.h
+
immintrin.h
 - + Floating Point - SSE - Swizzle - - - - Shuffle single-precision (32-bit) floating-point elements in "a" using the control in "imm8", and store the results in "dst". + AVX + Elementary Math Functions + + + Compute the base-10 logarithm of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst". -SELECT4(src, control){ - CASE(control[1:0]) - 0: tmp[31:0] := src[31:0] - 1: tmp[31:0] := src[63:32] - 2: tmp[31:0] := src[95:64] - 3: tmp[31:0] := src[127:96] - ESAC - RETURN tmp[31:0] -} - -dst[31:0] := SELECT4(a[127:0], imm8[1:0]) -dst[63:32] := SELECT4(a[127:0], imm8[3:2]) -dst[95:64] := SELECT4(b[127:0], imm8[5:4]) -dst[127:96] := SELECT4(b[127:0], imm8[7:6]) +FOR j := 0 to 3 + i := j*64 + dst[i+63:i] := LOG(a[i+63:i]) / LOG(10.0) +ENDFOR +dst[MAX:256] := 0 - -
xmmintrin.h
+
immintrin.h
 - + Floating Point - SSE - Swizzle - - - Unpack and interleave single-precision (32-bit) floating-point elements from the high half "a" and "b", and store the results in "dst". + AVX + Elementary Math Functions + + + Compute the base-10 logarithm of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst". -INTERLEAVE_HIGH_DWORDS(src1[127:0], src2[127:0]){ - dst[31:0] := src1[95:64] - dst[63:32] := src2[95:64] - dst[95:64] := src1[127:96] - dst[127:96] := src2[127:96] - RETURN dst[127:0] -} - -dst[127:0] := INTERLEAVE_HIGH_DWORDS(a[127:0], b[127:0]) +FOR j := 0 to 7 + i := j*32 + dst[i+31:i] := LOG(a[i+31:i]) / LOG(10.0) +ENDFOR +dst[MAX:256] := 0 - -
xmmintrin.h
+
immintrin.h
 - + Floating Point - SSE - Swizzle - - - Unpack and interleave single-precision (32-bit) floating-point elements from the low half of "a" and "b", and store the results in "dst". + AVX + Elementary Math Functions + + + Compute the natural logarithm of one plus packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst". -INTERLEAVE_DWORDS(src1[127:0], src2[127:0]){ - dst[31:0] := src1[31:0] - dst[63:32] := src2[31:0] - dst[95:64] := src1[63:32] - dst[127:96] := src2[63:32] - RETURN dst[127:0] -} - -dst[127:0] := INTERLEAVE_DWORDS(a[127:0], b[127:0]) +FOR j := 0 to 3 + i := j*64 + dst[i+63:i] := LOG(1.0 + a[i+63:i]) +ENDFOR +dst[MAX:256] := 0 - -
xmmintrin.h
+
immintrin.h
 - + Floating Point - SSE - Move - - - Move the upper 2 single-precision (32-bit) floating-point elements from "b" to the lower 2 elements of "dst", and copy the upper 2 elements from "a" to the upper 2 elements of "dst". + AVX + Elementary Math Functions + + + Compute the natural logarithm of one plus packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst". -dst[31:0] := b[95:64] -dst[63:32] := b[127:96] -dst[95:64] := a[95:64] -dst[127:96] := a[127:96] +FOR j := 0 to 7 + i := j*32 + dst[i+31:i] := LOG(1.0 + a[i+31:i]) +ENDFOR +dst[MAX:256] := 0 - -
xmmintrin.h
+
immintrin.h
 - + Floating Point - SSE - Move - - - Move the lower 2 single-precision (32-bit) floating-point elements from "b" to the upper 2 elements of "dst", and copy the lower 2 elements from "a" to the lower 2 elements of "dst". + AVX + Elementary Math Functions + + + Compute the base-2 logarithm of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst". -dst[31:0] := a[31:0] -dst[63:32] := a[63:32] -dst[95:64] := b[31:0] -dst[127:96] := b[63:32] +FOR j := 0 to 3 + i := j*64 + dst[i+63:i] := LOG(a[i+63:i]) / LOG(2.0) +ENDFOR +dst[MAX:256] := 0 - -
xmmintrin.h
+
immintrin.h
 - + Floating Point - SSE - Miscellaneous - - Set each bit of mask "dst" based on the most significant bit of the corresponding packed single-precision (32-bit) floating-point element in "a". + AVX + Elementary Math Functions + + + Compute the base-2 logarithm of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst". -FOR j := 0 to 3 +FOR j := 0 to 7 i := j*32 - IF a[i+31] - dst[j] := 1 - ELSE - dst[j] := 0 - FI + dst[i+31:i] := LOG(a[i+31:i]) / LOG(2.0) ENDFOR -dst[MAX:4] := 0 +dst[MAX:256] := 0 - -
xmmintrin.h
- - - SSE - General Support - - - Allocate "size" bytes of memory, aligned to the alignment specified in "align", and return a pointer to the allocated memory. "_mm_free" should be used to free memory that is allocated with "_mm_malloc". -
xmmintrin.h
- - - SSE - General Support - - Free aligned memory that was allocated with "_mm_malloc". -
xmmintrin.h
+
immintrin.h
 - - - SSE2 - General Support - - Provide a hint to the processor that the code sequence is a spin-wait loop. This can help improve the performance and power consumption of spin-wait loops. - -
emmintrin.h
+ + Floating Point + AVX + Elementary Math Functions + + + Convert the exponent of each packed double-precision (64-bit) floating-point element in "a" to a double-precision floating-point number representing the integer exponent, and store the results in "dst". This intrinsic essentially calculates "floor(log2(x))" for each element. + FOR j := 0 to 3 + i := j*64 + dst[i+63:i] := ConvertExpFP64(a[i+63:i]) +ENDFOR +dst[MAX:256] := 0 + +
immintrin.h
 - - SSE2 - General Support - - Invalidate and flush the cache line that contains "p" from all levels of the cache hierarchy. - -
emmintrin.h
+ + Floating Point + AVX + Elementary Math Functions + + + Convert the exponent of each packed single-precision (32-bit) floating-point element in "a" to a single-precision floating-point number representing the integer exponent, and store the results in "dst". This intrinsic essentially calculates "floor(log2(x))" for each element. + FOR j := 0 to 7 + i := j*32 + dst[i+31:i] := ConvertExpFP32(a[i+31:i]) +ENDFOR +dst[MAX:256] := 0 + +
immintrin.h
 - - SSE2 - General Support - - Perform a serializing operation on all load-from-memory instructions that were issued prior to this instruction. Guarantees that every load instruction that precedes, in program order, is globally visible before any load instruction which follows the fence in program order. - -
emmintrin.h
+ + Floating Point + AVX + Elementary Math Functions + + + + Compute the exponential value of packed double-precision (64-bit) floating-point elements in "a" raised by packed elements in "b", and store the results in "dst". + +FOR j := 0 to 3 + i := j*64 + dst[i+63:i] := POW(a[i+63:i], b[i+63:i]) +ENDFOR +dst[MAX:256] := 0 + +
immintrin.h
 - - SSE2 - General Support - - Perform a serializing operation on all load-from-memory and store-to-memory instructions that were issued prior to this instruction. Guarantees that every memory access that precedes, in program order, the memory fence instruction is globally visible before any memory instruction which follows the fence in program order. - -
emmintrin.h
+ + Floating Point + AVX + Elementary Math Functions + + + + Compute the exponential value of packed single-precision (32-bit) floating-point elements in "a" raised by packed elements in "b", and store the results in "dst". + +FOR j := 0 to 7 + i := j*32 + dst[i+31:i] := POW(a[i+31:i], b[i+31:i]) +ENDFOR +dst[MAX:256] := 0 + +
immintrin.h
 - + Integer - SSE2 + AVX Arithmetic - - - Add packed 8-bit integers in "a" and "b", and store the results in "dst". - -FOR j := 0 to 15 - i := j*8 - dst[i+7:i] := a[i+7:i] + b[i+7:i] + + + + Divide packed 8-bit integers in "a" by packed elements in "b", and store the remainders as packed 32-bit integers in "dst". + FOR j := 0 to 31 + i := 8*j + dst[i+7:i] := REMAINDER(a[i+7:i] / b[i+7:i]) ENDFOR +dst[MAX:256] := 0 - -
emmintrin.h
+
immintrin.h
 - + Integer - SSE2 + AVX Arithmetic - - - Add packed 16-bit integers in "a" and "b", and store the results in "dst". - -FOR j := 0 to 7 - i := j*16 - dst[i+15:i] := a[i+15:i] + b[i+15:i] + + + + Divide packed 16-bit integers in "a" by packed elements in "b", and store the remainders as packed 32-bit integers in "dst". + FOR j := 0 to 15 + i := 16*j + dst[i+15:i] := REMAINDER(a[i+15:i] / b[i+15:i]) ENDFOR +dst[MAX:256] := 0 - -
emmintrin.h
+
immintrin.h
 - + Integer - SSE2 + AVX Arithmetic - - - Add packed 32-bit integers in "a" and "b", and store the results in "dst". - -FOR j := 0 to 3 - i := j*32 - dst[i+31:i] := a[i+31:i] + b[i+31:i] + + + + Divide packed 32-bit integers in "a" by packed elements in "b", and store the remainders as packed 32-bit integers in "dst". + FOR j := 0 to 7 + i := 32*j + dst[i+31:i] := REMAINDER(a[i+31:i] / b[i+31:i]) ENDFOR +dst[MAX:256] := 0 - -
emmintrin.h
+
immintrin.h
 - + Integer - SSE2 + AVX Arithmetic - - - Add 64-bit integers "a" and "b", and store the result in "dst". - -dst[63:0] := a[63:0] + b[63:0] + + + + Divide packed 64-bit integers in "a" by packed elements in "b", and store the remainders as packed 32-bit integers in "dst". + FOR j := 0 to 3 + i := 64*j + dst[i+63:i] := REMAINDER(a[i+63:i] / b[i+63:i]) +ENDFOR +dst[MAX:256] := 0 - -
emmintrin.h
+
immintrin.h
 - + Integer - SSE2 + AVX Arithmetic - - - Add packed 64-bit integers in "a" and "b", and store the results in "dst". - -FOR j := 0 to 1 - i := j*64 - dst[i+63:i] := a[i+63:i] + b[i+63:i] + + + + Divide packed unsigned 8-bit integers in "a" by packed elements in "b", and store the remainders as packed unsigned 32-bit integers in "dst". + FOR j := 0 to 31 + i := 8*j + dst[i+7:i] := REMAINDER(a[i+7:i] / b[i+7:i]) ENDFOR +dst[MAX:256] := 0 - -
emmintrin.h
+
immintrin.h
 - + Integer - SSE2 + AVX Arithmetic - - - Add packed 8-bit integers in "a" and "b" using saturation, and store the results in "dst". - -FOR j := 0 to 15 - i := j*8 - dst[i+7:i] := Saturate_To_Int8( a[i+7:i] + b[i+7:i] ) + + + + Divide packed unsigned 16-bit integers in "a" by packed elements in "b", and store the remainders as packed unsigned 32-bit integers in "dst". + FOR j := 0 to 15 + i := 16*j + dst[i+15:i] := REMAINDER(a[i+15:i] / b[i+15:i]) ENDFOR +dst[MAX:256] := 0 - -
emmintrin.h
+
immintrin.h
 - + Integer - SSE2 + AVX Arithmetic - - - Add packed 16-bit integers in "a" and "b" using saturation, and store the results in "dst". - -FOR j := 0 to 7 - i := j*16 - dst[i+15:i] := Saturate_To_Int16( a[i+15:i] + b[i+15:i] ) + + + + Divide packed unsigned 32-bit integers in "a" by packed elements in "b", and store the remainders as packed unsigned 32-bit integers in "dst". + FOR j := 0 to 7 + i := 32*j + dst[i+31:i] := REMAINDER(a[i+31:i] / b[i+31:i]) ENDFOR +dst[MAX:256] := 0 - -
emmintrin.h
+
immintrin.h
 - + Integer - SSE2 + AVX Arithmetic - - - Add packed unsigned 8-bit integers in "a" and "b" using saturation, and store the results in "dst". + + + + Divide packed unsigned 64-bit integers in "a" by packed elements in "b", and store the remainders as packed unsigned 32-bit integers in "dst". + FOR j := 0 to 3 + i := 64*j + dst[i+63:i] := REMAINDER(a[i+63:i] / b[i+63:i]) +ENDFOR +dst[MAX:256] := 0 + +
immintrin.h
+ + + Floating Point + AVX + Trigonometry + + + Compute the sine of packed double-precision (64-bit) floating-point elements in "a" expressed in radians, and store the results in "dst". -FOR j := 0 to 15 - i := j*8 - dst[i+7:i] := Saturate_To_UnsignedInt8( a[i+7:i] + b[i+7:i] ) +FOR j := 0 to 3 + i := j*64 + dst[i+63:i] := SIN(a[i+63:i]) ENDFOR +dst[MAX:256] := 0 - -
emmintrin.h
+
immintrin.h
 - - Integer - SSE2 - Arithmetic - - - Add packed unsigned 16-bit integers in "a" and "b" using saturation, and store the results in "dst". + + Floating Point + AVX + Trigonometry + + + Compute the sine of packed single-precision (32-bit) floating-point elements in "a" expressed in radians, and store the results in "dst". FOR j := 0 to 7 - i := j*16 - dst[i+15:i] := Saturate_To_UnsignedInt16( a[i+15:i] + b[i+15:i] ) + i := j*32 + dst[i+31:i] := SIN(a[i+31:i]) ENDFOR +dst[MAX:256] := 0 - -
emmintrin.h
+
immintrin.h
 - - Integer - SSE2 - Probability/Statistics - - - Average packed unsigned 8-bit integers in "a" and "b", and store the results in "dst". + + Floating Point + AVX + Trigonometry + + + + Compute the sine and cosine of packed double-precision (64-bit) floating-point elements in "a" expressed in radians, store the sine in "dst", and store the cosine into memory at "mem_addr". -FOR j := 0 to 15 - i := j*8 - dst[i+7:i] := (a[i+7:i] + b[i+7:i] + 1) >> 1 +FOR j := 0 to 3 + i := j*64 + dst[i+63:i] := SIN(a[i+63:i]) + MEM[mem_addr+i+63:mem_addr+i] := COS(a[i+63:i]) ENDFOR +dst[MAX:256] := 0 - -
emmintrin.h
+
immintrin.h
 - - Integer - SSE2 - Probability/Statistics - - - Average packed unsigned 16-bit integers in "a" and "b", and store the results in "dst". + + Floating Point + AVX + Trigonometry + + + + Compute the sine and cosine of packed single-precision (32-bit) floating-point elements in "a" expressed in radians, store the sine in "dst", and store the cosine into memory at "mem_addr". FOR j := 0 to 7 - i := j*16 - dst[i+15:i] := (a[i+15:i] + b[i+15:i] + 1) >> 1 + i := j*32 + dst[i+31:i] := SIN(a[i+31:i]) + MEM[mem_addr+i+31:mem_addr+i] := COS(a[i+31:i]) ENDFOR +dst[MAX:256] := 0 - -
emmintrin.h
+
immintrin.h
 - - Integer - SSE2 - Arithmetic - - - Multiply packed signed 16-bit integers in "a" and "b", producing intermediate signed 32-bit integers. Horizontally add adjacent pairs of intermediate 32-bit integers, and pack the results in "dst". - + + Floating Point + AVX + Trigonometry + + + Compute the sine of packed double-precision (64-bit) floating-point elements in "a" expressed in degrees, and store the results in "dst". + FOR j := 0 to 3 + i := j*64 + dst[i+63:i] := SIND(a[i+63:i]) +ENDFOR +dst[MAX:256] := 0 + +
immintrin.h
+ + + Floating Point + AVX + Trigonometry + + + Compute the sine of packed single-precision (32-bit) floating-point elements in "a" expressed in degrees, and store the results in "dst". + FOR j := 0 to 7 + i := j*32 + dst[i+31:i] := SIND(a[i+31:i]) +ENDFOR +dst[MAX:256] := 0 + +
immintrin.h
+ + + Floating Point + AVX + Trigonometry + + + Compute the hyperbolic sine of packed double-precision (64-bit) floating-point elements in "a" expressed in radians, and store the results in "dst". FOR j := 0 to 3 - i := j*32 - dst[i+31:i] := a[i+31:i+16]*b[i+31:i+16] + a[i+15:i]*b[i+15:i] + i := j*64 + dst[i+63:i] := SINH(a[i+63:i]) ENDFOR +dst[MAX:256] := 0 - -
emmintrin.h
+
immintrin.h
 - - Integer - SSE2 - Special Math Functions - - - Compare packed 16-bit integers in "a" and "b", and store packed maximum values in "dst". + + Floating Point + AVX + Trigonometry + + + Compute the hyperbolic sine of packed single-precision (32-bit) floating-point elements in "a" expressed in radians, and store the results in "dst". FOR j := 0 to 7 - i := j*16 - IF a[i+15:i] > b[i+15:i] - dst[i+15:i] := a[i+15:i] - ELSE - dst[i+15:i] := b[i+15:i] - FI + i := j*32 + dst[i+31:i] := SINH(a[i+31:i]) ENDFOR +dst[MAX:256] := 0 - -
emmintrin.h
+
immintrin.h
 - - Integer - SSE2 + + Floating Point + AVX Special Math Functions - - - Compare packed unsigned 8-bit integers in "a" and "b", and store packed maximum values in "dst". + + + Round the packed double-precision (64-bit) floating-point elements in "a" up to an integer value, and store the results as packed double-precision floating-point elements in "dst". This intrinsic may generate the "roundpd"/"vroundpd" instruction. -FOR j := 0 to 15 - i := j*8 - IF a[i+7:i] > b[i+7:i] - dst[i+7:i] := a[i+7:i] - ELSE - dst[i+7:i] := b[i+7:i] - FI +FOR j := 0 to 3 + i := j*64 + dst[i+63:i] := CEIL(a[i+63:i]) ENDFOR +dst[MAX:256] := 0 - -
emmintrin.h
+
immintrin.h
 - - Integer - SSE2 + + Floating Point + AVX Special Math Functions - - - Compare packed 16-bit integers in "a" and "b", and store packed minimum values in "dst". + + + Round the packed single-precision (32-bit) floating-point elements in "a" up to an integer value, and store the results as packed single-precision floating-point elements in "dst". This intrinsic may generate the "roundps"/"vroundps" instruction. FOR j := 0 to 7 - i := j*16 - IF a[i+15:i] < b[i+15:i] - dst[i+15:i] := a[i+15:i] - ELSE - dst[i+15:i] := b[i+15:i] - FI + i := j*32 + dst[i+31:i] := CEIL(a[i+31:i]) ENDFOR +dst[MAX:256] := 0 - -
emmintrin.h
+
immintrin.h
 - - Integer - SSE2 + + Floating Point + AVX Special Math Functions - - - Compare packed unsigned 8-bit integers in "a" and "b", and store packed minimum values in "dst". + + + Round the packed double-precision (64-bit) floating-point elements in "a" down to an integer value, and store the results as packed double-precision floating-point elements in "dst". This intrinsic may generate the "roundpd"/"vroundpd" instruction. -FOR j := 0 to 15 - i := j*8 - IF a[i+7:i] < b[i+7:i] - dst[i+7:i] := a[i+7:i] - ELSE - dst[i+7:i] := b[i+7:i] - FI +FOR j := 0 to 3 + i := j*64 + dst[i+63:i] := FLOOR(a[i+63:i]) ENDFOR +dst[MAX:256] := 0 - -
emmintrin.h
+
immintrin.h
 - - Integer - SSE2 - Arithmetic - - - Multiply the packed 16-bit integers in "a" and "b", producing intermediate 32-bit integers, and store the high 16 bits of the intermediate integers in "dst". + + Floating Point + AVX + Special Math Functions + + + Round the packed single-precision (32-bit) floating-point elements in "a" down to an integer value, and store the results as packed single-precision floating-point elements in "dst". This intrinsic may generate the "roundps"/"vroundps" instruction. FOR j := 0 to 7 - i := j*16 - tmp[31:0] := a[i+15:i] * b[i+15:i] - dst[i+15:i] := tmp[31:16] + i := j*32 + dst[i+31:i] := FLOOR(a[i+31:i]) ENDFOR +dst[MAX:256] := 0 - -
emmintrin.h
+
immintrin.h
 - - Integer - SSE2 - Arithmetic - - - Multiply the packed unsigned 16-bit integers in "a" and "b", producing intermediate 32-bit integers, and store the high 16 bits of the intermediate integers in "dst". + + Floating Point + AVX + Special Math Functions + + + Round the packed double-precision (64-bit) floating-point elements in "a" to the nearest integer value, and store the results as packed double-precision floating-point elements in "dst". This intrinsic may generate the "roundpd"/"vroundpd" instruction. -FOR j := 0 to 7 - i := j*16 - tmp[31:0] := a[i+15:i] * b[i+15:i] - dst[i+15:i] := tmp[31:16] +FOR j := 0 to 3 + i := j*64 + dst[i+63:i] := ROUND(a[i+63:i]) ENDFOR +dst[MAX:256] := 0 - -
emmintrin.h
+
immintrin.h
 - - Integer - SSE2 - Arithmetic - - - Multiply the packed 16-bit integers in "a" and "b", producing intermediate 32-bit integers, and store the low 16 bits of the intermediate integers in "dst". + + Floating Point + AVX + Special Math Functions + + + Round the packed single-precision (32-bit) floating-point elements in "a" to the nearest integer value, and store the results as packed single-precision floating-point elements in "dst". This intrinsic may generate the "roundps"/"vroundps" instruction. FOR j := 0 to 7 - i := j*16 - tmp[31:0] := a[i+15:i] * b[i+15:i] - dst[i+15:i] := tmp[15:0] + i := j*32 + dst[i+31:i] := ROUND(a[i+31:i]) ENDFOR +dst[MAX:256] := 0 - -
emmintrin.h
+
immintrin.h
 - - Integer - SSE2 - Arithmetic - - - Multiply the low unsigned 32-bit integers from "a" and "b", and store the unsigned 64-bit result in "dst". + + Floating Point + AVX + Elementary Math Functions + + + Compute the square root of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst". Note that this intrinsic is less efficient than "_mm_sqrt_pd". -dst[63:0] := a[31:0] * b[31:0] +FOR j := 0 to 3 + i := j*64 + dst[i+63:i] := SQRT(a[i+63:i]) +ENDFOR +dst[MAX:256] := 0 - -
emmintrin.h
+
immintrin.h
 - - Integer - SSE2 - Arithmetic - - - Multiply the low unsigned 32-bit integers from each packed 64-bit element in "a" and "b", and store the unsigned 64-bit results in "dst". + + Floating Point + AVX + Elementary Math Functions + + + Compute the square root of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst". Note that this intrinsic is less efficient than "_mm_sqrt_ps". -FOR j := 0 to 1 +FOR j := 0 to 7 + i := j*32 + dst[i+31:i] := SQRT(a[i+31:i]) +ENDFOR +dst[MAX:256] := 0 + +
immintrin.h
+ + + Floating Point + AVX + Trigonometry + + + Compute the tangent of packed double-precision (64-bit) floating-point elements in "a" expressed in radians, and store the results in "dst". + +FOR j := 0 to 3 i := j*64 - dst[i+63:i] := a[i+31:i] * b[i+31:i] + dst[i+63:i] := TAN(a[i+63:i]) ENDFOR +dst[MAX:256] := 0 - -
emmintrin.h
+
immintrin.h
 - - Integer - SSE2 - Arithmetic - Miscellaneous - - - Compute the absolute differences of packed unsigned 8-bit integers in "a" and "b", then horizontally sum each consecutive 8 differences to produce two unsigned 16-bit integers, and pack these unsigned 16-bit integers in the low 16 bits of 64-bit elements in "dst". + + Floating Point + AVX + Trigonometry + + + Compute the tangent of packed single-precision (32-bit) floating-point elements in "a" expressed in radians, and store the results in "dst". -FOR j := 0 to 15 - i := j*8 - tmp[i+7:i] := ABS(a[i+7:i] - b[i+7:i]) +FOR j := 0 to 7 + i := j*32 + dst[i+31:i] := TAN(a[i+31:i]) ENDFOR -FOR j := 0 to 1 +dst[MAX:256] := 0 + +
immintrin.h
+ + + Floating Point + AVX + Trigonometry + + + Compute the tangent of packed double-precision (64-bit) floating-point elements in "a" expressed in degrees, and store the results in "dst". + FOR j := 0 to 3 i := j*64 - dst[i+15:i] := tmp[i+7:i] + tmp[i+15:i+8] + tmp[i+23:i+16] + tmp[i+31:i+24] + - tmp[i+39:i+32] + tmp[i+47:i+40] + tmp[i+55:i+48] + tmp[i+63:i+56] - dst[i+63:i+16] := 0 + dst[i+63:i] := TAND(a[i+63:i]) ENDFOR +dst[MAX:256] := 0 - -
emmintrin.h
+
immintrin.h
 - - Integer - SSE2 - Arithmetic - - - Subtract packed 8-bit integers in "b" from packed 8-bit integers in "a", and store the results in "dst". + + Floating Point + AVX + Trigonometry + + + Compute the tangent of packed single-precision (32-bit) floating-point elements in "a" expressed in degrees, and store the results in "dst". + FOR j := 0 to 7 + i := j*32 + dst[i+31:i] := TAND(a[i+31:i]) +ENDFOR +dst[MAX:256] := 0 + +
immintrin.h
+ + + Floating Point + AVX + Trigonometry + + + Compute the hyperbolic tangent of packed double-precision (64-bit) floating-point elements in "a" expressed in radians, and store the results in "dst". -FOR j := 0 to 15 - i := j*8 - dst[i+7:i] := a[i+7:i] - b[i+7:i] +FOR j := 0 to 3 + i := j*64 + dst[i+63:i] := TANH(a[i+63:i]) ENDFOR +dst[MAX:256] := 0 - -
emmintrin.h
+
immintrin.h
 - - Integer - SSE2 - Arithmetic - - - Subtract packed 16-bit integers in "b" from packed 16-bit integers in "a", and store the results in "dst". + + Floating Point + AVX + Trigonometry + + + Compute the hyperbolic tangent of packed single-precision (32-bit) floating-point elements in "a" expressed in radians, and store the results in "dst". FOR j := 0 to 7 - i := j*16 - dst[i+15:i] := a[i+15:i] - b[i+15:i] + i := j*32 + dst[i+31:i] := TANH(a[i+31:i]) ENDFOR +dst[MAX:256] := 0 - -
emmintrin.h
+
immintrin.h
 - - Integer - SSE2 - Arithmetic - - - Subtract packed 32-bit integers in "b" from packed 32-bit integers in "a", and store the results in "dst". - -FOR j := 0 to 3 + + Floating Point + AVX + Miscellaneous + + + Truncate the packed double-precision (64-bit) floating-point elements in "a", and store the results as packed double-precision floating-point elements in "dst". This intrinsic may generate the "roundpd"/"vroundpd" instruction. + FOR j := 0 to 3 + i := j*64 + dst[i+63:i] := TRUNCATE(a[i+63:i]) +ENDFOR +dst[MAX:256] := 0 + +
immintrin.h
+ + + Floating Point + AVX + Miscellaneous + + + Truncate the packed single-precision (32-bit) floating-point elements in "a", and store the results as packed single-precision floating-point elements in "dst". This intrinsic may generate the "roundps"/"vroundps" instruction. + FOR j := 0 to 7 i := j*32 - dst[i+31:i] := a[i+31:i] - b[i+31:i] + dst[i+31:i] := TRUNCATE(a[i+31:i]) ENDFOR +dst[MAX:256] := 0 - -
emmintrin.h
+
immintrin.h
 - + Integer - SSE2 + AVX Arithmetic - - - Subtract 64-bit integer "b" from 64-bit integer "a", and store the result in "dst". - -dst[63:0] := a[63:0] - b[63:0] + + + + Divide packed unsigned 32-bit integers in "a" by packed elements in "b", and store the truncated results in "dst". + FOR j := 0 to 7 + i := 32*j + dst[i+31:i] := TRUNCATE(a[i+31:i] / b[i+31:i]) +ENDFOR +dst[MAX:256] := 0 - -
emmintrin.h
+
immintrin.h
 - + Integer - SSE2 + AVX Arithmetic - - - Subtract packed 64-bit integers in "b" from packed 64-bit integers in "a", and store the results in "dst". - -FOR j := 0 to 1 - i := j*64 - dst[i+63:i] := a[i+63:i] - b[i+63:i] + + + + + Divide packed unsigned 32-bit integers in "a" by packed elements in "b", store the truncated results in "dst", and store the remainders as packed unsigned 32-bit integers into memory at "mem_addr". + FOR j := 0 to 7 + i := 32*j + dst[i+31:i] := TRUNCATE(a[i+31:i] / b[i+31:i]) + MEM[mem_addr+i+31:mem_addr+i] := REMAINDER(a[i+31:i] / b[i+31:i]) ENDFOR +dst[MAX:256] := 0 - -
emmintrin.h
+
immintrin.h
 - + Integer - SSE2 + AVX Arithmetic - - - Subtract packed 8-bit integers in "b" from packed 8-bit integers in "a" using saturation, and store the results in "dst". - -FOR j := 0 to 15 - i := j*8 - dst[i+7:i] := Saturate_To_Int8(a[i+7:i] - b[i+7:i]) + + + + Divide packed unsigned 32-bit integers in "a" by packed elements in "b", and store the remainders as packed unsigned 32-bit integers in "dst". + FOR j := 0 to 7 + i := 32*j + dst[i+31:i] := REMAINDER(a[i+31:i] / b[i+31:i]) ENDFOR +dst[MAX:256] := 0 - -
emmintrin.h
+
immintrin.h
 - - Integer - SSE2 - Arithmetic - - - Subtract packed 16-bit integers in "b" from packed 16-bit integers in "a" using saturation, and store the results in "dst". + + Floating Point + AVX + Convert + + + Copy the lower single-precision (32-bit) floating-point element of "a" to "dst". -FOR j := 0 to 7 - i := j*16 - dst[i+15:i] := Saturate_To_Int16(a[i+15:i] - b[i+15:i]) -ENDFOR +dst[31:0] := a[31:0] - -
emmintrin.h
+ +
immintrin.h
 - - Integer - SSE2 - Arithmetic - - - Subtract packed unsigned 8-bit integers in "b" from packed unsigned 8-bit integers in "a" using saturation, and store the results in "dst". + + Floating Point + AVX + Convert + + + Copy the lower double-precision (64-bit) floating-point element of "a" to "dst". -FOR j := 0 to 15 - i := j*8 - dst[i+7:i] := Saturate_To_UnsignedInt8(a[i+7:i] - b[i+7:i]) -ENDFOR +dst[63:0] := a[63:0] - -
emmintrin.h
+ +
immintrin.h
 - + Integer - SSE2 - Arithmetic - - - Subtract packed unsigned 16-bit integers in "b" from packed unsigned 16-bit integers in "a" using saturation, and store the results in "dst". + AVX + Convert + + + Copy the lower 32-bit integer in "a" to "dst". -FOR j := 0 to 7 - i := j*16 - dst[i+15:i] := Saturate_To_UnsignedInt16(a[i+15:i] - b[i+15:i]) -ENDFOR +dst[31:0] := a[31:0] - -
emmintrin.h
+ +
immintrin.h
 - + Integer - SSE2 - Shift - - - Shift "a" left by "imm8" bytes while shifting in zeros, and store the results in "dst". + AVX2 + Swizzle + + + + Extract an 8-bit integer from "a", selected with "index", and store the result in "dst". -tmp := imm8[7:0] -IF tmp > 15 - tmp := 16 -FI -dst[127:0] := a[127:0] << (tmp*8) +dst[7:0] := (a[255:0] >> (index[4:0] * 8))[7:0] - -
emmintrin.h
+
immintrin.h
 - + Integer - SSE2 - Shift - - - Shift "a" left by "imm8" bytes while shifting in zeros, and store the results in "dst". + AVX2 + Swizzle + + + + Extract a 16-bit integer from "a", selected with "index", and store the result in "dst". -tmp := imm8[7:0] -IF tmp > 15 - tmp := 16 -FI -dst[127:0] := a[127:0] << (tmp*8) +dst[15:0] := (a[255:0] >> (index[3:0] * 16))[15:0] - -
emmintrin.h
+
immintrin.h
 - + Integer - SSE2 - Shift - - - Shift "a" right by "imm8" bytes while shifting in zeros, and store the results in "dst". + AVX2 + Special Math Functions + + + Compute the absolute value of packed signed 8-bit integers in "a", and store the unsigned results in "dst". -tmp := imm8[7:0] -IF tmp > 15 - tmp := 16 -FI -dst[127:0] := a[127:0] >> (tmp*8) +FOR j := 0 to 31 + i := j*8 + dst[i+7:i] := ABS(a[i+7:i]) +ENDFOR +dst[MAX:256] := 0 - -
emmintrin.h
+ +
immintrin.h
 - + Integer - SSE2 - Shift - - - Shift packed 16-bit integers in "a" left by "imm8" while shifting in zeros, and store the results in "dst". + AVX2 + Special Math Functions + + + Compute the absolute value of packed signed 16-bit integers in "a", and store the unsigned results in "dst". -FOR j := 0 to 7 +FOR j := 0 to 15 i := j*16 - IF imm8[7:0] > 15 - dst[i+15:i] := 0 - ELSE - dst[i+15:i] := ZeroExtend(a[i+15:i] << imm8[7:0]) - FI + dst[i+15:i] := ABS(a[i+15:i]) ENDFOR +dst[MAX:256] := 0 - -
emmintrin.h
+ +
immintrin.h
 - + Integer - SSE2 - Shift - - - Shift packed 16-bit integers in "a" left by "count" while shifting in zeros, and store the results in "dst". + AVX2 + Special Math Functions + + + Compute the absolute value of packed signed 32-bit integers in "a", and store the unsigned results in "dst". FOR j := 0 to 7 - i := j*16 - IF count[63:0] > 15 - dst[i+15:i] := 0 - ELSE - dst[i+15:i] := ZeroExtend(a[i+15:i] << count[63:0]) - FI + i := j*32 + dst[i+31:i] := ABS(a[i+31:i]) ENDFOR +dst[MAX:256] := 0 - -
emmintrin.h
+ +
immintrin.h
 - + Integer - SSE2 - Shift - - - Shift packed 32-bit integers in "a" left by "imm8" while shifting in zeros, and store the results in "dst". + AVX2 + Arithmetic + + + + Add packed 8-bit integers in "a" and "b", and store the results in "dst". -FOR j := 0 to 3 - i := j*32 - IF imm8[7:0] > 31 - dst[i+31:i] := 0 - ELSE - dst[i+31:i] := ZeroExtend(a[i+31:i] << imm8[7:0]) - FI +FOR j := 0 to 31 + i := j*8 + dst[i+7:i] := a[i+7:i] + b[i+7:i] ENDFOR +dst[MAX:256] := 0 - -
emmintrin.h
+ +
immintrin.h
 - + Integer - SSE2 - Shift - - - Shift packed 32-bit integers in "a" left by "count" while shifting in zeros, and store the results in "dst". + AVX2 + Arithmetic + + + + Add packed 16-bit integers in "a" and "b", and store the results in "dst". -FOR j := 0 to 3 +FOR j := 0 to 15 + i := j*16 + dst[i+15:i] := a[i+15:i] + b[i+15:i] +ENDFOR +dst[MAX:256] := 0 + + +
immintrin.h
+ + + Integer + AVX2 + Arithmetic + + + + Add packed 32-bit integers in "a" and "b", and store the results in "dst". + +FOR j := 0 to 7 i := j*32 - IF count[63:0] > 31 - dst[i+31:i] := 0 - ELSE - dst[i+31:i] := ZeroExtend(a[i+31:i] << count[63:0]) - FI + dst[i+31:i] := a[i+31:i] + b[i+31:i] ENDFOR +dst[MAX:256] := 0 - -
emmintrin.h
+ +
immintrin.h
 - + Integer - SSE2 - Shift - - - Shift packed 64-bit integers in "a" left by "imm8" while shifting in zeros, and store the results in "dst". + AVX2 + Arithmetic + + + + Add packed 64-bit integers in "a" and "b", and store the results in "dst". -FOR j := 0 to 1 +FOR j := 0 to 3 i := j*64 - IF imm8[7:0] > 63 - dst[i+63:i] := 0 - ELSE - dst[i+63:i] := ZeroExtend(a[i+63:i] << imm8[7:0]) - FI + dst[i+63:i] := a[i+63:i] + b[i+63:i] ENDFOR +dst[MAX:256] := 0 - -
emmintrin.h
+ +
immintrin.h
 - + Integer - SSE2 - Shift - - - Shift packed 64-bit integers in "a" left by "count" while shifting in zeros, and store the results in "dst". + AVX2 + Arithmetic + + + + Add packed 8-bit integers in "a" and "b" using saturation, and store the results in "dst". -FOR j := 0 to 1 - i := j*64 - IF count[63:0] > 63 - dst[i+63:i] := 0 - ELSE - dst[i+63:i] := ZeroExtend(a[i+63:i] << count[63:0]) - FI +FOR j := 0 to 31 + i := j*8 + dst[i+7:i] := Saturate8( a[i+7:i] + b[i+7:i] ) ENDFOR +dst[MAX:256] := 0 - -
emmintrin.h
+ +
immintrin.h
 - + Integer - SSE2 - Shift - - - Shift packed 16-bit integers in "a" right by "imm8" while shifting in sign bits, and store the results in "dst". + AVX2 + Arithmetic + + + + Add packed 16-bit integers in "a" and "b" using saturation, and store the results in "dst". -FOR j := 0 to 7 +FOR j := 0 to 15 i := j*16 - IF imm8[7:0] > 15 - dst[i+15:i] := SignBit - ELSE - dst[i+15:i] := SignExtend(a[i+15:i] >> imm8[7:0]) - FI + dst[i+15:i] := Saturate16( a[i+15:i] + b[i+15:i] ) ENDFOR +dst[MAX:256] := 0 - -
emmintrin.h
+ +
immintrin.h
 - + Integer - SSE2 - Shift - - - Shift packed 16-bit integers in "a" right by "count" while shifting in sign bits, and store the results in "dst". + AVX2 + Arithmetic + + + + Add packed unsigned 8-bit integers in "a" and "b" using saturation, and store the results in "dst". -FOR j := 0 to 7 - i := j*16 - IF count[63:0] > 15 - dst[i+15:i] := SignBit - ELSE - dst[i+15:i] := SignExtend(a[i+15:i] >> count[63:0]) - FI +FOR j := 0 to 31 + i := j*8 + dst[i+7:i] := SaturateU8( a[i+7:i] + b[i+7:i] ) ENDFOR +dst[MAX:256] := 0 - -
emmintrin.h
+ +
immintrin.h
 - + Integer - SSE2 - Shift - - - Shift packed 32-bit integers in "a" right by "imm8" while shifting in sign bits, and store the results in "dst". + AVX2 + Arithmetic + + + + Add packed unsigned 16-bit integers in "a" and "b" using saturation, and store the results in "dst". -FOR j := 0 to 3 - i := j*32 - IF imm8[7:0] > 31 - dst[i+31:i] := SignBit - ELSE - dst[i+31:i] := SignExtend(a[i+31:i] >> imm8[7:0]) - FI +FOR j := 0 to 15 + i := j*16 + dst[i+15:i] := SaturateU16( a[i+15:i] + b[i+15:i] ) ENDFOR +dst[MAX:256] := 0 - -
emmintrin.h
+ +
immintrin.h
 - + Integer - SSE2 - Shift - - - Shift packed 32-bit integers in "a" right by "count" while shifting in sign bits, and store the results in "dst". + AVX2 + Miscellaneous + + + + + Concatenate pairs of 16-byte blocks in "a" and "b" into a 32-byte temporary result, shift the result right by "imm8" bytes, and store the low 16 bytes in "dst". -FOR j := 0 to 3 - i := j*32 - IF count[63:0] > 31 - dst[i+31:i] := SignBit - ELSE - dst[i+31:i] := SignExtend(a[i+31:i] >> count[63:0]) - FI +FOR j := 0 to 1 + i := j*128 + tmp[255:0] := ((a[i+127:i] << 128)[255:0] OR b[i+127:i]) >> (imm8*8) + dst[i+127:i] := tmp[127:0] ENDFOR +dst[MAX:256] := 0 - -
emmintrin.h
+ +
immintrin.h
 - + Integer - SSE2 - Shift - - - Shift "a" right by "imm8" bytes while shifting in zeros, and store the results in "dst". + AVX2 + Logical + + + + Compute the bitwise AND of 256 bits (representing integer data) in "a" and "b", and store the result in "dst". -tmp := imm8[7:0] -IF tmp > 15 - tmp := 16 -FI -dst[127:0] := a[127:0] >> (tmp*8) +dst[255:0] := (a[255:0] AND b[255:0]) +dst[MAX:256] := 0 - -
emmintrin.h
+ +
immintrin.h
 - + Integer - SSE2 - Shift - - - Shift packed 16-bit integers in "a" right by "imm8" while shifting in zeros, and store the results in "dst". + AVX2 + Logical + + + + Compute the bitwise NOT of 256 bits (representing integer data) in "a" and then AND with "b", and store the result in "dst". -FOR j := 0 to 7 - i := j*16 - IF imm8[7:0] > 15 - dst[i+15:i] := 0 - ELSE - dst[i+15:i] := ZeroExtend(a[i+15:i] >> imm8[7:0]) - FI +dst[255:0] := ((NOT a[255:0]) AND b[255:0]) +dst[MAX:256] := 0 + + +
immintrin.h
+ + + Integer + AVX2 + Probability/Statistics + + + + Average packed unsigned 8-bit integers in "a" and "b", and store the results in "dst". + +FOR j := 0 to 31 + i := j*8 + dst[i+7:i] := (a[i+7:i] + b[i+7:i] + 1) >> 1 ENDFOR +dst[MAX:256] := 0 - -
emmintrin.h
+ +
immintrin.h
 - + Integer - SSE2 - Shift - - - Shift packed 16-bit integers in "a" right by "count" while shifting in zeros, and store the results in "dst". + AVX2 + Probability/Statistics + + + + Average packed unsigned 16-bit integers in "a" and "b", and store the results in "dst". -FOR j := 0 to 7 +FOR j := 0 to 15 i := j*16 - IF count[63:0] > 15 - dst[i+15:i] := 0 - ELSE - dst[i+15:i] := ZeroExtend(a[i+15:i] >> count[63:0]) - FI + dst[i+15:i] := (a[i+15:i] + b[i+15:i] + 1) >> 1 ENDFOR +dst[MAX:256] := 0 - -
emmintrin.h
+ +
immintrin.h
 - + Integer - SSE2 - Shift - - - Shift packed 32-bit integers in "a" right by "imm8" while shifting in zeros, and store the results in "dst". + AVX2 + Swizzle + + + + + Blend packed 16-bit integers from "a" and "b" within 128-bit lanes using control mask "imm8", and store the results in "dst". -FOR j := 0 to 3 - i := j*32 - IF imm8[7:0] > 31 - dst[i+31:i] := 0 +FOR j := 0 to 15 + i := j*16 + IF imm8[j%8] + dst[i+15:i] := b[i+15:i] ELSE - dst[i+31:i] := ZeroExtend(a[i+31:i] >> imm8[7:0]) + dst[i+15:i] := a[i+15:i] FI ENDFOR +dst[MAX:256] := 0 - -
emmintrin.h
+ +
immintrin.h
 - + Integer - SSE2 - Shift - - - Shift packed 32-bit integers in "a" right by "count" while shifting in zeros, and store the results in "dst". + AVX2 + Swizzle + + + + + Blend packed 32-bit integers from "a" and "b" using control mask "imm8", and store the results in "dst". FOR j := 0 to 3 i := j*32 - IF count[63:0] > 31 - dst[i+31:i] := 0 + IF imm8[j] + dst[i+31:i] := b[i+31:i] ELSE - dst[i+31:i] := ZeroExtend(a[i+31:i] >> count[63:0]) + dst[i+31:i] := a[i+31:i] FI ENDFOR +dst[MAX:128] := 0 - -
emmintrin.h
+ +
immintrin.h
 - + Integer - SSE2 - Shift - - - Shift packed 64-bit integers in "a" right by "imm8" while shifting in zeros, and store the results in "dst". + AVX2 + Swizzle + + + + + Blend packed 32-bit integers from "a" and "b" using control mask "imm8", and store the results in "dst". -FOR j := 0 to 1 - i := j*64 - IF imm8[7:0] > 63 - dst[i+63:i] := 0 +FOR j := 0 to 7 + i := j*32 + IF imm8[j] + dst[i+31:i] := b[i+31:i] ELSE - dst[i+63:i] := ZeroExtend(a[i+63:i] >> imm8[7:0]) + dst[i+31:i] := a[i+31:i] FI ENDFOR +dst[MAX:256] := 0 - -
emmintrin.h
+ +
immintrin.h
 - + Integer - SSE2 - Shift - - - Shift packed 64-bit integers in "a" right by "count" while shifting in zeros, and store the results in "dst". + AVX2 + Swizzle + + + + + Blend packed 8-bit integers from "a" and "b" using "mask", and store the results in "dst". -FOR j := 0 to 1 - i := j*64 - IF count[63:0] > 63 - dst[i+63:i] := 0 +FOR j := 0 to 31 + i := j*8 + IF mask[i+7] + dst[i+7:i] := b[i+7:i] ELSE - dst[i+63:i] := ZeroExtend(a[i+63:i] >> count[63:0]) + dst[i+7:i] := a[i+7:i] FI ENDFOR +dst[MAX:256] := 0 - -
emmintrin.h
+ +
immintrin.h
 - + Integer - SSE2 - Logical - - - Compute the bitwise AND of 128 bits (representing integer data) in "a" and "b", and store the result in "dst". + AVX2 + Swizzle + + + Broadcast the low packed 8-bit integer from "a" to all elements of "dst". -dst[127:0] := (a[127:0] AND b[127:0]) +FOR j := 0 to 15 + i := j*8 + dst[i+7:i] := a[7:0] +ENDFOR +dst[MAX:128] := 0 - -
emmintrin.h
+ +
immintrin.h
 - + Integer - SSE2 - Logical - - - Compute the bitwise NOT of 128 bits (representing integer data) in "a" and then AND with "b", and store the result in "dst". + AVX2 + Swizzle + + + Broadcast the low packed 8-bit integer from "a" to all elements of "dst". -dst[127:0] := ((NOT a[127:0]) AND b[127:0]) +FOR j := 0 to 31 + i := j*8 + dst[i+7:i] := a[7:0] +ENDFOR +dst[MAX:256] := 0 - -
emmintrin.h
+ +
immintrin.h
 - + Integer - SSE2 - Logical - - - Compute the bitwise OR of 128 bits (representing integer data) in "a" and "b", and store the result in "dst". + AVX2 + Swizzle + + + Broadcast the low packed 32-bit integer from "a" to all elements of "dst". -dst[127:0] := (a[127:0] OR b[127:0]) +FOR j := 0 to 3 + i := j*32 + dst[i+31:i] := a[31:0] +ENDFOR +dst[MAX:128] := 0 - -
emmintrin.h
+ +
immintrin.h
 - + Integer - SSE2 - Logical - - - Compute the bitwise XOR of 128 bits (representing integer data) in "a" and "b", and store the result in "dst". + AVX2 + Swizzle + + + Broadcast the low packed 32-bit integer from "a" to all elements of "dst". -dst[127:0] := (a[127:0] XOR b[127:0]) +FOR j := 0 to 7 + i := j*32 + dst[i+31:i] := a[31:0] +ENDFOR +dst[MAX:256] := 0 - -
emmintrin.h
+ +
immintrin.h
 - + Integer - SSE2 + AVX2 + Swizzle + + + Broadcast the low packed 64-bit integer from "a" to all elements of "dst". + +FOR j := 0 to 1 + i := j*64 + dst[i+63:i] := a[63:0] +ENDFOR +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Integer + AVX2 + Swizzle + + + Broadcast the low packed 64-bit integer from "a" to all elements of "dst". + +FOR j := 0 to 3 + i := j*64 + dst[i+63:i] := a[63:0] +ENDFOR +dst[MAX:256] := 0 + + +
immintrin.h
+ + + Floating Point + AVX2 + Swizzle + + + Broadcast the low double-precision (64-bit) floating-point element from "a" to all elements of "dst". + +FOR j := 0 to 1 + i := j*64 + dst[i+63:i] := a[63:0] +ENDFOR +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Floating Point + AVX2 + Swizzle + + + Broadcast the low double-precision (64-bit) floating-point element from "a" to all elements of "dst". + +FOR j := 0 to 3 + i := j*64 + dst[i+63:i] := a[63:0] +ENDFOR +dst[MAX:256] := 0 + + +
immintrin.h
+ + + Integer + AVX2 + Swizzle + + + Broadcast 128 bits of integer data from "a" to all 128-bit lanes in "dst". + +dst[127:0] := a[127:0] +dst[255:128] := a[127:0] +dst[MAX:256] := 0 + + +
immintrin.h
+ + + Integer + AVX2 + Swizzle + + + Broadcast 128 bits of integer data from "a" to all 128-bit lanes in "dst". + +dst[127:0] := a[127:0] +dst[255:128] := a[127:0] +dst[MAX:256] := 0 + + +
immintrin.h
+ + + Floating Point + AVX2 + Swizzle + + + Broadcast the low single-precision (32-bit) floating-point element from "a" to all elements of "dst". + +FOR j := 0 to 3 + i := j*32 + dst[i+31:i] := a[31:0] +ENDFOR +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Floating Point + AVX2 + Swizzle + + + Broadcast the low single-precision (32-bit) floating-point element from "a" to all elements of "dst". + +FOR j := 0 to 7 + i := j*32 + dst[i+31:i] := a[31:0] +ENDFOR +dst[MAX:256] := 0 + + +
immintrin.h
+ + + Integer + AVX2 + Swizzle + + + Broadcast the low packed 16-bit integer from "a" to all elements of "dst". + +FOR j := 0 to 7 + i := j*16 + dst[i+15:i] := a[15:0] +ENDFOR +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Integer + AVX2 + Swizzle + + + Broadcast the low packed 16-bit integer from "a" to all elements of "dst". + +FOR j := 0 to 15 + i := j*16 + dst[i+15:i] := a[15:0] +ENDFOR +dst[MAX:256] := 0 + + +
immintrin.h
+ + + Integer + AVX2 Compare - - + + + Compare packed 8-bit integers in "a" and "b" for equality, and store the results in "dst". -FOR j := 0 to 15 +FOR j := 0 to 31 i := j*8 dst[i+7:i] := ( a[i+7:i] == b[i+7:i] ) ? 0xFF : 0 ENDFOR +dst[MAX:256] := 0 - -
emmintrin.h
+ +
immintrin.h
 - + Integer - SSE2 + AVX2 Compare - - + + + Compare packed 16-bit integers in "a" and "b" for equality, and store the results in "dst". -FOR j := 0 to 7 +FOR j := 0 to 15 i := j*16 dst[i+15:i] := ( a[i+15:i] == b[i+15:i] ) ? 0xFFFF : 0 ENDFOR +dst[MAX:256] := 0 - -
emmintrin.h
+ +
immintrin.h
 - + Integer - SSE2 + AVX2 Compare - - + + + Compare packed 32-bit integers in "a" and "b" for equality, and store the results in "dst". -FOR j := 0 to 3 +FOR j := 0 to 7 i := j*32 dst[i+31:i] := ( a[i+31:i] == b[i+31:i] ) ? 0xFFFFFFFF : 0 ENDFOR +dst[MAX:256] := 0 - -
emmintrin.h
+ +
immintrin.h
 - + Integer - SSE2 + AVX2 Compare - - - Compare packed 8-bit integers in "a" and "b" for greater-than, and store the results in "dst". + + + + Compare packed 64-bit integers in "a" and "b" for equality, and store the results in "dst". -FOR j := 0 to 15 +FOR j := 0 to 3 + i := j*64 + dst[i+63:i] := ( a[i+63:i] == b[i+63:i] ) ? 0xFFFFFFFFFFFFFFFF : 0 +ENDFOR +dst[MAX:256] := 0 + + +
immintrin.h
+ + + Integer + AVX2 + Compare + + + + Compare packed signed 8-bit integers in "a" and "b" for greater-than, and store the results in "dst". + +FOR j := 0 to 31 i := j*8 dst[i+7:i] := ( a[i+7:i] > b[i+7:i] ) ? 0xFF : 0 ENDFOR +dst[MAX:256] := 0 - -
emmintrin.h
+ +
immintrin.h
 - + Integer - SSE2 + AVX2 Compare - - - Compare packed 16-bit integers in "a" and "b" for greater-than, and store the results in "dst". + + + + Compare packed signed 16-bit integers in "a" and "b" for greater-than, and store the results in "dst". -FOR j := 0 to 7 +FOR j := 0 to 15 i := j*16 dst[i+15:i] := ( a[i+15:i] > b[i+15:i] ) ? 0xFFFF : 0 ENDFOR +dst[MAX:256] := 0 - -
emmintrin.h
+ +
immintrin.h
 - + Integer - SSE2 + AVX2 Compare - - - Compare packed 32-bit integers in "a" and "b" for greater-than, and store the results in "dst". + + + + Compare packed signed 32-bit integers in "a" and "b" for greater-than, and store the results in "dst". -FOR j := 0 to 3 +FOR j := 0 to 7 i := j*32 dst[i+31:i] := ( a[i+31:i] > b[i+31:i] ) ? 0xFFFFFFFF : 0 ENDFOR +dst[MAX:256] := 0 - -
emmintrin.h
+ +
immintrin.h
 - + Integer - SSE2 + AVX2 Compare - - - Compare packed 8-bit integers in "a" and "b" for less-than, and store the results in "dst". Note: This intrinsic emits the pcmpgtb instruction with the order of the operands switched. + + + + Compare packed signed 64-bit integers in "a" and "b" for greater-than, and store the results in "dst". -FOR j := 0 to 15 - i := j*8 - dst[i+7:i] := ( a[i+7:i] < b[i+7:i] ) ? 0xFF : 0 +FOR j := 0 to 3 + i := j*64 + dst[i+63:i] := ( a[i+63:i] > b[i+63:i] ) ? 0xFFFFFFFFFFFFFFFF : 0 ENDFOR +dst[MAX:256] := 0 - -
emmintrin.h
+ +
immintrin.h
 - + Integer - SSE2 - Compare - - - Compare packed 16-bit integers in "a" and "b" for less-than, and store the results in "dst". Note: This intrinsic emits the pcmpgtw instruction with the order of the operands switched. + AVX2 + Convert + + + Sign extend packed 16-bit integers in "a" to packed 32-bit integers, and store the results in "dst". -FOR j := 0 to 7 - i := j*16 - dst[i+15:i] := ( a[i+15:i] < b[i+15:i] ) ? 0xFFFF : 0 +FOR j:= 0 to 7 + i := 32*j + k := 16*j + dst[i+31:i] := SignExtend32(a[k+15:k]) ENDFOR +dst[MAX:256] := 0 - -
emmintrin.h
+ +
immintrin.h
 - + Integer - SSE2 - Compare - - - Compare packed 32-bit integers in "a" and "b" for less-than, and store the results in "dst". Note: This intrinsic emits the pcmpgtd instruction with the order of the operands switched. + AVX2 + Convert + + + Sign extend packed 16-bit integers in "a" to packed 64-bit integers, and store the results in "dst". -FOR j := 0 to 3 - i := j*32 - dst[i+31:i] := ( a[i+31:i] < b[i+31:i] ) ? 0xFFFFFFFF : 0 +FOR j:= 0 to 3 + i := 64*j + k := 16*j + dst[i+63:i] := SignExtend64(a[k+15:k]) ENDFOR +dst[MAX:256] := 0 - -
emmintrin.h
+ +
immintrin.h
 - - Floating Point + Integer - SSE2 + AVX2 Convert - - Convert packed 32-bit integers in "a" to packed double-precision (64-bit) floating-point elements, and store the results in "dst". + + + Sign extend packed 32-bit integers in "a" to packed 64-bit integers, and store the results in "dst". -FOR j := 0 to 1 - i := j*32 - m := j*64 - dst[m+63:m] := Convert_Int32_To_FP64(a[i+31:i]) +FOR j:= 0 to 3 + i := 64*j + k := 32*j + dst[i+63:i] := SignExtend64(a[k+31:k]) ENDFOR +dst[MAX:256] := 0 - -
emmintrin.h
+ +
immintrin.h
 - - Floating Point - SSE2 + + Integer + AVX2 Convert - - - Convert the 32-bit integer "b" to a double-precision (64-bit) floating-point element, store the result in the lower element of "dst", and copy the upper element from "a" to the upper element of "dst". + + + Sign extend packed 8-bit integers in "a" to packed 16-bit integers, and store the results in "dst". -dst[63:0] := Convert_Int32_To_FP64(b[31:0]) -dst[127:64] := a[127:64] -dst[MAX:128] := 0 +FOR j := 0 to 15 + i := j*8 + l := j*16 + dst[l+15:l] := SignExtend16(a[i+7:i]) +ENDFOR +dst[MAX:256] := 0 - -
emmintrin.h
+ +
immintrin.h
 - - Floating Point + Integer - SSE2 + AVX2 Convert - - - Convert the 64-bit integer "b" to a double-precision (64-bit) floating-point element, store the result in the lower element of "dst", and copy the upper element from "a" to the upper element of "dst". + + + Sign extend packed 8-bit integers in "a" to packed 32-bit integers, and store the results in "dst". -dst[63:0] := Convert_Int64_To_FP64(b[63:0]) -dst[127:64] := a[127:64] -dst[MAX:128] := 0 +FOR j := 0 to 7 + i := 32*j + k := 8*j + dst[i+31:i] := SignExtend32(a[k+7:k]) +ENDFOR +dst[MAX:256] := 0 - -
emmintrin.h
+ +
immintrin.h
 - - Floating Point + Integer - SSE2 + AVX2 Convert - - - Convert the 64-bit integer "b" to a double-precision (64-bit) floating-point element, store the result in the lower element of "dst", and copy the upper element from "a" to the upper element of "dst". + + + Sign extend packed 8-bit integers in the low 8 bytes of "a" to packed 64-bit integers, and store the results in "dst". -dst[63:0] := Convert_Int64_To_FP64(b[63:0]) -dst[127:64] := a[127:64] -dst[MAX:128] := 0 +FOR j := 0 to 3 + i := 64*j + k := 8*j + dst[i+63:i] := SignExtend64(a[k+7:k]) +ENDFOR +dst[MAX:256] := 0 - -
emmintrin.h
+ +
immintrin.h
 - - Floating Point + Integer - SSE2 + AVX2 Convert - - Convert packed 32-bit integers in "a" to packed single-precision (32-bit) floating-point elements, and store the results in "dst". + + + Zero extend packed unsigned 16-bit integers in "a" to packed 32-bit integers, and store the results in "dst". -FOR j := 0 to 3 +FOR j := 0 to 7 i := 32*j - dst[i+31:i] := Convert_Int32_To_FP32(a[i+31:i]) + k := 16*j + dst[i+31:i] := ZeroExtend32(a[k+15:k]) ENDFOR +dst[MAX:256] := 0 - -
emmintrin.h
+ +
immintrin.h
 - - Floating Point - SSE2 + + Integer + AVX2 Convert - - Convert packed 32-bit integers in "a" to packed double-precision (64-bit) floating-point elements, and store the results in "dst". + + + Zero extend packed unsigned 16-bit integers in "a" to packed 64-bit integers, and store the results in "dst". -FOR j := 0 to 1 - i := j*32 - m := j*64 - dst[m+63:m] := Convert_Int32_To_FP64(a[i+31:i]) +FOR j:= 0 to 3 + i := 64*j + k := 16*j + dst[i+63:i] := ZeroExtend64(a[k+15:k]) ENDFOR +dst[MAX:256] := 0 - -
emmintrin.h
+ +
immintrin.h
 - + Integer - SSE2 + AVX2 Convert - - Copy 32-bit integer "a" to the lower elements of "dst", and zero the upper elements of "dst". + + + Zero extend packed unsigned 32-bit integers in "a" to packed 64-bit integers, and store the results in "dst". -dst[31:0] := a[31:0] -dst[127:32] := 0 +FOR j:= 0 to 3 + i := 64*j + k := 32*j + dst[i+63:i] := ZeroExtend64(a[k+31:k]) +ENDFOR +dst[MAX:256] := 0 - -
emmintrin.h
+ +
immintrin.h
 - + Integer - SSE2 + AVX2 Convert - - Copy 64-bit integer "a" to the lower element of "dst", and zero the upper element. + + + Zero extend packed unsigned 8-bit integers in "a" to packed 16-bit integers, and store the results in "dst". -dst[63:0] := a[63:0] -dst[127:64] := 0 +FOR j := 0 to 15 + i := j*8 + l := j*16 + dst[l+15:l] := ZeroExtend16(a[i+7:i]) +ENDFOR +dst[MAX:256] := 0 - -
emmintrin.h
+ +
immintrin.h
 - + Integer - SSE2 + AVX2 Convert - - Copy 64-bit integer "a" to the lower element of "dst", and zero the upper element. + + + Zero extend packed unsigned 8-bit integers in "a" to packed 32-bit integers, and store the results in "dst". -dst[63:0] := a[63:0] -dst[127:64] := 0 +FOR j := 0 to 7 + i := 32*j + k := 8*j + dst[i+31:i] := ZeroExtend32(a[k+7:k]) +ENDFOR +dst[MAX:256] := 0 - -
emmintrin.h
+ +
immintrin.h
 - + Integer - SSE2 + AVX2 Convert - - Copy the lower 32-bit integer in "a" to "dst". + + + Zero extend packed unsigned 8-bit integers in the low 8 byte sof "a" to packed 64-bit integers, and store the results in "dst". -dst[31:0] := a[31:0] +FOR j := 0 to 3 + i := 64*j + k := 8*j + dst[i+63:i] := ZeroExtend64(a[k+7:k]) +ENDFOR +dst[MAX:256] := 0 - -
emmintrin.h
+ +
immintrin.h
 - + Integer - SSE2 - Convert - - Copy the lower 64-bit integer in "a" to "dst". + AVX2 + Swizzle + + + + Extract 128 bits (composed of integer data) from "a", selected with "imm8", and store the result in "dst". -dst[63:0] := a[63:0] +CASE imm8[0] OF +0: dst[127:0] := a[127:0] +1: dst[127:0] := a[255:128] +ESAC +dst[MAX:128] := 0 - -
emmintrin.h
+ +
immintrin.h
 - + Integer - SSE2 - Convert - - Copy the lower 64-bit integer in "a" to "dst". + AVX2 + Arithmetic + + + + Horizontally add adjacent pairs of 16-bit integers in "a" and "b", and pack the signed 16-bit results in "dst". -dst[63:0] := a[63:0] +dst[15:0] := a[31:16] + a[15:0] +dst[31:16] := a[63:48] + a[47:32] +dst[47:32] := a[95:80] + a[79:64] +dst[63:48] := a[127:112] + a[111:96] +dst[79:64] := b[31:16] + b[15:0] +dst[95:80] := b[63:48] + b[47:32] +dst[111:96] := b[95:80] + b[79:64] +dst[127:112] := b[127:112] + b[111:96] +dst[143:128] := a[159:144] + a[143:128] +dst[159:144] := a[191:176] + a[175:160] +dst[175:160] := a[223:208] + a[207:192] +dst[191:176] := a[255:240] + a[239:224] +dst[207:192] := b[159:144] + b[143:128] +dst[223:208] := b[191:176] + b[175:160] +dst[239:224] := b[223:208] + b[207:192] +dst[255:240] := b[255:240] + b[239:224] +dst[MAX:256] := 0 - -
emmintrin.h
+ +
immintrin.h
 - + Integer - SSE2 - Set - - - Set packed 64-bit integers in "dst" with the supplied values. + AVX2 + Arithmetic + + + + Horizontally add adjacent pairs of 32-bit integers in "a" and "b", and pack the signed 32-bit results in "dst". -dst[63:0] := e0 -dst[127:64] := e1 +dst[31:0] := a[63:32] + a[31:0] +dst[63:32] := a[127:96] + a[95:64] +dst[95:64] := b[63:32] + b[31:0] +dst[127:96] := b[127:96] + b[95:64] +dst[159:128] := a[191:160] + a[159:128] +dst[191:160] := a[255:224] + a[223:192] +dst[223:192] := b[191:160] + b[159:128] +dst[255:224] := b[255:224] + b[223:192] +dst[MAX:256] := 0 -
emmintrin.h
+ +
immintrin.h
 - + Integer - SSE2 - Set - - - Set packed 64-bit integers in "dst" with the supplied values. + AVX2 + Arithmetic + + + + Horizontally add adjacent pairs of signed 16-bit integers in "a" and "b" using saturation, and pack the signed 16-bit results in "dst". -dst[63:0] := e0 -dst[127:64] := e1 +dst[15:0] := Saturate16(a[31:16] + a[15:0]) +dst[31:16] := Saturate16(a[63:48] + a[47:32]) +dst[47:32] := Saturate16(a[95:80] + a[79:64]) +dst[63:48] := Saturate16(a[127:112] + a[111:96]) +dst[79:64] := Saturate16(b[31:16] + b[15:0]) +dst[95:80] := Saturate16(b[63:48] + b[47:32]) +dst[111:96] := Saturate16(b[95:80] + b[79:64]) +dst[127:112] := Saturate16(b[127:112] + b[111:96]) +dst[143:128] := Saturate16(a[159:144] + a[143:128]) +dst[159:144] := Saturate16(a[191:176] + a[175:160]) +dst[175:160] := Saturate16(a[223:208] + a[207:192]) +dst[191:176] := Saturate16(a[255:240] + a[239:224]) +dst[207:192] := Saturate16(b[159:144] + b[143:128]) +dst[223:208] := Saturate16(b[191:176] + b[175:160]) +dst[239:224] := Saturate16(b[223:208] + b[207:192]) +dst[255:240] := Saturate16(b[255:240] + b[239:224]) +dst[MAX:256] := 0 -
emmintrin.h
+ +
immintrin.h
 - + Integer - SSE2 - Set - - - - - Set packed 32-bit integers in "dst" with the supplied values. + AVX2 + Arithmetic + + + + Horizontally subtract adjacent pairs of 16-bit integers in "a" and "b", and pack the signed 16-bit results in "dst". -dst[31:0] := e0 -dst[63:32] := e1 -dst[95:64] := e2 -dst[127:96] := e3 +dst[15:0] := a[15:0] - a[31:16] +dst[31:16] := a[47:32] - a[63:48] +dst[47:32] := a[79:64] - a[95:80] +dst[63:48] := a[111:96] - a[127:112] +dst[79:64] := b[15:0] - b[31:16] +dst[95:80] := b[47:32] - b[63:48] +dst[111:96] := b[79:64] - b[95:80] +dst[127:112] := b[111:96] - b[127:112] +dst[143:128] := a[143:128] - a[159:144] +dst[159:144] := a[175:160] - a[191:176] +dst[175:160] := a[207:192] - a[223:208] +dst[191:176] := a[239:224] - a[255:240] +dst[207:192] := b[143:128] - b[159:144] +dst[223:208] := b[175:160] - b[191:176] +dst[239:224] := b[207:192] - b[223:208] +dst[255:240] := b[239:224] - b[255:240] +dst[MAX:256] := 0 -
emmintrin.h
+ +
immintrin.h
 - + Integer - SSE2 - Set - - - - - - - - - Set packed 16-bit integers in "dst" with the supplied values. + AVX2 + Arithmetic + + + + Horizontally subtract adjacent pairs of 32-bit integers in "a" and "b", and pack the signed 32-bit results in "dst". -dst[15:0] := e0 -dst[31:16] := e1 -dst[47:32] := e2 -dst[63:48] := e3 -dst[79:64] := e4 -dst[95:80] := e5 -dst[111:96] := e6 -dst[127:112] := e7 +dst[31:0] := a[31:0] - a[63:32] +dst[63:32] := a[95:64] - a[127:96] +dst[95:64] := b[31:0] - b[63:32] +dst[127:96] := b[95:64] - b[127:96] +dst[159:128] := a[159:128] - a[191:160] +dst[191:160] := a[223:192] - a[255:224] +dst[223:192] := b[159:128] - b[191:160] +dst[255:224] := b[223:192] - b[255:224] +dst[MAX:256] := 0 -
emmintrin.h
+ +
immintrin.h
 - + Integer - SSE2 - Set - - - - - - - - - - - - - - - - - Set packed 8-bit integers in "dst" with the supplied values in reverse order. + AVX2 + Arithmetic + + + + Horizontally subtract adjacent pairs of signed 16-bit integers in "a" and "b" using saturation, and pack the signed 16-bit results in "dst". -dst[7:0] := e0 -dst[15:8] := e1 -dst[23:16] := e2 -dst[31:24] := e3 -dst[39:32] := e4 -dst[47:40] := e5 -dst[55:48] := e6 -dst[63:56] := e7 -dst[71:64] := e8 -dst[79:72] := e9 -dst[87:80] := e10 -dst[95:88] := e11 -dst[103:96] := e12 -dst[111:104] := e13 -dst[119:112] := e14 -dst[127:120] := e15 +dst[15:0] := Saturate16(a[15:0] - a[31:16]) +dst[31:16] := Saturate16(a[47:32] - a[63:48]) +dst[47:32] := Saturate16(a[79:64] - a[95:80]) +dst[63:48] := Saturate16(a[111:96] - a[127:112]) +dst[79:64] := Saturate16(b[15:0] - b[31:16]) +dst[95:80] := Saturate16(b[47:32] - b[63:48]) +dst[111:96] := Saturate16(b[79:64] - b[95:80]) +dst[127:112] := Saturate16(b[111:96] - b[127:112]) +dst[143:128] := Saturate16(a[143:128] - a[159:144]) +dst[159:144] := Saturate16(a[175:160] - a[191:176]) +dst[175:160] := Saturate16(a[207:192] - a[223:208]) +dst[191:176] := Saturate16(a[239:224] - a[255:240]) +dst[207:192] := Saturate16(b[143:128] - b[159:144]) +dst[223:208] := Saturate16(b[175:160] - b[191:176]) +dst[239:224] := Saturate16(b[207:192] - b[223:208]) +dst[255:240] := Saturate16(b[239:224] - b[255:240]) +dst[MAX:256] := 0 -
emmintrin.h
+ +
immintrin.h
 - - Integer - SSE2 - Set - - Broadcast 64-bit integer "a" to all elements of "dst". + + Floating Point + AVX2 + Load + + + + + Gather double-precision (64-bit) floating-point elements from memory using 32-bit indices. 64-bit elements are loaded from addresses starting at "base_addr" and offset by each 32-bit element in "vindex" (each index is scaled by the factor in "scale"). Gathered elements are merged into "dst". "scale" should be 1, 2, 4 or 8. FOR j := 0 to 1 i := j*64 - dst[i+63:i] := a[63:0] + m := j*32 + addr := base_addr + SignExtend64(vindex[m+31:m]) * ZeroExtend64(scale) * 8 + dst[i+63:i] := MEM[addr+63:addr] ENDFOR +dst[MAX:128] := 0 -
emmintrin.h
+ +
immintrin.h
 - - Integer - SSE2 - Set - - Broadcast 64-bit integer "a" to all elements of "dst". This intrinsic may generate the "vpbroadcastq". + + Floating Point + AVX2 + Load + + + + + Gather double-precision (64-bit) floating-point elements from memory using 32-bit indices. 64-bit elements are loaded from addresses starting at "base_addr" and offset by each 32-bit element in "vindex" (each index is scaled by the factor in "scale"). Gathered elements are merged into "dst". "scale" should be 1, 2, 4 or 8. -FOR j := 0 to 1 +FOR j := 0 to 3 i := j*64 - dst[i+63:i] := a[63:0] + m := j*32 + addr := base_addr + SignExtend64(vindex[m+31:m]) * ZeroExtend64(scale) * 8 + dst[i+63:i] := MEM[addr+63:addr] ENDFOR +dst[MAX:256] := 0 -
emmintrin.h
+ +
immintrin.h
 - - Integer - SSE2 - Set - - Broadcast 32-bit integer "a" to all elements of "dst". This intrinsic may generate "vpbroadcastd". + + Floating Point + AVX2 + Load + + + + + Gather single-precision (32-bit) floating-point elements from memory using 32-bit indices. 32-bit elements are loaded from addresses starting at "base_addr" and offset by each 32-bit element in "vindex" (each index is scaled by the factor in "scale"). Gathered elements are merged into "dst". "scale" should be 1, 2, 4 or 8. FOR j := 0 to 3 i := j*32 - dst[i+31:i] := a[31:0] + m := j*32 + addr := base_addr + SignExtend64(vindex[m+31:m]) * ZeroExtend64(scale) * 8 + dst[i+31:i] := MEM[addr+31:addr] ENDFOR +dst[MAX:128] := 0 -
emmintrin.h
+ +
immintrin.h
 - - Integer - SSE2 - Set - - Broadcast 16-bit integer "a" to all all elements of "dst". This intrinsic may generate "vpbroadcastw". + + Floating Point + AVX2 + Load + + + + + Gather single-precision (32-bit) floating-point elements from memory using 32-bit indices. 32-bit elements are loaded from addresses starting at "base_addr" and offset by each 32-bit element in "vindex" (each index is scaled by the factor in "scale"). Gathered elements are merged into "dst". "scale" should be 1, 2, 4 or 8. FOR j := 0 to 7 - i := j*16 - dst[i+15:i] := a[15:0] + i := j*32 + m := j*32 + addr := base_addr + SignExtend64(vindex[m+31:m]) * ZeroExtend64(scale) * 8 + dst[i+31:i] := MEM[addr+31:addr] ENDFOR +dst[MAX:256] := 0 -
emmintrin.h
+ +
immintrin.h
 - + Integer - SSE2 - Set - - Broadcast 8-bit integer "a" to all elements of "dst". This intrinsic may generate "vpbroadcastb". + AVX2 + Load + + + + + Gather 32-bit integers from memory using 32-bit indices. 32-bit elements are loaded from addresses starting at "base_addr" and offset by each 32-bit element in "vindex" (each index is scaled by the factor in "scale"). Gathered elements are merged into "dst". "scale" should be 1, 2, 4 or 8. -FOR j := 0 to 15 - i := j*8 - dst[i+7:i] := a[7:0] +FOR j := 0 to 3 + i := j*32 + m := j*32 + addr := base_addr + SignExtend64(vindex[m+31:m]) * ZeroExtend64(scale) * 8 + dst[i+31:i] := MEM[addr+31:addr] ENDFOR +dst[MAX:128] := 0 -
emmintrin.h
+ +
immintrin.h
 - + Integer - SSE2 - Set - - - Set packed 64-bit integers in "dst" with the supplied values in reverse order. + AVX2 + Load + + + + + Gather 32-bit integers from memory using 32-bit indices. 32-bit elements are loaded from addresses starting at "base_addr" and offset by each 32-bit element in "vindex" (each index is scaled by the factor in "scale"). Gathered elements are merged into "dst". "scale" should be 1, 2, 4 or 8. -dst[63:0] := e1 -dst[127:64] := e0 +FOR j := 0 to 7 + i := j*32 + m := j*32 + addr := base_addr + SignExtend64(vindex[m+31:m]) * ZeroExtend64(scale) * 8 + dst[i+31:i] := MEM[addr+31:addr] +ENDFOR +dst[MAX:256] := 0 -
emmintrin.h
+ +
immintrin.h
 - + Integer - SSE2 - Set - - - - - Set packed 32-bit integers in "dst" with the supplied values in reverse order. + AVX2 + Load + + + + + Gather 64-bit integers from memory using 32-bit indices. 64-bit elements are loaded from addresses starting at "base_addr" and offset by each 32-bit element in "vindex" (each index is scaled by the factor in "scale"). Gathered elements are merged into "dst". "scale" should be 1, 2, 4 or 8. -dst[31:0] := e3 -dst[63:32] := e2 -dst[95:64] := e1 -dst[127:96] := e0 +FOR j := 0 to 1 + i := j*64 + m := j*32 + addr := base_addr + SignExtend64(vindex[m+31:m]) * ZeroExtend64(scale) * 8 + dst[i+63:i] := MEM[addr+63:addr] +ENDFOR +dst[MAX:128] := 0 -
emmintrin.h
+ +
immintrin.h
 - + Integer - SSE2 - Set - - - - - - - - - Set packed 16-bit integers in "dst" with the supplied values in reverse order. + AVX2 + Load + + + + + Gather 64-bit integers from memory using 32-bit indices. 64-bit elements are loaded from addresses starting at "base_addr" and offset by each 32-bit element in "vindex" (each index is scaled by the factor in "scale"). Gathered elements are merged into "dst". "scale" should be 1, 2, 4 or 8. -dst[15:0] := e7 -dst[31:16] := e6 -dst[47:32] := e5 -dst[63:48] := e4 -dst[79:64] := e3 -dst[95:80] := e2 -dst[111:96] := e1 -dst[127:112] := e0 +FOR j := 0 to 3 + i := j*64 + m := j*32 + addr := base_addr + SignExtend64(vindex[m+31:m]) * ZeroExtend64(scale) * 8 + dst[i+63:i] := MEM[addr+63:addr] +ENDFOR +dst[MAX:256] := 0 -
emmintrin.h
+ +
immintrin.h
 - - Integer - SSE2 - Set - - - - - - - - - - - - - - - - - Set packed 8-bit integers in "dst" with the supplied values in reverse order. + + Floating Point + AVX2 + Load + + + + + Gather double-precision (64-bit) floating-point elements from memory using 64-bit indices. 64-bit elements are loaded from addresses starting at "base_addr" and offset by each 64-bit element in "vindex" (each index is scaled by the factor in "scale"). Gathered elements are merged into "dst". "scale" should be 1, 2, 4 or 8. -dst[7:0] := e15 -dst[15:8] := e14 -dst[23:16] := e13 -dst[31:24] := e12 -dst[39:32] := e11 -dst[47:40] := e10 -dst[55:48] := e9 -dst[63:56] := e8 -dst[71:64] := e7 -dst[79:72] := e6 -dst[87:80] := e5 -dst[95:88] := e4 -dst[103:96] := e3 -dst[111:104] := e2 -dst[119:112] := e1 -dst[127:120] := e0 +FOR j := 0 to 1 + i := j*64 + m := j*64 + addr := base_addr + vindex[m+63:m] * ZeroExtend64(scale) * 8 + dst[i+63:i] := MEM[addr+63:addr] +ENDFOR +dst[MAX:128] := 0 -
emmintrin.h
+ +
immintrin.h
 - - Integer - SSE2 - Set - Return vector of type __m128i with all elements set to zero. + + Floating Point + AVX2 + Load + + + + + Gather double-precision (64-bit) floating-point elements from memory using 64-bit indices. 64-bit elements are loaded from addresses starting at "base_addr" and offset by each 64-bit element in "vindex" (each index is scaled by the factor in "scale"). Gathered elements are merged into "dst". "scale" should be 1, 2, 4 or 8. -dst[MAX:0] := 0 +FOR j := 0 to 3 + i := j*64 + m := j*64 + addr := base_addr + vindex[m+63:m] * ZeroExtend64(scale) * 8 + dst[i+63:i] := MEM[addr+63:addr] +ENDFOR +dst[MAX:256] := 0 - -
emmintrin.h
+ +
immintrin.h
 - - Integer - SSE2 + + Floating Point + AVX2 Load - - Load 64-bit integer from memory into the first element of "dst". + + + + + Gather single-precision (32-bit) floating-point elements from memory using 64-bit indices. 32-bit elements are loaded from addresses starting at "base_addr" and offset by each 64-bit element in "vindex" (each index is scaled by the factor in "scale"). Gathered elements are merged into "dst". "scale" should be 1, 2, 4 or 8. -dst[63:0] := MEM[mem_addr+63:mem_addr] +FOR j := 0 to 1 + i := j*32 + m := j*64 + addr := base_addr + vindex[m+63:m] * ZeroExtend64(scale) * 8 + dst[i+31:i] := MEM[addr+31:addr] +ENDFOR dst[MAX:64] := 0 - -
emmintrin.h
+ +
immintrin.h
 - - Integer - SSE2 + + Floating Point + AVX2 Load - - Load 128-bits of integer data from memory into "dst". - "mem_addr" must be aligned on a 16-byte boundary or a general-protection exception may be generated. + + + + + Gather single-precision (32-bit) floating-point elements from memory using 64-bit indices. 32-bit elements are loaded from addresses starting at "base_addr" and offset by each 64-bit element in "vindex" (each index is scaled by the factor in "scale"). Gathered elements are merged into "dst". "scale" should be 1, 2, 4 or 8. -dst[127:0] := MEM[mem_addr+127:mem_addr] +FOR j := 0 to 3 + i := j*32 + m := j*64 + addr := base_addr + vindex[m+63:m] * ZeroExtend64(scale) * 8 + dst[i+31:i] := MEM[addr+31:addr] +ENDFOR +dst[MAX:128] := 0 - -
emmintrin.h
+ +
immintrin.h
 - + Integer - SSE2 + AVX2 Load - - Load 128-bits of integer data from memory into "dst". - "mem_addr" does not need to be aligned on any particular boundary. - - -dst[127:0] := MEM[mem_addr+127:mem_addr] - - -
emmintrin.h
- - - Integer - SSE2 - Store - - - - Conditionally store 8-bit integer elements from "a" into memory using "mask" (elements are not stored when the highest bit is not set in the corresponding element) and a non-temporal memory hint. "mem_addr" does not need to be aligned on any particular boundary. + + + + + Gather 32-bit integers from memory using 64-bit indices. 32-bit elements are loaded from addresses starting at "base_addr" and offset by each 64-bit element in "vindex" (each index is scaled by the factor in "scale"). Gathered elements are merged into "dst". "scale" should be 1, 2, 4 or 8. -FOR j := 0 to 15 - i := j*8 - IF mask[i+7] - MEM[mem_addr+i+7:mem_addr+i] := a[i+7:i] - FI +FOR j := 0 to 1 + i := j*32 + m := j*64 + addr := base_addr + vindex[m+63:m] * ZeroExtend64(scale) * 8 + dst[i+31:i] := MEM[addr+31:addr] ENDFOR +dst[MAX:64] := 0 - -
emmintrin.h
+ +
immintrin.h
 - + Integer - SSE2 - Store - - - Store 128-bits of integer data from "a" into memory. - "mem_addr" must be aligned on a 16-byte boundary or a general-protection exception may be generated. + AVX2 + Load + + + + + Gather 32-bit integers from memory using 64-bit indices. 32-bit elements are loaded from addresses starting at "base_addr" and offset by each 64-bit element in "vindex" (each index is scaled by the factor in "scale"). Gathered elements are merged into "dst". "scale" should be 1, 2, 4 or 8. -MEM[mem_addr+127:mem_addr] := a[127:0] +FOR j := 0 to 3 + i := j*32 + m := j*64 + addr := base_addr + vindex[m+63:m] * ZeroExtend64(scale) * 8 + dst[i+31:i] := MEM[addr+31:addr] +ENDFOR +dst[MAX:128] := 0 - -
emmintrin.h
+ +
immintrin.h
 - + Integer - SSE2 - Store - - - Store 128-bits of integer data from "a" into memory. - "mem_addr" does not need to be aligned on any particular boundary. - + AVX2 + Load + + + + + Gather 64-bit integers from memory using 64-bit indices. 64-bit elements are loaded from addresses starting at "base_addr" and offset by each 64-bit element in "vindex" (each index is scaled by the factor in "scale"). Gathered elements are merged into "dst". "scale" should be 1, 2, 4 or 8. -MEM[mem_addr+127:mem_addr] := a[127:0] +FOR j := 0 to 1 + i := j*64 + m := j*64 + addr := base_addr + vindex[m+63:m] * ZeroExtend64(scale) * 8 + dst[i+63:i] := MEM[addr+63:addr] +ENDFOR +dst[MAX:128] := 0 - -
emmintrin.h
+ +
immintrin.h
 - + Integer - SSE2 - Store - - - Store 64-bit integer from the first element of "a" into memory. + AVX2 + Load + + + + + Gather 64-bit integers from memory using 64-bit indices. 64-bit elements are loaded from addresses starting at "base_addr" and offset by each 64-bit element in "vindex" (each index is scaled by the factor in "scale"). Gathered elements are merged into "dst". "scale" should be 1, 2, 4 or 8. -MEM[mem_addr+63:mem_addr] := a[63:0] +FOR j := 0 to 3 + i := j*64 + m := j*64 + addr := base_addr + vindex[m+63:m] * ZeroExtend64(scale) * 8 + dst[i+63:i] := MEM[addr+63:addr] +ENDFOR +dst[MAX:256] := 0 - -
emmintrin.h
+ +
immintrin.h
 - + Integer - SSE2 - Store - - - Store 128-bits of integer data from "a" into memory using a non-temporal memory hint. - "mem_addr" must be aligned on a 16-byte boundary or a general-protection exception may be generated. + AVX2 + Swizzle + + + + + Copy "a" to "dst", then insert 128 bits (composed of integer data) from "b" into "dst" at the location specified by "imm8". -MEM[mem_addr+127:mem_addr] := a[127:0] +dst[255:0] := a[255:0] +CASE (imm8[0]) OF +0: dst[127:0] := b[127:0] +1: dst[255:128] := b[127:0] +ESAC +dst[MAX:256] := 0 - -
emmintrin.h
+ +
immintrin.h
 - + Integer - SSE2 - Store - - - Store 32-bit integer "a" into memory using a non-temporal hint to minimize cache pollution. If the cache line containing address "mem_addr" is already in the cache, the cache will be updated. + AVX2 + Arithmetic + + + + Multiply packed signed 16-bit integers in "a" and "b", producing intermediate signed 32-bit integers. Horizontally add adjacent pairs of intermediate 32-bit integers, and pack the results in "dst". -MEM[mem_addr+31:mem_addr] := a[31:0] +FOR j := 0 to 7 + i := j*32 + dst[i+31:i] := SignExtend32(a[i+31:i+16]*b[i+31:i+16]) + SignExtend32(a[i+15:i]*b[i+15:i]) +ENDFOR +dst[MAX:256] := 0 - -
emmintrin.h
+ +
immintrin.h
 - + Integer - SSE2 - Store - - - Store 64-bit integer "a" into memory using a non-temporal hint to minimize cache pollution. If the cache line containing address "mem_addr" is already in the cache, the cache will be updated. + AVX2 + Arithmetic + + + + Vertically multiply each unsigned 8-bit integer from "a" with the corresponding signed 8-bit integer from "b", producing intermediate signed 16-bit integers. Horizontally add adjacent pairs of intermediate signed 16-bit integers, and pack the saturated results in "dst". -MEM[mem_addr+63:mem_addr] := a[63:0] +FOR j := 0 to 15 + i := j*16 + dst[i+15:i] := Saturate16( a[i+15:i+8]*b[i+15:i+8] + a[i+7:i]*b[i+7:i] ) +ENDFOR +dst[MAX:256] := 0 - -
emmintrin.h
+ +
immintrin.h
 - - Integer - SSE2 - Miscellaneous - - Copy the lower 64-bit integer in "a" to "dst". + + Floating Point + AVX2 + Load + + + + + + + Gather double-precision (64-bit) floating-point elements from memory using 32-bit indices. 64-bit elements are loaded from addresses starting at "base_addr" and offset by each 32-bit element in "vindex" (each index is scaled by the factor in "scale"). Gathered elements are merged into "dst" using "mask" (elements are copied from "src" when the highest bit is not set in the corresponding element). "scale" should be 1, 2, 4 or 8. -dst[63:0] := a[63:0] +FOR j := 0 to 1 + i := j*64 + m := j*32 + IF mask[i+63] + addr := base_addr + SignExtend64(vindex[m+31:m]) * ZeroExtend64(scale) * 8 + dst[i+63:i] := MEM[addr+63:addr] + ELSE + dst[i+63:i] := src[i+63:i] + FI +ENDFOR +mask[MAX:128] := 0 +dst[MAX:128] := 0 - -
emmintrin.h
+ +
immintrin.h
 - - Integer - SSE2 - Move - - Copy the 64-bit integer "a" to the lower element of "dst", and zero the upper element. + + Floating Point + AVX2 + Load + + + + + + + Gather double-precision (64-bit) floating-point elements from memory using 32-bit indices. 64-bit elements are loaded from addresses starting at "base_addr" and offset by each 32-bit element in "vindex" (each index is scaled by the factor in "scale"). Gathered elements are merged into "dst" using "mask" (elements are copied from "src" when the highest bit is not set in the corresponding element). "scale" should be 1, 2, 4 or 8. -dst[63:0] := a[63:0] -dst[127:64] := 0 +FOR j := 0 to 3 + i := j*64 + m := j*32 + IF mask[i+63] + addr := base_addr + SignExtend64(vindex[m+31:m]) * ZeroExtend64(scale) * 8 + dst[i+63:i] := MEM[addr+63:addr] + ELSE + dst[i+63:i] := src[i+63:i] + FI +ENDFOR +mask[MAX:256] := 0 +dst[MAX:256] := 0 - -
emmintrin.h
+ +
immintrin.h
 - - Integer - SSE2 - Move - - Copy the lower 64-bit integer in "a" to the lower element of "dst", and zero the upper element. + + Floating Point + AVX2 + Load + + + + + + + Gather single-precision (32-bit) floating-point elements from memory using 32-bit indices. 32-bit elements are loaded from addresses starting at "base_addr" and offset by each 32-bit element in "vindex" (each index is scaled by the factor in "scale"). Gathered elements are merged into "dst" using "mask" (elements are copied from "src" when the highest bit is not set in the corresponding element). "scale" should be 1, 2, 4 or 8. -dst[63:0] := a[63:0] -dst[127:64] := 0 +FOR j := 0 to 3 + i := j*32 + m := j*32 + IF mask[i+31] + addr := base_addr + SignExtend64(vindex[m+31:m]) * ZeroExtend64(scale) * 8 + dst[i+31:i] := MEM[addr+31:addr] + ELSE + dst[i+31:i] := src[i+31:i] + FI +ENDFOR +mask[MAX:128] := 0 +dst[MAX:128] := 0 - -
emmintrin.h
- - - Integer - SSE2 - Miscellaneous - - - Convert packed 16-bit integers from "a" and "b" to packed 8-bit integers using signed saturation, and store the results in "dst". - - -dst[7:0] := Saturate_Int16_To_Int8 (a[15:0]) -dst[15:8] := Saturate_Int16_To_Int8 (a[31:16]) -dst[23:16] := Saturate_Int16_To_Int8 (a[47:32]) -dst[31:24] := Saturate_Int16_To_Int8 (a[63:48]) -dst[39:32] := Saturate_Int16_To_Int8 (a[79:64]) -dst[47:40] := Saturate_Int16_To_Int8 (a[95:80]) -dst[55:48] := Saturate_Int16_To_Int8 (a[111:96]) -dst[63:56] := Saturate_Int16_To_Int8 (a[127:112]) -dst[71:64] := Saturate_Int16_To_Int8 (b[15:0]) -dst[79:72] := Saturate_Int16_To_Int8 (b[31:16]) -dst[87:80] := Saturate_Int16_To_Int8 (b[47:32]) -dst[95:88] := Saturate_Int16_To_Int8 (b[63:48]) -dst[103:96] := Saturate_Int16_To_Int8 (b[79:64]) -dst[111:104] := Saturate_Int16_To_Int8 (b[95:80]) -dst[119:112] := Saturate_Int16_To_Int8 (b[111:96]) -dst[127:120] := Saturate_Int16_To_Int8 (b[127:112]) - - -
emmintrin.h
+ +
immintrin.h
 - - Integer - SSE2 - Miscellaneous - - - Convert packed 32-bit integers from "a" and "b" to packed 16-bit integers using signed saturation, and store the results in "dst". + + Floating Point + AVX2 + Load + + + + + + + Gather single-precision (32-bit) floating-point elements from memory using 32-bit indices. 32-bit elements are loaded from addresses starting at "base_addr" and offset by each 32-bit element in "vindex" (each index is scaled by the factor in "scale"). Gathered elements are merged into "dst" using "mask" (elements are copied from "src" when the highest bit is not set in the corresponding element). "scale" should be 1, 2, 4 or 8. -dst[15:0] := Saturate_Int32_To_Int16 (a[31:0]) -dst[31:16] := Saturate_Int32_To_Int16 (a[63:32]) -dst[47:32] := Saturate_Int32_To_Int16 (a[95:64]) -dst[63:48] := Saturate_Int32_To_Int16 (a[127:96]) -dst[79:64] := Saturate_Int32_To_Int16 (b[31:0]) -dst[95:80] := Saturate_Int32_To_Int16 (b[63:32]) -dst[111:96] := Saturate_Int32_To_Int16 (b[95:64]) -dst[127:112] := Saturate_Int32_To_Int16 (b[127:96]) +FOR j := 0 to 7 + i := j*32 + m := j*32 + IF mask[i+31] + addr := base_addr + SignExtend64(vindex[m+31:m]) * ZeroExtend64(scale) * 8 + dst[i+31:i] := MEM[addr+31:addr] + ELSE + dst[i+31:i] := src[i+31:i] + FI +ENDFOR +mask[MAX:256] := 0 +dst[MAX:256] := 0 - -
emmintrin.h
- - - Integer - SSE2 - Miscellaneous - - - Convert packed 16-bit integers from "a" and "b" to packed 8-bit integers using unsigned saturation, and store the results in "dst". - -dst[7:0] := Saturate_Int16_To_UnsignedInt8 (a[15:0]) -dst[15:8] := Saturate_Int16_To_UnsignedInt8 (a[31:16]) -dst[23:16] := Saturate_Int16_To_UnsignedInt8 (a[47:32]) -dst[31:24] := Saturate_Int16_To_UnsignedInt8 (a[63:48]) -dst[39:32] := Saturate_Int16_To_UnsignedInt8 (a[79:64]) -dst[47:40] := Saturate_Int16_To_UnsignedInt8 (a[95:80]) -dst[55:48] := Saturate_Int16_To_UnsignedInt8 (a[111:96]) -dst[63:56] := Saturate_Int16_To_UnsignedInt8 (a[127:112]) -dst[71:64] := Saturate_Int16_To_UnsignedInt8 (b[15:0]) -dst[79:72] := Saturate_Int16_To_UnsignedInt8 (b[31:16]) -dst[87:80] := Saturate_Int16_To_UnsignedInt8 (b[47:32]) -dst[95:88] := Saturate_Int16_To_UnsignedInt8 (b[63:48]) -dst[103:96] := Saturate_Int16_To_UnsignedInt8 (b[79:64]) -dst[111:104] := Saturate_Int16_To_UnsignedInt8 (b[95:80]) -dst[119:112] := Saturate_Int16_To_UnsignedInt8 (b[111:96]) -dst[127:120] := Saturate_Int16_To_UnsignedInt8 (b[127:112]) - - -
emmintrin.h
+ +
immintrin.h
 - + Integer - SSE2 - Swizzle - - - Extract a 16-bit integer from "a", selected with "imm8", and store the result in the lower element of "dst". + AVX2 + Load + + + + + + + Gather 32-bit integers from memory using 32-bit indices. 32-bit elements are loaded from addresses starting at "base_addr" and offset by each 32-bit element in "vindex" (each index is scaled by the factor in "scale"). Gathered elements are merged into "dst" using "mask" (elements are copied from "src" when the highest bit is not set in the corresponding element). "scale" should be 1, 2, 4 or 8. -dst[15:0] := (a[127:0] >> (imm8[2:0] * 16))[15:0] -dst[31:16] := 0 +FOR j := 0 to 3 + i := j*32 + m := j*32 + IF mask[i+31] + addr := base_addr + SignExtend64(vindex[m+31:m]) * ZeroExtend64(scale) * 8 + dst[i+31:i] := MEM[addr+31:addr] + ELSE + dst[i+31:i] := src[i+31:i] + FI +ENDFOR +mask[MAX:128] := 0 +dst[MAX:128] := 0 - -
emmintrin.h
+ +
immintrin.h
 - + Integer - SSE2 - Swizzle - - - - Copy "a" to "dst", and insert the 16-bit integer "i" into "dst" at the location specified by "imm8". + AVX2 + Load + + + + + + + Gather 32-bit integers from memory using 32-bit indices. 32-bit elements are loaded from addresses starting at "base_addr" and offset by each 32-bit element in "vindex" (each index is scaled by the factor in "scale"). Gathered elements are merged into "dst" using "mask" (elements are copied from "src" when the highest bit is not set in the corresponding element). "scale" should be 1, 2, 4 or 8. -dst[127:0] := a[127:0] -sel := imm8[2:0]*16 -dst[sel+15:sel] := i[15:0] +FOR j := 0 to 7 + i := j*32 + m := j*32 + IF mask[i+31] + addr := base_addr + SignExtend64(vindex[m+31:m]) * ZeroExtend64(scale) * 8 + dst[i+31:i] := MEM[addr+31:addr] + ELSE + dst[i+31:i] := src[i+31:i] + FI +ENDFOR +mask[MAX:256] := 0 +dst[MAX:256] := 0 - -
emmintrin.h
+ +
immintrin.h
 - + Integer - SSE2 - Miscellaneous - - Create mask from the most significant bit of each 8-bit element in "a", and store the result in "dst". + AVX2 + Load + + + + + + + Gather 64-bit integers from memory using 32-bit indices. 64-bit elements are loaded from addresses starting at "base_addr" and offset by each 32-bit element in "vindex" (each index is scaled by the factor in "scale"). Gathered elements are merged into "dst" using "mask" (elements are copied from "src" when the highest bit is not set in the corresponding element). "scale" should be 1, 2, 4 or 8. -FOR j := 0 to 15 - i := j*8 - dst[j] := a[i+7] +FOR j := 0 to 1 + i := j*64 + m := j*32 + IF mask[i+63] + addr := base_addr + SignExtend64(vindex[m+31:m]) * ZeroExtend64(scale) * 8 + dst[i+63:i] := MEM[addr+63:addr] + ELSE + dst[i+63:i] := src[i+63:i] + FI ENDFOR -dst[MAX:16] := 0 +mask[MAX:128] := 0 +dst[MAX:128] := 0 - -
emmintrin.h
+ +
immintrin.h
 - + Integer - SSE2 - Swizzle - - - Shuffle 32-bit integers in "a" using the control in "imm8", and store the results in "dst". + AVX2 + Load + + + + + + + Gather 64-bit integers from memory using 32-bit indices. 64-bit elements are loaded from addresses starting at "base_addr" and offset by each 32-bit element in "vindex" (each index is scaled by the factor in "scale"). Gathered elements are merged into "dst" using "mask" (elements are copied from "src" when the highest bit is not set in the corresponding element). "scale" should be 1, 2, 4 or 8. -SELECT4(src, control){ - CASE(control[1:0]) - 0: tmp[31:0] := src[31:0] - 1: tmp[31:0] := src[63:32] - 2: tmp[31:0] := src[95:64] - 3: tmp[31:0] := src[127:96] - ESAC - RETURN tmp[31:0] -} - -dst[31:0] := SELECT4(a[127:0], imm8[1:0]) -dst[63:32] := SELECT4(a[127:0], imm8[3:2]) -dst[95:64] := SELECT4(a[127:0], imm8[5:4]) -dst[127:96] := SELECT4(a[127:0], imm8[7:6]) +FOR j := 0 to 3 + i := j*64 + m := j*32 + IF mask[i+63] + addr := base_addr + SignExtend64(vindex[m+31:m]) * ZeroExtend64(scale) * 8 + dst[i+63:i] := MEM[addr+63:addr] + ELSE + dst[i+63:i] := src[i+63:i] + FI +ENDFOR +mask[MAX:256] := 0 +dst[MAX:256] := 0 - -
emmintrin.h
+ +
immintrin.h
 - - Integer - SSE2 - Swizzle - - - Shuffle 16-bit integers in the high 64 bits of "a" using the control in "imm8". Store the results in the high 64 bits of "dst", with the low 64 bits being copied from from "a" to "dst". + + Floating Point + AVX2 + Load + + + + + + + Gather double-precision (64-bit) floating-point elements from memory using 64-bit indices. 64-bit elements are loaded from addresses starting at "base_addr" and offset by each 64-bit element in "vindex" (each index is scaled by the factor in "scale"). Gathered elements are merged into "dst" using "mask" (elements are copied from "src" when the highest bit is not set in the corresponding element). "scale" should be 1, 2, 4 or 8. -dst[63:0] := a[63:0] -dst[79:64] := (a >> (imm8[1:0] * 16))[79:64] -dst[95:80] := (a >> (imm8[3:2] * 16))[79:64] -dst[111:96] := (a >> (imm8[5:4] * 16))[79:64] -dst[127:112] := (a >> (imm8[7:6] * 16))[79:64] +FOR j := 0 to 1 + i := j*64 + m := j*64 + IF mask[i+63] + addr := base_addr + vindex[m+63:m] * ZeroExtend64(scale) * 8 + dst[i+63:i] := MEM[addr+63:addr] + ELSE + dst[i+63:i] := src[i+63:i] + FI +ENDFOR +mask[MAX:128] := 0 +dst[MAX:128] := 0 - -
emmintrin.h
+ +
immintrin.h
 - - Integer - SSE2 - Swizzle - - - Shuffle 16-bit integers in the low 64 bits of "a" using the control in "imm8". Store the results in the low 64 bits of "dst", with the high 64 bits being copied from from "a" to "dst". + + Floating Point + AVX2 + Load + + + + + + + Gather double-precision (64-bit) floating-point elements from memory using 64-bit indices. 64-bit elements are loaded from addresses starting at "base_addr" and offset by each 64-bit element in "vindex" (each index is scaled by the factor in "scale"). Gathered elements are merged into "dst" using "mask" (elements are copied from "src" when the highest bit is not set in the corresponding element). "scale" should be 1, 2, 4 or 8. -dst[15:0] := (a >> (imm8[1:0] * 16))[15:0] -dst[31:16] := (a >> (imm8[3:2] * 16))[15:0] -dst[47:32] := (a >> (imm8[5:4] * 16))[15:0] -dst[63:48] := (a >> (imm8[7:6] * 16))[15:0] -dst[127:64] := a[127:64] +FOR j := 0 to 3 + i := j*64 + m := j*64 + IF mask[i+63] + addr := base_addr + vindex[m+63:m] * ZeroExtend64(scale) * 8 + dst[i+63:i] := MEM[addr+63:addr] + ELSE + dst[i+63:i] := src[i+63:i] + FI +ENDFOR +mask[MAX:256] := 0 +dst[MAX:256] := 0 - -
emmintrin.h
+ +
immintrin.h
 - - Integer - SSE2 - Swizzle - - - Unpack and interleave 8-bit integers from the high half of "a" and "b", and store the results in "dst". + + Floating Point + AVX2 + Load + + + + + + + Gather single-precision (32-bit) floating-point elements from memory using 64-bit indices. 32-bit elements are loaded from addresses starting at "base_addr" and offset by each 64-bit element in "vindex" (each index is scaled by the factor in "scale"). Gathered elements are merged into "dst" using "mask" (elements are copied from "src" when the highest bit is not set in the corresponding element). "scale" should be 1, 2, 4 or 8. -INTERLEAVE_HIGH_BYTES(src1[127:0], src2[127:0]){ - dst[7:0] := src1[71:64] - dst[15:8] := src2[71:64] - dst[23:16] := src1[79:72] - dst[31:24] := src2[79:72] - dst[39:32] := src1[87:80] - dst[47:40] := src2[87:80] - dst[55:48] := src1[95:88] - dst[63:56] := src2[95:88] - dst[71:64] := src1[103:96] - dst[79:72] := src2[103:96] - dst[87:80] := src1[111:104] - dst[95:88] := src2[111:104] - dst[103:96] := src1[119:112] - dst[111:104] := src2[119:112] - dst[119:112] := src1[127:120] - dst[127:120] := src2[127:120] - RETURN dst[127:0] -} - -dst[127:0] := INTERLEAVE_HIGH_BYTES(a[127:0], b[127:0]) +FOR j := 0 to 1 + i := j*32 + m := j*64 + IF mask[i+31] + addr := base_addr + vindex[m+63:m] * ZeroExtend64(scale) * 8 + dst[i+31:i] := MEM[addr+31:addr] + ELSE + dst[i+31:i] := src[i+31:i] + FI +ENDFOR +mask[MAX:64] := 0 +dst[MAX:64] := 0 - -
emmintrin.h
+ +
immintrin.h
 - - Integer - SSE2 - Swizzle - - - Unpack and interleave 16-bit integers from the high half of "a" and "b", and store the results in "dst". + + Floating Point + AVX2 + Load + + + + + + + Gather single-precision (32-bit) floating-point elements from memory using 64-bit indices. 32-bit elements are loaded from addresses starting at "base_addr" and offset by each 64-bit element in "vindex" (each index is scaled by the factor in "scale"). Gathered elements are merged into "dst" using "mask" (elements are copied from "src" when the highest bit is not set in the corresponding element). "scale" should be 1, 2, 4 or 8. -INTERLEAVE_HIGH_WORDS(src1[127:0], src2[127:0]){ - dst[15:0] := src1[79:64] - dst[31:16] := src2[79:64] - dst[47:32] := src1[95:80] - dst[63:48] := src2[95:80] - dst[79:64] := src1[111:96] - dst[95:80] := src2[111:96] - dst[111:96] := src1[127:112] - dst[127:112] := src2[127:112] - RETURN dst[127:0] -} - -dst[127:0] := INTERLEAVE_HIGH_WORDS(a[127:0], b[127:0]) +FOR j := 0 to 3 + i := j*32 + m := j*64 + IF mask[i+31] + addr := base_addr + vindex[m+63:m] * ZeroExtend64(scale) * 8 + dst[i+31:i] := MEM[addr+31:addr] + ELSE + dst[i+31:i] := src[i+31:i] + FI +ENDFOR +mask[MAX:128] := 0 +dst[MAX:128] := 0 - -
emmintrin.h
+ +
immintrin.h
 - + Integer - SSE2 - Swizzle - - - Unpack and interleave 32-bit integers from the high half of "a" and "b", and store the results in "dst". + AVX2 + Load + + + + + + + Gather 32-bit integers from memory using 64-bit indices. 32-bit elements are loaded from addresses starting at "base_addr" and offset by each 64-bit element in "vindex" (each index is scaled by the factor in "scale"). Gathered elements are merged into "dst" using "mask" (elements are copied from "src" when the highest bit is not set in the corresponding element). "scale" should be 1, 2, 4 or 8. -INTERLEAVE_HIGH_DWORDS(src1[127:0], src2[127:0]){ - dst[31:0] := src1[95:64] - dst[63:32] := src2[95:64] - dst[95:64] := src1[127:96] - dst[127:96] := src2[127:96] - RETURN dst[127:0] -} - -dst[127:0] := INTERLEAVE_HIGH_DWORDS(a[127:0], b[127:0]) +FOR j := 0 to 1 + i := j*32 + m := j*64 + IF mask[i+31] + addr := base_addr + vindex[m+63:m] * ZeroExtend64(scale) * 8 + dst[i+31:i] := MEM[addr+31:addr] + ELSE + dst[i+31:i] := src[i+31:i] + FI +ENDFOR +mask[MAX:64] := 0 +dst[MAX:64] := 0 - -
emmintrin.h
+ +
immintrin.h
 - + Integer - SSE2 - Swizzle - - - Unpack and interleave 64-bit integers from the high half of "a" and "b", and store the results in "dst". + AVX2 + Load + + + + + + + Gather 32-bit integers from memory using 64-bit indices. 32-bit elements are loaded from addresses starting at "base_addr" and offset by each 64-bit element in "vindex" (each index is scaled by the factor in "scale"). Gathered elements are merged into "dst" using "mask" (elements are copied from "src" when the highest bit is not set in the corresponding element). "scale" should be 1, 2, 4 or 8. -INTERLEAVE_HIGH_QWORDS(src1[127:0], src2[127:0]){ - dst[63:0] := src1[127:64] - dst[127:64] := src2[127:64] - RETURN dst[127:0] -} - -dst[127:0] := INTERLEAVE_HIGH_QWORDS(a[127:0], b[127:0]) +FOR j := 0 to 3 + i := j*32 + m := j*64 + IF mask[i+31] + addr := base_addr + vindex[m+63:m] * ZeroExtend64(scale) * 8 + dst[i+31:i] := MEM[addr+31:addr] + ELSE + dst[i+31:i] := src[i+31:i] + FI +ENDFOR +mask[MAX:128] := 0 +dst[MAX:128] := 0 - -
emmintrin.h
+ +
immintrin.h
 - + Integer - SSE2 - Swizzle - - - Unpack and interleave 8-bit integers from the low half of "a" and "b", and store the results in "dst". + AVX2 + Load + + + + + + + Gather 64-bit integers from memory using 64-bit indices. 64-bit elements are loaded from addresses starting at "base_addr" and offset by each 64-bit element in "vindex" (each index is scaled by the factor in "scale"). Gathered elements are merged into "dst" using "mask" (elements are copied from "src" when the highest bit is not set in the corresponding element). "scale" should be 1, 2, 4 or 8. -INTERLEAVE_BYTES(src1[127:0], src2[127:0]){ - dst[7:0] := src1[7:0] - dst[15:8] := src2[7:0] - dst[23:16] := src1[15:8] - dst[31:24] := src2[15:8] - dst[39:32] := src1[23:16] - dst[47:40] := src2[23:16] - dst[55:48] := src1[31:24] - dst[63:56] := src2[31:24] - dst[71:64] := src1[39:32] - dst[79:72] := src2[39:32] - dst[87:80] := src1[47:40] - dst[95:88] := src2[47:40] - dst[103:96] := src1[55:48] - dst[111:104] := src2[55:48] - dst[119:112] := src1[63:56] - dst[127:120] := src2[63:56] - RETURN dst[127:0] -} - -dst[127:0] := INTERLEAVE_BYTES(a[127:0], b[127:0]) +FOR j := 0 to 1 + i := j*64 + m := j*64 + IF mask[i+63] + addr := base_addr + vindex[m+63:m] * ZeroExtend64(scale) * 8 + dst[i+63:i] := MEM[addr+63:addr] + ELSE + dst[i+63:i] := src[i+63:i] + FI +ENDFOR +mask[MAX:128] := 0 +dst[MAX:128] := 0 - -
emmintrin.h
+ +
immintrin.h
 - + Integer - SSE2 - Swizzle - - - Unpack and interleave 16-bit integers from the low half of "a" and "b", and store the results in "dst". + AVX2 + Load + + + + + + + Gather 64-bit integers from memory using 64-bit indices. 64-bit elements are loaded from addresses starting at "base_addr" and offset by each 64-bit element in "vindex" (each index is scaled by the factor in "scale"). Gathered elements are merged into "dst" using "mask" (elements are copied from "src" when the highest bit is not set in the corresponding element). "scale" should be 1, 2, 4 or 8. -INTERLEAVE_WORDS(src1[127:0], src2[127:0]){ - dst[15:0] := src1[15:0] - dst[31:16] := src2[15:0] - dst[47:32] := src1[31:16] - dst[63:48] := src2[31:16] - dst[79:64] := src1[47:32] - dst[95:80] := src2[47:32] - dst[111:96] := src1[63:48] - dst[127:112] := src2[63:48] - RETURN dst[127:0] -} - -dst[127:0] := INTERLEAVE_WORDS(a[127:0], b[127:0]) +FOR j := 0 to 3 + i := j*64 + m := j*64 + IF mask[i+63] + addr := base_addr + vindex[m+63:m] * ZeroExtend64(scale) * 8 + dst[i+63:i] := MEM[addr+63:addr] + ELSE + dst[i+63:i] := src[i+63:i] + FI +ENDFOR +mask[MAX:256] := 0 +dst[MAX:256] := 0 - -
emmintrin.h
+ +
immintrin.h
 - + Integer - SSE2 - Swizzle - - - Unpack and interleave 32-bit integers from the low half of "a" and "b", and store the results in "dst". + AVX2 + Load + + + + Load packed 32-bit integers from memory into "dst" using "mask" (elements are zeroed out when the highest bit is not set in the corresponding element). -INTERLEAVE_DWORDS(src1[127:0], src2[127:0]){ - dst[31:0] := src1[31:0] - dst[63:32] := src2[31:0] - dst[95:64] := src1[63:32] - dst[127:96] := src2[63:32] - RETURN dst[127:0] -} - -dst[127:0] := INTERLEAVE_DWORDS(a[127:0], b[127:0]) +FOR j := 0 to 3 + i := j*32 + IF mask[i+31] + dst[i+31:i] := MEM[mem_addr+i+31:mem_addr+i] + ELSE + dst[i+31:i] := 0 + FI +ENDFOR +dst[MAX:128] := 0 - -
emmintrin.h
+ +
immintrin.h
 - + Integer - SSE2 - Swizzle - - - Unpack and interleave 64-bit integers from the low half of "a" and "b", and store the results in "dst". + AVX2 + Load + + + + Load packed 32-bit integers from memory into "dst" using "mask" (elements are zeroed out when the highest bit is not set in the corresponding element). -INTERLEAVE_QWORDS(src1[127:0], src2[127:0]){ - dst[63:0] := src1[63:0] - dst[127:64] := src2[63:0] - RETURN dst[127:0] -} - -dst[127:0] := INTERLEAVE_QWORDS(a[127:0], b[127:0]) +FOR j := 0 to 7 + i := j*32 + IF mask[i+31] + dst[i+31:i] := MEM[mem_addr+i+31:mem_addr+i] + ELSE + dst[i+31:i] := 0 + FI +ENDFOR +dst[MAX:256] := 0 - -
emmintrin.h
+ +
immintrin.h
 - - Floating Point - SSE2 - Arithmetic - - - Add the lower double-precision (64-bit) floating-point element in "a" and "b", store the result in the lower element of "dst", and copy the upper element from "a" to the upper element of "dst". + + Integer + AVX2 + Load + + + + Load packed 64-bit integers from memory into "dst" using "mask" (elements are zeroed out when the highest bit is not set in the corresponding element). -dst[63:0] := a[63:0] + b[63:0] -dst[127:64] := a[127:64] +FOR j := 0 to 1 + i := j*64 + IF mask[i+63] + dst[i+63:i] := MEM[mem_addr+i+63:mem_addr+i] + ELSE + dst[i+63:i] := 0 + FI +ENDFOR +dst[MAX:128] := 0 - -
emmintrin.h
+ +
immintrin.h
 - - Floating Point - SSE2 - Arithmetic - - - Add packed double-precision (64-bit) floating-point elements in "a" and "b", and store the results in "dst". + + Integer + AVX2 + Load + + + + Load packed 64-bit integers from memory into "dst" using "mask" (elements are zeroed out when the highest bit is not set in the corresponding element). -FOR j := 0 to 1 +FOR j := 0 to 3 i := j*64 - dst[i+63:i] := a[i+63:i] + b[i+63:i] + IF mask[i+63] + dst[i+63:i] := MEM[mem_addr+i+63:mem_addr+i] + ELSE + dst[i+63:i] := 0 + FI ENDFOR +dst[MAX:256] := 0 - -
emmintrin.h
+ +
immintrin.h
 - - Floating Point - SSE2 - Arithmetic - - - Divide the lower double-precision (64-bit) floating-point element in "a" by the lower double-precision (64-bit) floating-point element in "b", store the result in the lower element of "dst", and copy the upper element from "a" to the upper element of "dst". + + Integer + AVX2 + Store + + + + + Store packed 32-bit integers from "a" into memory using "mask" (elements are not stored when the highest bit is not set in the corresponding element). -dst[63:0] := a[63:0] 0 b[63:0] -dst[127:64] := a[127:64] +FOR j := 0 to 3 + i := j*32 + IF mask[i+31] + MEM[mem_addr+i+31:mem_addr+i] := a[i+31:i] + FI +ENDFOR - -
emmintrin.h
+ +
immintrin.h
 - - Floating Point - SSE2 - Arithmetic - - - Divide packed double-precision (64-bit) floating-point elements in "a" by packed elements in "b", and store the results in "dst". + + Integer + AVX2 + Store + + + + + Store packed 32-bit integers from "a" into memory using "mask" (elements are not stored when the highest bit is not set in the corresponding element). -FOR j := 0 to 1 - i := 64*j - dst[i+63:i] := a[i+63:i] / b[i+63:i] +FOR j := 0 to 7 + i := j*32 + IF mask[i+31] + MEM[mem_addr+i+31:mem_addr+i] := a[i+31:i] + FI ENDFOR - -
emmintrin.h
+ +
immintrin.h
 - - Floating Point - SSE2 - Special Math Functions - - - Compare the lower double-precision (64-bit) floating-point elements in "a" and "b", store the maximum value in the lower element of "dst", and copy the upper element from "a" to the upper element of "dst". + + Integer + AVX2 + Store + + + + + Store packed 64-bit integers from "a" into memory using "mask" (elements are not stored when the highest bit is not set in the corresponding element). -dst[63:0] := MAX(a[63:0], b[63:0]) -dst[127:64] := a[127:64] +FOR j := 0 to 1 + i := j*64 + IF mask[i+63] + MEM[mem_addr+i+63:mem_addr+i] := a[i+63:i] + FI +ENDFOR - -
emmintrin.h
+ +
immintrin.h
 - - Floating Point - SSE2 - Special Math Functions - - - Compare packed double-precision (64-bit) floating-point elements in "a" and "b", and store packed maximum values in "dst". + + Integer + AVX2 + Store + + + + + Store packed 64-bit integers from "a" into memory using "mask" (elements are not stored when the highest bit is not set in the corresponding element). -FOR j := 0 to 1 +FOR j := 0 to 3 i := j*64 - dst[i+63:i] := MAX(a[i+63:i], b[i+63:i]) + IF mask[i+63] + MEM[mem_addr+i+63:mem_addr+i] := a[i+63:i] + FI ENDFOR - -
emmintrin.h
+ +
immintrin.h
 - - Floating Point - SSE2 + + Integer + AVX2 Special Math Functions - - - Compare the lower double-precision (64-bit) floating-point elements in "a" and "b", store the minimum value in the lower element of "dst", and copy the upper element from "a" to the upper element of "dst". + + + + Compare packed signed 8-bit integers in "a" and "b", and store packed maximum values in "dst". -dst[63:0] := MIN(a[63:0], b[63:0]) -dst[127:64] := a[127:64] +FOR j := 0 to 31 + i := j*8 + dst[i+7:i] := MAX(a[i+7:i], b[i+7:i]) +ENDFOR +dst[MAX:256] := 0 - -
emmintrin.h
+ +
immintrin.h
 - - Floating Point - SSE2 + + Integer + AVX2 Special Math Functions - - - Compare packed double-precision (64-bit) floating-point elements in "a" and "b", and store packed minimum values in "dst". - + + + + Compare packed signed 16-bit integers in "a" and "b", and store packed maximum values in "dst". -FOR j := 0 to 1 - i := j*64 - dst[i+63:i] := MIN(a[i+63:i], b[i+63:i]) +FOR j := 0 to 15 + i := j*16 + dst[i+15:i] := MAX(a[i+15:i], b[i+15:i]) ENDFOR +dst[MAX:256] := 0 - -
emmintrin.h
+ +
immintrin.h
 - - Floating Point - SSE2 - Arithmetic - - - Multiply the lower double-precision (64-bit) floating-point element in "a" and "b", store the result in the lower element of "dst", and copy the upper element from "a" to the upper element of "dst". + + Integer + AVX2 + Special Math Functions + + + + Compare packed signed 32-bit integers in "a" and "b", and store packed maximum values in "dst". -dst[63:0] := a[63:0] * b[63:0] -dst[127:64] := a[127:64] +FOR j := 0 to 7 + i := j*32 + dst[i+31:i] := MAX(a[i+31:i], b[i+31:i]) +ENDFOR +dst[MAX:256] := 0 - -
emmintrin.h
+ +
immintrin.h
 - - Floating Point - SSE2 - Arithmetic - - - Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", and store the results in "dst". - + + Integer + AVX2 + Special Math Functions + + + + Compare packed unsigned 8-bit integers in "a" and "b", and store packed maximum values in "dst". -FOR j := 0 to 1 - i := j*64 - dst[i+63:i] := a[i+63:i] * b[i+63:i] +FOR j := 0 to 31 + i := j*8 + dst[i+7:i] := MAX(a[i+7:i], b[i+7:i]) ENDFOR +dst[MAX:256] := 0 - -
emmintrin.h
+ +
immintrin.h
 - - Floating Point - SSE2 - Elementary Math Functions - - - Compute the square root of the lower double-precision (64-bit) floating-point element in "b", store the result in the lower element of "dst", and copy the upper element from "a" to the upper element of "dst". + + Integer + AVX2 + Special Math Functions + + + + Compare packed unsigned 16-bit integers in "a" and "b", and store packed maximum values in "dst". -dst[63:0] := SQRT(b[63:0]) -dst[127:64] := a[127:64] +FOR j := 0 to 15 + i := j*16 + dst[i+15:i] := MAX(a[i+15:i], b[i+15:i]) +ENDFOR +dst[MAX:256] := 0 - -
emmintrin.h
+ +
immintrin.h
 - - Floating Point - SSE2 - Elementary Math Functions - - Compute the square root of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst". + + Integer + AVX2 + Special Math Functions + + + + Compare packed unsigned 32-bit integers in "a" and "b", and store packed maximum values in "dst". -FOR j := 0 to 1 - i := j*64 - dst[i+63:i] := SQRT(a[i+63:i]) +FOR j := 0 to 7 + i := j*32 + dst[i+31:i] := MAX(a[i+31:i], b[i+31:i]) ENDFOR +dst[MAX:256] := 0 - -
emmintrin.h
+ +
immintrin.h
 - - Floating Point - SSE2 - Arithmetic - - - Subtract the lower double-precision (64-bit) floating-point element in "b" from the lower double-precision (64-bit) floating-point element in "a", store the result in the lower element of "dst", and copy the upper element from "a" to the upper element of "dst". + + Integer + AVX2 + Special Math Functions + + + + Compare packed signed 8-bit integers in "a" and "b", and store packed minimum values in "dst". -dst[63:0] := a[63:0] - b[63:0] -dst[127:64] := a[127:64] +FOR j := 0 to 31 + i := j*8 + dst[i+7:i] := MIN(a[i+7:i], b[i+7:i]) +ENDFOR +dst[MAX:256] := 0 - -
emmintrin.h
+ +
immintrin.h
 - - Floating Point - SSE2 - Arithmetic - - - Subtract packed double-precision (64-bit) floating-point elements in "b" from packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst". + + Integer + AVX2 + Special Math Functions + + + + Compare packed signed 16-bit integers in "a" and "b", and store packed minimum values in "dst". -FOR j := 0 to 1 - i := j*64 - dst[i+63:i] := a[i+63:i] - b[i+63:i] +FOR j := 0 to 15 + i := j*16 + dst[i+15:i] := MIN(a[i+15:i], b[i+15:i]) ENDFOR +dst[MAX:256] := 0 - -
emmintrin.h
+ +
immintrin.h
 - - Floating Point - SSE2 - Logical - - - Compute the bitwise AND of packed double-precision (64-bit) floating-point elements in "a" and "b", and store the results in "dst". + + Integer + AVX2 + Special Math Functions + + + + Compare packed signed 32-bit integers in "a" and "b", and store packed minimum values in "dst". -FOR j := 0 to 1 - i := j*64 - dst[i+63:i] := (a[i+63:i] AND b[i+63:i]) +FOR j := 0 to 7 + i := j*32 + dst[i+31:i] := MIN(a[i+31:i], b[i+31:i]) ENDFOR +dst[MAX:256] := 0 - -
emmintrin.h
+ +
immintrin.h
 - - Floating Point - SSE2 - Logical - - - Compute the bitwise NOT of packed double-precision (64-bit) floating-point elements in "a" and then AND with "b", and store the results in "dst". + + Integer + AVX2 + Special Math Functions + + + + Compare packed unsigned 8-bit integers in "a" and "b", and store packed minimum values in "dst". -FOR j := 0 to 1 - i := j*64 - dst[i+63:i] := ((NOT a[i+63:i]) AND b[i+63:i]) +FOR j := 0 to 31 + i := j*8 + dst[i+7:i] := MIN(a[i+7:i], b[i+7:i]) ENDFOR +dst[MAX:256] := 0 - -
emmintrin.h
+ +
immintrin.h
 - - Floating Point - SSE2 - Logical - - - Compute the bitwise OR of packed double-precision (64-bit) floating-point elements in "a" and "b", and store the results in "dst". - + + Integer + AVX2 + Special Math Functions + + + + Compare packed unsigned 16-bit integers in "a" and "b", and store packed minimum values in "dst". -FOR j := 0 to 1 - i := j*64 - dst[i+63:i] := a[i+63:i] BITWISE OR b[i+63:i] +FOR j := 0 to 15 + i := j*16 + dst[i+15:i] := MIN(a[i+15:i], b[i+15:i]) ENDFOR +dst[MAX:256] := 0 - -
emmintrin.h
+ +
immintrin.h
 - - Floating Point - SSE2 - Logical - - - Compute the bitwise XOR of packed double-precision (64-bit) floating-point elements in "a" and "b", and store the results in "dst". - + + Integer + AVX2 + Special Math Functions + + + + Compare packed unsigned 32-bit integers in "a" and "b", and store packed minimum values in "dst". -FOR j := 0 to 1 - i := j*64 - dst[i+63:i] := a[i+63:i] XOR b[i+63:i] +FOR j := 0 to 7 + i := j*32 + dst[i+31:i] := MIN(a[i+31:i], b[i+31:i]) ENDFOR +dst[MAX:256] := 0 - -
emmintrin.h
+ +
immintrin.h
 - - Floating Point - SSE2 - Compare - - - Compare the lower double-precision (64-bit) floating-point elements in "a" and "b" for equality, store the result in the lower element of "dst", and copy the upper element from "a" to the upper element of "dst". + + Integer + AVX2 + Miscellaneous + + + Create mask from the most significant bit of each 8-bit element in "a", and store the result in "dst". -dst[63:0] := (a[63:0] == b[63:0]) ? 0xFFFFFFFFFFFFFFFF : 0 -dst[127:64] := a[127:64] +FOR j := 0 to 31 + i := j*8 + dst[j] := a[i+7] +ENDFOR - -
emmintrin.h
+ +
immintrin.h
 - - Floating Point - SSE2 - Compare - - - Compare the lower double-precision (64-bit) floating-point elements in "a" and "b" for less-than, store the result in the lower element of "dst", and copy the upper element from "a" to the upper element of "dst". + + Integer + AVX2 + Miscellaneous + + + + + Compute the sum of absolute differences (SADs) of quadruplets of unsigned 8-bit integers in "a" compared to those in "b", and store the 16-bit results in "dst". + Eight SADs are performed for each 128-bit lane using one quadruplet from "b" and eight quadruplets from "a". One quadruplet is selected from "b" starting at on the offset specified in "imm8". Eight quadruplets are formed from sequential 8-bit integers selected from "a" starting at the offset specified in "imm8". -dst[63:0] := (a[63:0] < b[63:0]) ? 0xFFFFFFFFFFFFFFFF : 0 -dst[127:64] := a[127:64] +DEFINE MPSADBW(a[127:0], b[127:0], imm8[2:0]) { + a_offset := imm8[2]*32 + b_offset := imm8[1:0]*32 + FOR j := 0 to 7 + i := j*8 + k := a_offset+i + l := b_offset + tmp[i*2+15:i*2] := ABS(Signed(a[k+7:k] - b[l+7:l])) + ABS(Signed(a[k+15:k+8] - b[l+15:l+8])) + \ + ABS(Signed(a[k+23:k+16] - b[l+23:l+16])) + ABS(Signed(a[k+31:k+24] - b[l+31:l+24])) + ENDFOR + RETURN tmp[127:0] +} +dst[127:0] := MPSADBW(a[127:0], b[127:0], imm8[2:0]) +dst[255:128] := MPSADBW(a[255:128], b[255:128], imm8[5:3]) +dst[MAX:256] := 0 - -
emmintrin.h
+ +
immintrin.h
 - - Floating Point - SSE2 - Compare - - - Compare the lower double-precision (64-bit) floating-point elements in "a" and "b" for less-than-or-equal, store the result in the lower element of "dst", and copy the upper element from "a" to the upper element of "dst". + + Integer + AVX2 + Arithmetic + + + + Multiply the low signed 32-bit integers from each packed 64-bit element in "a" and "b", and store the signed 64-bit results in "dst". -dst[63:0] := (a[63:0] <= b[63:0]) ? 0xFFFFFFFFFFFFFFFF : 0 -dst[127:64] := a[127:64] +FOR j := 0 to 3 + i := j*64 + dst[i+63:i] := SignExtend64(a[i+31:i]) * SignExtend64(b[i+31:i]) +ENDFOR +dst[MAX:256] := 0 - -
emmintrin.h
+ +
immintrin.h
 - - Floating Point - SSE2 - Compare - - - Compare the lower double-precision (64-bit) floating-point elements in "a" and "b" for greater-than, store the result in the lower element of "dst", and copy the upper element from "a" to the upper element of "dst". + + Integer + AVX2 + Arithmetic + + + + Multiply the low unsigned 32-bit integers from each packed 64-bit element in "a" and "b", and store the unsigned 64-bit results in "dst". -dst[63:0] := (a[63:0] > b[63:0]) ? 0xFFFFFFFFFFFFFFFF : 0 -dst[127:64] := a[127:64] +FOR j := 0 to 3 + i := j*64 + dst[i+63:i] := a[i+31:i] * b[i+31:i] +ENDFOR +dst[MAX:256] := 0 - -
emmintrin.h
+ +
immintrin.h
 - - Floating Point - SSE2 - Compare - - - Compare the lower double-precision (64-bit) floating-point elements in "a" and "b" for greater-than-or-equal, store the result in the lower element of "dst", and copy the upper element from "a" to the upper element of "dst". + + Integer + AVX2 + Arithmetic + + + + Multiply the packed signed 16-bit integers in "a" and "b", producing intermediate 32-bit integers, and store the high 16 bits of the intermediate integers in "dst". -dst[63:0] := (a[63:0] >= b[63:0]) ? 0xFFFFFFFFFFFFFFFF : 0 -dst[127:64] := a[127:64] +FOR j := 0 to 15 + i := j*16 + tmp[31:0] := SignExtend32(a[i+15:i]) * SignExtend32(b[i+15:i]) + dst[i+15:i] := tmp[31:16] +ENDFOR +dst[MAX:256] := 0 - -
emmintrin.h
+ +
immintrin.h
 - - Floating Point - SSE2 - Compare - - - Compare the lower double-precision (64-bit) floating-point elements in "a" and "b" to see if neither is NaN, store the result in the lower element of "dst", and copy the upper element from "a" to the upper element of "dst". + + Integer + AVX2 + Arithmetic + + + + Multiply the packed unsigned 16-bit integers in "a" and "b", producing intermediate 32-bit integers, and store the high 16 bits of the intermediate integers in "dst". -dst[63:0] := (a[63:0] != NaN AND b[63:0] != NaN) ? 0xFFFFFFFFFFFFFFFF : 0 -dst[127:64] := a[127:64] +FOR j := 0 to 15 + i := j*16 + tmp[31:0] := a[i+15:i] * b[i+15:i] + dst[i+15:i] := tmp[31:16] +ENDFOR +dst[MAX:256] := 0 - -
emmintrin.h
+ +
immintrin.h
 - - Floating Point - SSE2 - Compare - - - Compare the lower double-precision (64-bit) floating-point elements in "a" and "b" to see if either is NaN, store the result in the lower element of "dst", and copy the upper element from "a" to the upper element of "dst". + + Integer + AVX2 + Arithmetic + + + + Multiply packed signed 16-bit integers in "a" and "b", producing intermediate signed 32-bit integers. Truncate each intermediate integer to the 18 most significant bits, round by adding 1, and store bits [16:1] to "dst". -dst[63:0] := (a[63:0] == NaN OR b[63:0] == NaN) ? 0xFFFFFFFFFFFFFFFF : 0 -dst[127:64] := a[127:64] +FOR j := 0 to 15 + i := j*16 + tmp[31:0] := ((SignExtend32(a[i+15:i]) * SignExtend32(b[i+15:i])) >> 14) + 1 + dst[i+15:i] := tmp[16:1] +ENDFOR +dst[MAX:256] := 0 - -
emmintrin.h
+ +
immintrin.h
 - - Floating Point - SSE2 - Compare - - - Compare the lower double-precision (64-bit) floating-point elements in "a" and "b" for not-equal, store the result in the lower element of "dst", and copy the upper element from "a" to the upper element of "dst". + + Integer + AVX2 + Arithmetic + + + + Multiply the packed signed 16-bit integers in "a" and "b", producing intermediate 32-bit integers, and store the low 16 bits of the intermediate integers in "dst". -dst[63:0] := (a[63:0] != b[63:0]) ? 0xFFFFFFFFFFFFFFFF : 0 -dst[127:64] := a[127:64] +FOR j := 0 to 15 + i := j*16 + tmp[31:0] := SignExtend32(a[i+15:i]) * SignExtend32(b[i+15:i]) + dst[i+15:i] := tmp[15:0] +ENDFOR +dst[MAX:256] := 0 - -
emmintrin.h
+ +
immintrin.h
 - - Floating Point - SSE2 - Compare - - - Compare the lower double-precision (64-bit) floating-point elements in "a" and "b" for not-less-than, store the result in the lower element of "dst", and copy the upper element from "a" to the upper element of "dst". + + Integer + AVX2 + Arithmetic + + + + Multiply the packed signed 32-bit integers in "a" and "b", producing intermediate 64-bit integers, and store the low 32 bits of the intermediate integers in "dst". -dst[63:0] := !(a[63:0] < b[63:0]) ? 0xFFFFFFFFFFFFFFFF : 0 -dst[127:64] := a[127:64] +FOR j := 0 to 7 + i := j*32 + tmp[63:0] := a[i+31:i] * b[i+31:i] + dst[i+31:i] := tmp[31:0] +ENDFOR +dst[MAX:256] := 0 - -
emmintrin.h
+ +
immintrin.h
 - - Floating Point - SSE2 - Compare - - - Compare the lower double-precision (64-bit) floating-point elements in "a" and "b" for not-less-than-or-equal, store the result in the lower element of "dst", and copy the upper element from "a" to the upper element of "dst". + + Integer + AVX2 + Logical + + + + Compute the bitwise OR of 256 bits (representing integer data) in "a" and "b", and store the result in "dst". -dst[63:0] := !(a[63:0] <= b[63:0]) ? 0xFFFFFFFFFFFFFFFF : 0 -dst[127:64] := a[127:64] +dst[255:0] := (a[255:0] OR b[255:0]) +dst[MAX:256] := 0 - -
emmintrin.h
+ +
immintrin.h
 - - Floating Point - SSE2 - Compare - - - Compare the lower double-precision (64-bit) floating-point elements in "a" and "b" for not-greater-than, store the result in the lower element of "dst", and copy the upper element from "a" to the upper element of "dst". + + Integer + AVX2 + Miscellaneous + + + + Convert packed signed 16-bit integers from "a" and "b" to packed 8-bit integers using signed saturation, and store the results in "dst". + +dst[7:0] := Saturate8(a[15:0]) +dst[15:8] := Saturate8(a[31:16]) +dst[23:16] := Saturate8(a[47:32]) +dst[31:24] := Saturate8(a[63:48]) +dst[39:32] := Saturate8(a[79:64]) +dst[47:40] := Saturate8(a[95:80]) +dst[55:48] := Saturate8(a[111:96]) +dst[63:56] := Saturate8(a[127:112]) +dst[71:64] := Saturate8(b[15:0]) +dst[79:72] := Saturate8(b[31:16]) +dst[87:80] := Saturate8(b[47:32]) +dst[95:88] := Saturate8(b[63:48]) +dst[103:96] := Saturate8(b[79:64]) +dst[111:104] := Saturate8(b[95:80]) +dst[119:112] := Saturate8(b[111:96]) +dst[127:120] := Saturate8(b[127:112]) +dst[135:128] := Saturate8(a[143:128]) +dst[143:136] := Saturate8(a[159:144]) +dst[151:144] := Saturate8(a[175:160]) +dst[159:152] := Saturate8(a[191:176]) +dst[167:160] := Saturate8(a[207:192]) +dst[175:168] := Saturate8(a[223:208]) +dst[183:176] := Saturate8(a[239:224]) +dst[191:184] := Saturate8(a[255:240]) +dst[199:192] := Saturate8(b[143:128]) +dst[207:200] := Saturate8(b[159:144]) +dst[215:208] := Saturate8(b[175:160]) +dst[223:216] := Saturate8(b[191:176]) +dst[231:224] := Saturate8(b[207:192]) +dst[239:232] := Saturate8(b[223:208]) +dst[247:240] := Saturate8(b[239:224]) +dst[255:248] := Saturate8(b[255:240]) +dst[MAX:256] := 0 + + +
immintrin.h
+ + + Integer + AVX2 + Miscellaneous + + + + Convert packed signed 32-bit integers from "a" and "b" to packed 16-bit integers using signed saturation, and store the results in "dst". -dst[63:0] := !(a[63:0] > b[63:0]) ? 0xFFFFFFFFFFFFFFFF : 0 -dst[127:64] := a[127:64] +dst[15:0] := Saturate16(a[31:0]) +dst[31:16] := Saturate16(a[63:32]) +dst[47:32] := Saturate16(a[95:64]) +dst[63:48] := Saturate16(a[127:96]) +dst[79:64] := Saturate16(b[31:0]) +dst[95:80] := Saturate16(b[63:32]) +dst[111:96] := Saturate16(b[95:64]) +dst[127:112] := Saturate16(b[127:96]) +dst[143:128] := Saturate16(a[159:128]) +dst[159:144] := Saturate16(a[191:160]) +dst[175:160] := Saturate16(a[223:192]) +dst[191:176] := Saturate16(a[255:224]) +dst[207:192] := Saturate16(b[159:128]) +dst[223:208] := Saturate16(b[191:160]) +dst[239:224] := Saturate16(b[223:192]) +dst[255:240] := Saturate16(b[255:224]) +dst[MAX:256] := 0 - -
emmintrin.h
+ +
immintrin.h
 - - Floating Point - SSE2 - Compare - - - Compare the lower double-precision (64-bit) floating-point elements in "a" and "b" for not-greater-than-or-equal, store the result in the lower element of "dst", and copy the upper element from "a" to the upper element of "dst". + + Integer + AVX2 + Miscellaneous + + + + Convert packed signed 16-bit integers from "a" and "b" to packed 8-bit integers using unsigned saturation, and store the results in "dst". + +dst[7:0] := SaturateU8(a[15:0]) +dst[15:8] := SaturateU8(a[31:16]) +dst[23:16] := SaturateU8(a[47:32]) +dst[31:24] := SaturateU8(a[63:48]) +dst[39:32] := SaturateU8(a[79:64]) +dst[47:40] := SaturateU8(a[95:80]) +dst[55:48] := SaturateU8(a[111:96]) +dst[63:56] := SaturateU8(a[127:112]) +dst[71:64] := SaturateU8(b[15:0]) +dst[79:72] := SaturateU8(b[31:16]) +dst[87:80] := SaturateU8(b[47:32]) +dst[95:88] := SaturateU8(b[63:48]) +dst[103:96] := SaturateU8(b[79:64]) +dst[111:104] := SaturateU8(b[95:80]) +dst[119:112] := SaturateU8(b[111:96]) +dst[127:120] := SaturateU8(b[127:112]) +dst[135:128] := SaturateU8(a[143:128]) +dst[143:136] := SaturateU8(a[159:144]) +dst[151:144] := SaturateU8(a[175:160]) +dst[159:152] := SaturateU8(a[191:176]) +dst[167:160] := SaturateU8(a[207:192]) +dst[175:168] := SaturateU8(a[223:208]) +dst[183:176] := SaturateU8(a[239:224]) +dst[191:184] := SaturateU8(a[255:240]) +dst[199:192] := SaturateU8(b[143:128]) +dst[207:200] := SaturateU8(b[159:144]) +dst[215:208] := SaturateU8(b[175:160]) +dst[223:216] := SaturateU8(b[191:176]) +dst[231:224] := SaturateU8(b[207:192]) +dst[239:232] := SaturateU8(b[223:208]) +dst[247:240] := SaturateU8(b[239:224]) +dst[255:248] := SaturateU8(b[255:240]) +dst[MAX:256] := 0 + + +
immintrin.h
+ + + Integer + AVX2 + Miscellaneous + + + + Convert packed signed 32-bit integers from "a" and "b" to packed 16-bit integers using unsigned saturation, and store the results in "dst". -dst[63:0] := !(a[63:0] >= b[63:0]) ? 0xFFFFFFFFFFFFFFFF : 0 -dst[127:64] := a[127:64] +dst[15:0] := SaturateU16(a[31:0]) +dst[31:16] := SaturateU16(a[63:32]) +dst[47:32] := SaturateU16(a[95:64]) +dst[63:48] := SaturateU16(a[127:96]) +dst[79:64] := SaturateU16(b[31:0]) +dst[95:80] := SaturateU16(b[63:32]) +dst[111:96] := SaturateU16(b[95:64]) +dst[127:112] := SaturateU16(b[127:96]) +dst[143:128] := SaturateU16(a[159:128]) +dst[159:144] := SaturateU16(a[191:160]) +dst[175:160] := SaturateU16(a[223:192]) +dst[191:176] := SaturateU16(a[255:224]) +dst[207:192] := SaturateU16(b[159:128]) +dst[223:208] := SaturateU16(b[191:160]) +dst[239:224] := SaturateU16(b[223:192]) +dst[255:240] := SaturateU16(b[255:224]) +dst[MAX:256] := 0 - -
emmintrin.h
+ +
immintrin.h
 - - Floating Point - SSE2 - Compare - - - Compare packed double-precision (64-bit) floating-point elements in "a" and "b" for equality, and store the results in "dst". - -FOR j := 0 to 1 - i := j*64 - dst[i+63:i] := (a[i+63:i] == b[i+63:i]) ? 0xFFFFFFFFFFFFFFFF : 0 -ENDFOR + + Integer + AVX2 + Swizzle + + + + + Shuffle 128-bits (composed of integer data) selected by "imm8" from "a" and "b", and store the results in "dst". + +DEFINE SELECT4(src1, src2, control) { + CASE(control[1:0]) OF + 0: tmp[127:0] := src1[127:0] + 1: tmp[127:0] := src1[255:128] + 2: tmp[127:0] := src2[127:0] + 3: tmp[127:0] := src2[255:128] + ESAC + IF control[3] + tmp[127:0] := 0 + FI + RETURN tmp[127:0] +} +dst[127:0] := SELECT4(a[255:0], b[255:0], imm8[3:0]) +dst[255:128] := SELECT4(a[255:0], b[255:0], imm8[7:4]) +dst[MAX:256] := 0 - -
emmintrin.h
+ +
immintrin.h
 - - Floating Point - SSE2 - Compare - - - Compare packed double-precision (64-bit) floating-point elements in "a" and "b" for less-than, and store the results in "dst". + + Integer + AVX2 + Swizzle + + + + Shuffle 64-bit integers in "a" across lanes using the control in "imm8", and store the results in "dst". -FOR j := 0 to 1 - i := j*64 - dst[i+63:i] := (a[i+63:i] < b[i+63:i]) ? 0xFFFFFFFFFFFFFFFF : 0 -ENDFOR +DEFINE SELECT4(src, control) { + CASE(control[1:0]) OF + 0: tmp[63:0] := src[63:0] + 1: tmp[63:0] := src[127:64] + 2: tmp[63:0] := src[191:128] + 3: tmp[63:0] := src[255:192] + ESAC + RETURN tmp[63:0] +} +dst[63:0] := SELECT4(a[255:0], imm8[1:0]) +dst[127:64] := SELECT4(a[255:0], imm8[3:2]) +dst[191:128] := SELECT4(a[255:0], imm8[5:4]) +dst[255:192] := SELECT4(a[255:0], imm8[7:6]) +dst[MAX:256] := 0 - -
emmintrin.h
+ +
immintrin.h
 - + Floating Point - SSE2 - Compare - - - Compare packed double-precision (64-bit) floating-point elements in "a" and "b" for less-than-or-equal, and store the results in "dst". + AVX2 + Swizzle + + + + Shuffle double-precision (64-bit) floating-point elements in "a" across lanes using the control in "imm8", and store the results in "dst". -FOR j := 0 to 1 - i := j*64 - dst[i+63:i] := (a[i+63:i] <= b[i+63:i]) ? 0xFFFFFFFFFFFFFFFF : 0 -ENDFOR +DEFINE SELECT4(src, control) { + CASE(control[1:0]) OF + 0: tmp[63:0] := src[63:0] + 1: tmp[63:0] := src[127:64] + 2: tmp[63:0] := src[191:128] + 3: tmp[63:0] := src[255:192] + ESAC + RETURN tmp[63:0] +} +dst[63:0] := SELECT4(a[255:0], imm8[1:0]) +dst[127:64] := SELECT4(a[255:0], imm8[3:2]) +dst[191:128] := SELECT4(a[255:0], imm8[5:4]) +dst[255:192] := SELECT4(a[255:0], imm8[7:6]) +dst[MAX:256] := 0 - -
emmintrin.h
+ +
immintrin.h
 - - Floating Point - SSE2 - Compare - - - Compare packed double-precision (64-bit) floating-point elements in "a" and "b" for greater-than, and store the results in "dst". + + Integer + AVX2 + Swizzle + + + + Shuffle 32-bit integers in "a" across lanes using the corresponding index in "idx", and store the results in "dst". -FOR j := 0 to 1 - i := j*64 - dst[i+63:i] := (a[i+63:i] > b[i+63:i]) ? 0xFFFFFFFFFFFFFFFF : 0 -ENDFOR +FOR j := 0 to 7 + i := j*32 + id := idx[i+2:i]*32 + dst[i+31:i] := a[id+31:id] +ENDFOR +dst[MAX:256] := 0 - -
emmintrin.h
+ +
immintrin.h
 - + Floating Point - SSE2 - Compare - - - Compare packed double-precision (64-bit) floating-point elements in "a" and "b" for greater-than-or-equal, and store the results in "dst". + AVX2 + Swizzle + + + + Shuffle single-precision (32-bit) floating-point elements in "a" across lanes using the corresponding index in "idx". -FOR j := 0 to 1 - i := j*64 - dst[i+63:i] := (a[i+63:i] >= b[i+63:i]) ? 0xFFFFFFFFFFFFFFFF : 0 -ENDFOR +FOR j := 0 to 7 + i := j*32 + id := idx[i+2:i]*32 + dst[i+31:i] := a[id+31:id] +ENDFOR +dst[MAX:256] := 0 - -
emmintrin.h
+ +
immintrin.h
 - - Floating Point - SSE2 - Compare - - - Compare packed double-precision (64-bit) floating-point elements in "a" and "b" to see if neither is NaN, and store the results in "dst". + + Integer + AVX2 + Arithmetic + + + + Compute the absolute differences of packed unsigned 8-bit integers in "a" and "b", then horizontally sum each consecutive 8 differences to produce four unsigned 16-bit integers, and pack these unsigned 16-bit integers in the low 16 bits of 64-bit elements in "dst". -FOR j := 0 to 1 +FOR j := 0 to 31 + i := j*8 + tmp[i+7:i] := ABS(a[i+7:i] - b[i+7:i]) +ENDFOR +FOR j := 0 to 3 i := j*64 - dst[i+63:i] := (a[i+63:i] != NaN AND b[i+63:i] != NaN) ? 0xFFFFFFFFFFFFFFFF : 0 -ENDFOR + dst[i+15:i] := tmp[i+7:i] + tmp[i+15:i+8] + tmp[i+23:i+16] + tmp[i+31:i+24] + \ + tmp[i+39:i+32] + tmp[i+47:i+40] + tmp[i+55:i+48] + tmp[i+63:i+56] + dst[i+63:i+16] := 0 +ENDFOR +dst[MAX:256] := 0 - -
emmintrin.h
+ +
immintrin.h
 - - Floating Point - SSE2 - Compare - - - Compare packed double-precision (64-bit) floating-point elements in "a" and "b" to see if either is NaN, and store the results in "dst". + + Integer + AVX2 + Swizzle + + + + Shuffle 32-bit integers in "a" within 128-bit lanes using the control in "imm8", and store the results in "dst". -FOR j := 0 to 1 - i := j*64 - dst[i+63:i] := (a[i+63:i] == NaN OR b[i+63:i] == NaN) ? 0xFFFFFFFFFFFFFFFF : 0 -ENDFOR +DEFINE SELECT4(src, control) { + CASE(control[1:0]) OF + 0: tmp[31:0] := src[31:0] + 1: tmp[31:0] := src[63:32] + 2: tmp[31:0] := src[95:64] + 3: tmp[31:0] := src[127:96] + ESAC + RETURN tmp[31:0] +} +dst[31:0] := SELECT4(a[127:0], imm8[1:0]) +dst[63:32] := SELECT4(a[127:0], imm8[3:2]) +dst[95:64] := SELECT4(a[127:0], imm8[5:4]) +dst[127:96] := SELECT4(a[127:0], imm8[7:6]) +dst[159:128] := SELECT4(a[255:128], imm8[1:0]) +dst[191:160] := SELECT4(a[255:128], imm8[3:2]) +dst[223:192] := SELECT4(a[255:128], imm8[5:4]) +dst[255:224] := SELECT4(a[255:128], imm8[7:6]) +dst[MAX:256] := 0 - -
emmintrin.h
+ +
immintrin.h
 - - Floating Point - SSE2 - Compare - - - Compare packed double-precision (64-bit) floating-point elements in "a" and "b" for not-equal, and store the results in "dst". + + Integer + AVX2 + Swizzle + + + + Shuffle 8-bit integers in "a" within 128-bit lanes according to shuffle control mask in the corresponding 8-bit element of "b", and store the results in "dst". -FOR j := 0 to 1 - i := j*64 - dst[i+63:i] := (a[i+63:i] != b[i+63:i]) ? 0xFFFFFFFFFFFFFFFF : 0 -ENDFOR +FOR j := 0 to 15 + i := j*8 + IF b[i+7] == 1 + dst[i+7:i] := 0 + ELSE + index[3:0] := b[i+3:i] + dst[i+7:i] := a[index*8+7:index*8] + FI + IF b[128+i+7] == 1 + dst[128+i+7:128+i] := 0 + ELSE + index[3:0] := b[128+i+3:128+i] + dst[128+i+7:128+i] := a[128+index*8+7:128+index*8] + FI +ENDFOR +dst[MAX:256] := 0 - -
emmintrin.h
+ +
immintrin.h
 - - Floating Point - SSE2 - Compare - - - Compare packed double-precision (64-bit) floating-point elements in "a" and "b" for not-less-than, and store the results in "dst". + + Integer + AVX2 + Swizzle + + + + Shuffle 16-bit integers in the high 64 bits of 128-bit lanes of "a" using the control in "imm8". Store the results in the high 64 bits of 128-bit lanes of "dst", with the low 64 bits of 128-bit lanes being copied from from "a" to "dst". -FOR j := 0 to 1 - i := j*64 - dst[i+63:i] := !(a[i+63:i] < b[i+63:i]) ? 0xFFFFFFFFFFFFFFFF : 0 -ENDFOR +dst[63:0] := a[63:0] +dst[79:64] := (a >> (imm8[1:0] * 16))[79:64] +dst[95:80] := (a >> (imm8[3:2] * 16))[79:64] +dst[111:96] := (a >> (imm8[5:4] * 16))[79:64] +dst[127:112] := (a >> (imm8[7:6] * 16))[79:64] +dst[191:128] := a[191:128] +dst[207:192] := (a >> (imm8[1:0] * 16))[207:192] +dst[223:208] := (a >> (imm8[3:2] * 16))[207:192] +dst[239:224] := (a >> (imm8[5:4] * 16))[207:192] +dst[255:240] := (a >> (imm8[7:6] * 16))[207:192] +dst[MAX:256] := 0 - -
emmintrin.h
+ +
immintrin.h
 - - Floating Point - SSE2 - Compare - - - Compare packed double-precision (64-bit) floating-point elements in "a" and "b" for not-less-than-or-equal, and store the results in "dst". + + Integer + AVX2 + Swizzle + + + + Shuffle 16-bit integers in the low 64 bits of 128-bit lanes of "a" using the control in "imm8". Store the results in the low 64 bits of 128-bit lanes of "dst", with the high 64 bits of 128-bit lanes being copied from from "a" to "dst". -FOR j := 0 to 1 - i := j*64 - dst[i+63:i] := !(a[i+63:i] <= b[i+63:i]) ? 0xFFFFFFFFFFFFFFFF : 0 -ENDFOR +dst[15:0] := (a >> (imm8[1:0] * 16))[15:0] +dst[31:16] := (a >> (imm8[3:2] * 16))[15:0] +dst[47:32] := (a >> (imm8[5:4] * 16))[15:0] +dst[63:48] := (a >> (imm8[7:6] * 16))[15:0] +dst[127:64] := a[127:64] +dst[143:128] := (a >> (imm8[1:0] * 16))[143:128] +dst[159:144] := (a >> (imm8[3:2] * 16))[143:128] +dst[175:160] := (a >> (imm8[5:4] * 16))[143:128] +dst[191:176] := (a >> (imm8[7:6] * 16))[143:128] +dst[255:192] := a[255:192] +dst[MAX:256] := 0 - -
emmintrin.h
+ +
immintrin.h
 - - Floating Point - SSE2 - Compare - - - Compare packed double-precision (64-bit) floating-point elements in "a" and "b" for not-greater-than, and store the results in "dst". + + Integer + AVX2 + Arithmetic + + + + Negate packed signed 8-bit integers in "a" when the corresponding signed 8-bit integer in "b" is negative, and store the results in "dst". Element in "dst" are zeroed out when the corresponding element in "b" is zero. -FOR j := 0 to 1 - i := j*64 - dst[i+63:i] := !(a[i+63:i] > b[i+63:i]) ? 0xFFFFFFFFFFFFFFFF : 0 -ENDFOR +FOR j := 0 to 31 + i := j*8 + IF b[i+7:i] < 0 + dst[i+7:i] := -(a[i+7:i]) + ELSE IF b[i+7:i] == 0 + dst[i+7:i] := 0 + ELSE + dst[i+7:i] := a[i+7:i] + FI +ENDFOR +dst[MAX:256] := 0 - -
emmintrin.h
+ +
immintrin.h
 - - Floating Point - SSE2 - Compare - - - Compare packed double-precision (64-bit) floating-point elements in "a" and "b" for not-greater-than-or-equal, and store the results in "dst". + + Integer + AVX2 + Arithmetic + + + + Negate packed signed 16-bit integers in "a" when the corresponding signed 16-bit integer in "b" is negative, and store the results in "dst". Element in "dst" are zeroed out when the corresponding element in "b" is zero. -FOR j := 0 to 1 - i := j*64 - dst[i+63:i] := !(a[i+63:i] >= b[i+63:i]) ? 0xFFFFFFFFFFFFFFFF : 0 -ENDFOR +FOR j := 0 to 15 + i := j*16 + IF b[i+15:i] < 0 + dst[i+15:i] := -(a[i+15:i]) + ELSE IF b[i+15:i] == 0 + dst[i+15:i] := 0 + ELSE + dst[i+15:i] := a[i+15:i] + FI +ENDFOR +dst[MAX:256] := 0 - -
emmintrin.h
+ +
immintrin.h
 - - Floating Point - SSE2 - Compare - - - Compare the lower double-precision (64-bit) floating-point element in "a" and "b" for equality, and return the boolean result (0 or 1). + + Integer + AVX2 + Arithmetic + + + + Negate packed signed 32-bit integers in "a" when the corresponding signed 32-bit integer in "b" is negative, and store the results in "dst". Element in "dst" are zeroed out when the corresponding element in "b" is zero. -RETURN ( a[63:0] == b[63:0] ) ? 1 : 0 +FOR j := 0 to 7 + i := j*32 + IF b[i+31:i] < 0 + dst[i+31:i] := -(a[i+31:i]) + ELSE IF b[i+31:i] == 0 + dst[i+31:i] := 0 + ELSE + dst[i+31:i] := a[i+31:i] + FI +ENDFOR +dst[MAX:256] := 0 - -
emmintrin.h
+ +
immintrin.h
 - - Floating Point - SSE2 - Compare - - - Compare the lower double-precision (64-bit) floating-point element in "a" and "b" for less-than, and return the boolean result (0 or 1). + + Integer + AVX2 + Shift + + + + Shift 128-bit lanes in "a" left by "imm8" bytes while shifting in zeros, and store the results in "dst". -RETURN ( a[63:0] < b[63:0] ) ? 1 : 0 +tmp := imm8[7:0] +IF tmp > 15 + tmp := 16 +FI +dst[127:0] := a[127:0] << (tmp*8) +dst[255:128] := a[255:128] << (tmp*8) +dst[MAX:256] := 0 - -
emmintrin.h
+ +
immintrin.h
 - - Floating Point - SSE2 - Compare - - - Compare the lower double-precision (64-bit) floating-point element in "a" and "b" for less-than-or-equal, and return the boolean result (0 or 1). + + Integer + AVX2 + Shift + + + + Shift 128-bit lanes in "a" left by "imm8" bytes while shifting in zeros, and store the results in "dst". -RETURN ( a[63:0] <= b[63:0] ) ? 1 : 0 +tmp := imm8[7:0] +IF tmp > 15 + tmp := 16 +FI +dst[127:0] := a[127:0] << (tmp*8) +dst[255:128] := a[255:128] << (tmp*8) +dst[MAX:256] := 0 - -
emmintrin.h
+ +
immintrin.h
 - - Floating Point - SSE2 - Compare - - - Compare the lower double-precision (64-bit) floating-point element in "a" and "b" for greater-than, and return the boolean result (0 or 1). + + Integer + AVX2 + Shift + + + + Shift packed 16-bit integers in "a" left by "count" while shifting in zeros, and store the results in "dst". -RETURN ( a[63:0] > b[63:0] ) ? 1 : 0 +FOR j := 0 to 15 + i := j*16 + IF count[63:0] > 15 + dst[i+15:i] := 0 + ELSE + dst[i+15:i] := ZeroExtend16(a[i+15:i] << count[63:0]) + FI +ENDFOR +dst[MAX:256] := 0 - -
emmintrin.h
+ +
immintrin.h
 - - Floating Point - SSE2 - Compare - - - Compare the lower double-precision (64-bit) floating-point element in "a" and "b" for greater-than-or-equal, and return the boolean result (0 or 1). + + Integer + AVX2 + Shift + + + + Shift packed 16-bit integers in "a" left by "imm8" while shifting in zeros, and store the results in "dst". -RETURN ( a[63:0] >= b[63:0] ) ? 1 : 0 +FOR j := 0 to 15 + i := j*16 + IF imm8[7:0] > 15 + dst[i+15:i] := 0 + ELSE + dst[i+15:i] := ZeroExtend16(a[i+15:i] << imm8[7:0]) + FI +ENDFOR +dst[MAX:256] := 0 - -
emmintrin.h
+ +
immintrin.h
 - - Floating Point - SSE2 - Compare - - - Compare the lower double-precision (64-bit) floating-point element in "a" and "b" for not-equal, and return the boolean result (0 or 1). + + Integer + AVX2 + Shift + + + + Shift packed 32-bit integers in "a" left by "count" while shifting in zeros, and store the results in "dst". -RETURN ( a[63:0] != b[63:0] ) ? 1 : 0 +FOR j := 0 to 7 + i := j*32 + IF count[63:0] > 31 + dst[i+31:i] := 0 + ELSE + dst[i+31:i] := ZeroExtend32(a[i+31:i] << count[63:0]) + FI +ENDFOR +dst[MAX:256] := 0 - -
emmintrin.h
+ +
immintrin.h
 - - Floating Point - SSE2 - Compare - - - Compare the lower double-precision (64-bit) floating-point element in "a" and "b" for equality, and return the boolean result (0 or 1). This instruction will not signal an exception for QNaNs. + + Integer + AVX2 + Shift + + + + Shift packed 32-bit integers in "a" left by "imm8" while shifting in zeros, and store the results in "dst". -RETURN ( a[63:0] == b[63:0] ) ? 1 : 0 +FOR j := 0 to 7 + i := j*32 + IF imm8[7:0] > 31 + dst[i+31:i] := 0 + ELSE + dst[i+31:i] := ZeroExtend32(a[i+31:i] << imm8[7:0]) + FI +ENDFOR +dst[MAX:256] := 0 - -
emmintrin.h
+ +
immintrin.h
 - - Floating Point - SSE2 - Compare - - - Compare the lower double-precision (64-bit) floating-point element in "a" and "b" for less-than, and return the boolean result (0 or 1). This instruction will not signal an exception for QNaNs. + + Integer + AVX2 + Shift + + + + Shift packed 64-bit integers in "a" left by "count" while shifting in zeros, and store the results in "dst". -RETURN ( a[63:0] < b[63:0] ) ? 1 : 0 +FOR j := 0 to 3 + i := j*64 + IF count[63:0] > 63 + dst[i+63:i] := 0 + ELSE + dst[i+63:i] := ZeroExtend64(a[i+63:i] << count[63:0]) + FI +ENDFOR +dst[MAX:256] := 0 - -
emmintrin.h
+ +
immintrin.h
 - - Floating Point - SSE2 - Compare - - - Compare the lower double-precision (64-bit) floating-point element in "a" and "b" for less-than-or-equal, and return the boolean result (0 or 1). This instruction will not signal an exception for QNaNs. + + Integer + AVX2 + Shift + + + + Shift packed 64-bit integers in "a" left by "imm8" while shifting in zeros, and store the results in "dst". -RETURN ( a[63:0] <= b[63:0] ) ? 1 : 0 +FOR j := 0 to 3 + i := j*64 + IF imm8[7:0] > 63 + dst[i+63:i] := 0 + ELSE + dst[i+63:i] := ZeroExtend64(a[i+63:i] << imm8[7:0]) + FI +ENDFOR +dst[MAX:256] := 0 - -
emmintrin.h
+ +
immintrin.h
 - - Floating Point - SSE2 - Compare - - - Compare the lower double-precision (64-bit) floating-point element in "a" and "b" for greater-than, and return the boolean result (0 or 1). This instruction will not signal an exception for QNaNs. + + Integer + AVX2 + Shift + + + + Shift packed 32-bit integers in "a" left by the amount specified by the corresponding element in "count" while shifting in zeros, and store the results in "dst". -RETURN ( a[63:0] > b[63:0] ) ? 1 : 0 +FOR j := 0 to 3 + i := j*32 + IF count[i+31:i] < 32 + dst[i+31:i] := ZeroExtend32(a[i+31:i] << count[i+31:i]) + ELSE + dst[i+31:i] := 0 + FI +ENDFOR +dst[MAX:128] := 0 - -
emmintrin.h
+ +
immintrin.h
 - - Floating Point - SSE2 - Compare - - - Compare the lower double-precision (64-bit) floating-point element in "a" and "b" for greater-than-or-equal, and return the boolean result (0 or 1). This instruction will not signal an exception for QNaNs. + + Integer + AVX2 + Shift + + + + Shift packed 32-bit integers in "a" left by the amount specified by the corresponding element in "count" while shifting in zeros, and store the results in "dst". -RETURN ( a[63:0] >= b[63:0] ) ? 1 : 0 +FOR j := 0 to 7 + i := j*32 + IF count[i+31:i] < 32 + dst[i+31:i] := ZeroExtend32(a[i+31:i] << count[i+31:i]) + ELSE + dst[i+31:i] := 0 + FI +ENDFOR +dst[MAX:256] := 0 - -
emmintrin.h
+ +
immintrin.h
 - - Floating Point - SSE2 - Compare - - - Compare the lower double-precision (64-bit) floating-point element in "a" and "b" for not-equal, and return the boolean result (0 or 1). This instruction will not signal an exception for QNaNs. + + Integer + AVX2 + Shift + + + + Shift packed 64-bit integers in "a" left by the amount specified by the corresponding element in "count" while shifting in zeros, and store the results in "dst". -RETURN ( a[63:0] != b[63:0] ) ? 1 : 0 +FOR j := 0 to 1 + i := j*64 + IF count[i+63:i] < 64 + dst[i+63:i] := ZeroExtend64(a[i+63:i] << count[i+63:i]) + ELSE + dst[i+63:i] := 0 + FI +ENDFOR +dst[MAX:128] := 0 - -
emmintrin.h
+ +
immintrin.h
 - - Floating Point - SSE2 - Convert - - Convert packed double-precision (64-bit) floating-point elements in "a" to packed single-precision (32-bit) floating-point elements, and store the results in "dst". + + Integer + AVX2 + Shift + + + + Shift packed 64-bit integers in "a" left by the amount specified by the corresponding element in "count" while shifting in zeros, and store the results in "dst". -FOR j := 0 to 1 - i := 32*j - k := 64*j - dst[i+31:i] := Convert_FP64_To_FP32(a[k+63:k]) +FOR j := 0 to 3 + i := j*64 + IF count[i+63:i] < 64 + dst[i+63:i] := ZeroExtend64(a[i+63:i] << count[i+63:i]) + ELSE + dst[i+63:i] := 0 + FI ENDFOR +dst[MAX:256] := 0 - -
emmintrin.h
+ +
immintrin.h
 - - Floating Point - SSE2 - Convert - - Convert packed single-precision (32-bit) floating-point elements in "a" to packed double-precision (64-bit) floating-point elements, and store the results in "dst". + + Integer + AVX2 + Shift + + + + Shift packed 16-bit integers in "a" right by "count" while shifting in sign bits, and store the results in "dst". -FOR j := 0 to 1 - i := 64*j - k := 32*j - dst[i+63:i] := Convert_FP32_To_FP64(a[k+31:k]) +FOR j := 0 to 15 + i := j*16 + IF count[63:0] > 15 + dst[i+15:i] := (a[i+15] ? 0xFFFF : 0x0) + ELSE + dst[i+15:i] := SignExtend16(a[i+15:i] >> count[63:0]) + FI ENDFOR +dst[MAX:256] := 0 - -
emmintrin.h
+ +
immintrin.h
 - - Floating Point + Integer - SSE2 - Convert - - Convert packed double-precision (64-bit) floating-point elements in "a" to packed 32-bit integers, and store the results in "dst". + AVX2 + Shift + + + + Shift packed 16-bit integers in "a" right by "imm8" while shifting in sign bits, and store the results in "dst". -FOR j := 0 to 1 - i := 32*j - k := 64*j - dst[i+31:i] := Convert_FP64_To_Int32(a[k+63:k]) +FOR j := 0 to 15 + i := j*16 + IF imm8[7:0] > 15 + dst[i+15:i] := (a[i+15] ? 0xFFFF : 0x0) + ELSE + dst[i+15:i] := SignExtend16(a[i+15:i] >> imm8[7:0]) + FI ENDFOR +dst[MAX:256] := 0 - -
emmintrin.h
+ +
immintrin.h
 - - Floating Point + Integer - SSE2 - Convert - - Convert the lower double-precision (64-bit) floating-point element in "a" to a 32-bit integer, and store the result in "dst". + AVX2 + Shift + + + + Shift packed 32-bit integers in "a" right by "count" while shifting in sign bits, and store the results in "dst". -dst[31:0] := Convert_FP64_To_Int32(a[63:0]) +FOR j := 0 to 7 + i := j*32 + IF count[63:0] > 31 + dst[i+31:i] := (a[i+31] ? 0xFFFFFFFF : 0x0) + ELSE + dst[i+31:i] := SignExtend32(a[i+31:i] >> count[63:0]) + FI +ENDFOR +dst[MAX:256] := 0 - -
emmintrin.h
+ +
immintrin.h
 - - Floating Point + Integer - SSE2 - Convert - - Convert the lower double-precision (64-bit) floating-point element in "a" to a 64-bit integer, and store the result in "dst". + AVX2 + Shift + + + + Shift packed 32-bit integers in "a" right by "imm8" while shifting in sign bits, and store the results in "dst". -dst[63:0] := Convert_FP64_To_Int64(a[63:0]) +FOR j := 0 to 7 + i := j*32 + IF imm8[7:0] > 31 + dst[i+31:i] := (a[i+31] ? 0xFFFFFFFF : 0x0) + ELSE + dst[i+31:i] := SignExtend32(a[i+31:i] >> imm8[7:0]) + FI +ENDFOR +dst[MAX:256] := 0 - -
emmintrin.h
+ +
immintrin.h
 - - Floating Point + Integer - SSE2 - Convert - - Convert the lower double-precision (64-bit) floating-point element in "a" to a 64-bit integer, and store the result in "dst". + AVX2 + Shift + + + + Shift packed 32-bit integers in "a" right by the amount specified by the corresponding element in "count" while shifting in sign bits, and store the results in "dst". -dst[63:0] := Convert_FP64_To_Int64(a[63:0]) +FOR j := 0 to 3 + i := j*32 + IF count[i+31:i] < 32 + dst[i+31:i] := SignExtend32(a[i+31:i] >> count[i+31:i]) + ELSE + dst[i+31:i] := (a[i+31] ? 0xFFFFFFFF : 0) + FI +ENDFOR +dst[MAX:128] := 0 - -
emmintrin.h
+ +
immintrin.h
 - - Floating Point - SSE2 - Convert - - - Convert the lower double-precision (64-bit) floating-point element in "b" to a single-precision (32-bit) floating-point element, store the result in the lower element of "dst", and copy the upper element from "a" to the upper element of "dst". - + + Integer + AVX2 + Shift + + + + Shift packed 32-bit integers in "a" right by the amount specified by the corresponding element in "count" while shifting in sign bits, and store the results in "dst". -dst[31:0] := Convert_FP64_To_FP32(b[63:0]) -dst[127:32] := a[127:31] -dst[MAX:64] := 0 +FOR j := 0 to 7 + i := j*32 + IF count[i+31:i] < 32 + dst[i+31:i] := SignExtend32(a[i+31:i] >> count[i+31:i]) + ELSE + dst[i+31:i] := (a[i+31] ? 0xFFFFFFFF : 0) + FI +ENDFOR +dst[MAX:256] := 0 - -
emmintrin.h
+ +
immintrin.h
 - - Floating Point - SSE2 - Convert - - Copy the lower double-precision (64-bit) floating-point element of "a" to "dst". - dst[63:0] := a[63:0] - -
emmintrin.h
+ + Integer + AVX2 + Shift + + + + Shift 128-bit lanes in "a" right by "imm8" bytes while shifting in zeros, and store the results in "dst". + +tmp := imm8[7:0] +IF tmp > 15 + tmp := 16 +FI +dst[127:0] := a[127:0] >> (tmp*8) +dst[255:128] := a[255:128] >> (tmp*8) +dst[MAX:256] := 0 + + +
immintrin.h
 - - Floating Point - SSE2 - Convert - - - Convert the lower single-precision (32-bit) floating-point element in "b" to a double-precision (64-bit) floating-point element, store the result in the lower element of "dst", and copy the upper element from "a" to the upper element of "dst". - + + Integer + AVX2 + Shift + + + + Shift 128-bit lanes in "a" right by "imm8" bytes while shifting in zeros, and store the results in "dst". -dst[63:0] := Convert_FP32_To_FP64(b[31:0]) -dst[127:64] := a[127:64] -dst[MAX:64] := 0 +tmp := imm8[7:0] +IF tmp > 15 + tmp := 16 +FI +dst[127:0] := a[127:0] >> (tmp*8) +dst[255:128] := a[255:128] >> (tmp*8) +dst[MAX:256] := 0 - -
emmintrin.h
+ +
immintrin.h
 - - Floating Point + Integer - SSE2 - Convert - - Convert packed double-precision (64-bit) floating-point elements in "a" to packed 32-bit integers with truncation, and store the results in "dst". + AVX2 + Shift + + + + Shift packed 16-bit integers in "a" right by "count" while shifting in zeros, and store the results in "dst". -FOR j := 0 to 1 - i := 32*j - k := 64*j - dst[i+31:i] := Convert_FP64_To_Int32_Truncate(a[k+63:k]) +FOR j := 0 to 15 + i := j*16 + IF count[63:0] > 15 + dst[i+15:i] := 0 + ELSE + dst[i+15:i] := ZeroExtend16(a[i+15:i] >> count[63:0]) + FI ENDFOR +dst[MAX:256] := 0 - -
emmintrin.h
+ +
immintrin.h
 - - Floating Point + Integer - SSE2 - Convert - - Convert the lower double-precision (64-bit) floating-point element in "a" to a 32-bit integer with truncation, and store the result in "dst". + AVX2 + Shift + + + + Shift packed 16-bit integers in "a" right by "imm8" while shifting in zeros, and store the results in "dst". -dst[31:0] := Convert_FP64_To_Int32_Truncate(a[63:0]) +FOR j := 0 to 15 + i := j*16 + IF imm8[7:0] > 15 + dst[i+15:i] := 0 + ELSE + dst[i+15:i] := ZeroExtend16(a[i+15:i] >> imm8[7:0]) + FI +ENDFOR +dst[MAX:256] := 0 - -
emmintrin.h
+ +
immintrin.h
 - - Floating Point + Integer - SSE2 - Convert - - Convert the lower double-precision (64-bit) floating-point element in "a" to a 64-bit integer with truncation, and store the result in "dst". + AVX2 + Shift + + + + Shift packed 32-bit integers in "a" right by "count" while shifting in zeros, and store the results in "dst". -dst[63:0] := Convert_FP64_To_Int64_Truncate(a[63:0]) +FOR j := 0 to 7 + i := j*32 + IF count[63:0] > 31 + dst[i+31:i] := 0 + ELSE + dst[i+31:i] := ZeroExtend32(a[i+31:i] >> count[63:0]) + FI +ENDFOR +dst[MAX:256] := 0 - -
emmintrin.h
+ +
immintrin.h
 - - Floating Point + Integer - SSE2 - Convert - - Convert the lower double-precision (64-bit) floating-point element in "a" to a 64-bit integer with truncation, and store the result in "dst". + AVX2 + Shift + + + + Shift packed 32-bit integers in "a" right by "imm8" while shifting in zeros, and store the results in "dst". -dst[63:0] := Convert_FP64_To_Int64_Truncate(a[63:0]) +FOR j := 0 to 7 + i := j*32 + IF imm8[7:0] > 31 + dst[i+31:i] := 0 + ELSE + dst[i+31:i] := ZeroExtend32(a[i+31:i] >> imm8[7:0]) + FI +ENDFOR +dst[MAX:256] := 0 - -
emmintrin.h
+ +
immintrin.h
 - - Floating Point + Integer - SSE2 - Convert - - Convert packed single-precision (32-bit) floating-point elements in "a" to packed 32-bit integers, and store the results in "dst". + AVX2 + Shift + + + + Shift packed 64-bit integers in "a" right by "count" while shifting in zeros, and store the results in "dst". FOR j := 0 to 3 - i := 32*j - dst[i+31:i] := Convert_FP32_To_Int32(a[i+31:i]) + i := j*64 + IF count[63:0] > 63 + dst[i+63:i] := 0 + ELSE + dst[i+63:i] := ZeroExtend64(a[i+63:i] >> count[63:0]) + FI ENDFOR +dst[MAX:256] := 0 - -
emmintrin.h
+ +
immintrin.h
 - - Floating Point + Integer - SSE2 - Convert - - Convert packed single-precision (32-bit) floating-point elements in "a" to packed 32-bit integers with truncation, and store the results in "dst". + AVX2 + Shift + + + + Shift packed 64-bit integers in "a" right by "imm8" while shifting in zeros, and store the results in "dst". FOR j := 0 to 3 - i := 32*j - dst[i+31:i] := Convert_FP32_To_Int32_Truncate(a[i+31:i]) + i := j*64 + IF imm8[7:0] > 63 + dst[i+63:i] := 0 + ELSE + dst[i+63:i] := ZeroExtend64(a[i+63:i] >> imm8[7:0]) + FI ENDFOR +dst[MAX:256] := 0 - -
emmintrin.h
+ +
immintrin.h
 - - Floating Point + Integer - SSE2 - Convert - - Convert packed double-precision (64-bit) floating-point elements in "a" to packed 32-bit integers, and store the results in "dst". + AVX2 + Shift + + + + Shift packed 32-bit integers in "a" right by the amount specified by the corresponding element in "count" while shifting in zeros, and store the results in "dst". -FOR j := 0 to 1 - i := 32*j - k := 64*j - dst[i+31:i] := Convert_FP64_To_Int32(a[k+63:k]) +FOR j := 0 to 3 + i := j*32 + IF count[i+31:i] < 32 + dst[i+31:i] := ZeroExtend32(a[i+31:i] >> count[i+31:i]) + ELSE + dst[i+31:i] := 0 + FI ENDFOR +dst[MAX:128] := 0 - -
emmintrin.h
+ +
immintrin.h
 - - Floating Point + Integer - SSE2 - Convert - - Convert packed double-precision (64-bit) floating-point elements in "a" to packed 32-bit integers with truncation, and store the results in "dst". - -FOR j := 0 to 1 - i := 32*j - k := 64*j - dst[i+31:i] := Convert_FP64_To_Int32_Truncate(a[k+63:k]) -ENDFOR - - -
emmintrin.h
- - - Floating Point - SSE2 - Set - - Copy double-precision (64-bit) floating-point element "a" to the lower element of "dst", and zero the upper element. + AVX2 + Shift + + + + Shift packed 32-bit integers in "a" right by the amount specified by the corresponding element in "count" while shifting in zeros, and store the results in "dst". -dst[63:0] := a[63:0] -dst[127:64] := 0 +FOR j := 0 to 7 + i := j*32 + IF count[i+31:i] < 32 + dst[i+31:i] := ZeroExtend32(a[i+31:i] >> count[i+31:i]) + ELSE + dst[i+31:i] := 0 + FI +ENDFOR +dst[MAX:256] := 0 -
emmintrin.h
+ +
immintrin.h
 - - Floating Point - SSE2 - Set - - Broadcast double-precision (64-bit) floating-point value "a" to all elements of "dst". + + Integer + AVX2 + Shift + + + + Shift packed 64-bit integers in "a" right by the amount specified by the corresponding element in "count" while shifting in zeros, and store the results in "dst". FOR j := 0 to 1 i := j*64 - dst[i+63:i] := a[63:0] + IF count[i+63:i] < 64 + dst[i+63:i] := ZeroExtend64(a[i+63:i] >> count[i+63:i]) + ELSE + dst[i+63:i] := 0 + FI ENDFOR +dst[MAX:128] := 0 -
emmintrin.h
+ +
immintrin.h
 - - Floating Point - SSE2 - Set - - Broadcast double-precision (64-bit) floating-point value "a" to all elements of "dst". + + Integer + AVX2 + Shift + + + + Shift packed 64-bit integers in "a" right by the amount specified by the corresponding element in "count" while shifting in zeros, and store the results in "dst". -FOR j := 0 to 1 +FOR j := 0 to 3 i := j*64 - dst[i+63:i] := a[63:0] + IF count[i+63:i] < 64 + dst[i+63:i] := ZeroExtend64(a[i+63:i] >> count[i+63:i]) + ELSE + dst[i+63:i] := 0 + FI ENDFOR +dst[MAX:256] := 0 -
emmintrin.h
- - - Floating Point - SSE2 - Set - - - Set packed double-precision (64-bit) floating-point elements in "dst" with the supplied values. - -dst[63:0] := e0 -dst[127:64] := e1 - -
emmintrin.h
- - - Floating Point - SSE2 - Set - - - Set packed double-precision (64-bit) floating-point elements in "dst" with the supplied values in reverse order. - -dst[63:0] := e1 -dst[127:64] := e0 - -
emmintrin.h
- - - Floating Point - SSE2 - Set - - Return vector of type __m128d with all elements set to zero. - -dst[MAX:0] := 0 - - -
emmintrin.h
- - - Floating Point - SSE2 - Load - - Load 128-bits (composed of 2 packed double-precision (64-bit) floating-point elements) from memory into "dst". - "mem_addr" must be aligned on a 16-byte boundary or a general-protection exception may be generated. - -dst[127:0] := MEM[mem_addr+127:mem_addr] - - -
emmintrin.h
+ +
immintrin.h
 - - Floating Point - SSE2 + + Integer + AVX2 Load - - Load a double-precision (64-bit) floating-point element from memory into both elements of "dst". + + + Load 256-bits of integer data from memory into "dst" using a non-temporal memory hint. + "mem_addr" must be aligned on a 32-byte boundary or a general-protection exception may be generated. -dst[63:0] := MEM[mem_addr+63:mem_addr] -dst[127:64] := MEM[mem_addr+63:mem_addr] +dst[255:0] := MEM[mem_addr+255:mem_addr] +dst[MAX:256] := 0 - -
emmintrin.h
+ +
immintrin.h
 - - - - - Floating Point - SSE2 - Load - - Load a double-precision (64-bit) floating-point element from memory into both elements of "dst". + + Integer + AVX2 + Arithmetic + + + + Subtract packed 8-bit integers in "b" from packed 8-bit integers in "a", and store the results in "dst". -dst[63:0] := MEM[mem_addr+63:mem_addr] -dst[127:64] := MEM[mem_addr+63:mem_addr] +FOR j := 0 to 31 + i := j*8 + dst[i+7:i] := a[i+7:i] - b[i+7:i] +ENDFOR +dst[MAX:256] := 0 - -
emmintrin.h
+ +
immintrin.h
 - - Floating Point - SSE2 - Load - - Load 2 double-precision (64-bit) floating-point elements from memory into "dst" in reverse order. mem_addr must be aligned on a 16-byte boundary or a general-protection exception may be generated. + + Integer + AVX2 + Arithmetic + + + + Subtract packed 16-bit integers in "b" from packed 16-bit integers in "a", and store the results in "dst". -dst[63:0] := MEM[mem_addr+127:mem_addr+64] -dst[127:64] := MEM[mem_addr+63:mem_addr] +FOR j := 0 to 15 + i := j*16 + dst[i+15:i] := a[i+15:i] - b[i+15:i] +ENDFOR +dst[MAX:256] := 0 - -
emmintrin.h
+ +
immintrin.h
 - - Floating Point - SSE2 - Load - - Load 128-bits (composed of 2 packed double-precision (64-bit) floating-point elements) from memory into "dst". - "mem_addr" does not need to be aligned on any particular boundary. + + Integer + AVX2 + Arithmetic + + + + Subtract packed 32-bit integers in "b" from packed 32-bit integers in "a", and store the results in "dst". -dst[127:0] := MEM[mem_addr+127:mem_addr] +FOR j := 0 to 7 + i := j*32 + dst[i+31:i] := a[i+31:i] - b[i+31:i] +ENDFOR +dst[MAX:256] := 0 - -
emmintrin.h
+ +
immintrin.h
 - - Floating Point - SSE2 - Load - - Load a double-precision (64-bit) floating-point element from memory into the lower of "dst", and zero the upper element. "mem_addr" does not need to be aligned on any particular boundary. + + Integer + AVX2 + Arithmetic + + + + Subtract packed 64-bit integers in "b" from packed 64-bit integers in "a", and store the results in "dst". -dst[63:0] := MEM[mem_addr+63:mem_addr] -dst[127:64] := 0 +FOR j := 0 to 3 + i := j*64 + dst[i+63:i] := a[i+63:i] - b[i+63:i] +ENDFOR +dst[MAX:256] := 0 - -
emmintrin.h
+ +
immintrin.h
 - - Floating Point - SSE2 - Load - - - Load a double-precision (64-bit) floating-point element from memory into the upper element of "dst", and copy the lower element from "a" to "dst". "mem_addr" does not need to be aligned on any particular boundary. + + Integer + AVX2 + Arithmetic + + + + Subtract packed signed 8-bit integers in "b" from packed 8-bit integers in "a" using saturation, and store the results in "dst". -dst[63:0] := a[63:0] -dst[127:64] := MEM[mem_addr+63:mem_addr] +FOR j := 0 to 31 + i := j*8 + dst[i+7:i] := Saturate8(a[i+7:i] - b[i+7:i]) +ENDFOR +dst[MAX:256] := 0 - -
emmintrin.h
+ +
immintrin.h
 - - Floating Point - SSE2 - Load - - - Load a double-precision (64-bit) floating-point element from memory into the lower element of "dst", and copy the upper element from "a" to "dst". "mem_addr" does not need to be aligned on any particular boundary. + + Integer + AVX2 + Arithmetic + + + + Subtract packed signed 16-bit integers in "b" from packed 16-bit integers in "a" using saturation, and store the results in "dst". -dst[63:0] := MEM[mem_addr+63:mem_addr] -dst[127:64] := a[127:64] +FOR j := 0 to 15 + i := j*16 + dst[i+15:i] := Saturate16(a[i+15:i] - b[i+15:i]) +ENDFOR +dst[MAX:256] := 0 - -
emmintrin.h
+ +
immintrin.h
 - - Floating Point - SSE2 - Store - - - Store 128-bits (composed of 2 packed double-precision (64-bit) floating-point elements) from "a" into memory using a non-temporal memory hint. - "mem_addr" must be aligned on a 16-byte boundary or a general-protection exception may be generated. + + Integer + AVX2 + Arithmetic + + + + Subtract packed unsigned 8-bit integers in "b" from packed unsigned 8-bit integers in "a" using saturation, and store the results in "dst". -MEM[mem_addr+127:mem_addr] := a[127:0] +FOR j := 0 to 31 + i := j*8 + dst[i+7:i] := SaturateU8(a[i+7:i] - b[i+7:i]) +ENDFOR +dst[MAX:256] := 0 - -
emmintrin.h
+ +
immintrin.h
 - - Floating Point - SSE2 - Store - - - Store the lower double-precision (64-bit) floating-point element from "a" into memory. "mem_addr" does not need to be aligned on any particular boundary. + + Integer + AVX2 + Arithmetic + + + + Subtract packed unsigned 16-bit integers in "b" from packed unsigned 16-bit integers in "a" using saturation, and store the results in "dst". -MEM[mem_addr+63:mem_addr] := a[63:0] +FOR j := 0 to 15 + i := j*16 + dst[i+15:i] := SaturateU16(a[i+15:i] - b[i+15:i]) +ENDFOR +dst[MAX:256] := 0 - -
emmintrin.h
+ +
immintrin.h
 - - Floating Point - SSE2 - Store - - - Store the lower double-precision (64-bit) floating-point element from "a" into 2 contiguous elements in memory. "mem_addr" must be aligned on a 16-byte boundary or a general-protection exception may be generated. + + Integer + AVX2 + Logical + + + + Compute the bitwise XOR of 256 bits (representing integer data) in "a" and "b", and store the result in "dst". -MEM[mem_addr+63:mem_addr] := a[63:0] -MEM[mem_addr+127:mem_addr+64] := a[63:0] +dst[255:0] := (a[255:0] XOR b[255:0]) +dst[MAX:256] := 0 -
emmintrin.h
+ +
immintrin.h
 - - Floating Point - SSE2 - Store - - - Store the lower double-precision (64-bit) floating-point element from "a" into 2 contiguous elements in memory. "mem_addr" must be aligned on a 16-byte boundary or a general-protection exception may be generated. + + Integer + AVX2 + Swizzle + + + + Unpack and interleave 8-bit integers from the high half of each 128-bit lane in "a" and "b", and store the results in "dst". -MEM[mem_addr+63:mem_addr] := a[63:0] -MEM[mem_addr+127:mem_addr+64] := a[63:0] +DEFINE INTERLEAVE_HIGH_BYTES(src1[127:0], src2[127:0]) { + dst[7:0] := src1[71:64] + dst[15:8] := src2[71:64] + dst[23:16] := src1[79:72] + dst[31:24] := src2[79:72] + dst[39:32] := src1[87:80] + dst[47:40] := src2[87:80] + dst[55:48] := src1[95:88] + dst[63:56] := src2[95:88] + dst[71:64] := src1[103:96] + dst[79:72] := src2[103:96] + dst[87:80] := src1[111:104] + dst[95:88] := src2[111:104] + dst[103:96] := src1[119:112] + dst[111:104] := src2[119:112] + dst[119:112] := src1[127:120] + dst[127:120] := src2[127:120] + RETURN dst[127:0] +} +dst[127:0] := INTERLEAVE_HIGH_BYTES(a[127:0], b[127:0]) +dst[255:128] := INTERLEAVE_HIGH_BYTES(a[255:128], b[255:128]) +dst[MAX:256] := 0 -
emmintrin.h
+ +
immintrin.h
 - - Floating Point - SSE2 - Store - - - Store 128-bits (composed of 2 packed double-precision (64-bit) floating-point elements) from "a" into memory. - "mem_addr" must be aligned on a 16-byte boundary or a general-protection exception may be generated. + + Integer + AVX2 + Swizzle + + + + Unpack and interleave 16-bit integers from the high half of each 128-bit lane in "a" and "b", and store the results in "dst". -MEM[mem_addr+127:mem_addr] := a[127:0] +DEFINE INTERLEAVE_HIGH_WORDS(src1[127:0], src2[127:0]) { + dst[15:0] := src1[79:64] + dst[31:16] := src2[79:64] + dst[47:32] := src1[95:80] + dst[63:48] := src2[95:80] + dst[79:64] := src1[111:96] + dst[95:80] := src2[111:96] + dst[111:96] := src1[127:112] + dst[127:112] := src2[127:112] + RETURN dst[127:0] +} +dst[127:0] := INTERLEAVE_HIGH_WORDS(a[127:0], b[127:0]) +dst[255:128] := INTERLEAVE_HIGH_WORDS(a[255:128], b[255:128]) +dst[MAX:256] := 0 - -
emmintrin.h
+ +
immintrin.h
 - - Floating Point - SSE2 - Store - - - Store 128-bits (composed of 2 packed double-precision (64-bit) floating-point elements) from "a" into memory. - "mem_addr" does not need to be aligned on any particular boundary. + + Integer + AVX2 + Swizzle + + + + Unpack and interleave 32-bit integers from the high half of each 128-bit lane in "a" and "b", and store the results in "dst". -MEM[mem_addr+127:mem_addr] := a[127:0] +DEFINE INTERLEAVE_HIGH_DWORDS(src1[127:0], src2[127:0]) { + dst[31:0] := src1[95:64] + dst[63:32] := src2[95:64] + dst[95:64] := src1[127:96] + dst[127:96] := src2[127:96] + RETURN dst[127:0] +} +dst[127:0] := INTERLEAVE_HIGH_DWORDS(a[127:0], b[127:0]) +dst[255:128] := INTERLEAVE_HIGH_DWORDS(a[255:128], b[255:128]) +dst[MAX:256] := 0 - -
emmintrin.h
+ +
immintrin.h
 - - Floating Point - SSE2 - Store - - - Store 2 double-precision (64-bit) floating-point elements from "a" into memory in reverse order. - "mem_addr" must be aligned on a 16-byte boundary or a general-protection exception may be generated. + + Integer + AVX2 + Swizzle + + + + Unpack and interleave 64-bit integers from the high half of each 128-bit lane in "a" and "b", and store the results in "dst". -MEM[mem_addr+63:mem_addr] := a[127:64] -MEM[mem_addr+127:mem_addr+64] := a[63:0] +DEFINE INTERLEAVE_HIGH_QWORDS(src1[127:0], src2[127:0]) { + dst[63:0] := src1[127:64] + dst[127:64] := src2[127:64] + RETURN dst[127:0] +} +dst[127:0] := INTERLEAVE_HIGH_QWORDS(a[127:0], b[127:0]) +dst[255:128] := INTERLEAVE_HIGH_QWORDS(a[255:128], b[255:128]) +dst[MAX:256] := 0 -
emmintrin.h
+ +
immintrin.h
 - - Floating Point - SSE2 - Store - - - Store the upper double-precision (64-bit) floating-point element from "a" into memory. + + Integer + AVX2 + Swizzle + + + + Unpack and interleave 8-bit integers from the low half of each 128-bit lane in "a" and "b", and store the results in "dst". -MEM[mem_addr+63:mem_addr] := a[127:64] +DEFINE INTERLEAVE_BYTES(src1[127:0], src2[127:0]) { + dst[7:0] := src1[7:0] + dst[15:8] := src2[7:0] + dst[23:16] := src1[15:8] + dst[31:24] := src2[15:8] + dst[39:32] := src1[23:16] + dst[47:40] := src2[23:16] + dst[55:48] := src1[31:24] + dst[63:56] := src2[31:24] + dst[71:64] := src1[39:32] + dst[79:72] := src2[39:32] + dst[87:80] := src1[47:40] + dst[95:88] := src2[47:40] + dst[103:96] := src1[55:48] + dst[111:104] := src2[55:48] + dst[119:112] := src1[63:56] + dst[127:120] := src2[63:56] + RETURN dst[127:0] +} +dst[127:0] := INTERLEAVE_BYTES(a[127:0], b[127:0]) +dst[255:128] := INTERLEAVE_BYTES(a[255:128], b[255:128]) +dst[MAX:256] := 0 - -
emmintrin.h
+ +
immintrin.h
 - - Floating Point - SSE2 - Store - - - Store the lower double-precision (64-bit) floating-point element from "a" into memory. + + Integer + AVX2 + Swizzle + + + + Unpack and interleave 16-bit integers from the low half of each 128-bit lane in "a" and "b", and store the results in "dst". -MEM[mem_addr+63:mem_addr] := a[63:0] +DEFINE INTERLEAVE_WORDS(src1[127:0], src2[127:0]) { + dst[15:0] := src1[15:0] + dst[31:16] := src2[15:0] + dst[47:32] := src1[31:16] + dst[63:48] := src2[31:16] + dst[79:64] := src1[47:32] + dst[95:80] := src2[47:32] + dst[111:96] := src1[63:48] + dst[127:112] := src2[63:48] + RETURN dst[127:0] +} +dst[127:0] := INTERLEAVE_WORDS(a[127:0], b[127:0]) +dst[255:128] := INTERLEAVE_WORDS(a[255:128], b[255:128]) +dst[MAX:256] := 0 - -
emmintrin.h
+ +
immintrin.h
 - - Floating Point - SSE2 + + Integer + AVX2 Swizzle - - - Unpack and interleave double-precision (64-bit) floating-point elements from the high half of "a" and "b", and store the results in "dst". + + + + Unpack and interleave 32-bit integers from the low half of each 128-bit lane in "a" and "b", and store the results in "dst". -INTERLEAVE_HIGH_QWORDS(src1[127:0], src2[127:0]){ - dst[63:0] := src1[127:64] - dst[127:64] := src2[127:64] - RETURN dst[127:0] -} - -dst[127:0] := INTERLEAVE_HIGH_QWORDS(a[127:0], b[127:0]) +DEFINE INTERLEAVE_DWORDS(src1[127:0], src2[127:0]) { + dst[31:0] := src1[31:0] + dst[63:32] := src2[31:0] + dst[95:64] := src1[63:32] + dst[127:96] := src2[63:32] + RETURN dst[127:0] +} +dst[127:0] := INTERLEAVE_DWORDS(a[127:0], b[127:0]) +dst[255:128] := INTERLEAVE_DWORDS(a[255:128], b[255:128]) +dst[MAX:256] := 0 - -
emmintrin.h
+ +
immintrin.h
 - - Floating Point - SSE2 + + Integer + AVX2 Swizzle - - - Unpack and interleave double-precision (64-bit) floating-point elements from the low half of "a" and "b", and store the results in "dst". + + + + Unpack and interleave 64-bit integers from the low half of each 128-bit lane in "a" and "b", and store the results in "dst". -INTERLEAVE_QWORDS(src1[127:0], src2[127:0]){ +DEFINE INTERLEAVE_QWORDS(src1[127:0], src2[127:0]) { dst[63:0] := src1[63:0] dst[127:64] := src2[63:0] RETURN dst[127:0] } - dst[127:0] := INTERLEAVE_QWORDS(a[127:0], b[127:0]) +dst[255:128] := INTERLEAVE_QWORDS(a[255:128], b[255:128]) +dst[MAX:256] := 0 - -
emmintrin.h
+ +
immintrin.h
 - - Floating Point - SSE2 + + Mask + AVX512BW Miscellaneous - - Set each bit of mask "dst" based on the most significant bit of the corresponding packed double-precision (64-bit) floating-point element in "a". + + + + Unpack and interleave 32 bits from masks "a" and "b", and store the 64-bit result in "dst". -FOR j := 0 to 1 - i := j*64 - IF a[i+63] - dst[j] := 1 - ELSE - dst[j] := 0 - FI -ENDFOR -dst[MAX:2] := 0 +dst[31:0] := b[31:0] +dst[63:32] := a[31:0] +dst[MAX:64] := 0 - -
emmintrin.h
+ +
immintrin.h
 - - Floating Point - SSE2 - Swizzle - - - - Shuffle double-precision (64-bit) floating-point elements using the control in "imm8", and store the results in "dst". + + Mask + AVX512BW + Miscellaneous + + + + Unpack and interleave 16 bits from masks "a" and "b", and store the 32-bit result in "dst". -dst[63:0] := (imm8[0] == 0) ? a[63:0] : a[127:64] -dst[127:64] := (imm8[1] == 0) ? b[63:0] : b[127:64] +dst[15:0] := b[15:0] +dst[31:16] := a[15:0] +dst[MAX:32] := 0 - -
emmintrin.h
+ +
immintrin.h
 - - Floating Point - SSE2 - Move - - - Move the lower double-precision (64-bit) floating-point element from "b" to the lower element of "dst", and copy the upper element from "a" to the upper element of "dst". + + Integer + AVX512VL + AVX512BW + Miscellaneous + + + + + Compute the sum of absolute differences (SADs) of quadruplets of unsigned 8-bit integers in "a" compared to those in "b", and store the 16-bit results in "dst". + Four SADs are performed on four 8-bit quadruplets for each 64-bit lane. The first two SADs use the lower 8-bit quadruplet of the lane from "a", and the last two SADs use the uppper 8-bit quadruplet of the lane from "a". Quadruplets from "b" are selected from within 128-bit lanes according to the control in "imm8", and each SAD in each 64-bit lane uses the selected quadruplet at 8-bit offsets. -dst[63:0] := b[63:0] -dst[127:64] := a[127:64] +FOR i := 0 to 1 + tmp.m128[i].dword[0] := b.m128[i].dword[ imm8[1:0] ] + tmp.m128[i].dword[1] := b.m128[i].dword[ imm8[3:2] ] + tmp.m128[i].dword[2] := b.m128[i].dword[ imm8[5:4] ] + tmp.m128[i].dword[3] := b.m128[i].dword[ imm8[7:6] ] +ENDFOR +FOR j := 0 to 3 + i := j*64 + dst[i+15:i] := ABS(a[i+7:i] - tmp[i+7:i]) + ABS(a[i+15:i+8] - tmp[i+15:i+8]) +\ + ABS(a[i+23:i+16] - tmp[i+23:i+16]) + ABS(a[i+31:i+24] - tmp[i+31:i+24]) + + dst[i+31:i+16] := ABS(a[i+7:i] - tmp[i+15:i+8]) + ABS(a[i+15:i+8] - tmp[i+23:i+16]) +\ + ABS(a[i+23:i+16] - tmp[i+31:i+24]) + ABS(a[i+31:i+24] - tmp[i+39:i+32]) + + dst[i+47:i+32] := ABS(a[i+39:i+32] - tmp[i+23:i+16]) + ABS(a[i+47:i+40] - tmp[i+31:i+24]) +\ + ABS(a[i+55:i+48] - tmp[i+39:i+32]) + ABS(a[i+63:i+56] - tmp[i+47:i+40]) + + dst[i+63:i+48] := ABS(a[i+39:i+32] - tmp[i+31:i+24]) + ABS(a[i+47:i+40] - tmp[i+39:i+32]) +\ + ABS(a[i+55:i+48] - tmp[i+47:i+40]) + ABS(a[i+63:i+56] - tmp[i+55:i+48]) +ENDFOR +dst[MAX:256] := 0 - -
emmintrin.h
- - - Floating Point - SSE2 - Cast - - Cast vector of type __m128d to type __m128. This intrinsic is only used for compilation and does not generate any instructions, thus it has zero latency. -
emmintrin.h
+ +
immintrin.h
 - - Floating Point + Integer - SSE2 - Cast - - Cast vector of type __m128d to type __m128i. This intrinsic is only used for compilation and does not generate any instructions, thus it has zero latency. -
emmintrin.h
- - - Floating Point - SSE2 - Cast - - Cast vector of type __m128 to type __m128d. This intrinsic is only used for compilation and does not generate any instructions, thus it has zero latency. -
emmintrin.h
+ AVX512VL + AVX512BW + Miscellaneous + + + + + + + Compute the sum of absolute differences (SADs) of quadruplets of unsigned 8-bit integers in "a" compared to those in "b", and store the 16-bit results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Four SADs are performed on four 8-bit quadruplets for each 64-bit lane. The first two SADs use the lower 8-bit quadruplet of the lane from "a", and the last two SADs use the uppper 8-bit quadruplet of the lane from "a". Quadruplets from "b" are selected from within 128-bit lanes according to the control in "imm8", and each SAD in each 64-bit lane uses the selected quadruplet at 8-bit offsets. + +FOR i := 0 to 1 + tmp.m128[i].dword[0] := b.m128[i].dword[ imm8[1:0] ] + tmp.m128[i].dword[1] := b.m128[i].dword[ imm8[3:2] ] + tmp.m128[i].dword[2] := b.m128[i].dword[ imm8[5:4] ] + tmp.m128[i].dword[3] := b.m128[i].dword[ imm8[7:6] ] +ENDFOR +FOR j := 0 to 3 + i := j*64 + tmp_dst[i+15:i] := ABS(a[i+7:i] - tmp[i+7:i]) + ABS(a[i+15:i+8] - tmp[i+15:i+8]) +\ + ABS(a[i+23:i+16] - tmp[i+23:i+16]) + ABS(a[i+31:i+24] - tmp[i+31:i+24]) + + tmp_dst[i+31:i+16] := ABS(a[i+7:i] - tmp[i+15:i+8]) + ABS(a[i+15:i+8] - tmp[i+23:i+16]) +\ + ABS(a[i+23:i+16] - tmp[i+31:i+24]) + ABS(a[i+31:i+24] - tmp[i+39:i+32]) + + tmp_dst[i+47:i+32] := ABS(a[i+39:i+32] - tmp[i+23:i+16]) + ABS(a[i+47:i+40] - tmp[i+31:i+24]) +\ + ABS(a[i+55:i+48] - tmp[i+39:i+32]) + ABS(a[i+63:i+56] - tmp[i+47:i+40]) + + tmp_dst[i+63:i+48] := ABS(a[i+39:i+32] - tmp[i+31:i+24]) + ABS(a[i+47:i+40] - tmp[i+39:i+32]) +\ + ABS(a[i+55:i+48] - tmp[i+47:i+40]) + ABS(a[i+63:i+56] - tmp[i+55:i+48]) +ENDFOR +FOR j := 0 to 15 + i := j*16 + IF k[j] + dst[i+15:i] := tmp_dst[i+15:i] + ELSE + dst[i+15:i] := src[i+15:i] + FI +ENDFOR +dst[MAX:256] := 0 + + +
immintrin.h
 - - Floating Point + Integer - SSE2 - Cast - - Cast vector of type __m128 to type __m128i. This intrinsic is only used for compilation and does not generate any instructions, thus it has zero latency. -
emmintrin.h
- - - Floating Point - SSE2 - Cast - - Cast vector of type __m128i to type __m128d. This intrinsic is only used for compilation and does not generate any instructions, thus it has zero latency. -
emmintrin.h
- - - Floating Point - SSE2 - Cast - - Cast vector of type __m128i to type __m128. This intrinsic is only used for compilation and does not generate any instructions, thus it has zero latency. -
emmintrin.h
+ AVX512VL + AVX512BW + Miscellaneous + + + + + + Compute the sum of absolute differences (SADs) of quadruplets of unsigned 8-bit integers in "a" compared to those in "b", and store the 16-bit results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Four SADs are performed on four 8-bit quadruplets for each 64-bit lane. The first two SADs use the lower 8-bit quadruplet of the lane from "a", and the last two SADs use the uppper 8-bit quadruplet of the lane from "a". Quadruplets from "b" are selected from within 128-bit lanes according to the control in "imm8", and each SAD in each 64-bit lane uses the selected quadruplet at 8-bit offsets. + +FOR i := 0 to 1 + tmp.m128[i].dword[0] := b.m128[i].dword[ imm8[1:0] ] + tmp.m128[i].dword[1] := b.m128[i].dword[ imm8[3:2] ] + tmp.m128[i].dword[2] := b.m128[i].dword[ imm8[5:4] ] + tmp.m128[i].dword[3] := b.m128[i].dword[ imm8[7:6] ] +ENDFOR +FOR j := 0 to 3 + i := j*64 + tmp_dst[i+15:i] := ABS(a[i+7:i] - tmp[i+7:i]) + ABS(a[i+15:i+8] - tmp[i+15:i+8]) +\ + ABS(a[i+23:i+16] - tmp[i+23:i+16]) + ABS(a[i+31:i+24] - tmp[i+31:i+24]) + + tmp_dst[i+31:i+16] := ABS(a[i+7:i] - tmp[i+15:i+8]) + ABS(a[i+15:i+8] - tmp[i+23:i+16]) +\ + ABS(a[i+23:i+16] - tmp[i+31:i+24]) + ABS(a[i+31:i+24] - tmp[i+39:i+32]) + + tmp_dst[i+47:i+32] := ABS(a[i+39:i+32] - tmp[i+23:i+16]) + ABS(a[i+47:i+40] - tmp[i+31:i+24]) +\ + ABS(a[i+55:i+48] - tmp[i+39:i+32]) + ABS(a[i+63:i+56] - tmp[i+47:i+40]) + + tmp_dst[i+63:i+48] := ABS(a[i+39:i+32] - tmp[i+31:i+24]) + ABS(a[i+47:i+40] - tmp[i+39:i+32]) +\ + ABS(a[i+55:i+48] - tmp[i+47:i+40]) + ABS(a[i+63:i+56] - tmp[i+55:i+48]) +ENDFOR +FOR j := 0 to 15 + i := j*16 + IF k[j] + dst[i+15:i] := tmp_dst[i+15:i] + ELSE + dst[i+15:i] := 0 + FI +ENDFOR +dst[MAX:256] := 0 + + +
immintrin.h
 - - Floating Point - Convert - - Convert the half-precision (16-bit) floating-point value "a" to a single-precision (32-bit) floating-point value, and store the result in "dst". + + Integer + AVX512BW + Miscellaneous + + + + + Compute the sum of absolute differences (SADs) of quadruplets of unsigned 8-bit integers in "a" compared to those in "b", and store the 16-bit results in "dst". + Four SADs are performed on four 8-bit quadruplets for each 64-bit lane. The first two SADs use the lower 8-bit quadruplet of the lane from "a", and the last two SADs use the uppper 8-bit quadruplet of the lane from "a". Quadruplets from "b" are selected from within 128-bit lanes according to the control in "imm8", and each SAD in each 64-bit lane uses the selected quadruplet at 8-bit offsets. -dst[31:0] := Convert_FP16_To_FP32(a[15:0]) +FOR i := 0 to 3 + tmp.m128[i].dword[0] := b.m128[i].dword[ imm8[1:0] ] + tmp.m128[i].dword[1] := b.m128[i].dword[ imm8[3:2] ] + tmp.m128[i].dword[2] := b.m128[i].dword[ imm8[5:4] ] + tmp.m128[i].dword[3] := b.m128[i].dword[ imm8[7:6] ] +ENDFOR +FOR j := 0 to 7 + i := j*64 + dst[i+15:i] := ABS(a[i+7:i] - tmp[i+7:i]) + ABS(a[i+15:i+8] - tmp[i+15:i+8]) +\ + ABS(a[i+23:i+16] - tmp[i+23:i+16]) + ABS(a[i+31:i+24] - tmp[i+31:i+24]) + + dst[i+31:i+16] := ABS(a[i+7:i] - tmp[i+15:i+8]) + ABS(a[i+15:i+8] - tmp[i+23:i+16]) +\ + ABS(a[i+23:i+16] - tmp[i+31:i+24]) + ABS(a[i+31:i+24] - tmp[i+39:i+32]) + + dst[i+47:i+32] := ABS(a[i+39:i+32] - tmp[i+23:i+16]) + ABS(a[i+47:i+40] - tmp[i+31:i+24]) +\ + ABS(a[i+55:i+48] - tmp[i+39:i+32]) + ABS(a[i+63:i+56] - tmp[i+47:i+40]) + + dst[i+63:i+48] := ABS(a[i+39:i+32] - tmp[i+31:i+24]) + ABS(a[i+47:i+40] - tmp[i+39:i+32]) +\ + ABS(a[i+55:i+48] - tmp[i+47:i+40]) + ABS(a[i+63:i+56] - tmp[i+55:i+48]) +ENDFOR +dst[MAX:512] := 0 -
emmintrin.h
+ +
immintrin.h
 - - Floating Point - Convert - - - Convert the single-precision (32-bit) floating-point value "a" to a half-precision (16-bit) floating-point value, and store the result in "dst". + + Integer + AVX512BW + Miscellaneous + + + + + + + Compute the sum of absolute differences (SADs) of quadruplets of unsigned 8-bit integers in "a" compared to those in "b", and store the 16-bit results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Four SADs are performed on four 8-bit quadruplets for each 64-bit lane. The first two SADs use the lower 8-bit quadruplet of the lane from "a", and the last two SADs use the uppper 8-bit quadruplet of the lane from "a". Quadruplets from "b" are selected from within 128-bit lanes according to the control in "imm8", and each SAD in each 64-bit lane uses the selected quadruplet at 8-bit offsets. -dst[15:0] := Convert_FP32_To_FP16(a[31:0]) +FOR i := 0 to 3 + tmp.m128[i].dword[0] := b.m128[i].dword[ imm8[1:0] ] + tmp.m128[i].dword[1] := b.m128[i].dword[ imm8[3:2] ] + tmp.m128[i].dword[2] := b.m128[i].dword[ imm8[5:4] ] + tmp.m128[i].dword[3] := b.m128[i].dword[ imm8[7:6] ] +ENDFOR +FOR j := 0 to 7 + i := j*64 + tmp_dst[i+15:i] := ABS(a[i+7:i] - tmp[i+7:i]) + ABS(a[i+15:i+8] - tmp[i+15:i+8]) +\ + ABS(a[i+23:i+16] - tmp[i+23:i+16]) + ABS(a[i+31:i+24] - tmp[i+31:i+24]) + + tmp_dst[i+31:i+16] := ABS(a[i+7:i] - tmp[i+15:i+8]) + ABS(a[i+15:i+8] - tmp[i+23:i+16]) +\ + ABS(a[i+23:i+16] - tmp[i+31:i+24]) + ABS(a[i+31:i+24] - tmp[i+39:i+32]) + + tmp_dst[i+47:i+32] := ABS(a[i+39:i+32] - tmp[i+23:i+16]) + ABS(a[i+47:i+40] - tmp[i+31:i+24]) +\ + ABS(a[i+55:i+48] - tmp[i+39:i+32]) + ABS(a[i+63:i+56] - tmp[i+47:i+40]) + + tmp_dst[i+63:i+48] := ABS(a[i+39:i+32] - tmp[i+31:i+24]) + ABS(a[i+47:i+40] - tmp[i+39:i+32]) +\ + ABS(a[i+55:i+48] - tmp[i+47:i+40]) + ABS(a[i+63:i+56] - tmp[i+55:i+48]) +ENDFOR +FOR j := 0 to 31 + i := j*16 + IF k[j] + dst[i+15:i] := tmp_dst[i+15:i] + ELSE + dst[i+15:i] := src[i+15:i] + FI +ENDFOR +dst[MAX:512] := 0 -
emmintrin.h
+ +
immintrin.h
 - - Floating Point - FP16C - Convert - - Convert packed half-precision (16-bit) floating-point elements in "a" to packed single-precision (32-bit) floating-point elements, and store the results in "dst". + + Integer + AVX512BW + Miscellaneous + + + + + + Compute the sum of absolute differences (SADs) of quadruplets of unsigned 8-bit integers in "a" compared to those in "b", and store the 16-bit results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Four SADs are performed on four 8-bit quadruplets for each 64-bit lane. The first two SADs use the lower 8-bit quadruplet of the lane from "a", and the last two SADs use the uppper 8-bit quadruplet of the lane from "a". Quadruplets from "b" are selected from within 128-bit lanes according to the control in "imm8", and each SAD in each 64-bit lane uses the selected quadruplet at 8-bit offsets. -FOR j := 0 to 3 - i := j*32 - m := j*16 - dst[i+31:i] := Convert_FP16_To_FP32(a[m+15:m]) +FOR i := 0 to 3 + tmp.m128[i].dword[0] := b.m128[i].dword[ imm8[1:0] ] + tmp.m128[i].dword[1] := b.m128[i].dword[ imm8[3:2] ] + tmp.m128[i].dword[2] := b.m128[i].dword[ imm8[5:4] ] + tmp.m128[i].dword[3] := b.m128[i].dword[ imm8[7:6] ] ENDFOR -dst[MAX:128] := 0 +FOR j := 0 to 7 + i := j*64 + tmp_dst[i+15:i] := ABS(a[i+7:i] - tmp[i+7:i]) + ABS(a[i+15:i+8] - tmp[i+15:i+8]) +\ + ABS(a[i+23:i+16] - tmp[i+23:i+16]) + ABS(a[i+31:i+24] - tmp[i+31:i+24]) + + tmp_dst[i+31:i+16] := ABS(a[i+7:i] - tmp[i+15:i+8]) + ABS(a[i+15:i+8] - tmp[i+23:i+16]) +\ + ABS(a[i+23:i+16] - tmp[i+31:i+24]) + ABS(a[i+31:i+24] - tmp[i+39:i+32]) + + tmp_dst[i+47:i+32] := ABS(a[i+39:i+32] - tmp[i+23:i+16]) + ABS(a[i+47:i+40] - tmp[i+31:i+24]) +\ + ABS(a[i+55:i+48] - tmp[i+39:i+32]) + ABS(a[i+63:i+56] - tmp[i+47:i+40]) + + tmp_dst[i+63:i+48] := ABS(a[i+39:i+32] - tmp[i+31:i+24]) + ABS(a[i+47:i+40] - tmp[i+39:i+32]) +\ + ABS(a[i+55:i+48] - tmp[i+47:i+40]) + ABS(a[i+63:i+56] - tmp[i+55:i+48]) +ENDFOR +FOR j := 0 to 31 + i := j*16 + IF k[j] + dst[i+15:i] := tmp_dst[i+15:i] + ELSE + dst[i+15:i] := 0 + FI +ENDFOR +dst[MAX:512] := 0 - -
emmintrin.h
+ +
immintrin.h
 - - Floating Point - FP16C - Convert - - - Convert packed single-precision (32-bit) floating-point elements in "a" to packed half-precision (16-bit) floating-point elements, and store the results in "dst". - [round_note] - + + Integer + AVX512VL + AVX512BW + Miscellaneous + + + + + Compute the sum of absolute differences (SADs) of quadruplets of unsigned 8-bit integers in "a" compared to those in "b", and store the 16-bit results in "dst". + Four SADs are performed on four 8-bit quadruplets for each 64-bit lane. The first two SADs use the lower 8-bit quadruplet of the lane from "a", and the last two SADs use the uppper 8-bit quadruplet of the lane from "a". Quadruplets from "b" are selected according to the control in "imm8", and each SAD in each 64-bit lane uses the selected quadruplet at 8-bit offsets. -FOR j := 0 to 3 - i := 16*j - l := 32*j - dst[i+15:i] := Convert_FP32_To_FP16FP(a[l+31:l]) +tmp.dword[0] := b.dword[ imm8[1:0] ] +tmp.dword[1] := b.dword[ imm8[3:2] ] +tmp.dword[2] := b.dword[ imm8[5:4] ] +tmp.dword[3] := b.dword[ imm8[7:6] ] +FOR j := 0 to 1 + i := j*64 + dst[i+15:i] := ABS(a[i+7:i] - tmp[i+7:i]) + ABS(a[i+15:i+8] - tmp[i+15:i+8]) +\ + ABS(a[i+23:i+16] - tmp[i+23:i+16]) + ABS(a[i+31:i+24] - tmp[i+31:i+24]) + + dst[i+31:i+16] := ABS(a[i+7:i] - tmp[i+15:i+8]) + ABS(a[i+15:i+8] - tmp[i+23:i+16]) +\ + ABS(a[i+23:i+16] - tmp[i+31:i+24]) + ABS(a[i+31:i+24] - tmp[i+39:i+32]) + + dst[i+47:i+32] := ABS(a[i+39:i+32] - tmp[i+23:i+16]) + ABS(a[i+47:i+40] - tmp[i+31:i+24]) +\ + ABS(a[i+55:i+48] - tmp[i+39:i+32]) + ABS(a[i+63:i+56] - tmp[i+47:i+40]) + + dst[i+63:i+48] := ABS(a[i+39:i+32] - tmp[i+31:i+24]) + ABS(a[i+47:i+40] - tmp[i+39:i+32]) +\ + ABS(a[i+55:i+48] - tmp[i+47:i+40]) + ABS(a[i+63:i+56] - tmp[i+55:i+48]) ENDFOR dst[MAX:128] := 0 - -
emmintrin.h
+ +
immintrin.h
 - - - Floating Point - SSE3 - Arithmetic - - - Alternatively add and subtract packed single-precision (32-bit) floating-point elements in "a" to/from packed elements in "b", and store the results in "dst". + + Integer + AVX512VL + AVX512BW + Miscellaneous + + + + + + + Compute the sum of absolute differences (SADs) of quadruplets of unsigned 8-bit integers in "a" compared to those in "b", and store the 16-bit results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Four SADs are performed on four 8-bit quadruplets for each 64-bit lane. The first two SADs use the lower 8-bit quadruplet of the lane from "a", and the last two SADs use the uppper 8-bit quadruplet of the lane from "a". Quadruplets from "b" are selected according to the control in "imm8", and each SAD in each 64-bit lane uses the selected quadruplet at 8-bit offsets. -FOR j := 0 to 3 - i := j*32 - IF (j is even) - dst[i+31:i] := a[i+31:i] - b[i+31:i] +tmp.dword[0] := b.dword[ imm8[1:0] ] +tmp.dword[1] := b.dword[ imm8[3:2] ] +tmp.dword[2] := b.dword[ imm8[5:4] ] +tmp.dword[3] := b.dword[ imm8[7:6] ] +FOR j := 0 to 1 + i := j*64 + tmp_dst[i+15:i] := ABS(a[i+7:i] - tmp[i+7:i]) + ABS(a[i+15:i+8] - tmp[i+15:i+8]) +\ + ABS(a[i+23:i+16] - tmp[i+23:i+16]) + ABS(a[i+31:i+24] - tmp[i+31:i+24]) + + tmp_dst[i+31:i+16] := ABS(a[i+7:i] - tmp[i+15:i+8]) + ABS(a[i+15:i+8] - tmp[i+23:i+16]) +\ + ABS(a[i+23:i+16] - tmp[i+31:i+24]) + ABS(a[i+31:i+24] - tmp[i+39:i+32]) + + tmp_dst[i+47:i+32] := ABS(a[i+39:i+32] - tmp[i+23:i+16]) + ABS(a[i+47:i+40] - tmp[i+31:i+24]) +\ + ABS(a[i+55:i+48] - tmp[i+39:i+32]) + ABS(a[i+63:i+56] - tmp[i+47:i+40]) + + tmp_dst[i+63:i+48] := ABS(a[i+39:i+32] - tmp[i+31:i+24]) + ABS(a[i+47:i+40] - tmp[i+39:i+32]) +\ + ABS(a[i+55:i+48] - tmp[i+47:i+40]) + ABS(a[i+63:i+56] - tmp[i+55:i+48]) +ENDFOR +FOR j := 0 to 7 + i := j*16 + IF k[j] + dst[i+15:i] := tmp_dst[i+15:i] ELSE - dst[i+31:i] := a[i+31:i] + b[i+31:i] + dst[i+15:i] := src[i+15:i] FI ENDFOR +dst[MAX:128] := 0 - -
pmmintrin.h
+ +
immintrin.h
 - - Floating Point - SSE3 - Arithmetic - - - Alternatively add and subtract packed double-precision (64-bit) floating-point elements in "a" to/from packed elements in "b", and store the results in "dst". + + Integer + AVX512VL + AVX512BW + Miscellaneous + + + + + + Compute the sum of absolute differences (SADs) of quadruplets of unsigned 8-bit integers in "a" compared to those in "b", and store the 16-bit results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Four SADs are performed on four 8-bit quadruplets for each 64-bit lane. The first two SADs use the lower 8-bit quadruplet of the lane from "a", and the last two SADs use the uppper 8-bit quadruplet of the lane from "a". Quadruplets from "b" are selected according to the control in "imm8", and each SAD in each 64-bit lane uses the selected quadruplet at 8-bit offsets. +tmp.dword[0] := b.dword[ imm8[1:0] ] +tmp.dword[1] := b.dword[ imm8[3:2] ] +tmp.dword[2] := b.dword[ imm8[5:4] ] +tmp.dword[3] := b.dword[ imm8[7:6] ] FOR j := 0 to 1 i := j*64 - IF (j is even) - dst[i+63:i] := a[i+63:i] - b[i+63:i] + tmp_dst[i+15:i] := ABS(a[i+7:i] - tmp[i+7:i]) + ABS(a[i+15:i+8] - tmp[i+15:i+8]) +\ + ABS(a[i+23:i+16] - tmp[i+23:i+16]) + ABS(a[i+31:i+24] - tmp[i+31:i+24]) + + tmp_dst[i+31:i+16] := ABS(a[i+7:i] - tmp[i+15:i+8]) + ABS(a[i+15:i+8] - tmp[i+23:i+16]) +\ + ABS(a[i+23:i+16] - tmp[i+31:i+24]) + ABS(a[i+31:i+24] - tmp[i+39:i+32]) + + tmp_dst[i+47:i+32] := ABS(a[i+39:i+32] - tmp[i+23:i+16]) + ABS(a[i+47:i+40] - tmp[i+31:i+24]) +\ + ABS(a[i+55:i+48] - tmp[i+39:i+32]) + ABS(a[i+63:i+56] - tmp[i+47:i+40]) + + tmp_dst[i+63:i+48] := ABS(a[i+39:i+32] - tmp[i+31:i+24]) + ABS(a[i+47:i+40] - tmp[i+39:i+32]) +\ + ABS(a[i+55:i+48] - tmp[i+47:i+40]) + ABS(a[i+63:i+56] - tmp[i+55:i+48]) +ENDFOR +FOR j := 0 to 7 + i := j*16 + IF k[j] + dst[i+15:i] := tmp_dst[i+15:i] ELSE - dst[i+63:i] := a[i+63:i] + b[i+63:i] + dst[i+15:i] := 0 FI ENDFOR +dst[MAX:128] := 0 - -
pmmintrin.h
- - - Floating Point - SSE3 - Arithmetic - - - Horizontally add adjacent pairs of double-precision (64-bit) floating-point elements in "a" and "b", and pack the results in "dst". - -dst[63:0] := a[127:64] + a[63:0] -dst[127:64] := b[127:64] + b[63:0] - - -
pmmintrin.h
+ +
immintrin.h
 - - Floating Point - SSE3 - Arithmetic - - - Horizontally add adjacent pairs of single-precision (32-bit) floating-point elements in "a" and "b", and pack the results in "dst". + + Integer + AVX512VL + AVX512BW + Load + + + + + Load packed 16-bit integers from memory into "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + "mem_addr" does not need to be aligned on any particular boundary. -dst[31:0] := a[63:32] + a[31:0] -dst[63:32] := a[127:96] + a[95:64] -dst[95:64] := b[63:32] + b[31:0] -dst[127:96] := b[127:96] + b[95:64] +FOR j := 0 to 15 + i := j*16 + IF k[j] + dst[i+15:i] := MEM[mem_addr+i+15:mem_addr+i] + ELSE + dst[i+15:i] := src[i+15:i] + FI +ENDFOR +dst[MAX:256] := 0 - -
pmmintrin.h
+ +
immintrin.h
 - - Floating Point - SSE3 - Arithmetic - - - Horizontally subtract adjacent pairs of double-precision (64-bit) floating-point elements in "a" and "b", and pack the results in "dst". + + Integer + AVX512VL + AVX512BW + Move + + + + + Move packed 16-bit integers from "a" into "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -dst[63:0] := a[63:0] - a[127:64] -dst[127:64] := b[63:0] - b[127:64] +FOR j := 0 to 15 + i := j*16 + IF k[j] + dst[i+15:i] := a[i+15:i] + ELSE + dst[i+15:i] := src[i+15:i] + FI +ENDFOR +dst[MAX:256] := 0 - -
pmmintrin.h
+ +
immintrin.h
 - - Floating Point - SSE3 - Arithmetic - - - Horizontally add adjacent pairs of single-precision (32-bit) floating-point elements in "a" and "b", and pack the results in "dst". + + Integer + AVX512VL + AVX512BW + Store + + + + + Store packed 16-bit integers from "a" into memory using writemask "k". + "mem_addr" does not need to be aligned on any particular boundary. -dst[31:0] := a[31:0] - a[63:32] -dst[63:32] := a[95:64] - a[127:96] -dst[95:64] := b[31:0] - b[63:32] -dst[127:96] := b[95:64] - b[127:96] +FOR j := 0 to 15 + i := j*16 + IF k[j] + MEM[mem_addr+i+15:mem_addr+i] := a[i+15:i] + FI +ENDFOR - -
pmmintrin.h
+ +
immintrin.h
 - + Integer - SSE3 + AVX512VL + AVX512BW Load - - Load 128-bits of integer data from unaligned memory into "dst". This intrinsic may perform better than "_mm_loadu_si128" when the data crosses a cache line boundary. + + + + Load packed 16-bit integers from memory into "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + "mem_addr" does not need to be aligned on any particular boundary. -dst[127:0] := MEM[mem_addr+127:mem_addr] +FOR j := 0 to 15 + i := j*16 + IF k[j] + dst[i+15:i] := MEM[mem_addr+i+15:mem_addr+i] + ELSE + dst[i+15:i] := 0 + FI +ENDFOR +dst[MAX:256] := 0 - -
pmmintrin.h
- - - MONITOR - General Support - - - - Arm address monitoring hardware using the address specified in "p". A store to an address within the specified address range triggers the monitoring hardware. Specify optional extensions in "extensions", and optional hints in "hints". - -
pmmintrin.h
+ +
immintrin.h
 - - Floating Point - SSE3 + + Integer + AVX512VL + AVX512BW Move - - Duplicate the low double-precision (64-bit) floating-point element from "a", and store the results in "dst". - + + + + Move packed 16-bit integers from "a" into "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -tmp[63:0] := a[63:0] -tmp[127:64] := a[63:0] +FOR j := 0 to 15 + i := j*16 + IF k[j] + dst[i+15:i] := a[i+15:i] + ELSE + dst[i+15:i] := 0 + FI +ENDFOR +dst[MAX:256] := 0 - -
pmmintrin.h
+ +
immintrin.h
 - - Floating Point - SSE3 + + Integer + AVX512BW Load - - Load a double-precision (64-bit) floating-point element from memory into both elements of "dst". - - -tmp[63:0] := MEM[mem_addr+63:mem_addr] -tmp[127:64] := MEM[mem_addr+63:mem_addr] - - -
pmmintrin.h
- - - Floating Point - SSE3 - Move - - Duplicate odd-indexed single-precision (32-bit) floating-point elements from "a", and store the results in "dst". - + + + + + Load packed 16-bit integers from memory into "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + "mem_addr" does not need to be aligned on any particular boundary. -dst[31:0] := a[63:32] -dst[63:32] := a[63:32] -dst[95:64] := a[127:96] -dst[127:96] := a[127:96] +FOR j := 0 to 31 + i := j*16 + IF k[j] + dst[i+15:i] := MEM[mem_addr+i+15:mem_addr+i] + ELSE + dst[i+15:i] := src[i+15:i] + FI +ENDFOR +dst[MAX:512] := 0 - -
pmmintrin.h
+ +
immintrin.h
 - - Floating Point - SSE3 + + Integer + AVX512BW Move - - Duplicate even-indexed single-precision (32-bit) floating-point elements from "a", and store the results in "dst". - + + + + + Move packed 16-bit integers from "a" into "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -dst[31:0] := a[31:0] -dst[63:32] := a[31:0] -dst[95:64] := a[95:64] -dst[127:96] := a[95:64] +FOR j := 0 to 31 + i := j*16 + IF k[j] + dst[i+15:i] := a[i+15:i] + ELSE + dst[i+15:i] := src[i+15:i] + FI +ENDFOR +dst[MAX:512] := 0 - -
pmmintrin.h
- - - MONITOR - General Support - - - Hint to the processor that it can enter an implementation-dependent-optimized state while waiting for an event or store operation to the address range specified by MONITOR. - -
pmmintrin.h
+ +
immintrin.h
 - - + Integer - SSSE3 - Special Math Functions - - Compute the absolute value of packed 8-bit integers in "a", and store the unsigned results in "dst". + AVX512BW + Store + + + + + Store packed 16-bit integers from "a" into memory using writemask "k". + "mem_addr" does not need to be aligned on any particular boundary. -FOR j := 0 to 7 - i := j*8 - dst[i+7:i] := ABS(a[i+7:i]) +FOR j := 0 to 31 + i := j*16 + IF k[j] + MEM[mem_addr+i+15:mem_addr+i] := a[i+15:i] + FI ENDFOR - -
tmmintrin.h
+ +
immintrin.h
 - + Integer - SSSE3 - Special Math Functions - - Compute the absolute value of packed 8-bit integers in "a", and store the unsigned results in "dst". + AVX512BW + Load + + + + Load packed 16-bit integers from memory into "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + "mem_addr" does not need to be aligned on any particular boundary. -FOR j := 0 to 15 - i := j*8 - dst[i+7:i] := ABS(a[i+7:i]) +FOR j := 0 to 31 + i := j*16 + IF k[j] + dst[i+15:i] := MEM[mem_addr+i+15:mem_addr+i] + ELSE + dst[i+15:i] := 0 + FI ENDFOR +dst[MAX:512] := 0 - -
tmmintrin.h
+ +
immintrin.h
 - + Integer - SSSE3 - Special Math Functions - - Compute the absolute value of packed 16-bit integers in "a", and store the unsigned results in "dst". + AVX512BW + Move + + + + Move packed 16-bit integers from "a" into "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 3 +FOR j := 0 to 31 i := j*16 - dst[i+15:i] := ABS(a[i+15:i]) + IF k[j] + dst[i+15:i] := a[i+15:i] + ELSE + dst[i+15:i] := 0 + FI ENDFOR +dst[MAX:512] := 0 - -
tmmintrin.h
+ +
immintrin.h
 - + Integer - SSSE3 - Special Math Functions - - Compute the absolute value of packed 16-bit integers in "a", and store the unsigned results in "dst". + AVX512VL + AVX512BW + Load + + + + + Load packed 16-bit integers from memory into "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + "mem_addr" does not need to be aligned on any particular boundary. FOR j := 0 to 7 i := j*16 - dst[i+15:i] := ABS(a[i+15:i]) + IF k[j] + dst[i+15:i] := MEM[mem_addr+i+15:mem_addr+i] + ELSE + dst[i+15:i] := src[i+15:i] + FI ENDFOR +dst[MAX:128] := 0 - -
tmmintrin.h
+ +
immintrin.h
 - + Integer - SSSE3 - Special Math Functions - - Compute the absolute value of packed 32-bit integers in "a", and store the unsigned results in "dst". + AVX512VL + AVX512BW + Move + + + + + Move packed 16-bit integers from "a" into "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 1 - i := j*32 - dst[i+31:i] := ABS(a[i+31:i]) +FOR j := 0 to 7 + i := j*16 + IF k[j] + dst[i+15:i] := a[i+15:i] + ELSE + dst[i+15:i] := src[i+15:i] + FI ENDFOR +dst[MAX:128] := 0 - -
tmmintrin.h
+ +
immintrin.h
 - + Integer - SSSE3 - Special Math Functions - - Compute the absolute value of packed 32-bit integers in "a", and store the unsigned results in "dst". + AVX512VL + AVX512BW + Store + + + + + Store packed 16-bit integers from "a" into memory using writemask "k". + "mem_addr" does not need to be aligned on any particular boundary. -FOR j := 0 to 3 - i := j*32 - dst[i+31:i] := ABS(a[i+31:i]) +FOR j := 0 to 7 + i := j*16 + IF k[j] + MEM[mem_addr+i+15:mem_addr+i] := a[i+15:i] + FI ENDFOR - -
tmmintrin.h
+ +
immintrin.h
 - + Integer - SSSE3 - Swizzle - - - Shuffle packed 8-bit integers in "a" according to shuffle control mask in the corresponding 8-bit element of "b", and store the results in "dst". + AVX512VL + AVX512BW + Load + + + + Load packed 16-bit integers from memory into "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + "mem_addr" does not need to be aligned on any particular boundary. -FOR j := 0 to 15 - i := j*8 - IF b[i+7] == 1 - dst[i+7:i] := 0 +FOR j := 0 to 7 + i := j*16 + IF k[j] + dst[i+15:i] := MEM[mem_addr+i+15:mem_addr+i] ELSE - index[3:0] := b[i+3:i] - dst[i+7:i] := a[index*8+7:index*8] + dst[i+15:i] := 0 FI ENDFOR +dst[MAX:128] := 0 - -
tmmintrin.h
+ +
immintrin.h
 - + Integer - SSSE3 - Swizzle - - - Shuffle packed 8-bit integers in "a" according to shuffle control mask in the corresponding 8-bit element of "b", and store the results in "dst". + AVX512VL + AVX512BW + Move + + + + Move packed 16-bit integers from "a" into "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). FOR j := 0 to 7 - i := j*8 - IF b[i+7] == 1 - dst[i+7:i] := 0 + i := j*16 + IF k[j] + dst[i+15:i] := a[i+15:i] ELSE - index[2:0] := b[i+2:i] - dst[i+7:i] := a[index*8+7:index*8] + dst[i+15:i] := 0 FI ENDFOR +dst[MAX:128] := 0 - -
tmmintrin.h
+ +
immintrin.h
 - + Integer - SSSE3 - Miscellaneous - - - - Concatenate 16-byte blocks in "a" and "b" into a 32-byte temporary result, shift the result right by "count" bytes, and store the low 16 bytes in "dst". + AVX512VL + AVX512BW + Load + + + + + Load packed 8-bit integers from memory into "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + "mem_addr" does not need to be aligned on any particular boundary. -tmp[255:0] := ((a[127:0] << 128) OR b[127:0]) >> (count[7:0]*8) -dst[127:0] := tmp[127:0] +FOR j := 0 to 31 + i := j*8 + IF k[j] + dst[i+7:i] := MEM[mem_addr+i+7:mem_addr+i] + ELSE + dst[i+7:i] := src[i+7:i] + FI +ENDFOR +dst[MAX:256] := 0 - -
tmmintrin.h
+ +
immintrin.h
 - + Integer - SSSE3 - Miscellaneous - - - - Concatenate 8-byte blocks in "a" and "b" into a 16-byte temporary result, shift the result right by "count" bytes, and store the low 16 bytes in "dst". + AVX512VL + AVX512BW + Move + + + + + Move packed 8-bit integers from "a" into "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -tmp[127:0] := ((a[63:0] << 64) OR b[63:0]) >> (count[7:0]*8) -dst[63:0] := tmp[63:0] - - -
tmmintrin.h
- - +FOR j := 0 to 31 + i := j*8 + IF k[j] + dst[i+7:i] := a[i+7:i] + ELSE + dst[i+7:i] := src[i+7:i] + FI +ENDFOR +dst[MAX:256] := 0 + + +
immintrin.h
+ + Integer - SSSE3 - Arithmetic - - - Horizontally add adjacent pairs of 16-bit integers in "a" and "b", and pack the signed 16-bit results in "dst". + AVX512VL + AVX512BW + Store + + + + + Store packed 8-bit integers from "a" into memory using writemask "k". + "mem_addr" does not need to be aligned on any particular boundary. -dst[15:0] := a[31:16] + a[15:0] -dst[31:16] := a[63:48] + a[47:32] -dst[47:32] := a[95:80] + a[79:64] -dst[63:48] := a[127:112] + a[111:96] -dst[79:64] := b[31:16] + b[15:0] -dst[95:80] := b[63:48] + b[47:32] -dst[111:96] := b[95:80] + b[79:64] -dst[127:112] := b[127:112] + b[111:96] +FOR j := 0 to 31 + i := j*8 + IF k[j] + MEM[mem_addr+i+7:mem_addr+i] := a[i+7:i] + FI +ENDFOR - -
tmmintrin.h
+ +
immintrin.h
 - + Integer - SSSE3 - Arithmetic - - - Horizontally add adjacent pairs of 16-bit integers in "a" and "b" using saturation, and pack the signed 16-bit results in "dst". + AVX512VL + AVX512BW + Load + + + + Load packed 8-bit integers from memory into "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + "mem_addr" does not need to be aligned on any particular boundary. -dst[15:0]= Saturate_To_Int16(a[31:16] + a[15:0]) -dst[31:16] = Saturate_To_Int16(a[63:48] + a[47:32]) -dst[47:32] = Saturate_To_Int16(a[95:80] + a[79:64]) -dst[63:48] = Saturate_To_Int16(a[127:112] + a[111:96]) -dst[79:64] = Saturate_To_Int16(b[31:16] + b[15:0]) -dst[95:80] = Saturate_To_Int16(b[63:48] + b[47:32]) -dst[111:96] = Saturate_To_Int16(b[95:80] + b[79:64]) -dst[127:112] = Saturate_To_Int16(b[127:112] + b[111:96]) +FOR j := 0 to 31 + i := j*8 + IF k[j] + dst[i+7:i] := MEM[mem_addr+i+7:mem_addr+i] + ELSE + dst[i+7:i] := 0 + FI +ENDFOR +dst[MAX:256] := 0 - -
tmmintrin.h
+ +
immintrin.h
 - + Integer - SSSE3 - Arithmetic - - - Horizontally add adjacent pairs of 32-bit integers in "a" and "b", and pack the signed 32-bit results in "dst". + AVX512VL + AVX512BW + Move + + + + Move packed 8-bit integers from "a" into "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -dst[31:0] := a[63:32] + a[31:0] -dst[63:32] := a[127:96] + a[95:64] -dst[95:64] := b[63:32] + b[31:0] -dst[127:96] := b[127:96] + b[95:64] +FOR j := 0 to 31 + i := j*8 + IF k[j] + dst[i+7:i] := a[i+7:i] + ELSE + dst[i+7:i] := 0 + FI +ENDFOR +dst[MAX:256] := 0 - -
tmmintrin.h
+ +
immintrin.h
 - + Integer - SSSE3 - Arithmetic - - - Horizontally add adjacent pairs of 16-bit integers in "a" and "b", and pack the signed 16-bit results in "dst". + AVX512BW + Load + + + + + Load packed 8-bit integers from memory into "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + "mem_addr" does not need to be aligned on any particular boundary. -dst[15:0] := a[31:16] + a[15:0] -dst[31:16] := a[63:48] + a[47:32] -dst[47:32] := b[31:16] + b[15:0] -dst[63:48] := b[63:48] + b[47:32] +FOR j := 0 to 63 + i := j*8 + IF k[j] + dst[i+7:i] := MEM[mem_addr+i+7:mem_addr+i] + ELSE + dst[i+7:i] := src[i+7:i] + FI +ENDFOR +dst[MAX:512] := 0 - -
tmmintrin.h
+ +
immintrin.h
 - + Integer - SSSE3 - Arithmetic - - - Horizontally add adjacent pairs of 32-bit integers in "a" and "b", and pack the signed 32-bit results in "dst". + AVX512BW + Move + + + + + Move packed 8-bit integers from "a" into "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -dst[31:0] := a[63:32] + a[31:0] -dst[63:32] := b[63:32] + b[31:0] +FOR j := 0 to 63 + i := j*8 + IF k[j] + dst[i+7:i] := a[i+7:i] + ELSE + dst[i+7:i] := src[i+7:i] + FI +ENDFOR +dst[MAX:512] := 0 - -
tmmintrin.h
+ +
immintrin.h
 - + Integer - SSSE3 - Arithmetic - - - Horizontally add adjacent pairs of 16-bit integers in "a" and "b" using saturation, and pack the signed 16-bit results in "dst". + AVX512BW + Store + + + + + Store packed 8-bit integers from "a" into memory using writemask "k". + "mem_addr" does not need to be aligned on any particular boundary. -dst[15:0]= Saturate_To_Int16(a[31:16] + a[15:0]) -dst[31:16] = Saturate_To_Int16(a[63:48] + a[47:32]) -dst[47:32] = Saturate_To_Int16(b[31:16] + b[15:0]) -dst[63:48] = Saturate_To_Int16(b[63:48] + b[47:32]) +FOR j := 0 to 63 + i := j*8 + IF k[j] + MEM[mem_addr+i+7:mem_addr+i] := a[i+7:i] + FI +ENDFOR - -
tmmintrin.h
+ +
immintrin.h
 - + Integer - SSSE3 - Arithmetic - - - Horizontally subtract adjacent pairs of 16-bit integers in "a" and "b", and pack the signed 16-bit results in "dst". + AVX512BW + Load + + + + Load packed 8-bit integers from memory into "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + "mem_addr" does not need to be aligned on any particular boundary. -dst[15:0] := a[15:0] - a[31:16] -dst[31:16] := a[47:32] - a[63:48] -dst[47:32] := a[79:64] - a[95:80] -dst[63:48] := a[111:96] - a[127:112] -dst[79:64] := b[15:0] - b[31:16] -dst[95:80] := b[47:32] - b[63:48] -dst[111:96] := b[79:64] - b[95:80] -dst[127:112] := b[111:96] - b[127:112] +FOR j := 0 to 63 + i := j*8 + IF k[j] + dst[i+7:i] := MEM[mem_addr+i+7:mem_addr+i] + ELSE + dst[i+7:i] := 0 + FI +ENDFOR +dst[MAX:512] := 0 - -
tmmintrin.h
+ +
immintrin.h
 - + Integer - SSSE3 - Arithmetic - - - Horizontally subtract adjacent pairs of 16-bit integers in "a" and "b" using saturation, and pack the signed 16-bit results in "dst". + AVX512BW + Move + + + + Move packed 8-bit integers from "a" into "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -dst[15:0]= Saturate_To_Int16(a[15:0] - a[31:16]) -dst[31:16] = Saturate_To_Int16(a[47:32] - a[63:48]) -dst[47:32] = Saturate_To_Int16(a[79:64] - a[95:80]) -dst[63:48] = Saturate_To_Int16(a[111:96] - a[127:112]) -dst[79:64] = Saturate_To_Int16(b[15:0] - b[31:16]) -dst[95:80] = Saturate_To_Int16(b[47:32] - b[63:48]) -dst[111:96] = Saturate_To_Int16(b[79:64] - b[95:80]) -dst[127:112] = Saturate_To_Int16(b[111:96] - b[127:112]) +FOR j := 0 to 63 + i := j*8 + IF k[j] + dst[i+7:i] := a[i+7:i] + ELSE + dst[i+7:i] := 0 + FI +ENDFOR +dst[MAX:512] := 0 - -
tmmintrin.h
+ +
immintrin.h
 - + Integer - SSSE3 - Arithmetic - - - Horizontally subtract adjacent pairs of 32-bit integers in "a" and "b", and pack the signed 32-bit results in "dst". + AVX512VL + AVX512BW + Load + + + + + Load packed 8-bit integers from memory into "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + "mem_addr" does not need to be aligned on any particular boundary. -dst[31:0] := a[31:0] - a[63:32] -dst[63:32] := a[95:64] - a[127:96] -dst[95:64] := b[31:0] - b[63:32] -dst[127:96] := b[95:64] - b[127:96] +FOR j := 0 to 15 + i := j*8 + IF k[j] + dst[i+7:i] := MEM[mem_addr+i+7:mem_addr+i] + ELSE + dst[i+7:i] := src[i+7:i] + FI +ENDFOR +dst[MAX:128] := 0 - -
tmmintrin.h
+ +
immintrin.h
 - + Integer - SSSE3 - Arithmetic - - - Horizontally subtract adjacent pairs of 16-bit integers in "a" and "b", and pack the signed 16-bit results in "dst". + AVX512VL + AVX512BW + Move + + + + + Move packed 8-bit integers from "a" into "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -dst[15:0] := a[15:0] - a[31:16] -dst[31:16] := a[47:32] - a[63:48] -dst[47:32] := b[15:0] - b[31:16] -dst[63:48] := b[47:32] - b[63:48] +FOR j := 0 to 15 + i := j*8 + IF k[j] + dst[i+7:i] := a[i+7:i] + ELSE + dst[i+7:i] := src[i+7:i] + FI +ENDFOR +dst[MAX:128] := 0 - -
tmmintrin.h
+ +
immintrin.h
 - + Integer - SSSE3 - Arithmetic - - - Horizontally subtract adjacent pairs of 32-bit integers in "a" and "b", and pack the signed 32-bit results in "dst". + AVX512VL + AVX512BW + Store + + + + + Store packed 8-bit integers from "a" into memory using writemask "k". + "mem_addr" does not need to be aligned on any particular boundary. -dst[31:0] := a[31:0] - a[63:32] -dst[63:32] := b[31:0] - b[63:32] +FOR j := 0 to 15 + i := j*8 + IF k[j] + MEM[mem_addr+i+7:mem_addr+i] := a[i+7:i] + FI +ENDFOR - -
tmmintrin.h
+ +
immintrin.h
 - + Integer - SSSE3 - Arithmetic - - - Horizontally subtract adjacent pairs of 16-bit integers in "a" and "b" using saturation, and pack the signed 16-bit results in "dst". + AVX512VL + AVX512BW + Load + + + + Load packed 8-bit integers from memory into "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + "mem_addr" does not need to be aligned on any particular boundary. -dst[15:0]= Saturate_To_Int16(a[15:0] - a[31:16]) -dst[31:16] = Saturate_To_Int16(a[47:32] - a[63:48]) -dst[47:32] = Saturate_To_Int16(b[15:0] - b[31:16]) -dst[63:48] = Saturate_To_Int16(b[47:32] - b[63:48]) +FOR j := 0 to 15 + i := j*8 + IF k[j] + dst[i+7:i] := MEM[mem_addr+i+7:mem_addr+i] + ELSE + dst[i+7:i] := 0 + FI +ENDFOR +dst[MAX:128] := 0 - -
tmmintrin.h
+ +
immintrin.h
 - + Integer - SSSE3 - Arithmetic - - - Vertically multiply each unsigned 8-bit integer from "a" with the corresponding signed 8-bit integer from "b", producing intermediate signed 16-bit integers. Horizontally add adjacent pairs of intermediate signed 16-bit integers, and pack the saturated results in "dst". - + AVX512VL + AVX512BW + Move + + + + Move packed 8-bit integers from "a" into "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := j*16 - dst[i+15:i] := Saturate_To_Int16( a[i+15:i+8]*b[i+15:i+8] + a[i+7:i]*b[i+7:i] ) +FOR j := 0 to 15 + i := j*8 + IF k[j] + dst[i+7:i] := a[i+7:i] + ELSE + dst[i+7:i] := 0 + FI ENDFOR +dst[MAX:128] := 0 - -
tmmintrin.h
+ +
immintrin.h
 - + Integer - SSSE3 + AVX512VL + AVX512BW Arithmetic - - - Vertically multiply each unsigned 8-bit integer from "a" with the corresponding signed 8-bit integer from "b", producing intermediate signed 16-bit integers. Horizontally add adjacent pairs of intermediate signed 16-bit integers, and pack the saturated results in "dst". - + + + + + Compute the absolute value of packed signed 8-bit integers in "a", and store the unsigned results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*16 - dst[i+15:i] := Saturate_To_Int16( a[i+15:i+8]*b[i+15:i+8] + a[i+7:i]*b[i+7:i] ) +FOR j := 0 to 31 + i := j*8 + IF k[j] + dst[i+7:i] := ABS(a[i+7:i]) + ELSE + dst[i+7:i] := src[i+7:i] + FI ENDFOR +dst[MAX:256] := 0 - -
tmmintrin.h
+ +
immintrin.h
 - + Integer - SSSE3 + AVX512VL + AVX512BW Arithmetic - - - Multiply packed 16-bit integers in "a" and "b", producing intermediate signed 32-bit integers. Truncate each intermediate integer to the 18 most significant bits, round by adding 1, and store bits [16:1] to "dst". + + + + Compute the absolute value of packed signed 8-bit integers in "a", and store the unsigned results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := j*16 - tmp[31:0] := ((a[i+15:i] * b[i+15:i]) >> 14) + 1 - dst[i+15:i] := tmp[16:1] +FOR j := 0 to 31 + i := j*8 + IF k[j] + dst[i+7:i] := ABS(a[i+7:i]) + ELSE + dst[i+7:i] := 0 + FI ENDFOR +dst[MAX:256] := 0 - -
tmmintrin.h
+ +
immintrin.h
 - + Integer - SSSE3 + AVX512BW Arithmetic - - - Multiply packed 16-bit integers in "a" and "b", producing intermediate signed 32-bit integers. Truncate each intermediate integer to the 18 most significant bits, round by adding 1, and store bits [16:1] to "dst". + + + Compute the absolute value of packed signed 8-bit integers in "a", and store the unsigned results in "dst". -FOR j := 0 to 3 - i := j*16 - tmp[31:0] := ((a[i+15:i] * b[i+15:i]) >> 14) + 1 - dst[i+15:i] := tmp[16:1] +FOR j := 0 to 63 + i := j*8 + dst[i+7:i] := ABS(a[i+7:i]) ENDFOR +dst[MAX:512] := 0 - -
tmmintrin.h
+ +
immintrin.h
 - + Integer - SSSE3 + AVX512BW Arithmetic - - - Negate packed 8-bit integers in "a" when the corresponding signed 8-bit integer in "b" is negative, and store the results in "dst". Element in "dst" are zeroed out when the corresponding element in "b" is zero. + + + + + Compute the absolute value of packed signed 8-bit integers in "a", and store the unsigned results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 15 +FOR j := 0 to 63 i := j*8 - IF b[i+7:i] < 0 - dst[i+7:i] := NEG(a[i+7:i]) - ELSE IF b[i+7:i] = 0 - dst[i+7:i] := 0 + IF k[j] + dst[i+7:i] := ABS(a[i+7:i]) ELSE - dst[i+7:i] := a[i+7:i] + dst[i+7:i] := src[i+7:i] FI ENDFOR +dst[MAX:512] := 0 - -
tmmintrin.h
+ +
immintrin.h
 - + Integer - SSSE3 + AVX512BW Arithmetic - - - Negate packed 16-bit integers in "a" when the corresponding signed 16-bit integer in "b" is negative, and store the results in "dst". Element in "dst" are zeroed out when the corresponding element in "b" is zero. + + + + Compute the absolute value of packed signed 8-bit integers in "a", and store the unsigned results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := j*16 - IF b[i+15:i] < 0 - dst[i+15:i] := NEG(a[i+15:i]) - ELSE IF b[i+15:i] = 0 - dst[i+15:i] := 0 +FOR j := 0 to 63 + i := j*8 + IF k[j] + dst[i+7:i] := ABS(a[i+7:i]) ELSE - dst[i+15:i] := a[i+15:i] + dst[i+7:i] := 0 FI ENDFOR +dst[MAX:512] := 0 - -
tmmintrin.h
+ +
immintrin.h
 - + Integer - SSSE3 + AVX512VL + AVX512BW Arithmetic - - - Negate packed 32-bit integers in "a" when the corresponding signed 32-bit integer in "b" is negative, and store the results in "dst". Element in "dst" are zeroed out when the corresponding element in "b" is zero. + + + + + Compute the absolute value of packed signed 8-bit integers in "a", and store the unsigned results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*32 - IF b[i+31:i] < 0 - dst[i+31:i] := NEG(a[i+31:i]) - ELSE IF b[i+31:i] = 0 - dst[i+31:i] := 0 +FOR j := 0 to 15 + i := j*8 + IF k[j] + dst[i+7:i] := ABS(a[i+7:i]) ELSE - dst[i+31:i] := a[i+31:i] + dst[i+7:i] := src[i+7:i] FI ENDFOR +dst[MAX:128] := 0 - -
tmmintrin.h
+ +
immintrin.h
 - + Integer - SSSE3 + AVX512VL + AVX512BW Arithmetic - - - Negate packed 8-bit integers in "a" when the corresponding signed 8-bit integer in "b" is negative, and store the results in "dst". Element in "dst" are zeroed out when the corresponding element in "b" is zero. + + + + Compute the absolute value of packed signed 8-bit integers in "a", and store the unsigned results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 +FOR j := 0 to 15 i := j*8 - IF b[i+7:i] < 0 - dst[i+7:i] := NEG(a[i+7:i]) - ELSE IF b[i+7:i] = 0 - dst[i+7:i] := 0 + IF k[j] + dst[i+7:i] := ABS(a[i+7:i]) ELSE - dst[i+7:i] := a[i+7:i] + dst[i+7:i] := 0 FI ENDFOR +dst[MAX:128] := 0 - -
tmmintrin.h
+ +
immintrin.h
 - + Integer - SSSE3 + AVX512VL + AVX512BW Arithmetic - - - Negate packed 16-bit integers in "a" when the corresponding signed 16-bit integer in "b" is negative, and store the results in "dst". Element in "dst" are zeroed out when the corresponding element in "b" is zero. + + + + + Compute the absolute value of packed signed 16-bit integers in "a", and store the unsigned results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 3 +FOR j := 0 to 15 i := j*16 - IF b[i+15:i] < 0 - dst[i+15:i] := NEG(a[i+15:i]) - ELSE IF b[i+15:i] = 0 - dst[i+15:i] := 0 + IF k[j] + dst[i+15:i] := ABS(a[i+15:i]) ELSE - dst[i+15:i] := a[i+15:i] + dst[i+15:i] := src[i+15:i] FI ENDFOR +dst[MAX:256] := 0 - -
tmmintrin.h
+ +
immintrin.h
 - + Integer - SSSE3 + AVX512VL + AVX512BW Arithmetic - - - Negate packed 32-bit integers in "a" when the corresponding signed 32-bit integer in "b" is negative, and store the results in "dst". Element in "dst" are zeroed out when the corresponding element in "b" is zero. + + + + Compute the absolute value of packed signed 16-bit integers in "a", and store the unsigned results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 1 - i := j*32 - IF b[i+31:i] < 0 - dst[i+31:i] := NEG(a[i+31:i]) - ELSE IF b[i+31:i] = 0 - dst[i+31:i] := 0 +FOR j := 0 to 15 + i := j*16 + IF k[j] + dst[i+15:i] := ABS(a[i+15:i]) ELSE - dst[i+31:i] := a[i+31:i] + dst[i+15:i] := 0 FI ENDFOR +dst[MAX:256] := 0 - -
tmmintrin.h
+ +
immintrin.h
 - - - Floating Point - SSE4.1 - Swizzle - - - - Blend packed double-precision (64-bit) floating-point elements from "a" and "b" using control mask "imm8", and store the results in "dst". + + Integer + AVX512BW + Arithmetic + + + Compute the absolute value of packed signed 16-bit integers in "a", and store the unsigned results in "dst". -FOR j := 0 to 1 - i := j*64 - IF imm8[j%8] - dst[i+63:i] := b[i+63:i] - ELSE - dst[i+63:i] := a[i+63:i] - FI +FOR j := 0 to 31 + i := j*16 + dst[i+15:i] := ABS(a[i+15:i]) ENDFOR +dst[MAX:512] := 0 - -
smmintrin.h
+ +
immintrin.h
 - - Floating Point - SSE4.1 - Swizzle - - - - Blend packed single-precision (32-bit) floating-point elements from "a" and "b" using control mask "imm8", and store the results in "dst". + + Integer + AVX512BW + Arithmetic + + + + + Compute the absolute value of packed signed 16-bit integers in "a", and store the unsigned results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*32 - IF imm8[j%8] - dst[i+31:i] := b[i+31:i] +FOR j := 0 to 31 + i := j*16 + IF k[j] + dst[i+15:i] := ABS(a[i+15:i]) ELSE - dst[i+31:i] := a[i+31:i] + dst[i+15:i] := src[i+15:i] FI ENDFOR +dst[MAX:512] := 0 - -
smmintrin.h
+ +
immintrin.h
 - - Floating Point - SSE4.1 - Swizzle - - - - Blend packed double-precision (64-bit) floating-point elements from "a" and "b" using "mask", and store the results in "dst". + + Integer + AVX512BW + Arithmetic + + + + Compute the absolute value of packed signed 16-bit integers in "a", and store the unsigned results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 1 - i := j*64 - IF mask[i+63] - dst[i+63:i] := b[i+63:i] +FOR j := 0 to 31 + i := j*16 + IF k[j] + dst[i+15:i] := ABS(a[i+15:i]) ELSE - dst[i+63:i] := a[i+63:i] + dst[i+15:i] := 0 FI ENDFOR +dst[MAX:512] := 0 - -
smmintrin.h
+ +
immintrin.h
 - - Floating Point - SSE4.1 - Swizzle - - - - Blend packed single-precision (32-bit) floating-point elements from "a" and "b" using "mask", and store the results in "dst". + + Integer + AVX512VL + AVX512BW + Arithmetic + + + + + Compute the absolute value of packed signed 16-bit integers in "a", and store the unsigned results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*32 - IF mask[i+31] - dst[i+31:i] := b[i+31:i] +FOR j := 0 to 7 + i := j*16 + IF k[j] + dst[i+15:i] := ABS(a[i+15:i]) ELSE - dst[i+31:i] := a[i+31:i] + dst[i+15:i] := src[i+15:i] FI ENDFOR +dst[MAX:128] := 0 - -
smmintrin.h
+ +
immintrin.h
 - + Integer - SSE4.1 - Swizzle - - - - Blend packed 8-bit integers from "a" and "b" using "mask", and store the results in "dst". + AVX512VL + AVX512BW + Arithmetic + + + + Compute the absolute value of packed signed 16-bit integers in "a", and store the unsigned results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*8 - IF mask[i+7] - dst[i+7:i] := b[i+7:i] +FOR j := 0 to 7 + i := j*16 + IF k[j] + dst[i+15:i] := ABS(a[i+15:i]) ELSE - dst[i+7:i] := a[i+7:i] + dst[i+15:i] := 0 FI ENDFOR +dst[MAX:128] := 0 - -
smmintrin.h
+ +
immintrin.h
 - + Integer - SSE4.1 - Swizzle - - - - Blend packed 16-bit integers from "a" and "b" using control mask "imm8", and store the results in "dst". + AVX512VL + AVX512BW + Convert + Miscellaneous + + + + + + Convert packed signed 32-bit integers from "a" and "b" to packed 16-bit integers using signed saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 7 +tmp_dst[15:0] := Saturate16(a[31:0]) +tmp_dst[31:16] := Saturate16(a[63:32]) +tmp_dst[47:32] := Saturate16(a[95:64]) +tmp_dst[63:48] := Saturate16(a[127:96]) +tmp_dst[79:64] := Saturate16(b[31:0]) +tmp_dst[95:80] := Saturate16(b[63:32]) +tmp_dst[111:96] := Saturate16(b[95:64]) +tmp_dst[127:112] := Saturate16(b[127:96]) +tmp_dst[143:128] := Saturate16(a[159:128]) +tmp_dst[159:144] := Saturate16(a[191:160]) +tmp_dst[175:160] := Saturate16(a[223:192]) +tmp_dst[191:176] := Saturate16(a[255:224]) +tmp_dst[207:192] := Saturate16(b[159:128]) +tmp_dst[223:208] := Saturate16(b[191:160]) +tmp_dst[239:224] := Saturate16(b[223:192]) +tmp_dst[255:240] := Saturate16(b[255:224]) +FOR j := 0 to 15 i := j*16 - IF imm8[j%8] - dst[i+15:i] := b[i+15:i] + IF k[j] + dst[i+15:i] := tmp_dst[i+15:i] ELSE - dst[i+15:i] := a[i+15:i] + dst[i+15:i] := src[i+15:i] FI ENDFOR +dst[MAX:256] := 0 - -
smmintrin.h
- - - Floating Point - SSE4.1 - Arithmetic - - - - Conditionally multiply the packed double-precision (64-bit) floating-point elements in "a" and "b" using the high 4 bits in "imm8", sum the four products, and conditionally store the sum in "dst" using the low 4 bits of "imm8". - -DP(a[127:0], b[127:0], imm8[7:0]) { - FOR j := 0 to 1 - i := j*64 - IF imm8[(4+j)%8]] - temp[i+63:i] := a[i+63:i] * b[i+63:i] - ELSE - temp[i+63:i] := 0 - FI - ENDFOR - - sum[63:0] := temp[127:64] + temp[63:0] - - FOR j := 0 to 1 - i := j*64 - IF imm8[j%8] - tmpdst[i+63:i] := sum[63:0] - ELSE - tmpdst[i+63:i] := 0 - FI - ENDFOR - RETURN tmpdst[127:0] -} - -dst[127:0] := DP(a[127:0], b[127:0], imm8[7:0]) - - -
smmintrin.h
- - - Floating Point - SSE4.1 - Arithmetic - - - - Conditionally multiply the packed single-precision (32-bit) floating-point elements in "a" and "b" using the high 4 bits in "imm8", sum the four products, and conditionally store the sum in "dst" using the low 4 bits of "imm8". - -DP(a[127:0], b[127:0], imm8[7:0]) { - FOR j := 0 to 3 - i := j*32 - IF imm8[(4+j)%8] - temp[i+31:i] := a[i+31:i] * b[i+31:i] - ELSE - temp[i+31:i] := 0 - FI - ENDFOR - - sum[31:0] := (temp[127:96] + temp[95:64]) + (temp[63:32] + temp[31:0]) - - FOR j := 0 to 3 - i := j*32 - IF imm8[j%8] - tmpdst[i+31:i] := sum[31:0] - ELSE - tmpdst[i+31:i] := 0 - FI - ENDFOR - RETURN tmpdst[127:0] -} - -dst[127:0] := DP(a[127:0], b[127:0], imm8[7:0]) - - -
smmintrin.h
- - - Floating Point - SSE4.1 - Swizzle - - - Extract a single-precision (32-bit) floating-point element from "a", selected with "imm8", and store the result in "dst". - -dst[31:0] := (a[127:0] >> (imm8[1:0] * 32))[31:0] - - -
smmintrin.h
- - - Integer - SSE4.1 - Swizzle - - - Extract an 8-bit integer from "a", selected with "imm8", and store the result in the lower element of "dst". - -dst[7:0] := (a[127:0] >> (imm8[3:0] * 8))[7:0] -dst[31:8] := 0 - - -
smmintrin.h
+ +
immintrin.h
 - + Integer - SSE4.1 - Swizzle - - - Extract a 32-bit integer from "a", selected with "imm8", and store the result in "dst". + AVX512VL + AVX512BW + Convert + Miscellaneous + + + + + Convert packed signed 32-bit integers from "a" and "b" to packed 16-bit integers using signed saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -dst[31:0] := (a[127:0] >> (imm8[1:0] * 32))[31:0] +tmp_dst[15:0] := Saturate16(a[31:0]) +tmp_dst[31:16] := Saturate16(a[63:32]) +tmp_dst[47:32] := Saturate16(a[95:64]) +tmp_dst[63:48] := Saturate16(a[127:96]) +tmp_dst[79:64] := Saturate16(b[31:0]) +tmp_dst[95:80] := Saturate16(b[63:32]) +tmp_dst[111:96] := Saturate16(b[95:64]) +tmp_dst[127:112] := Saturate16(b[127:96]) +tmp_dst[143:128] := Saturate16(a[159:128]) +tmp_dst[159:144] := Saturate16(a[191:160]) +tmp_dst[175:160] := Saturate16(a[223:192]) +tmp_dst[191:176] := Saturate16(a[255:224]) +tmp_dst[207:192] := Saturate16(b[159:128]) +tmp_dst[223:208] := Saturate16(b[191:160]) +tmp_dst[239:224] := Saturate16(b[223:192]) +tmp_dst[255:240] := Saturate16(b[255:224]) +FOR j := 0 to 15 + i := j*16 + IF k[j] + dst[i+15:i] := tmp_dst[i+15:i] + ELSE + dst[i+15:i] := 0 + FI +ENDFOR +dst[MAX:256] := 0 - -
smmintrin.h
- - - Integer - SSE4.1 - Swizzle - - - Extract a 64-bit integer from "a", selected with "imm8", and store the result in "dst". - -dst[63:0] := (a[127:0] >> (imm8[0] * 64))[63:0] + +
immintrin.h
+ + + Integer + AVX512BW + Convert + Miscellaneous + + + + + + Convert packed signed 32-bit integers from "a" and "b" to packed 16-bit integers using signed saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +tmp_dst[15:0] := Saturate16(a[31:0]) +tmp_dst[31:16] := Saturate16(a[63:32]) +tmp_dst[47:32] := Saturate16(a[95:64]) +tmp_dst[63:48] := Saturate16(a[127:96]) +tmp_dst[79:64] := Saturate16(b[31:0]) +tmp_dst[95:80] := Saturate16(b[63:32]) +tmp_dst[111:96] := Saturate16(b[95:64]) +tmp_dst[127:112] := Saturate16(b[127:96]) +tmp_dst[143:128] := Saturate16(a[159:128]) +tmp_dst[159:144] := Saturate16(a[191:160]) +tmp_dst[175:160] := Saturate16(a[223:192]) +tmp_dst[191:176] := Saturate16(a[255:224]) +tmp_dst[207:192] := Saturate16(b[159:128]) +tmp_dst[223:208] := Saturate16(b[191:160]) +tmp_dst[239:224] := Saturate16(b[223:192]) +tmp_dst[255:240] := Saturate16(b[255:224]) +tmp_dst[271:256] := Saturate16(a[287:256]) +tmp_dst[287:272] := Saturate16(a[319:288]) +tmp_dst[303:288] := Saturate16(a[351:320]) +tmp_dst[319:304] := Saturate16(a[383:352]) +tmp_dst[335:320] := Saturate16(b[287:256]) +tmp_dst[351:336] := Saturate16(b[319:288]) +tmp_dst[367:352] := Saturate16(b[351:320]) +tmp_dst[383:368] := Saturate16(b[383:352]) +tmp_dst[399:384] := Saturate16(a[415:384]) +tmp_dst[415:400] := Saturate16(a[447:416]) +tmp_dst[431:416] := Saturate16(a[479:448]) +tmp_dst[447:432] := Saturate16(a[511:480]) +tmp_dst[463:448] := Saturate16(b[415:384]) +tmp_dst[479:464] := Saturate16(b[447:416]) +tmp_dst[495:480] := Saturate16(b[479:448]) +tmp_dst[511:496] := Saturate16(b[511:480]) +FOR j := 0 to 31 + i := j*16 + IF k[j] + dst[i+15:i] := tmp_dst[i+15:i] + ELSE + dst[i+15:i] := src[i+15:i] + FI +ENDFOR +dst[MAX:512] := 0 - -
smmintrin.h
- - - Floating Point - SSE4.1 - Swizzle - - - - Copy "a" to "tmp", then insert a single-precision (32-bit) floating-point element from "b" into "tmp" using the control in "imm8". Store "tmp" to "dst" using the mask in "imm8" (elements are zeroed out when the corresponding bit is set). - -tmp2[127:0] := a[127:0] -CASE (imm8[7:6]) of -0: tmp1[31:0] := b[31:0] -1: tmp1[31:0] := b[63:32] -2: tmp1[31:0] := b[95:64] -3: tmp1[31:0] := b[127:96] -ESAC -CASE (imm8[5:4]) of -0: tmp2[31:0] := tmp1[31:0] -1: tmp2[63:32] := tmp1[31:0] -2: tmp2[95:64] := tmp1[31:0] -3: tmp2[127:96] := tmp1[31:0] -ESAC -FOR j := 0 to 3 - i := j*32 - IF imm8[j%8] - dst[i+31:i] := 0 + +
immintrin.h
+ + + Integer + AVX512BW + Convert + Miscellaneous + + + + + Convert packed signed 32-bit integers from "a" and "b" to packed 16-bit integers using signed saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + +tmp_dst[15:0] := Saturate16(a[31:0]) +tmp_dst[31:16] := Saturate16(a[63:32]) +tmp_dst[47:32] := Saturate16(a[95:64]) +tmp_dst[63:48] := Saturate16(a[127:96]) +tmp_dst[79:64] := Saturate16(b[31:0]) +tmp_dst[95:80] := Saturate16(b[63:32]) +tmp_dst[111:96] := Saturate16(b[95:64]) +tmp_dst[127:112] := Saturate16(b[127:96]) +tmp_dst[143:128] := Saturate16(a[159:128]) +tmp_dst[159:144] := Saturate16(a[191:160]) +tmp_dst[175:160] := Saturate16(a[223:192]) +tmp_dst[191:176] := Saturate16(a[255:224]) +tmp_dst[207:192] := Saturate16(b[159:128]) +tmp_dst[223:208] := Saturate16(b[191:160]) +tmp_dst[239:224] := Saturate16(b[223:192]) +tmp_dst[255:240] := Saturate16(b[255:224]) +tmp_dst[271:256] := Saturate16(a[287:256]) +tmp_dst[287:272] := Saturate16(a[319:288]) +tmp_dst[303:288] := Saturate16(a[351:320]) +tmp_dst[319:304] := Saturate16(a[383:352]) +tmp_dst[335:320] := Saturate16(b[287:256]) +tmp_dst[351:336] := Saturate16(b[319:288]) +tmp_dst[367:352] := Saturate16(b[351:320]) +tmp_dst[383:368] := Saturate16(b[383:352]) +tmp_dst[399:384] := Saturate16(a[415:384]) +tmp_dst[415:400] := Saturate16(a[447:416]) +tmp_dst[431:416] := Saturate16(a[479:448]) +tmp_dst[447:432] := Saturate16(a[511:480]) +tmp_dst[463:448] := Saturate16(b[415:384]) +tmp_dst[479:464] := Saturate16(b[447:416]) +tmp_dst[495:480] := Saturate16(b[479:448]) +tmp_dst[511:496] := Saturate16(b[511:480]) +FOR j := 0 to 31 + i := j*16 + IF k[j] + dst[i+15:i] := tmp_dst[i+15:i] ELSE - dst[i+31:i] := tmp2[i+31:i] + dst[i+15:i] := 0 FI ENDFOR +dst[MAX:512] := 0 - -
smmintrin.h
+ +
immintrin.h
 - + Integer - SSE4.1 - Swizzle - - - - Copy "a" to "dst", and insert the lower 8-bit integer from "i" into "dst" at the location specified by "imm8". + AVX512BW + Convert + Miscellaneous + + + + Convert packed signed 32-bit integers from "a" and "b" to packed 16-bit integers using signed saturation, and store the results in "dst". -dst[127:0] := a[127:0] -sel := imm8[3:0]*8 -dst[sel+7:sel] := i[7:0] +dst[15:0] := Saturate16(a[31:0]) +dst[31:16] := Saturate16(a[63:32]) +dst[47:32] := Saturate16(a[95:64]) +dst[63:48] := Saturate16(a[127:96]) +dst[79:64] := Saturate16(b[31:0]) +dst[95:80] := Saturate16(b[63:32]) +dst[111:96] := Saturate16(b[95:64]) +dst[127:112] := Saturate16(b[127:96]) +dst[143:128] := Saturate16(a[159:128]) +dst[159:144] := Saturate16(a[191:160]) +dst[175:160] := Saturate16(a[223:192]) +dst[191:176] := Saturate16(a[255:224]) +dst[207:192] := Saturate16(b[159:128]) +dst[223:208] := Saturate16(b[191:160]) +dst[239:224] := Saturate16(b[223:192]) +dst[255:240] := Saturate16(b[255:224]) +dst[271:256] := Saturate16(a[287:256]) +dst[287:272] := Saturate16(a[319:288]) +dst[303:288] := Saturate16(a[351:320]) +dst[319:304] := Saturate16(a[383:352]) +dst[335:320] := Saturate16(b[287:256]) +dst[351:336] := Saturate16(b[319:288]) +dst[367:352] := Saturate16(b[351:320]) +dst[383:368] := Saturate16(b[383:352]) +dst[399:384] := Saturate16(a[415:384]) +dst[415:400] := Saturate16(a[447:416]) +dst[431:416] := Saturate16(a[479:448]) +dst[447:432] := Saturate16(a[511:480]) +dst[463:448] := Saturate16(b[415:384]) +dst[479:464] := Saturate16(b[447:416]) +dst[495:480] := Saturate16(b[479:448]) +dst[511:496] := Saturate16(b[511:480]) +dst[MAX:512] := 0 - -
smmintrin.h
+ +
immintrin.h
 - + Integer - SSE4.1 - Swizzle - - - - Copy "a" to "dst", and insert the 32-bit integer "i" into "dst" at the location specified by "imm8". + AVX512VL + AVX512BW + Convert + Miscellaneous + + + + + + Convert packed signed 32-bit integers from "a" and "b" to packed 16-bit integers using signed saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -dst[127:0] := a[127:0] -sel := imm8[1:0]*32 -dst[sel+31:sel] := i[31:0] +tmp_dst[15:0] := Saturate16(a[31:0]) +tmp_dst[31:16] := Saturate16(a[63:32]) +tmp_dst[47:32] := Saturate16(a[95:64]) +tmp_dst[63:48] := Saturate16(a[127:96]) +tmp_dst[79:64] := Saturate16(b[31:0]) +tmp_dst[95:80] := Saturate16(b[63:32]) +tmp_dst[111:96] := Saturate16(b[95:64]) +tmp_dst[127:112] := Saturate16(b[127:96]) +FOR j := 0 to 7 + i := j*16 + IF k[j] + dst[i+15:i] := tmp_dst[i+15:i] + ELSE + dst[i+15:i] := src[i+15:i] + FI +ENDFOR +dst[MAX:128] := 0 - -
smmintrin.h
+ +
immintrin.h
 - + Integer - SSE4.1 - Swizzle - - - - Copy "a" to "dst", and insert the 64-bit integer "i" into "dst" at the location specified by "imm8". + AVX512VL + AVX512BW + Convert + Miscellaneous + + + + + Convert packed signed 32-bit integers from "a" and "b" to packed 16-bit integers using signed saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -dst[127:0] := a[127:0] -sel := imm8[0]*64 -dst[sel+63:sel] := i[63:0] +tmp_dst[15:0] := Saturate16(a[31:0]) +tmp_dst[31:16] := Saturate16(a[63:32]) +tmp_dst[47:32] := Saturate16(a[95:64]) +tmp_dst[63:48] := Saturate16(a[127:96]) +tmp_dst[79:64] := Saturate16(b[31:0]) +tmp_dst[95:80] := Saturate16(b[63:32]) +tmp_dst[111:96] := Saturate16(b[95:64]) +tmp_dst[127:112] := Saturate16(b[127:96]) +FOR j := 0 to 7 + i := j*16 + IF k[j] + dst[i+15:i] := tmp_dst[i+15:i] + ELSE + dst[i+15:i] := 0 + FI +ENDFOR +dst[MAX:128] := 0 - -
smmintrin.h
- - - Integer - SSE4.1 - Special Math Functions - - - Compare packed 8-bit integers in "a" and "b", and store packed maximum values in "dst". - -FOR j := 0 to 15 + +
immintrin.h
+ + + Integer + AVX512VL + AVX512BW + Convert + Miscellaneous + + + + + + Convert packed signed 16-bit integers from "a" and "b" to packed 8-bit integers using signed saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +tmp_dst[7:0] := Saturate8(a[15:0]) +tmp_dst[15:8] := Saturate8(a[31:16]) +tmp_dst[23:16] := Saturate8(a[47:32]) +tmp_dst[31:24] := Saturate8(a[63:48]) +tmp_dst[39:32] := Saturate8(a[79:64]) +tmp_dst[47:40] := Saturate8(a[95:80]) +tmp_dst[55:48] := Saturate8(a[111:96]) +tmp_dst[63:56] := Saturate8(a[127:112]) +tmp_dst[71:64] := Saturate8(b[15:0]) +tmp_dst[79:72] := Saturate8(b[31:16]) +tmp_dst[87:80] := Saturate8(b[47:32]) +tmp_dst[95:88] := Saturate8(b[63:48]) +tmp_dst[103:96] := Saturate8(b[79:64]) +tmp_dst[111:104] := Saturate8(b[95:80]) +tmp_dst[119:112] := Saturate8(b[111:96]) +tmp_dst[127:120] := Saturate8(b[127:112]) +tmp_dst[135:128] := Saturate8(a[143:128]) +tmp_dst[143:136] := Saturate8(a[159:144]) +tmp_dst[151:144] := Saturate8(a[175:160]) +tmp_dst[159:152] := Saturate8(a[191:176]) +tmp_dst[167:160] := Saturate8(a[207:192]) +tmp_dst[175:168] := Saturate8(a[223:208]) +tmp_dst[183:176] := Saturate8(a[239:224]) +tmp_dst[191:184] := Saturate8(a[255:240]) +tmp_dst[199:192] := Saturate8(b[143:128]) +tmp_dst[207:200] := Saturate8(b[159:144]) +tmp_dst[215:208] := Saturate8(b[175:160]) +tmp_dst[223:216] := Saturate8(b[191:176]) +tmp_dst[231:224] := Saturate8(b[207:192]) +tmp_dst[239:232] := Saturate8(b[223:208]) +tmp_dst[247:240] := Saturate8(b[239:224]) +tmp_dst[255:248] := Saturate8(b[255:240]) +FOR j := 0 to 31 i := j*8 - IF a[i+7:i] > b[i+7:i] - dst[i+7:i] := a[i+7:i] + IF k[j] + dst[i+7:i] := tmp_dst[i+7:i] ELSE - dst[i+7:i] := b[i+7:i] + dst[i+7:i] := src[i+7:i] FI ENDFOR +dst[MAX:256] := 0 - -
smmintrin.h
+ +
immintrin.h
 - + Integer - SSE4.1 - Special Math Functions - - - Compare packed 32-bit integers in "a" and "b", and store packed maximum values in "dst". + AVX512VL + AVX512BW + Convert + Miscellaneous + + + + + Convert packed signed 16-bit integers from "a" and "b" to packed 8-bit integers using signed saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*32 - IF a[i+31:i] > b[i+31:i] - dst[i+31:i] := a[i+31:i] +tmp_dst[7:0] := Saturate8(a[15:0]) +tmp_dst[15:8] := Saturate8(a[31:16]) +tmp_dst[23:16] := Saturate8(a[47:32]) +tmp_dst[31:24] := Saturate8(a[63:48]) +tmp_dst[39:32] := Saturate8(a[79:64]) +tmp_dst[47:40] := Saturate8(a[95:80]) +tmp_dst[55:48] := Saturate8(a[111:96]) +tmp_dst[63:56] := Saturate8(a[127:112]) +tmp_dst[71:64] := Saturate8(b[15:0]) +tmp_dst[79:72] := Saturate8(b[31:16]) +tmp_dst[87:80] := Saturate8(b[47:32]) +tmp_dst[95:88] := Saturate8(b[63:48]) +tmp_dst[103:96] := Saturate8(b[79:64]) +tmp_dst[111:104] := Saturate8(b[95:80]) +tmp_dst[119:112] := Saturate8(b[111:96]) +tmp_dst[127:120] := Saturate8(b[127:112]) +tmp_dst[135:128] := Saturate8(a[143:128]) +tmp_dst[143:136] := Saturate8(a[159:144]) +tmp_dst[151:144] := Saturate8(a[175:160]) +tmp_dst[159:152] := Saturate8(a[191:176]) +tmp_dst[167:160] := Saturate8(a[207:192]) +tmp_dst[175:168] := Saturate8(a[223:208]) +tmp_dst[183:176] := Saturate8(a[239:224]) +tmp_dst[191:184] := Saturate8(a[255:240]) +tmp_dst[199:192] := Saturate8(b[143:128]) +tmp_dst[207:200] := Saturate8(b[159:144]) +tmp_dst[215:208] := Saturate8(b[175:160]) +tmp_dst[223:216] := Saturate8(b[191:176]) +tmp_dst[231:224] := Saturate8(b[207:192]) +tmp_dst[239:232] := Saturate8(b[223:208]) +tmp_dst[247:240] := Saturate8(b[239:224]) +tmp_dst[255:248] := Saturate8(b[255:240]) +FOR j := 0 to 31 + i := j*8 + IF k[j] + dst[i+7:i] := tmp_dst[i+7:i] ELSE - dst[i+31:i] := b[i+31:i] + dst[i+7:i] := 0 FI ENDFOR +dst[MAX:256] := 0 - -
smmintrin.h
- - - Integer - SSE4.1 - Special Math Functions - - - Compare packed unsigned 32-bit integers in "a" and "b", and store packed maximum values in "dst". - -FOR j := 0 to 3 - i := j*32 - IF a[i+31:i] > b[i+31:i] - dst[i+31:i] := a[i+31:i] + +
immintrin.h
+ + + Integer + AVX512BW + Convert + Miscellaneous + + + + + + Convert packed signed 16-bit integers from "a" and "b" to packed 8-bit integers using signed saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +tmp_dst[7:0] := Saturate8(a[15:0]) +tmp_dst[15:8] := Saturate8(a[31:16]) +tmp_dst[23:16] := Saturate8(a[47:32]) +tmp_dst[31:24] := Saturate8(a[63:48]) +tmp_dst[39:32] := Saturate8(a[79:64]) +tmp_dst[47:40] := Saturate8(a[95:80]) +tmp_dst[55:48] := Saturate8(a[111:96]) +tmp_dst[63:56] := Saturate8(a[127:112]) +tmp_dst[71:64] := Saturate8(b[15:0]) +tmp_dst[79:72] := Saturate8(b[31:16]) +tmp_dst[87:80] := Saturate8(b[47:32]) +tmp_dst[95:88] := Saturate8(b[63:48]) +tmp_dst[103:96] := Saturate8(b[79:64]) +tmp_dst[111:104] := Saturate8(b[95:80]) +tmp_dst[119:112] := Saturate8(b[111:96]) +tmp_dst[127:120] := Saturate8(b[127:112]) +tmp_dst[135:128] := Saturate8(a[143:128]) +tmp_dst[143:136] := Saturate8(a[159:144]) +tmp_dst[151:144] := Saturate8(a[175:160]) +tmp_dst[159:152] := Saturate8(a[191:176]) +tmp_dst[167:160] := Saturate8(a[207:192]) +tmp_dst[175:168] := Saturate8(a[223:208]) +tmp_dst[183:176] := Saturate8(a[239:224]) +tmp_dst[191:184] := Saturate8(a[255:240]) +tmp_dst[199:192] := Saturate8(b[143:128]) +tmp_dst[207:200] := Saturate8(b[159:144]) +tmp_dst[215:208] := Saturate8(b[175:160]) +tmp_dst[223:216] := Saturate8(b[191:176]) +tmp_dst[231:224] := Saturate8(b[207:192]) +tmp_dst[239:232] := Saturate8(b[223:208]) +tmp_dst[247:240] := Saturate8(b[239:224]) +tmp_dst[255:248] := Saturate8(b[255:240]) +tmp_dst[263:256] := Saturate8(a[271:256]) +tmp_dst[271:264] := Saturate8(a[287:272]) +tmp_dst[279:272] := Saturate8(a[303:288]) +tmp_dst[287:280] := Saturate8(a[319:304]) +tmp_dst[295:288] := Saturate8(a[335:320]) +tmp_dst[303:296] := Saturate8(a[351:336]) +tmp_dst[311:304] := Saturate8(a[367:352]) +tmp_dst[319:312] := Saturate8(a[383:368]) +tmp_dst[327:320] := Saturate8(b[271:256]) +tmp_dst[335:328] := Saturate8(b[287:272]) +tmp_dst[343:336] := Saturate8(b[303:288]) +tmp_dst[351:344] := Saturate8(b[319:304]) +tmp_dst[359:352] := Saturate8(b[335:320]) +tmp_dst[367:360] := Saturate8(b[351:336]) +tmp_dst[375:368] := Saturate8(b[367:352]) +tmp_dst[383:376] := Saturate8(b[383:368]) +tmp_dst[391:384] := Saturate8(a[399:384]) +tmp_dst[399:392] := Saturate8(a[415:400]) +tmp_dst[407:400] := Saturate8(a[431:416]) +tmp_dst[415:408] := Saturate8(a[447:432]) +tmp_dst[423:416] := Saturate8(a[463:448]) +tmp_dst[431:424] := Saturate8(a[479:464]) +tmp_dst[439:432] := Saturate8(a[495:480]) +tmp_dst[447:440] := Saturate8(a[511:496]) +tmp_dst[455:448] := Saturate8(b[399:384]) +tmp_dst[463:456] := Saturate8(b[415:400]) +tmp_dst[471:464] := Saturate8(b[431:416]) +tmp_dst[479:472] := Saturate8(b[447:432]) +tmp_dst[487:480] := Saturate8(b[463:448]) +tmp_dst[495:488] := Saturate8(b[479:464]) +tmp_dst[503:496] := Saturate8(b[495:480]) +tmp_dst[511:504] := Saturate8(b[511:496]) +FOR j := 0 to 63 + i := j*8 + IF k[j] + dst[i+7:i] := tmp_dst[i+7:i] ELSE - dst[i+31:i] := b[i+31:i] + dst[i+7:i] := src[i+7:i] FI ENDFOR +dst[MAX:512] := 0 - -
smmintrin.h
- - - Integer - SSE4.1 - Special Math Functions - - - Compare packed unsigned 16-bit integers in "a" and "b", and store packed maximum values in "dst". - -FOR j := 0 to 7 - i := j*16 - IF a[i+15:i] > b[i+15:i] - dst[i+15:i] := a[i+15:i] + +
immintrin.h
+ + + Integer + AVX512BW + Convert + Miscellaneous + + + + + Convert packed signed 16-bit integers from "a" and "b" to packed 8-bit integers using signed saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + +tmp_dst[7:0] := Saturate8(a[15:0]) +tmp_dst[15:8] := Saturate8(a[31:16]) +tmp_dst[23:16] := Saturate8(a[47:32]) +tmp_dst[31:24] := Saturate8(a[63:48]) +tmp_dst[39:32] := Saturate8(a[79:64]) +tmp_dst[47:40] := Saturate8(a[95:80]) +tmp_dst[55:48] := Saturate8(a[111:96]) +tmp_dst[63:56] := Saturate8(a[127:112]) +tmp_dst[71:64] := Saturate8(b[15:0]) +tmp_dst[79:72] := Saturate8(b[31:16]) +tmp_dst[87:80] := Saturate8(b[47:32]) +tmp_dst[95:88] := Saturate8(b[63:48]) +tmp_dst[103:96] := Saturate8(b[79:64]) +tmp_dst[111:104] := Saturate8(b[95:80]) +tmp_dst[119:112] := Saturate8(b[111:96]) +tmp_dst[127:120] := Saturate8(b[127:112]) +tmp_dst[135:128] := Saturate8(a[143:128]) +tmp_dst[143:136] := Saturate8(a[159:144]) +tmp_dst[151:144] := Saturate8(a[175:160]) +tmp_dst[159:152] := Saturate8(a[191:176]) +tmp_dst[167:160] := Saturate8(a[207:192]) +tmp_dst[175:168] := Saturate8(a[223:208]) +tmp_dst[183:176] := Saturate8(a[239:224]) +tmp_dst[191:184] := Saturate8(a[255:240]) +tmp_dst[199:192] := Saturate8(b[143:128]) +tmp_dst[207:200] := Saturate8(b[159:144]) +tmp_dst[215:208] := Saturate8(b[175:160]) +tmp_dst[223:216] := Saturate8(b[191:176]) +tmp_dst[231:224] := Saturate8(b[207:192]) +tmp_dst[239:232] := Saturate8(b[223:208]) +tmp_dst[247:240] := Saturate8(b[239:224]) +tmp_dst[255:248] := Saturate8(b[255:240]) +tmp_dst[263:256] := Saturate8(a[271:256]) +tmp_dst[271:264] := Saturate8(a[287:272]) +tmp_dst[279:272] := Saturate8(a[303:288]) +tmp_dst[287:280] := Saturate8(a[319:304]) +tmp_dst[295:288] := Saturate8(a[335:320]) +tmp_dst[303:296] := Saturate8(a[351:336]) +tmp_dst[311:304] := Saturate8(a[367:352]) +tmp_dst[319:312] := Saturate8(a[383:368]) +tmp_dst[327:320] := Saturate8(b[271:256]) +tmp_dst[335:328] := Saturate8(b[287:272]) +tmp_dst[343:336] := Saturate8(b[303:288]) +tmp_dst[351:344] := Saturate8(b[319:304]) +tmp_dst[359:352] := Saturate8(b[335:320]) +tmp_dst[367:360] := Saturate8(b[351:336]) +tmp_dst[375:368] := Saturate8(b[367:352]) +tmp_dst[383:376] := Saturate8(b[383:368]) +tmp_dst[391:384] := Saturate8(a[399:384]) +tmp_dst[399:392] := Saturate8(a[415:400]) +tmp_dst[407:400] := Saturate8(a[431:416]) +tmp_dst[415:408] := Saturate8(a[447:432]) +tmp_dst[423:416] := Saturate8(a[463:448]) +tmp_dst[431:424] := Saturate8(a[479:464]) +tmp_dst[439:432] := Saturate8(a[495:480]) +tmp_dst[447:440] := Saturate8(a[511:496]) +tmp_dst[455:448] := Saturate8(b[399:384]) +tmp_dst[463:456] := Saturate8(b[415:400]) +tmp_dst[471:464] := Saturate8(b[431:416]) +tmp_dst[479:472] := Saturate8(b[447:432]) +tmp_dst[487:480] := Saturate8(b[463:448]) +tmp_dst[495:488] := Saturate8(b[479:464]) +tmp_dst[503:496] := Saturate8(b[495:480]) +tmp_dst[511:504] := Saturate8(b[511:496]) +FOR j := 0 to 63 + i := j*8 + IF k[j] + dst[i+7:i] := tmp_dst[i+7:i] ELSE - dst[i+15:i] := b[i+15:i] + dst[i+7:i] := 0 FI ENDFOR +dst[MAX:512] := 0 - -
smmintrin.h
- - - Integer - SSE4.1 - Special Math Functions - - - Compare packed 8-bit integers in "a" and "b", and store packed minimum values in "dst". - + +
immintrin.h
+ + + Integer + AVX512BW + Convert + Miscellaneous + + + + Convert packed signed 16-bit integers from "a" and "b" to packed 8-bit integers using signed saturation, and store the results in "dst". + +dst[7:0] := Saturate8(a[15:0]) +dst[15:8] := Saturate8(a[31:16]) +dst[23:16] := Saturate8(a[47:32]) +dst[31:24] := Saturate8(a[63:48]) +dst[39:32] := Saturate8(a[79:64]) +dst[47:40] := Saturate8(a[95:80]) +dst[55:48] := Saturate8(a[111:96]) +dst[63:56] := Saturate8(a[127:112]) +dst[71:64] := Saturate8(b[15:0]) +dst[79:72] := Saturate8(b[31:16]) +dst[87:80] := Saturate8(b[47:32]) +dst[95:88] := Saturate8(b[63:48]) +dst[103:96] := Saturate8(b[79:64]) +dst[111:104] := Saturate8(b[95:80]) +dst[119:112] := Saturate8(b[111:96]) +dst[127:120] := Saturate8(b[127:112]) +dst[135:128] := Saturate8(a[143:128]) +dst[143:136] := Saturate8(a[159:144]) +dst[151:144] := Saturate8(a[175:160]) +dst[159:152] := Saturate8(a[191:176]) +dst[167:160] := Saturate8(a[207:192]) +dst[175:168] := Saturate8(a[223:208]) +dst[183:176] := Saturate8(a[239:224]) +dst[191:184] := Saturate8(a[255:240]) +dst[199:192] := Saturate8(b[143:128]) +dst[207:200] := Saturate8(b[159:144]) +dst[215:208] := Saturate8(b[175:160]) +dst[223:216] := Saturate8(b[191:176]) +dst[231:224] := Saturate8(b[207:192]) +dst[239:232] := Saturate8(b[223:208]) +dst[247:240] := Saturate8(b[239:224]) +dst[255:248] := Saturate8(b[255:240]) +dst[263:256] := Saturate8(a[271:256]) +dst[271:264] := Saturate8(a[287:272]) +dst[279:272] := Saturate8(a[303:288]) +dst[287:280] := Saturate8(a[319:304]) +dst[295:288] := Saturate8(a[335:320]) +dst[303:296] := Saturate8(a[351:336]) +dst[311:304] := Saturate8(a[367:352]) +dst[319:312] := Saturate8(a[383:368]) +dst[327:320] := Saturate8(b[271:256]) +dst[335:328] := Saturate8(b[287:272]) +dst[343:336] := Saturate8(b[303:288]) +dst[351:344] := Saturate8(b[319:304]) +dst[359:352] := Saturate8(b[335:320]) +dst[367:360] := Saturate8(b[351:336]) +dst[375:368] := Saturate8(b[367:352]) +dst[383:376] := Saturate8(b[383:368]) +dst[391:384] := Saturate8(a[399:384]) +dst[399:392] := Saturate8(a[415:400]) +dst[407:400] := Saturate8(a[431:416]) +dst[415:408] := Saturate8(a[447:432]) +dst[423:416] := Saturate8(a[463:448]) +dst[431:424] := Saturate8(a[479:464]) +dst[439:432] := Saturate8(a[495:480]) +dst[447:440] := Saturate8(a[511:496]) +dst[455:448] := Saturate8(b[399:384]) +dst[463:456] := Saturate8(b[415:400]) +dst[471:464] := Saturate8(b[431:416]) +dst[479:472] := Saturate8(b[447:432]) +dst[487:480] := Saturate8(b[463:448]) +dst[495:488] := Saturate8(b[479:464]) +dst[503:496] := Saturate8(b[495:480]) +dst[511:504] := Saturate8(b[511:496]) +dst[MAX:512] := 0 + + +
immintrin.h
+ + + Integer + AVX512VL + AVX512BW + Convert + Miscellaneous + + + + + + Convert packed signed 16-bit integers from "a" and "b" to packed 8-bit integers using signed saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +tmp_dst[7:0] := Saturate8(a[15:0]) +tmp_dst[15:8] := Saturate8(a[31:16]) +tmp_dst[23:16] := Saturate8(a[47:32]) +tmp_dst[31:24] := Saturate8(a[63:48]) +tmp_dst[39:32] := Saturate8(a[79:64]) +tmp_dst[47:40] := Saturate8(a[95:80]) +tmp_dst[55:48] := Saturate8(a[111:96]) +tmp_dst[63:56] := Saturate8(a[127:112]) +tmp_dst[71:64] := Saturate8(b[15:0]) +tmp_dst[79:72] := Saturate8(b[31:16]) +tmp_dst[87:80] := Saturate8(b[47:32]) +tmp_dst[95:88] := Saturate8(b[63:48]) +tmp_dst[103:96] := Saturate8(b[79:64]) +tmp_dst[111:104] := Saturate8(b[95:80]) +tmp_dst[119:112] := Saturate8(b[111:96]) +tmp_dst[127:120] := Saturate8(b[127:112]) FOR j := 0 to 15 i := j*8 - IF a[i+7:i] < b[i+7:i] - dst[i+7:i] := a[i+7:i] + IF k[j] + dst[i+7:i] := tmp_dst[i+7:i] ELSE - dst[i+7:i] := b[i+7:i] + dst[i+7:i] := src[i+7:i] FI ENDFOR +dst[MAX:128] := 0 - -
smmintrin.h
+ +
immintrin.h
 - + Integer - SSE4.1 - Special Math Functions - - - Compare packed 32-bit integers in "a" and "b", and store packed minimum values in "dst". + AVX512VL + AVX512BW + Convert + Miscellaneous + + + + + Convert packed signed 16-bit integers from "a" and "b" to packed 8-bit integers using signed saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*32 - IF a[i+31:i] < b[i+31:i] - dst[i+31:i] := a[i+31:i] +tmp_dst[7:0] := Saturate8(a[15:0]) +tmp_dst[15:8] := Saturate8(a[31:16]) +tmp_dst[23:16] := Saturate8(a[47:32]) +tmp_dst[31:24] := Saturate8(a[63:48]) +tmp_dst[39:32] := Saturate8(a[79:64]) +tmp_dst[47:40] := Saturate8(a[95:80]) +tmp_dst[55:48] := Saturate8(a[111:96]) +tmp_dst[63:56] := Saturate8(a[127:112]) +tmp_dst[71:64] := Saturate8(b[15:0]) +tmp_dst[79:72] := Saturate8(b[31:16]) +tmp_dst[87:80] := Saturate8(b[47:32]) +tmp_dst[95:88] := Saturate8(b[63:48]) +tmp_dst[103:96] := Saturate8(b[79:64]) +tmp_dst[111:104] := Saturate8(b[95:80]) +tmp_dst[119:112] := Saturate8(b[111:96]) +tmp_dst[127:120] := Saturate8(b[127:112]) +FOR j := 0 to 15 + i := j*8 + IF k[j] + dst[i+7:i] := tmp_dst[i+7:i] ELSE - dst[i+31:i] := b[i+31:i] + dst[i+7:i] := 0 FI ENDFOR +dst[MAX:128] := 0 - -
smmintrin.h
+ +
immintrin.h
 - + Integer - SSE4.1 - Special Math Functions - - - Compare packed unsigned 32-bit integers in "a" and "b", and store packed minimum values in "dst". + AVX512VL + AVX512BW + Convert + Miscellaneous + + + + + + Convert packed signed 32-bit integers from "a" and "b" to packed 16-bit integers using unsigned saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*32 - IF a[i+31:i] < b[i+31:i] - dst[i+31:i] := a[i+31:i] +tmp_dst[15:0] := SaturateU16(a[31:0]) +tmp_dst[31:16] := SaturateU16(a[63:32]) +tmp_dst[47:32] := SaturateU16(a[95:64]) +tmp_dst[63:48] := SaturateU16(a[127:96]) +tmp_dst[79:64] := SaturateU16(b[31:0]) +tmp_dst[95:80] := SaturateU16(b[63:32]) +tmp_dst[111:96] := SaturateU16(b[95:64]) +tmp_dst[127:112] := SaturateU16(b[127:96]) +tmp_dst[143:128] := SaturateU16(a[159:128]) +tmp_dst[159:144] := SaturateU16(a[191:160]) +tmp_dst[175:160] := SaturateU16(a[223:192]) +tmp_dst[191:176] := SaturateU16(a[255:224]) +tmp_dst[207:192] := SaturateU16(b[159:128]) +tmp_dst[223:208] := SaturateU16(b[191:160]) +tmp_dst[239:224] := SaturateU16(b[223:192]) +tmp_dst[255:240] := SaturateU16(b[255:224]) +FOR j := 0 to 15 + i := j*16 + IF k[j] + dst[i+15:i] := tmp_dst[i+15:i] ELSE - dst[i+31:i] := b[i+31:i] + dst[i+15:i] := src[i+15:i] FI ENDFOR +dst[MAX:256] := 0 - -
smmintrin.h
+ +
immintrin.h
 - + Integer - SSE4.1 - Special Math Functions - - - Compare packed unsigned 16-bit integers in "a" and "b", and store packed minimum values in "dst". + AVX512VL + AVX512BW + Convert + Miscellaneous + + + + + Convert packed signed 32-bit integers from "a" and "b" to packed 16-bit integers using unsigned saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 +tmp_dst[15:0] := SaturateU16(a[31:0]) +tmp_dst[31:16] := SaturateU16(a[63:32]) +tmp_dst[47:32] := SaturateU16(a[95:64]) +tmp_dst[63:48] := SaturateU16(a[127:96]) +tmp_dst[79:64] := SaturateU16(b[31:0]) +tmp_dst[95:80] := SaturateU16(b[63:32]) +tmp_dst[111:96] := SaturateU16(b[95:64]) +tmp_dst[127:112] := SaturateU16(b[127:96]) +tmp_dst[143:128] := SaturateU16(a[159:128]) +tmp_dst[159:144] := SaturateU16(a[191:160]) +tmp_dst[175:160] := SaturateU16(a[223:192]) +tmp_dst[191:176] := SaturateU16(a[255:224]) +tmp_dst[207:192] := SaturateU16(b[159:128]) +tmp_dst[223:208] := SaturateU16(b[191:160]) +tmp_dst[239:224] := SaturateU16(b[223:192]) +tmp_dst[255:240] := SaturateU16(b[255:224]) +FOR j := 0 to 15 i := j*16 - IF a[i+15:i] < b[i+15:i] - dst[i+15:i] := a[i+15:i] + IF k[j] + dst[i+15:i] := tmp_dst[i+15:i] ELSE - dst[i+15:i] := b[i+15:i] + dst[i+15:i] := 0 FI ENDFOR +dst[MAX:256] := 0 - -
smmintrin.h
+ +
immintrin.h
+ + + Integer + AVX512BW + Convert + Miscellaneous + + + + + + Convert packed signed 32-bit integers from "a" and "b" to packed 16-bit integers using unsigned saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +tmp_dst[15:0] := SaturateU16(a[31:0]) +tmp_dst[31:16] := SaturateU16(a[63:32]) +tmp_dst[47:32] := SaturateU16(a[95:64]) +tmp_dst[63:48] := SaturateU16(a[127:96]) +tmp_dst[79:64] := SaturateU16(b[31:0]) +tmp_dst[95:80] := SaturateU16(b[63:32]) +tmp_dst[111:96] := SaturateU16(b[95:64]) +tmp_dst[127:112] := SaturateU16(b[127:96]) +tmp_dst[143:128] := SaturateU16(a[159:128]) +tmp_dst[159:144] := SaturateU16(a[191:160]) +tmp_dst[175:160] := SaturateU16(a[223:192]) +tmp_dst[191:176] := SaturateU16(a[255:224]) +tmp_dst[207:192] := SaturateU16(b[159:128]) +tmp_dst[223:208] := SaturateU16(b[191:160]) +tmp_dst[239:224] := SaturateU16(b[223:192]) +tmp_dst[255:240] := SaturateU16(b[255:224]) +tmp_dst[271:256] := SaturateU16(a[287:256]) +tmp_dst[287:272] := SaturateU16(a[319:288]) +tmp_dst[303:288] := SaturateU16(a[351:320]) +tmp_dst[319:304] := SaturateU16(a[383:352]) +tmp_dst[335:320] := SaturateU16(b[287:256]) +tmp_dst[351:336] := SaturateU16(b[319:288]) +tmp_dst[367:352] := SaturateU16(b[351:320]) +tmp_dst[383:368] := SaturateU16(b[383:352]) +tmp_dst[399:384] := SaturateU16(a[415:384]) +tmp_dst[415:400] := SaturateU16(a[447:416]) +tmp_dst[431:416] := SaturateU16(a[479:448]) +tmp_dst[447:432] := SaturateU16(a[511:480]) +tmp_dst[463:448] := SaturateU16(b[415:384]) +tmp_dst[479:464] := SaturateU16(b[447:416]) +tmp_dst[495:480] := SaturateU16(b[479:448]) +tmp_dst[511:496] := SaturateU16(b[511:480]) +FOR j := 0 to 31 + i := j*16 + IF k[j] + dst[i+15:i] := tmp_dst[i+15:i] + ELSE + dst[i+15:i] := src[i+15:i] + FI +ENDFOR +dst[MAX:512] := 0 + + +
immintrin.h
+ + + Integer + AVX512BW + Convert + Miscellaneous + + + + + Convert packed signed 32-bit integers from "a" and "b" to packed 16-bit integers using unsigned saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + +tmp_dst[15:0] := SaturateU16(a[31:0]) +tmp_dst[31:16] := SaturateU16(a[63:32]) +tmp_dst[47:32] := SaturateU16(a[95:64]) +tmp_dst[63:48] := SaturateU16(a[127:96]) +tmp_dst[79:64] := SaturateU16(b[31:0]) +tmp_dst[95:80] := SaturateU16(b[63:32]) +tmp_dst[111:96] := SaturateU16(b[95:64]) +tmp_dst[127:112] := SaturateU16(b[127:96]) +tmp_dst[143:128] := SaturateU16(a[159:128]) +tmp_dst[159:144] := SaturateU16(a[191:160]) +tmp_dst[175:160] := SaturateU16(a[223:192]) +tmp_dst[191:176] := SaturateU16(a[255:224]) +tmp_dst[207:192] := SaturateU16(b[159:128]) +tmp_dst[223:208] := SaturateU16(b[191:160]) +tmp_dst[239:224] := SaturateU16(b[223:192]) +tmp_dst[255:240] := SaturateU16(b[255:224]) +tmp_dst[271:256] := SaturateU16(a[287:256]) +tmp_dst[287:272] := SaturateU16(a[319:288]) +tmp_dst[303:288] := SaturateU16(a[351:320]) +tmp_dst[319:304] := SaturateU16(a[383:352]) +tmp_dst[335:320] := SaturateU16(b[287:256]) +tmp_dst[351:336] := SaturateU16(b[319:288]) +tmp_dst[367:352] := SaturateU16(b[351:320]) +tmp_dst[383:368] := SaturateU16(b[383:352]) +tmp_dst[399:384] := SaturateU16(a[415:384]) +tmp_dst[415:400] := SaturateU16(a[447:416]) +tmp_dst[431:416] := SaturateU16(a[479:448]) +tmp_dst[447:432] := SaturateU16(a[511:480]) +tmp_dst[463:448] := SaturateU16(b[415:384]) +tmp_dst[479:464] := SaturateU16(b[447:416]) +tmp_dst[495:480] := SaturateU16(b[479:448]) +tmp_dst[511:496] := SaturateU16(b[511:480]) +FOR j := 0 to 31 + i := j*16 + IF k[j] + dst[i+15:i] := tmp_dst[i+15:i] + ELSE + dst[i+15:i] := 0 + FI +ENDFOR +dst[MAX:512] := 0 + + +
immintrin.h
 - + Integer - SSE4.1 + AVX512BW Convert Miscellaneous - - - Convert packed 32-bit integers from "a" and "b" to packed 16-bit integers using unsigned saturation, and store the results in "dst". + + + + Convert packed signed 32-bit integers from "a" and "b" to packed 16-bit integers using unsigned saturation, and store the results in "dst". -dst[15:0] := Saturate_Int32_To_UnsignedInt16 (a[31:0]) -dst[31:16] := Saturate_Int32_To_UnsignedInt16 (a[63:32]) -dst[47:32] := Saturate_Int32_To_UnsignedInt16 (a[95:64]) -dst[63:48] := Saturate_Int32_To_UnsignedInt16 (a[127:96]) -dst[79:64] := Saturate_Int32_To_UnsignedInt16 (b[31:0]) -dst[95:80] := Saturate_Int32_To_UnsignedInt16 (b[63:32]) -dst[111:96] := Saturate_Int32_To_UnsignedInt16 (b[95:64]) -dst[127:112] := Saturate_Int32_To_UnsignedInt16 (b[127:96]) +dst[15:0] := SaturateU16(a[31:0]) +dst[31:16] := SaturateU16(a[63:32]) +dst[47:32] := SaturateU16(a[95:64]) +dst[63:48] := SaturateU16(a[127:96]) +dst[79:64] := SaturateU16(b[31:0]) +dst[95:80] := SaturateU16(b[63:32]) +dst[111:96] := SaturateU16(b[95:64]) +dst[127:112] := SaturateU16(b[127:96]) +dst[143:128] := SaturateU16(a[159:128]) +dst[159:144] := SaturateU16(a[191:160]) +dst[175:160] := SaturateU16(a[223:192]) +dst[191:176] := SaturateU16(a[255:224]) +dst[207:192] := SaturateU16(b[159:128]) +dst[223:208] := SaturateU16(b[191:160]) +dst[239:224] := SaturateU16(b[223:192]) +dst[255:240] := SaturateU16(b[255:224]) +dst[271:256] := SaturateU16(a[287:256]) +dst[287:272] := SaturateU16(a[319:288]) +dst[303:288] := SaturateU16(a[351:320]) +dst[319:304] := SaturateU16(a[383:352]) +dst[335:320] := SaturateU16(b[287:256]) +dst[351:336] := SaturateU16(b[319:288]) +dst[367:352] := SaturateU16(b[351:320]) +dst[383:368] := SaturateU16(b[383:352]) +dst[399:384] := SaturateU16(a[415:384]) +dst[415:400] := SaturateU16(a[447:416]) +dst[431:416] := SaturateU16(a[479:448]) +dst[447:432] := SaturateU16(a[511:480]) +dst[463:448] := SaturateU16(b[415:384]) +dst[479:464] := SaturateU16(b[447:416]) +dst[495:480] := SaturateU16(b[479:448]) +dst[511:496] := SaturateU16(b[511:480]) +dst[MAX:512] := 0 - -
smmintrin.h
+ +
immintrin.h
 - + Integer - SSE4.1 - Compare - - - Compare packed 64-bit integers in "a" and "b" for equality, and store the results in "dst". + AVX512VL + AVX512BW + Convert + Miscellaneous + + + + + + Convert packed signed 32-bit integers from "a" and "b" to packed 16-bit integers using unsigned saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 1 - i := j*64 - dst[i+63:i] := ( a[i+63:i] == b[i+63:i] ) ? 0xFFFFFFFFFFFFFFFF : 0 +tmp_dst[15:0] := SaturateU16(a[31:0]) +tmp_dst[31:16] := SaturateU16(a[63:32]) +tmp_dst[47:32] := SaturateU16(a[95:64]) +tmp_dst[63:48] := SaturateU16(a[127:96]) +tmp_dst[79:64] := SaturateU16(b[31:0]) +tmp_dst[95:80] := SaturateU16(b[63:32]) +tmp_dst[111:96] := SaturateU16(b[95:64]) +tmp_dst[127:112] := SaturateU16(b[127:96]) +FOR j := 0 to 7 + i := j*16 + IF k[j] + dst[i+15:i] := tmp_dst[i+15:i] + ELSE + dst[i+15:i] := src[i+15:i] + FI ENDFOR +dst[MAX:128] := 0 - -
smmintrin.h
+ +
immintrin.h
 - + Integer - SSE4.1 + AVX512VL + AVX512BW Convert - - Sign extend packed 8-bit integers in "a" to packed 16-bit integers, and store the results in "dst". + Miscellaneous + + + + + Convert packed signed 32-bit integers from "a" and "b" to packed 16-bit integers using unsigned saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). +tmp_dst[15:0] := SaturateU16(a[31:0]) +tmp_dst[31:16] := SaturateU16(a[63:32]) +tmp_dst[47:32] := SaturateU16(a[95:64]) +tmp_dst[63:48] := SaturateU16(a[127:96]) +tmp_dst[79:64] := SaturateU16(b[31:0]) +tmp_dst[95:80] := SaturateU16(b[63:32]) +tmp_dst[111:96] := SaturateU16(b[95:64]) +tmp_dst[127:112] := SaturateU16(b[127:96]) FOR j := 0 to 7 + i := j*16 + IF k[j] + dst[i+15:i] := tmp_dst[i+15:i] + ELSE + dst[i+15:i] := 0 + FI +ENDFOR +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Integer + AVX512VL + AVX512BW + Convert + Miscellaneous + + + + + + Convert packed signed 16-bit integers from "a" and "b" to packed 8-bit integers using unsigned saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +tmp_dst[7:0] := SaturateU8(a[15:0]) +tmp_dst[15:8] := SaturateU8(a[31:16]) +tmp_dst[23:16] := SaturateU8(a[47:32]) +tmp_dst[31:24] := SaturateU8(a[63:48]) +tmp_dst[39:32] := SaturateU8(a[79:64]) +tmp_dst[47:40] := SaturateU8(a[95:80]) +tmp_dst[55:48] := SaturateU8(a[111:96]) +tmp_dst[63:56] := SaturateU8(a[127:112]) +tmp_dst[71:64] := SaturateU8(b[15:0]) +tmp_dst[79:72] := SaturateU8(b[31:16]) +tmp_dst[87:80] := SaturateU8(b[47:32]) +tmp_dst[95:88] := SaturateU8(b[63:48]) +tmp_dst[103:96] := SaturateU8(b[79:64]) +tmp_dst[111:104] := SaturateU8(b[95:80]) +tmp_dst[119:112] := SaturateU8(b[111:96]) +tmp_dst[127:120] := SaturateU8(b[127:112]) +tmp_dst[135:128] := SaturateU8(a[143:128]) +tmp_dst[143:136] := SaturateU8(a[159:144]) +tmp_dst[151:144] := SaturateU8(a[175:160]) +tmp_dst[159:152] := SaturateU8(a[191:176]) +tmp_dst[167:160] := SaturateU8(a[207:192]) +tmp_dst[175:168] := SaturateU8(a[223:208]) +tmp_dst[183:176] := SaturateU8(a[239:224]) +tmp_dst[191:184] := SaturateU8(a[255:240]) +tmp_dst[199:192] := SaturateU8(b[143:128]) +tmp_dst[207:200] := SaturateU8(b[159:144]) +tmp_dst[215:208] := SaturateU8(b[175:160]) +tmp_dst[223:216] := SaturateU8(b[191:176]) +tmp_dst[231:224] := SaturateU8(b[207:192]) +tmp_dst[239:232] := SaturateU8(b[223:208]) +tmp_dst[247:240] := SaturateU8(b[239:224]) +tmp_dst[255:248] := SaturateU8(b[255:240]) +FOR j := 0 to 31 i := j*8 - l := j*16 - dst[l+15:l] := SignExtend(a[i+7:i]) + IF k[j] + dst[i+7:i] := tmp_dst[i+7:i] + ELSE + dst[i+7:i] := src[i+7:i] + FI ENDFOR +dst[MAX:256] := 0 - -
smmintrin.h
+ +
immintrin.h
 - + Integer - SSE4.1 + AVX512VL + AVX512BW Convert - - Sign extend packed 8-bit integers in "a" to packed 32-bit integers, and store the results in "dst". + Miscellaneous + + + + + Convert packed signed 16-bit integers from "a" and "b" to packed 8-bit integers using unsigned saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := 32*j - k := 8*j - dst[i+31:i] := SignExtend(a[k+7:k]) +tmp_dst[7:0] := SaturateU8(a[15:0]) +tmp_dst[15:8] := SaturateU8(a[31:16]) +tmp_dst[23:16] := SaturateU8(a[47:32]) +tmp_dst[31:24] := SaturateU8(a[63:48]) +tmp_dst[39:32] := SaturateU8(a[79:64]) +tmp_dst[47:40] := SaturateU8(a[95:80]) +tmp_dst[55:48] := SaturateU8(a[111:96]) +tmp_dst[63:56] := SaturateU8(a[127:112]) +tmp_dst[71:64] := SaturateU8(b[15:0]) +tmp_dst[79:72] := SaturateU8(b[31:16]) +tmp_dst[87:80] := SaturateU8(b[47:32]) +tmp_dst[95:88] := SaturateU8(b[63:48]) +tmp_dst[103:96] := SaturateU8(b[79:64]) +tmp_dst[111:104] := SaturateU8(b[95:80]) +tmp_dst[119:112] := SaturateU8(b[111:96]) +tmp_dst[127:120] := SaturateU8(b[127:112]) +tmp_dst[135:128] := SaturateU8(a[143:128]) +tmp_dst[143:136] := SaturateU8(a[159:144]) +tmp_dst[151:144] := SaturateU8(a[175:160]) +tmp_dst[159:152] := SaturateU8(a[191:176]) +tmp_dst[167:160] := SaturateU8(a[207:192]) +tmp_dst[175:168] := SaturateU8(a[223:208]) +tmp_dst[183:176] := SaturateU8(a[239:224]) +tmp_dst[191:184] := SaturateU8(a[255:240]) +tmp_dst[199:192] := SaturateU8(b[143:128]) +tmp_dst[207:200] := SaturateU8(b[159:144]) +tmp_dst[215:208] := SaturateU8(b[175:160]) +tmp_dst[223:216] := SaturateU8(b[191:176]) +tmp_dst[231:224] := SaturateU8(b[207:192]) +tmp_dst[239:232] := SaturateU8(b[223:208]) +tmp_dst[247:240] := SaturateU8(b[239:224]) +tmp_dst[255:248] := SaturateU8(b[255:240]) +FOR j := 0 to 31 + i := j*8 + IF k[j] + dst[i+7:i] := tmp_dst[i+7:i] + ELSE + dst[i+7:i] := 0 + FI ENDFOR +dst[MAX:256] := 0 - -
smmintrin.h
- - - Integer - SSE4.1 - Convert - - Sign extend packed 8-bit integers in the low 8 bytes of "a" to packed 64-bit integers, and store the results in "dst". - -FOR j := 0 to 1 - i := 64*j - k := 8*j - dst[i+63:i] := SignExtend(a[k+7:k]) + +
immintrin.h
+ + + Integer + AVX512BW + Convert + Miscellaneous + + + + + + Convert packed signed 16-bit integers from "a" and "b" to packed 8-bit integers using unsigned saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +tmp_dst[7:0] := SaturateU8(a[15:0]) +tmp_dst[15:8] := SaturateU8(a[31:16]) +tmp_dst[23:16] := SaturateU8(a[47:32]) +tmp_dst[31:24] := SaturateU8(a[63:48]) +tmp_dst[39:32] := SaturateU8(a[79:64]) +tmp_dst[47:40] := SaturateU8(a[95:80]) +tmp_dst[55:48] := SaturateU8(a[111:96]) +tmp_dst[63:56] := SaturateU8(a[127:112]) +tmp_dst[71:64] := SaturateU8(b[15:0]) +tmp_dst[79:72] := SaturateU8(b[31:16]) +tmp_dst[87:80] := SaturateU8(b[47:32]) +tmp_dst[95:88] := SaturateU8(b[63:48]) +tmp_dst[103:96] := SaturateU8(b[79:64]) +tmp_dst[111:104] := SaturateU8(b[95:80]) +tmp_dst[119:112] := SaturateU8(b[111:96]) +tmp_dst[127:120] := SaturateU8(b[127:112]) +tmp_dst[135:128] := SaturateU8(a[143:128]) +tmp_dst[143:136] := SaturateU8(a[159:144]) +tmp_dst[151:144] := SaturateU8(a[175:160]) +tmp_dst[159:152] := SaturateU8(a[191:176]) +tmp_dst[167:160] := SaturateU8(a[207:192]) +tmp_dst[175:168] := SaturateU8(a[223:208]) +tmp_dst[183:176] := SaturateU8(a[239:224]) +tmp_dst[191:184] := SaturateU8(a[255:240]) +tmp_dst[199:192] := SaturateU8(b[143:128]) +tmp_dst[207:200] := SaturateU8(b[159:144]) +tmp_dst[215:208] := SaturateU8(b[175:160]) +tmp_dst[223:216] := SaturateU8(b[191:176]) +tmp_dst[231:224] := SaturateU8(b[207:192]) +tmp_dst[239:232] := SaturateU8(b[223:208]) +tmp_dst[247:240] := SaturateU8(b[239:224]) +tmp_dst[255:248] := SaturateU8(b[255:240]) +tmp_dst[263:256] := SaturateU8(a[271:256]) +tmp_dst[271:264] := SaturateU8(a[287:272]) +tmp_dst[279:272] := SaturateU8(a[303:288]) +tmp_dst[287:280] := SaturateU8(a[319:304]) +tmp_dst[295:288] := SaturateU8(a[335:320]) +tmp_dst[303:296] := SaturateU8(a[351:336]) +tmp_dst[311:304] := SaturateU8(a[367:352]) +tmp_dst[319:312] := SaturateU8(a[383:368]) +tmp_dst[327:320] := SaturateU8(b[271:256]) +tmp_dst[335:328] := SaturateU8(b[287:272]) +tmp_dst[343:336] := SaturateU8(b[303:288]) +tmp_dst[351:344] := SaturateU8(b[319:304]) +tmp_dst[359:352] := SaturateU8(b[335:320]) +tmp_dst[367:360] := SaturateU8(b[351:336]) +tmp_dst[375:368] := SaturateU8(b[367:352]) +tmp_dst[383:376] := SaturateU8(b[383:368]) +tmp_dst[391:384] := SaturateU8(a[399:384]) +tmp_dst[399:392] := SaturateU8(a[415:400]) +tmp_dst[407:400] := SaturateU8(a[431:416]) +tmp_dst[415:408] := SaturateU8(a[447:432]) +tmp_dst[423:416] := SaturateU8(a[463:448]) +tmp_dst[431:424] := SaturateU8(a[479:464]) +tmp_dst[439:432] := SaturateU8(a[495:480]) +tmp_dst[447:440] := SaturateU8(a[511:496]) +tmp_dst[455:448] := SaturateU8(b[399:384]) +tmp_dst[463:456] := SaturateU8(b[415:400]) +tmp_dst[471:464] := SaturateU8(b[431:416]) +tmp_dst[479:472] := SaturateU8(b[447:432]) +tmp_dst[487:480] := SaturateU8(b[463:448]) +tmp_dst[495:488] := SaturateU8(b[479:464]) +tmp_dst[503:496] := SaturateU8(b[495:480]) +tmp_dst[511:504] := SaturateU8(b[511:496]) +FOR j := 0 to 63 + i := j*8 + IF k[j] + dst[i+7:i] := tmp_dst[i+7:i] + ELSE + dst[i+7:i] := src[i+7:i] + FI ENDFOR +dst[MAX:512] := 0 - -
smmintrin.h
- - - Integer - SSE4.1 - Convert - - Sign extend packed 16-bit integers in "a" to packed 32-bit integers, and store the results in "dst". - -FOR j := 0 to 3 - i := 32*j - k := 16*j - dst[i+31:i] := SignExtend(a[k+15:k]) + +
immintrin.h
+ + + Integer + AVX512BW + Convert + Miscellaneous + + + + + Convert packed signed 16-bit integers from "a" and "b" to packed 8-bit integers using unsigned saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + +tmp_dst[7:0] := SaturateU8(a[15:0]) +tmp_dst[15:8] := SaturateU8(a[31:16]) +tmp_dst[23:16] := SaturateU8(a[47:32]) +tmp_dst[31:24] := SaturateU8(a[63:48]) +tmp_dst[39:32] := SaturateU8(a[79:64]) +tmp_dst[47:40] := SaturateU8(a[95:80]) +tmp_dst[55:48] := SaturateU8(a[111:96]) +tmp_dst[63:56] := SaturateU8(a[127:112]) +tmp_dst[71:64] := SaturateU8(b[15:0]) +tmp_dst[79:72] := SaturateU8(b[31:16]) +tmp_dst[87:80] := SaturateU8(b[47:32]) +tmp_dst[95:88] := SaturateU8(b[63:48]) +tmp_dst[103:96] := SaturateU8(b[79:64]) +tmp_dst[111:104] := SaturateU8(b[95:80]) +tmp_dst[119:112] := SaturateU8(b[111:96]) +tmp_dst[127:120] := SaturateU8(b[127:112]) +tmp_dst[135:128] := SaturateU8(a[143:128]) +tmp_dst[143:136] := SaturateU8(a[159:144]) +tmp_dst[151:144] := SaturateU8(a[175:160]) +tmp_dst[159:152] := SaturateU8(a[191:176]) +tmp_dst[167:160] := SaturateU8(a[207:192]) +tmp_dst[175:168] := SaturateU8(a[223:208]) +tmp_dst[183:176] := SaturateU8(a[239:224]) +tmp_dst[191:184] := SaturateU8(a[255:240]) +tmp_dst[199:192] := SaturateU8(b[143:128]) +tmp_dst[207:200] := SaturateU8(b[159:144]) +tmp_dst[215:208] := SaturateU8(b[175:160]) +tmp_dst[223:216] := SaturateU8(b[191:176]) +tmp_dst[231:224] := SaturateU8(b[207:192]) +tmp_dst[239:232] := SaturateU8(b[223:208]) +tmp_dst[247:240] := SaturateU8(b[239:224]) +tmp_dst[255:248] := SaturateU8(b[255:240]) +tmp_dst[263:256] := SaturateU8(a[271:256]) +tmp_dst[271:264] := SaturateU8(a[287:272]) +tmp_dst[279:272] := SaturateU8(a[303:288]) +tmp_dst[287:280] := SaturateU8(a[319:304]) +tmp_dst[295:288] := SaturateU8(a[335:320]) +tmp_dst[303:296] := SaturateU8(a[351:336]) +tmp_dst[311:304] := SaturateU8(a[367:352]) +tmp_dst[319:312] := SaturateU8(a[383:368]) +tmp_dst[327:320] := SaturateU8(b[271:256]) +tmp_dst[335:328] := SaturateU8(b[287:272]) +tmp_dst[343:336] := SaturateU8(b[303:288]) +tmp_dst[351:344] := SaturateU8(b[319:304]) +tmp_dst[359:352] := SaturateU8(b[335:320]) +tmp_dst[367:360] := SaturateU8(b[351:336]) +tmp_dst[375:368] := SaturateU8(b[367:352]) +tmp_dst[383:376] := SaturateU8(b[383:368]) +tmp_dst[391:384] := SaturateU8(a[399:384]) +tmp_dst[399:392] := SaturateU8(a[415:400]) +tmp_dst[407:400] := SaturateU8(a[431:416]) +tmp_dst[415:408] := SaturateU8(a[447:432]) +tmp_dst[423:416] := SaturateU8(a[463:448]) +tmp_dst[431:424] := SaturateU8(a[479:464]) +tmp_dst[439:432] := SaturateU8(a[495:480]) +tmp_dst[447:440] := SaturateU8(a[511:496]) +tmp_dst[455:448] := SaturateU8(b[399:384]) +tmp_dst[463:456] := SaturateU8(b[415:400]) +tmp_dst[471:464] := SaturateU8(b[431:416]) +tmp_dst[479:472] := SaturateU8(b[447:432]) +tmp_dst[487:480] := SaturateU8(b[463:448]) +tmp_dst[495:488] := SaturateU8(b[479:464]) +tmp_dst[503:496] := SaturateU8(b[495:480]) +tmp_dst[511:504] := SaturateU8(b[511:496]) +FOR j := 0 to 63 + i := j*8 + IF k[j] + dst[i+7:i] := tmp_dst[i+7:i] + ELSE + dst[i+7:i] := 0 + FI ENDFOR +dst[MAX:512] := 0 - -
smmintrin.h
- - - Integer - SSE4.1 - Convert - - Sign extend packed 16-bit integers in "a" to packed 64-bit integers, and store the results in "dst". - -FOR j := 0 to 1 - i := 64*j - k := 16*j - dst[i+63:i] := SignExtend(a[k+15:k]) + +
immintrin.h
+ + + Integer + AVX512BW + Convert + Miscellaneous + + + + Convert packed signed 16-bit integers from "a" and "b" to packed 8-bit integers using unsigned saturation, and store the results in "dst". + +dst[7:0] := SaturateU8(a[15:0]) +dst[15:8] := SaturateU8(a[31:16]) +dst[23:16] := SaturateU8(a[47:32]) +dst[31:24] := SaturateU8(a[63:48]) +dst[39:32] := SaturateU8(a[79:64]) +dst[47:40] := SaturateU8(a[95:80]) +dst[55:48] := SaturateU8(a[111:96]) +dst[63:56] := SaturateU8(a[127:112]) +dst[71:64] := SaturateU8(b[15:0]) +dst[79:72] := SaturateU8(b[31:16]) +dst[87:80] := SaturateU8(b[47:32]) +dst[95:88] := SaturateU8(b[63:48]) +dst[103:96] := SaturateU8(b[79:64]) +dst[111:104] := SaturateU8(b[95:80]) +dst[119:112] := SaturateU8(b[111:96]) +dst[127:120] := SaturateU8(b[127:112]) +dst[135:128] := SaturateU8(a[143:128]) +dst[143:136] := SaturateU8(a[159:144]) +dst[151:144] := SaturateU8(a[175:160]) +dst[159:152] := SaturateU8(a[191:176]) +dst[167:160] := SaturateU8(a[207:192]) +dst[175:168] := SaturateU8(a[223:208]) +dst[183:176] := SaturateU8(a[239:224]) +dst[191:184] := SaturateU8(a[255:240]) +dst[199:192] := SaturateU8(b[143:128]) +dst[207:200] := SaturateU8(b[159:144]) +dst[215:208] := SaturateU8(b[175:160]) +dst[223:216] := SaturateU8(b[191:176]) +dst[231:224] := SaturateU8(b[207:192]) +dst[239:232] := SaturateU8(b[223:208]) +dst[247:240] := SaturateU8(b[239:224]) +dst[255:248] := SaturateU8(b[255:240]) +dst[263:256] := SaturateU8(a[271:256]) +dst[271:264] := SaturateU8(a[287:272]) +dst[279:272] := SaturateU8(a[303:288]) +dst[287:280] := SaturateU8(a[319:304]) +dst[295:288] := SaturateU8(a[335:320]) +dst[303:296] := SaturateU8(a[351:336]) +dst[311:304] := SaturateU8(a[367:352]) +dst[319:312] := SaturateU8(a[383:368]) +dst[327:320] := SaturateU8(b[271:256]) +dst[335:328] := SaturateU8(b[287:272]) +dst[343:336] := SaturateU8(b[303:288]) +dst[351:344] := SaturateU8(b[319:304]) +dst[359:352] := SaturateU8(b[335:320]) +dst[367:360] := SaturateU8(b[351:336]) +dst[375:368] := SaturateU8(b[367:352]) +dst[383:376] := SaturateU8(b[383:368]) +dst[391:384] := SaturateU8(a[399:384]) +dst[399:392] := SaturateU8(a[415:400]) +dst[407:400] := SaturateU8(a[431:416]) +dst[415:408] := SaturateU8(a[447:432]) +dst[423:416] := SaturateU8(a[463:448]) +dst[431:424] := SaturateU8(a[479:464]) +dst[439:432] := SaturateU8(a[495:480]) +dst[447:440] := SaturateU8(a[511:496]) +dst[455:448] := SaturateU8(b[399:384]) +dst[463:456] := SaturateU8(b[415:400]) +dst[471:464] := SaturateU8(b[431:416]) +dst[479:472] := SaturateU8(b[447:432]) +dst[487:480] := SaturateU8(b[463:448]) +dst[495:488] := SaturateU8(b[479:464]) +dst[503:496] := SaturateU8(b[495:480]) +dst[511:504] := SaturateU8(b[511:496]) +dst[MAX:512] := 0 + + +
immintrin.h
+ + + Integer + AVX512VL + AVX512BW + Convert + Miscellaneous + + + + + + Convert packed signed 16-bit integers from "a" and "b" to packed 8-bit integers using unsigned saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +tmp_dst[7:0] := SaturateU8(a[15:0]) +tmp_dst[15:8] := SaturateU8(a[31:16]) +tmp_dst[23:16] := SaturateU8(a[47:32]) +tmp_dst[31:24] := SaturateU8(a[63:48]) +tmp_dst[39:32] := SaturateU8(a[79:64]) +tmp_dst[47:40] := SaturateU8(a[95:80]) +tmp_dst[55:48] := SaturateU8(a[111:96]) +tmp_dst[63:56] := SaturateU8(a[127:112]) +tmp_dst[71:64] := SaturateU8(b[15:0]) +tmp_dst[79:72] := SaturateU8(b[31:16]) +tmp_dst[87:80] := SaturateU8(b[47:32]) +tmp_dst[95:88] := SaturateU8(b[63:48]) +tmp_dst[103:96] := SaturateU8(b[79:64]) +tmp_dst[111:104] := SaturateU8(b[95:80]) +tmp_dst[119:112] := SaturateU8(b[111:96]) +tmp_dst[127:120] := SaturateU8(b[127:112]) +FOR j := 0 to 15 + i := j*8 + IF k[j] + dst[i+7:i] := tmp_dst[i+7:i] + ELSE + dst[i+7:i] := src[i+7:i] + FI ENDFOR +dst[MAX:128] := 0 - -
smmintrin.h
+ +
immintrin.h
 - + Integer - SSE4.1 + AVX512VL + AVX512BW Convert - - Sign extend packed 32-bit integers in "a" to packed 64-bit integers, and store the results in "dst". + Miscellaneous + + + + + Convert packed signed 16-bit integers from "a" and "b" to packed 8-bit integers using unsigned saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 1 - i := 64*j - k := 32*j - dst[i+63:i] := SignExtend(a[k+31:k]) +tmp_dst[7:0] := SaturateU8(a[15:0]) +tmp_dst[15:8] := SaturateU8(a[31:16]) +tmp_dst[23:16] := SaturateU8(a[47:32]) +tmp_dst[31:24] := SaturateU8(a[63:48]) +tmp_dst[39:32] := SaturateU8(a[79:64]) +tmp_dst[47:40] := SaturateU8(a[95:80]) +tmp_dst[55:48] := SaturateU8(a[111:96]) +tmp_dst[63:56] := SaturateU8(a[127:112]) +tmp_dst[71:64] := SaturateU8(b[15:0]) +tmp_dst[79:72] := SaturateU8(b[31:16]) +tmp_dst[87:80] := SaturateU8(b[47:32]) +tmp_dst[95:88] := SaturateU8(b[63:48]) +tmp_dst[103:96] := SaturateU8(b[79:64]) +tmp_dst[111:104] := SaturateU8(b[95:80]) +tmp_dst[119:112] := SaturateU8(b[111:96]) +tmp_dst[127:120] := SaturateU8(b[127:112]) +FOR j := 0 to 15 + i := j*8 + IF k[j] + dst[i+7:i] := tmp_dst[i+7:i] + ELSE + dst[i+7:i] := 0 + FI ENDFOR +dst[MAX:128] := 0 - -
smmintrin.h
+ +
immintrin.h
 - + Integer - SSE4.1 - Convert - - Zero extend packed unsigned 8-bit integers in "a" to packed 16-bit integers, and store the results in "dst". + AVX512VL + AVX512BW + Arithmetic + + + + + + Add packed 8-bit integers in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 7 +FOR j := 0 to 31 i := j*8 - l := j*16 - dst[l+15:l] := ZeroExtend(a[i+7:i]) + IF k[j] + dst[i+7:i] := a[i+7:i] + b[i+7:i] + ELSE + dst[i+7:i] := src[i+7:i] + FI ENDFOR +dst[MAX:256] := 0 - -
smmintrin.h
+ +
immintrin.h
 - + Integer - SSE4.1 - Convert - - Zero extend packed unsigned 8-bit integers in "a" to packed 32-bit integers, and store the results in "dst". - -FOR j := 0 to 3 - i := 32*j - k := 8*j - dst[i+31:i] := ZeroExtend(a[k+7:k]) + AVX512VL + AVX512BW + Arithmetic + + + + + Add packed 8-bit integers in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + +FOR j := 0 to 31 + i := j*8 + IF k[j] + dst[i+7:i] := a[i+7:i] + b[i+7:i] + ELSE + dst[i+7:i] := 0 + FI ENDFOR +dst[MAX:256] := 0 - -
smmintrin.h
+ +
immintrin.h
 - + Integer - SSE4.1 - Convert - - Zero extend packed unsigned 8-bit integers in the low 8 byte sof "a" to packed 64-bit integers, and store the results in "dst". + AVX512BW + Arithmetic + + + + Add packed 8-bit integers in "a" and "b", and store the results in "dst". -FOR j := 0 to 1 - i := 64*j - k := 8*j - dst[i+63:i] := ZeroExtend(a[k+7:k]) +FOR j := 0 to 63 + i := j*8 + dst[i+7:i] := a[i+7:i] + b[i+7:i] ENDFOR +dst[MAX:512] := 0 - -
smmintrin.h
+ +
immintrin.h
 - + Integer - SSE4.1 - Convert - - Zero extend packed unsigned 16-bit integers in "a" to packed 32-bit integers, and store the results in "dst". + AVX512BW + Arithmetic + + + + + + Add packed 8-bit integers in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := 32*j - k := 16*j - dst[i+31:i] := ZeroExtend(a[k+15:k]) +FOR j := 0 to 63 + i := j*8 + IF k[j] + dst[i+7:i] := a[i+7:i] + b[i+7:i] + ELSE + dst[i+7:i] := src[i+7:i] + FI ENDFOR +dst[MAX:512] := 0 - -
smmintrin.h
+ +
immintrin.h
 - + Integer - SSE4.1 - Convert - - Zero extend packed unsigned 16-bit integers in "a" to packed 64-bit integers, and store the results in "dst". + AVX512BW + Arithmetic + + + + + Add packed 8-bit integers in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 1 - i := 64*j - k := 16*j - dst[i+63:i] := ZeroExtend(a[k+15:k]) +FOR j := 0 to 63 + i := j*8 + IF k[j] + dst[i+7:i] := a[i+7:i] + b[i+7:i] + ELSE + dst[i+7:i] := 0 + FI ENDFOR +dst[MAX:512] := 0 - -
smmintrin.h
+ +
immintrin.h
 - + Integer - SSE4.1 - Convert - - Zero extend packed unsigned 32-bit integers in "a" to packed 64-bit integers, and store the results in "dst". + AVX512VL + AVX512BW + Arithmetic + + + + + + Add packed 8-bit integers in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 1 - i := 64*j - k := 32*j - dst[i+63:i] := ZeroExtend(a[k+31:k]) +FOR j := 0 to 15 + i := j*8 + IF k[j] + dst[i+7:i] := a[i+7:i] + b[i+7:i] + ELSE + dst[i+7:i] := src[i+7:i] + FI ENDFOR +dst[MAX:128] := 0 - -
smmintrin.h
+ +
immintrin.h
 - + Integer - SSE4.1 + AVX512VL + AVX512BW Arithmetic - - - Multiply the low 32-bit integers from each packed 64-bit element in "a" and "b", and store the signed 64-bit results in "dst". + + + + + Add packed 8-bit integers in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 1 - i := j*64 - dst[i+63:i] := a[i+31:i] * b[i+31:i] +FOR j := 0 to 15 + i := j*8 + IF k[j] + dst[i+7:i] := a[i+7:i] + b[i+7:i] + ELSE + dst[i+7:i] := 0 + FI ENDFOR +dst[MAX:128] := 0 - -
smmintrin.h
+ +
immintrin.h
 - + Integer - SSE4.1 + AVX512VL + AVX512BW Arithmetic - - - Multiply the packed 32-bit integers in "a" and "b", producing intermediate 64-bit integers, and store the low 32 bits of the intermediate integers in "dst". + + + + + + Add packed signed 8-bit integers in "a" and "b" using saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*32 - tmp[63:0] := a[i+31:i] * b[i+31:i] - dst[i+31:i] := tmp[31:0] +FOR j := 0 to 31 + i := j*8 + IF k[j] + dst[i+7:i] := Saturate8( a[i+7:i] + b[i+7:i] ) + ELSE + dst[i+7:i] := src[i+7:i] + FI ENDFOR +dst[MAX:256] := 0 - -
smmintrin.h
+ +
immintrin.h
 - + Integer - SSE4.1 - Logical - - - Compute the bitwise AND of 128 bits (representing integer data) in "a" and "b", and set "ZF" to 1 if the result is zero, otherwise set "ZF" to 0. Compute the bitwise NOT of "a" and then AND with "b", and set "CF" to 1 if the result is zero, otherwise set "CF" to 0. Return the "ZF" value. + AVX512VL + AVX512BW + Arithmetic + + + + + Add packed signed 8-bit integers in "a" and "b" using saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -IF (a[127:0] AND b[127:0] == 0) - ZF := 1 -ELSE - ZF := 0 -FI -IF ((NOT a[127:0]) AND b[127:0] == 0) - CF := 1 -ELSE - CF := 0 -FI -RETURN ZF +FOR j := 0 to 31 + i := j*8 + IF k[j] + dst[i+7:i] := Saturate8( a[i+7:i] + b[i+7:i] ) + ELSE + dst[i+7:i] := 0 + FI +ENDFOR +dst[MAX:256] := 0 - -
smmintrin.h
+ +
immintrin.h
 - + Integer - SSE4.1 - Logical - - - Compute the bitwise AND of 128 bits (representing integer data) in "a" and "b", and set "ZF" to 1 if the result is zero, otherwise set "ZF" to 0. Compute the bitwise NOT of "a" and then AND with "b", and set "CF" to 1 if the result is zero, otherwise set "CF" to 0. Return the "CF" value. + AVX512BW + Arithmetic + + + + Add packed signed 8-bit integers in "a" and "b" using saturation, and store the results in "dst". -IF (a[127:0] AND b[127:0] == 0) - ZF := 1 -ELSE - ZF := 0 -FI -IF ((NOT a[127:0]) AND b[127:0] == 0) - CF := 1 -ELSE - CF := 0 -FI -RETURN CF +FOR j := 0 to 63 + i := j*8 + dst[i+7:i] := Saturate8( a[i+7:i] + b[i+7:i] ) +ENDFOR +dst[MAX:512] := 0 - -
smmintrin.h
+ +
immintrin.h
 - + Integer - SSE4.1 - Logical - - - Compute the bitwise AND of 128 bits (representing integer data) in "a" and "b", and set "ZF" to 1 if the result is zero, otherwise set "ZF" to 0. Compute the bitwise NOT of "a" and then AND with "b", and set "CF" to 1 if the result is zero, otherwise set "CF" to 0. Return 1 if both the "ZF" and "CF" values are zero, otherwise return 0. + AVX512BW + Arithmetic + + + + + + Add packed signed 8-bit integers in "a" and "b" using saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -IF (a[127:0] AND b[127:0] == 0) - ZF := 1 -ELSE - ZF := 0 -FI -IF ((NOT a[127:0]) AND b[127:0] == 0) - CF := 1 -ELSE - CF := 0 -FI -IF (ZF == 0 && CF == 0) - RETURN 1 -ELSE - RETURN 0 -FI +FOR j := 0 to 63 + i := j*8 + IF k[j] + dst[i+7:i] := Saturate8( a[i+7:i] + b[i+7:i] ) + ELSE + dst[i+7:i] := src[i+7:i] + FI +ENDFOR +dst[MAX:512] := 0 - -
smmintrin.h
+ +
immintrin.h
 - + Integer - SSE4.1 - Logical - - - Compute the bitwise AND of 128 bits (representing integer data) in "a" and "mask", and return 1 if the result is zero, otherwise return 0. + AVX512BW + Arithmetic + + + + + Add packed signed 8-bit integers in "a" and "b" using saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -IF (a[127:0] AND mask[127:0] == 0) - ZF := 1 -ELSE - ZF := 0 -FI -RETURN ZF +FOR j := 0 to 63 + i := j*8 + IF k[j] + dst[i+7:i] := Saturate8( a[i+7:i] + b[i+7:i] ) + ELSE + dst[i+7:i] := 0 + FI +ENDFOR +dst[MAX:512] := 0 - -
smmintrin.h
+ +
immintrin.h
 - + Integer - SSE4.1 - Logical - - - Compute the bitwise AND of 128 bits (representing integer data) in "a" and "mask", and set "ZF" to 1 if the result is zero, otherwise set "ZF" to 0. Compute the bitwise NOT of "a" and then AND with "mask", and set "CF" to 1 if the result is zero, otherwise set "CF" to 0. Return 1 if both the "ZF" and "CF" values are zero, otherwise return 0. + AVX512VL + AVX512BW + Arithmetic + + + + + + Add packed signed 8-bit integers in "a" and "b" using saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -IF (a[127:0] AND mask[127:0] == 0) - ZF := 1 -ELSE - ZF := 0 -FI -IF ((NOT a[127:0]) AND mask[127:0] == 0) - CF := 1 -ELSE - CF := 0 -FI -IF (ZF == 0 && CF == 0) - RETURN 1 -ELSE - RETURN 0 -FI +FOR j := 0 to 15 + i := j*8 + IF k[j] + dst[i+7:i] := Saturate8( a[i+7:i] + b[i+7:i] ) + ELSE + dst[i+7:i] := src[i+7:i] + FI +ENDFOR +dst[MAX:128] := 0 - -
smmintrin.h
+ +
immintrin.h
 - + Integer - SSE4.1 - Logical - - Compute the bitwise NOT of "a" and then AND with a 128-bit vector containing all 1's, and return 1 if the result is zero, otherwise return 0. + AVX512VL + AVX512BW + Arithmetic + + + + + Add packed signed 8-bit integers in "a" and "b" using saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 127 - tmp[i] := 1 +FOR j := 0 to 15 + i := j*8 + IF k[j] + dst[i+7:i] := Saturate8( a[i+7:i] + b[i+7:i] ) + ELSE + dst[i+7:i] := 0 + FI ENDFOR - -IF ((NOT a[127:0]) AND tmp[127:0] == 0) - CF := 1 -ELSE - CF := 0 -FI -RETURN CF +dst[MAX:128] := 0 - - -
smmintrin.h
+ +
immintrin.h
 - - Floating Point - SSE4.1 - Special Math Functions - - - Round the packed double-precision (64-bit) floating-point elements in "a" using the "rounding" parameter, and store the results as packed double-precision floating-point elements in "dst". - [round_note] - + + Integer + AVX512VL + AVX512BW + Arithmetic + + + + + + Add packed signed 16-bit integers in "a" and "b" using saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 1 - i := j*64 - dst[i+63:i] := ROUND(a[i+63:i]) +FOR j := 0 to 15 + i := j*16 + IF k[j] + dst[i+15:i] := Saturate16( a[i+15:i] + b[i+15:i] ) + ELSE + dst[i+15:i] := src[i+15:i] + FI ENDFOR +dst[MAX:256] := 0 - -
smmintrin.h
+ +
immintrin.h
 - - Floating Point - SSE4.1 - Special Math Functions - - Round the packed double-precision (64-bit) floating-point elements in "a" down to an integer value, and store the results as packed double-precision floating-point elements in "dst". + + Integer + AVX512VL + AVX512BW + Arithmetic + + + + + Add packed signed 16-bit integers in "a" and "b" using saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 1 - i := j*64 - dst[i+63:i] := FLOOR(a[i+63:i]) +FOR j := 0 to 15 + i := j*16 + IF k[j] + dst[i+15:i] := Saturate16( a[i+15:i] + b[i+15:i] ) + ELSE + dst[i+15:i] := 0 + FI ENDFOR +dst[MAX:256] := 0 - -
smmintrin.h
+ +
immintrin.h
 - - Floating Point - SSE4.1 - Special Math Functions - - Round the packed double-precision (64-bit) floating-point elements in "a" up to an integer value, and store the results as packed double-precision floating-point elements in "dst". + + Integer + AVX512BW + Arithmetic + + + + Add packed signed 16-bit integers in "a" and "b" using saturation, and store the results in "dst". -FOR j := 0 to 1 - i := j*64 - dst[i+63:i] := CEIL(a[i+63:i]) +FOR j := 0 to 31 + i := j*16 + dst[i+15:i] := Saturate16( a[i+15:i] + b[i+15:i] ) ENDFOR +dst[MAX:512] := 0 - -
smmintrin.h
+ +
immintrin.h
 - - Floating Point - SSE4.1 - Special Math Functions - - - Round the packed single-precision (32-bit) floating-point elements in "a" using the "rounding" parameter, and store the results as packed single-precision floating-point elements in "dst". - [round_note] - + + Integer + AVX512BW + Arithmetic + + + + + + Add packed signed 16-bit integers in "a" and "b" using saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*32 - dst[i+31:i] := ROUND(a[i+31:i]) +FOR j := 0 to 31 + i := j*16 + IF k[j] + dst[i+15:i] := Saturate16( a[i+15:i] + b[i+15:i] ) + ELSE + dst[i+15:i] := src[i+15:i] + FI ENDFOR +dst[MAX:512] := 0 - -
smmintrin.h
+ +
immintrin.h
 - - Floating Point - SSE4.1 - Special Math Functions - - Round the packed single-precision (32-bit) floating-point elements in "a" down to an integer value, and store the results as packed single-precision floating-point elements in "dst". + + Integer + AVX512BW + Arithmetic + + + + + Add packed signed 16-bit integers in "a" and "b" using saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*32 - dst[i+31:i] := FLOOR(a[i+31:i]) +FOR j := 0 to 31 + i := j*16 + IF k[j] + dst[i+15:i] := Saturate16( a[i+15:i] + b[i+15:i] ) + ELSE + dst[i+15:i] := 0 + FI ENDFOR +dst[MAX:512] := 0 - -
smmintrin.h
+ +
immintrin.h
 - - Floating Point - SSE4.1 - Special Math Functions - - Round the packed single-precision (32-bit) floating-point elements in "a" up to an integer value, and store the results as packed single-precision floating-point elements in "dst". + + Integer + AVX512VL + AVX512BW + Arithmetic + + + + + + Add packed signed 16-bit integers in "a" and "b" using saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*32 - dst[i+31:i] := CEIL(a[i+31:i]) +FOR j := 0 to 7 + i := j*16 + IF k[j] + dst[i+15:i] := Saturate16( a[i+15:i] + b[i+15:i] ) + ELSE + dst[i+15:i] := src[i+15:i] + FI ENDFOR +dst[MAX:128] := 0 - -
smmintrin.h
+ +
immintrin.h
 - - Floating Point - SSE4.1 - Special Math Functions - - - - Round the lower double-precision (64-bit) floating-point element in "b" using the "rounding" parameter, store the result as a double-precision floating-point element in the lower element of "dst", and copy the upper element from "a" to the upper element of "dst". - [round_note] - + + Integer + AVX512VL + AVX512BW + Arithmetic + + + + + Add packed signed 16-bit integers in "a" and "b" using saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -dst[63:0] := ROUND(b[63:0]) -dst[127:64] := a[127:64] +FOR j := 0 to 7 + i := j*16 + IF k[j] + dst[i+15:i] := Saturate16( a[i+15:i] + b[i+15:i] ) + ELSE + dst[i+15:i] := 0 + FI +ENDFOR +dst[MAX:128] := 0 - -
smmintrin.h
+ +
immintrin.h
 - - Floating Point - SSE4.1 - Special Math Functions - - - Round the lower double-precision (64-bit) floating-point element in "b" down to an integer value, store the result as a double-precision floating-point element in the lower element of "dst", and copy the upper element from "a" to the upper element of "dst". + + Integer + AVX512VL + AVX512BW + Arithmetic + + + + + + Add packed unsigned 8-bit integers in "a" and "b" using saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -dst[63:0] := FLOOR(b[63:0]) -dst[127:64] := a[127:64] +FOR j := 0 to 31 + i := j*8 + IF k[j] + dst[i+7:i] := SaturateU8( a[i+7:i] + b[i+7:i] ) + ELSE + dst[i+7:i] := src[i+7:i] + FI +ENDFOR +dst[MAX:256] := 0 - -
smmintrin.h
+ +
immintrin.h
 - - Floating Point - SSE4.1 - Special Math Functions - - - Round the lower double-precision (64-bit) floating-point element in "b" up to an integer value, store the result as a double-precision floating-point element in the lower element of "dst", and copy the upper element from "a" to the upper element of "dst". + + Integer + AVX512VL + AVX512BW + Arithmetic + + + + + Add packed unsigned 8-bit integers in "a" and "b" using saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -dst[63:0] := CEIL(b[63:0]) -dst[127:64] := a[127:64] +FOR j := 0 to 31 + i := j*8 + IF k[j] + dst[i+7:i] := SaturateU8( a[i+7:i] + b[i+7:i] ) + ELSE + dst[i+7:i] := 0 + FI +ENDFOR +dst[MAX:256] := 0 - -
smmintrin.h
+ +
immintrin.h
 - - Floating Point - SSE4.1 - Special Math Functions - - - - Round the lower single-precision (32-bit) floating-point element in "b" using the "rounding" parameter, store the result as a single-precision floating-point element in the lower element of "dst", and copy the upper 3 packed elements from "a" to the upper elements of "dst". - [round_note] - + + Integer + AVX512BW + Arithmetic + + + + Add packed unsigned 8-bit integers in "a" and "b" using saturation, and store the results in "dst". -dst[31:0] := ROUND(b[31:0]) -dst[127:32] := a[127:32] +FOR j := 0 to 63 + i := j*8 + dst[i+7:i] := SaturateU8( a[i+7:i] + b[i+7:i] ) +ENDFOR +dst[MAX:512] := 0 - -
smmintrin.h
+ +
immintrin.h
 - - Floating Point - SSE4.1 - Special Math Functions - - - Round the lower single-precision (32-bit) floating-point element in "b" down to an integer value, store the result as a single-precision floating-point element in the lower element of "dst", and copy the upper 3 packed elements from "a" to the upper elements of "dst". + + Integer + AVX512BW + Arithmetic + + + + + + Add packed unsigned 8-bit integers in "a" and "b" using saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -dst[31:0] := FLOOR(b[31:0]) -dst[127:32] := a[127:32] +FOR j := 0 to 63 + i := j*8 + IF k[j] + dst[i+7:i] := SaturateU8( a[i+7:i] + b[i+7:i] ) + ELSE + dst[i+7:i] := src[i+7:i] + FI +ENDFOR +dst[MAX:512] := 0 - -
smmintrin.h
+ +
immintrin.h
 - - Floating Point - SSE4.1 - Special Math Functions - - - Round the lower single-precision (32-bit) floating-point element in "b" up to an integer value, store the result as a single-precision floating-point element in the lower element of "dst", and copy the upper 3 packed elements from "a" to the upper elements of "dst". + + Integer + AVX512BW + Arithmetic + + + + + Add packed unsigned 8-bit integers in "a" and "b" using saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -dst[31:0] := CEIL(b[31:0]) -dst[127:32] := a[127:32] +FOR j := 0 to 63 + i := j*8 + IF k[j] + dst[i+7:i] := SaturateU8( a[i+7:i] + b[i+7:i] ) + ELSE + dst[i+7:i] := 0 + FI +ENDFOR +dst[MAX:512] := 0 - -
smmintrin.h
+ +
immintrin.h
 - + Integer - SSE4.1 - Miscellaneous - - Horizontally compute the minimum amongst the packed unsigned 16-bit integers in "a", store the minimum and index in "dst", and zero the remaining bits in "dst". + AVX512VL + AVX512BW + Arithmetic + + + + + + Add packed unsigned 8-bit integers in "a" and "b" using saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -index[2:0] := 0 -min[15:0] := a[15:0] -FOR j := 0 to 7 - i := j*16 - IF a[i+15:i] < min[15:0] - index[2:0] := j - min[15:0] := a[i+15:i] +FOR j := 0 to 15 + i := j*8 + IF k[j] + dst[i+7:i] := SaturateU8( a[i+7:i] + b[i+7:i] ) + ELSE + dst[i+7:i] := src[i+7:i] FI ENDFOR -dst[15:0] := min[15:0] -dst[18:16] := index[2:0] -dst[127:19] := 0 +dst[MAX:128] := 0 - -
smmintrin.h
+ +
immintrin.h
 - + Integer - SSE4.1 + AVX512VL + AVX512BW Arithmetic - Miscellaneous - - - - Compute the sum of absolute differences (SADs) of quadruplets of unsigned 8-bit integers in "a" compared to those in "b", and store the 16-bit results in "dst". - Eight SADs are performed using one quadruplet from "b" and eight quadruplets from "a". One quadruplet is selected from "b" starting at on the offset specified in "imm8". Eight quadruplets are formed from sequential 8-bit integers selected from "a" starting at the offset specified in "imm8". + + + + + Add packed unsigned 8-bit integers in "a" and "b" using saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -MPSADBW(a[127:0], b[127:0], imm8[2:0]) { - a_offset := imm8[2]*32 - b_offset := imm8[1:0]*32 - FOR j := 0 to 7 - i := j*8 - k := a_offset+i - l := b_offset - tmp[i*2+15:i*2] := ABS(a[k+7:k] - b[l+7:l]) + ABS(a[k+15:k+8] - b[l+15:l+8]) + ABS(a[k+23:k+16] - b[l+23:l+16]) + ABS(a[k+31:k+24] - b[l+31:l+24]) - ENDFOR - RETURN tmp[127:0] -} - -dst[127:0] := MPSADBW(a[127:0], b[127:0], imm8[2:0]) +FOR j := 0 to 15 + i := j*8 + IF k[j] + dst[i+7:i] := SaturateU8( a[i+7:i] + b[i+7:i] ) + ELSE + dst[i+7:i] := 0 + FI +ENDFOR +dst[MAX:128] := 0 - -
smmintrin.h
+ +
immintrin.h
 - + Integer - SSE4.1 - Load - - Load 128-bits of integer data from memory into "dst" using a non-temporal memory hint. - "mem_addr" must be aligned on a 16-byte boundary or a general-protection exception may be generated. + AVX512VL + AVX512BW + Arithmetic + + + + + + Add packed unsigned 16-bit integers in "a" and "b" using saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -dst[127:0] := MEM[mem_addr+127:mem_addr] +FOR j := 0 to 15 + i := j*16 + IF k[j] + dst[i+15:i] := SaturateU16( a[i+15:i] + b[i+15:i] ) + ELSE + dst[i+15:i] := src[i+15:i] + FI +ENDFOR +dst[MAX:256] := 0 - -
smmintrin.h
+ +
immintrin.h
 - - - SSE4.2 - String Compare - - - - Compare packed strings with implicit lengths in "a" and "b" using the control in "imm8", and store the generated mask in "dst". - [strcmp_note] - + + Integer + AVX512VL + AVX512BW + Arithmetic + + + + + Add packed unsigned 16-bit integers in "a" and "b" using saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -size := (imm8[0] ? 16 : 8) // 8 or 16-bit characters -UpperBound := (128 / size) - 1 - -// compare all characters -aInvalid := 0 -bInvalid := 0 -FOR i := 0 to UpperBound - m := i*size - FOR j := 0 to UpperBound - n := j*size - BoolRes[i][j] := (a[m+size-1:m] == b[n+size-1:n]) - - // invalidate characters after EOS - IF a[m+size-1:m] == 0 - aInvalid := 1 - FI - IF b[n+size-1:n] == 0 - bInvalid := 1 - FI - - // override comparisons for invalid characters - CASE (imm8[3:2]) OF - 0: // equal any - IF (!aInvalid && bInvalid) - BoolRes[i][j] := 0 - ELSE IF (aInvalid && !bInvalid) - BoolRes[i][j] := 0 - ELSE If (aInvalid && bInvalid) - BoolRes[i][j] := 0 - FI - 1: // ranges - IF (!aInvalid && bInvalid) - BoolRes[i][j] := 0 - ELSE IF (aInvalid && !bInvalid) - BoolRes[i][j] := 0 - ELSE If (aInvalid && bInvalid) - BoolRes[i][j] := 0 - FI - 2: // equal each - IF (!aInvalid && bInvalid) - BoolRes[i][j] := 0 - ELSE IF (aInvalid && !bInvalid) - BoolRes[i][j] := 0 - ELSE If (aInvalid && bInvalid) - BoolRes[i][j] := 1 - FI - 3: // equal ordered - IF (!aInvalid && bInvalid) - BoolRes[i][j] := 0 - ELSE IF (aInvalid && !bInvalid) - BoolRes[i][j] := 1 - ELSE If (aInvalid && bInvalid) - BoolRes[i][j] := 1 - FI - ESAC - ENDFOR -ENDFOR - -// aggregate results -CASE (imm8[3:2]) OF -0: // equal any - IntRes1 := 0 - FOR i := 0 to UpperBound - FOR j := 0 to UpperBound - IntRes1[i] := IntRes1[i] OR BoolRes[i][j] - ENDFOR - ENDFOR -1: // ranges - IntRes1 := 0 - FOR i := 0 to UpperBound - FOR j := 0 to UpperBound, j += 2 - IntRes1[i] := IntRes1[i] OR (BoolRes[i][j] AND BoolRes[i][j+1]) - ENDFOR - ENDFOR -2: // equal each - IntRes1 := 0 - FOR i := 0 to UpperBound - IntRes1[i] := BoolRes[i][i] - ENDFOR -3: // equal ordered - IntRes1 := (imm8[0] ? 0xFF : 0xFFFF) - FOR i := 0 to UpperBound - k := i - FOR j := 0 to UpperBound-i - IntRes1[i] := IntRes1[i] AND BoolRes[k][j] - k++ - ENDFOR - ENDFOR -ESAC - -// optionally negate results -bInvalid := 0 -FOR i := 0 to UpperBound - IF imm8[4] - IF imm8[5] // only negate valid - IF b[n+size-1:n] == 0 - bInvalid := 1 - FI - IF bInvalid // invalid, don't negate - IntRes2[i] := IntRes1[i] - ELSE // valid, negate - IntRes2[i] := -1 XOR IntRes1[i] - FI - ELSE // negate all - IntRes2[i] := -1 XOR IntRes1[i] - FI - ELSE // don't negate - IntRes2[i] := IntRes1[i] +FOR j := 0 to 15 + i := j*16 + IF k[j] + dst[i+15:i] := SaturateU16( a[i+15:i] + b[i+15:i] ) + ELSE + dst[i+15:i] := 0 FI ENDFOR - -// output -IF imm8[6] // byte / word mask - FOR i := 0 to UpperBound - j := i*size - IF IntRes2[i] - dst[j+size-1:j] := (imm8[0] ? 0xFF : 0xFFFF) - ELSE - dst[j+size-1:j] := 0 - FI - ENDFOR -ELSE // bit mask - dst[UpperBound:0] := IntRes[UpperBound:0] - dst[127:UpperBound+1] := 0 -FI +dst[MAX:256] := 0 - -
nmmintrin.h
+ +
immintrin.h
 - - SSE4.2 - String Compare - - - - Compare packed strings with implicit lengths in "a" and "b" using the control in "imm8", and store the generated index in "dst". - [strcmp_note] - + + Integer + AVX512BW + Arithmetic + + + + Add packed unsigned 16-bit integers in "a" and "b" using saturation, and store the results in "dst". -size := (imm8[0] ? 16 : 8) // 8 or 16-bit characters -UpperBound := (128 / size) - 1 - -// compare all characters -aInvalid := 0 -bInvalid := 0 -FOR i := 0 to UpperBound - m := i*size - FOR j := 0 to UpperBound - n := j*size - BoolRes[i][j] := (a[m+size-1:m] == b[n+size-1:n]) - - // invalidate characters after EOS - IF a[m+size-1:m] == 0 - aInvalid := 1 - FI - IF b[n+size-1:n] == 0 - bInvalid := 1 - FI - - // override comparisons for invalid characters - CASE (imm8[3:2]) OF - 0: // equal any - IF (!aInvalid && bInvalid) - BoolRes[i][j] := 0 - ELSE IF (aInvalid && !bInvalid) - BoolRes[i][j] := 0 - ELSE If (aInvalid && bInvalid) - BoolRes[i][j] := 0 - FI - 1: // ranges - IF (!aInvalid && bInvalid) - BoolRes[i][j] := 0 - ELSE IF (aInvalid && !bInvalid) - BoolRes[i][j] := 0 - ELSE If (aInvalid && bInvalid) - BoolRes[i][j] := 0 - FI - 2: // equal each - IF (!aInvalid && bInvalid) - BoolRes[i][j] := 0 - ELSE IF (aInvalid && !bInvalid) - BoolRes[i][j] := 0 - ELSE If (aInvalid && bInvalid) - BoolRes[i][j] := 1 - FI - 3: // equal ordered - IF (!aInvalid && bInvalid) - BoolRes[i][j] := 0 - ELSE IF (aInvalid && !bInvalid) - BoolRes[i][j] := 1 - ELSE If (aInvalid && bInvalid) - BoolRes[i][j] := 1 - FI - ESAC - ENDFOR -ENDFOR - -// aggregate results -CASE (imm8[3:2]) OF -0: // equal any - IntRes1 := 0 - FOR i := 0 to UpperBound - FOR j := 0 to UpperBound - IntRes1[i] := IntRes1[i] OR BoolRes[i][j] - ENDFOR - ENDFOR -1: // ranges - IntRes1 := 0 - FOR i := 0 to UpperBound - FOR j := 0 to UpperBound, j += 2 - IntRes1[i] := IntRes1[i] OR (BoolRes[i][j] AND BoolRes[i][j+1]) - ENDFOR - ENDFOR -2: // equal each - IntRes1 := 0 - FOR i := 0 to UpperBound - IntRes1[i] := BoolRes[i][i] - ENDFOR -3: // equal ordered - IntRes1 := (imm8[0] ? 0xFF : 0xFFFF) - FOR i := 0 to UpperBound - k := i - FOR j := 0 to UpperBound-i - IntRes1[i] := IntRes1[i] AND BoolRes[k][j] - k++ - ENDFOR - ENDFOR -ESAC - -// optionally negate results -bInvalid := 0 -FOR i := 0 to UpperBound - IF imm8[4] - IF imm8[5] // only negate valid - IF b[n+size-1:n] == 0 - bInvalid := 1 - FI - IF bInvalid // invalid, don't negate - IntRes2[i] := IntRes1[i] - ELSE // valid, negate - IntRes2[i] := -1 XOR IntRes1[i] - FI - ELSE // negate all - IntRes2[i] := -1 XOR IntRes1[i] - FI - ELSE // don't negate - IntRes2[i] := IntRes1[i] - FI +FOR j := 0 to 31 + i := j*16 + dst[i+15:i] := SaturateU16( a[i+15:i] + b[i+15:i] ) ENDFOR - -// output -IF imm8[6] // most significant bit - tmp := UpperBound - dst := tmp - DO WHILE ((tmp >= 0) AND a[tmp] = 0) - tmp := tmp - 1 - dst := tmp - OD -ELSE // least significant bit - tmp := 0 - dst := tmp - DO WHILE ((tmp <= UpperBound) AND a[tmp] = 0) - tmp := tmp + 1 - dst := tmp - OD -FI +dst[MAX:512] := 0 - -
nmmintrin.h
+ +
immintrin.h
 - - SSE4.2 - String Compare - - - - Compare packed strings with implicit lengths in "a" and "b" using the control in "imm8", and returns 1 if any character in "b" was null, and 0 otherwise. - [strcmp_note] - + + Integer + AVX512BW + Arithmetic + + + + + + Add packed unsigned 16-bit integers in "a" and "b" using saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -size := (imm8[0] ? 16 : 8) // 8 or 16-bit characters -UpperBound := (128 / size) - 1 - -bInvalid := 0 -FOR j := 0 to UpperBound - n := j*size - IF b[n+size-1:n] == 0 - bInvalid := 1 +FOR j := 0 to 31 + i := j*16 + IF k[j] + dst[i+15:i] := SaturateU16( a[i+15:i] + b[i+15:i] ) + ELSE + dst[i+15:i] := src[i+15:i] FI ENDFOR - -dst := bInvalid +dst[MAX:512] := 0 - -
nmmintrin.h
+ +
immintrin.h
 - - SSE4.2 - String Compare - - - - Compare packed strings with implicit lengths in "a" and "b" using the control in "imm8", and returns 1 if the resulting mask was non-zero, and 0 otherwise. - [strcmp_note] - + + Integer + AVX512BW + Arithmetic + + + + + Add packed unsigned 16-bit integers in "a" and "b" using saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -size := (imm8[0] ? 16 : 8) // 8 or 16-bit characters -UpperBound := (128 / size) - 1 - -// compare all characters -aInvalid := 0 -bInvalid := 0 -FOR i := 0 to UpperBound - m := i*size - FOR j := 0 to UpperBound - n := j*size - BoolRes[i][j] := (a[m+size-1:m] == b[n+size-1:n]) - - // invalidate characters after EOS - IF a[m+size-1:m] == 0 - aInvalid := 1 - FI - IF b[n+size-1:n] == 0 - bInvalid := 1 - FI - - // override comparisons for invalid characters - CASE (imm8[3:2]) OF - 0: // equal any - IF (!aInvalid && bInvalid) - BoolRes[i][j] := 0 - ELSE IF (aInvalid && !bInvalid) - BoolRes[i][j] := 0 - ELSE If (aInvalid && bInvalid) - BoolRes[i][j] := 0 - FI - 1: // ranges - IF (!aInvalid && bInvalid) - BoolRes[i][j] := 0 - ELSE IF (aInvalid && !bInvalid) - BoolRes[i][j] := 0 - ELSE If (aInvalid && bInvalid) - BoolRes[i][j] := 0 - FI - 2: // equal each - IF (!aInvalid && bInvalid) - BoolRes[i][j] := 0 - ELSE IF (aInvalid && !bInvalid) - BoolRes[i][j] := 0 - ELSE If (aInvalid && bInvalid) - BoolRes[i][j] := 1 - FI - 3: // equal ordered - IF (!aInvalid && bInvalid) - BoolRes[i][j] := 0 - ELSE IF (aInvalid && !bInvalid) - BoolRes[i][j] := 1 - ELSE If (aInvalid && bInvalid) - BoolRes[i][j] := 1 - FI - ESAC - ENDFOR -ENDFOR - -// aggregate results -CASE (imm8[3:2]) OF -0: // equal any - IntRes1 := 0 - FOR i := 0 to UpperBound - FOR j := 0 to UpperBound - IntRes1[i] := IntRes1[i] OR BoolRes[i][j] - ENDFOR - ENDFOR -1: // ranges - IntRes1 := 0 - FOR i := 0 to UpperBound - FOR j := 0 to UpperBound, j += 2 - IntRes1[i] := IntRes1[i] OR (BoolRes[i][j] AND BoolRes[i][j+1]) - ENDFOR - ENDFOR -2: // equal each - IntRes1 := 0 - FOR i := 0 to UpperBound - IntRes1[i] := BoolRes[i][i] - ENDFOR -3: // equal ordered - IntRes1 := (imm8[0] ? 0xFF : 0xFFFF) - FOR i := 0 to UpperBound - k := i - FOR j := 0 to UpperBound-i - IntRes1[i] := IntRes1[i] AND BoolRes[k][j] - k++ - ENDFOR - ENDFOR -ESAC - -// optionally negate results -bInvalid := 0 -FOR i := 0 to UpperBound - IF imm8[4] - IF imm8[5] // only negate valid - IF b[n+size-1:n] == 0 - bInvalid := 1 - FI - IF bInvalid // invalid, don't negate - IntRes2[i] := IntRes1[i] - ELSE // valid, negate - IntRes2[i] := -1 XOR IntRes1[i] - FI - ELSE // negate all - IntRes2[i] := -1 XOR IntRes1[i] - FI - ELSE // don't negate - IntRes2[i] := IntRes1[i] +FOR j := 0 to 31 + i := j*16 + IF k[j] + dst[i+15:i] := SaturateU16( a[i+15:i] + b[i+15:i] ) + ELSE + dst[i+15:i] := 0 FI ENDFOR - -// output -dst := (IntRes2 != 0) +dst[MAX:512] := 0 - -
nmmintrin.h
+ +
immintrin.h
 - - SSE4.2 - String Compare - - - - Compare packed strings with implicit lengths in "a" and "b" using the control in "imm8", and returns 1 if any character in "a" was null, and 0 otherwise. - [strcmp_note] - + + Integer + AVX512VL + AVX512BW + Arithmetic + + + + + + Add packed unsigned 16-bit integers in "a" and "b" using saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -size := (imm8[0] ? 16 : 8) // 8 or 16-bit characters -UpperBound := (128 / size) - 1 - -aInvalid := 0 -FOR i := 0 to UpperBound - m := i*size - IF b[m+size-1:m] == 0 - aInvalid := 1 +FOR j := 0 to 7 + i := j*16 + IF k[j] + dst[i+15:i] := SaturateU16( a[i+15:i] + b[i+15:i] ) + ELSE + dst[i+15:i] := src[i+15:i] FI ENDFOR - -dst := aInvalid +dst[MAX:128] := 0 - -
nmmintrin.h
+ +
immintrin.h
 - - SSE4.2 - String Compare - - - - Compare packed strings with implicit lengths in "a" and "b" using the control in "imm8", and returns bit 0 of the resulting bit mask. - [strcmp_note] - + + Integer + AVX512VL + AVX512BW + Arithmetic + + + + + Add packed unsigned 16-bit integers in "a" and "b" using saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -size := (imm8[0] ? 16 : 8) // 8 or 16-bit characters -UpperBound := (128 / size) - 1 - -// compare all characters -aInvalid := 0 -bInvalid := 0 -FOR i := 0 to UpperBound - m := i*size - FOR j := 0 to UpperBound - n := j*size - BoolRes[i][j] := (a[m+size-1:m] == b[n+size-1:n]) - - // invalidate characters after EOS - IF a[m+size-1:m] == 0 - aInvalid := 1 - FI - IF b[n+size-1:n] == 0 - bInvalid := 1 - FI - - // override comparisons for invalid characters - CASE (imm8[3:2]) OF - 0: // equal any - IF (!aInvalid && bInvalid) - BoolRes[i][j] := 0 - ELSE IF (aInvalid && !bInvalid) - BoolRes[i][j] := 0 - ELSE If (aInvalid && bInvalid) - BoolRes[i][j] := 0 - FI - 1: // ranges - IF (!aInvalid && bInvalid) - BoolRes[i][j] := 0 - ELSE IF (aInvalid && !bInvalid) - BoolRes[i][j] := 0 - ELSE If (aInvalid && bInvalid) - BoolRes[i][j] := 0 - FI - 2: // equal each - IF (!aInvalid && bInvalid) - BoolRes[i][j] := 0 - ELSE IF (aInvalid && !bInvalid) - BoolRes[i][j] := 0 - ELSE If (aInvalid && bInvalid) - BoolRes[i][j] := 1 - FI - 3: // equal ordered - IF (!aInvalid && bInvalid) - BoolRes[i][j] := 0 - ELSE IF (aInvalid && !bInvalid) - BoolRes[i][j] := 1 - ELSE If (aInvalid && bInvalid) - BoolRes[i][j] := 1 - FI - ESAC - ENDFOR -ENDFOR - -// aggregate results -CASE (imm8[3:2]) OF -0: // equal any - IntRes1 := 0 - FOR i := 0 to UpperBound - FOR j := 0 to UpperBound - IntRes1[i] := IntRes1[i] OR BoolRes[i][j] - ENDFOR - ENDFOR -1: // ranges - IntRes1 := 0 - FOR i := 0 to UpperBound - FOR j := 0 to UpperBound, j += 2 - IntRes1[i] := IntRes1[i] OR (BoolRes[i][j] AND BoolRes[i][j+1]) - ENDFOR - ENDFOR -2: // equal each - IntRes1 := 0 - FOR i := 0 to UpperBound - IntRes1[i] := BoolRes[i][i] - ENDFOR -3: // equal ordered - IntRes1 := (imm8[0] ? 0xFF : 0xFFFF) - FOR i := 0 to UpperBound - k := i - FOR j := 0 to UpperBound-i - IntRes1[i] := IntRes1[i] AND BoolRes[k][j] - k++ - ENDFOR - ENDFOR -ESAC - -// optionally negate results -bInvalid := 0 -FOR i := 0 to UpperBound - IF imm8[4] - IF imm8[5] // only negate valid - IF b[n+size-1:n] == 0 - bInvalid := 1 - FI - IF bInvalid // invalid, don't negate - IntRes2[i] := IntRes1[i] - ELSE // valid, negate - IntRes2[i] := -1 XOR IntRes1[i] - FI - ELSE // negate all - IntRes2[i] := -1 XOR IntRes1[i] - FI - ELSE // don't negate - IntRes2[i] := IntRes1[i] +FOR j := 0 to 7 + i := j*16 + IF k[j] + dst[i+15:i] := SaturateU16( a[i+15:i] + b[i+15:i] ) + ELSE + dst[i+15:i] := 0 FI ENDFOR - -// output -dst := IntRes2[0] +dst[MAX:128] := 0 - -
nmmintrin.h
+ +
immintrin.h
 - - SSE4.2 - String Compare - - - - Compare packed strings with implicit lengths in "a" and "b" using the control in "imm8", and returns 1 if "b" did not contain a null character and the resulting mask was zero, and 0 otherwise. - [strcmp_note] - + + Integer + AVX512VL + AVX512BW + Arithmetic + + + + + + Add packed 16-bit integers in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -size := (imm8[0] ? 16 : 8) // 8 or 16-bit characters -UpperBound := (128 / size) - 1 - -// compare all characters -aInvalid := 0 -bInvalid := 0 -FOR i := 0 to UpperBound - m := i*size - FOR j := 0 to UpperBound - n := j*size - BoolRes[i][j] := (a[m+size-1:m] == b[n+size-1:n]) - - // invalidate characters after EOS - IF a[m+size-1:m] == 0 - aInvalid := 1 - FI - IF b[n+size-1:n] == 0 - bInvalid := 1 - FI - - // override comparisons for invalid characters - CASE (imm8[3:2]) OF - 0: // equal any - IF (!aInvalid && bInvalid) - BoolRes[i][j] := 0 - ELSE IF (aInvalid && !bInvalid) - BoolRes[i][j] := 0 - ELSE If (aInvalid && bInvalid) - BoolRes[i][j] := 0 - FI - 1: // ranges - IF (!aInvalid && bInvalid) - BoolRes[i][j] := 0 - ELSE IF (aInvalid && !bInvalid) - BoolRes[i][j] := 0 - ELSE If (aInvalid && bInvalid) - BoolRes[i][j] := 0 - FI - 2: // equal each - IF (!aInvalid && bInvalid) - BoolRes[i][j] := 0 - ELSE IF (aInvalid && !bInvalid) - BoolRes[i][j] := 0 - ELSE If (aInvalid && bInvalid) - BoolRes[i][j] := 1 - FI - 3: // equal ordered - IF (!aInvalid && bInvalid) - BoolRes[i][j] := 0 - ELSE IF (aInvalid && !bInvalid) - BoolRes[i][j] := 1 - ELSE If (aInvalid && bInvalid) - BoolRes[i][j] := 1 - FI - ESAC - ENDFOR -ENDFOR - -// aggregate results -CASE (imm8[3:2]) OF -0: // equal any - IntRes1 := 0 - FOR i := 0 to UpperBound - FOR j := 0 to UpperBound - IntRes1[i] := IntRes1[i] OR BoolRes[i][j] - ENDFOR - ENDFOR -1: // ranges - IntRes1 := 0 - FOR i := 0 to UpperBound - FOR j := 0 to UpperBound, j += 2 - IntRes1[i] := IntRes1[i] OR (BoolRes[i][j] AND BoolRes[i][j+1]) - ENDFOR - ENDFOR -2: // equal each - IntRes1 := 0 - FOR i := 0 to UpperBound - IntRes1[i] := BoolRes[i][i] - ENDFOR -3: // equal ordered - IntRes1 := (imm8[0] ? 0xFF : 0xFFFF) - FOR i := 0 to UpperBound - k := i - FOR j := 0 to UpperBound-i - IntRes1[i] := IntRes1[i] AND BoolRes[k][j] - k++ - ENDFOR - ENDFOR -ESAC - -// optionally negate results -bInvalid := 0 -FOR i := 0 to UpperBound - IF imm8[4] - IF imm8[5] // only negate valid - IF b[n+size-1:n] == 0 - bInvalid := 1 - FI - IF bInvalid // invalid, don't negate - IntRes2[i] := IntRes1[i] - ELSE // valid, negate - IntRes2[i] := -1 XOR IntRes1[i] - FI - ELSE // negate all - IntRes2[i] := -1 XOR IntRes1[i] - FI - ELSE // don't negate - IntRes2[i] := IntRes1[i] +FOR j := 0 to 15 + i := j*16 + IF k[j] + dst[i+15:i] := a[i+15:i] + b[i+15:i] + ELSE + dst[i+15:i] := src[i+15:i] FI ENDFOR - -// output -dst := (IntRes2 == 0) AND bInvalid +dst[MAX:256] := 0 - -
nmmintrin.h
+ +
immintrin.h
 - - SSE4.2 - String Compare - - - - - - Compare packed strings in "a" and "b" with lengths "la" and "lb" using the control in "imm8", and store the generated mask in "dst". - [strcmp_note] - + + Integer + AVX512VL + AVX512BW + Arithmetic + + + + + Add packed 16-bit integers in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -size := (imm8[0] ? 16 : 8) // 8 or 16-bit characters -UpperBound := (128 / size) - 1 - -// compare all characters -aInvalid := 0 -bInvalid := 0 -FOR i := 0 to UpperBound - m := i*size - FOR j := 0 to UpperBound - n := j*size - BoolRes[i][j] := (a[m+size-1:m] == b[n+size-1:n]) - - // invalidate characters after EOS - IF i == la - aInvalid := 1 - FI - IF j == lb - bInvalid := 1 - FI - - // override comparisons for invalid characters - CASE (imm8[3:2]) OF - 0: // equal any - IF (!aInvalid && bInvalid) - BoolRes[i][j] := 0 - ELSE IF (aInvalid && !bInvalid) - BoolRes[i][j] := 0 - ELSE If (aInvalid && bInvalid) - BoolRes[i][j] := 0 - FI - 1: // ranges - IF (!aInvalid && bInvalid) - BoolRes[i][j] := 0 - ELSE IF (aInvalid && !bInvalid) - BoolRes[i][j] := 0 - ELSE If (aInvalid && bInvalid) - BoolRes[i][j] := 0 - FI - 2: // equal each - IF (!aInvalid && bInvalid) - BoolRes[i][j] := 0 - ELSE IF (aInvalid && !bInvalid) - BoolRes[i][j] := 0 - ELSE If (aInvalid && bInvalid) - BoolRes[i][j] := 1 - FI - 3: // equal ordered - IF (!aInvalid && bInvalid) - BoolRes[i][j] := 0 - ELSE IF (aInvalid && !bInvalid) - BoolRes[i][j] := 1 - ELSE If (aInvalid && bInvalid) - BoolRes[i][j] := 1 - FI - ESAC - ENDFOR -ENDFOR - -// aggregate results -CASE (imm8[3:2]) OF -0: // equal any - IntRes1 := 0 - FOR i := 0 to UpperBound - FOR j := 0 to UpperBound - IntRes1[i] := IntRes1[i] OR BoolRes[i][j] - ENDFOR - ENDFOR -1: // ranges - IntRes1 := 0 - FOR i := 0 to UpperBound - FOR j := 0 to UpperBound, j += 2 - IntRes1[i] := IntRes1[i] OR (BoolRes[i][j] AND BoolRes[i][j+1]) - ENDFOR - ENDFOR -2: // equal each - IntRes1 := 0 - FOR i := 0 to UpperBound - IntRes1[i] := BoolRes[i][i] - ENDFOR -3: // equal ordered - IntRes1 := (imm8[0] ? 0xFF : 0xFFFF) - FOR i := 0 to UpperBound - k := i - FOR j := 0 to UpperBound-i - IntRes1[i] := IntRes1[i] AND BoolRes[k][j] - k++ - ENDFOR - ENDFOR -ESAC - -// optionally negate results -FOR i := 0 to UpperBound - IF imm8[4] - IF imm8[5] // only negate valid - IF i >= lb // invalid, don't negate - IntRes2[i] := IntRes1[i] - ELSE // valid, negate - IntRes2[i] := -1 XOR IntRes1[i] - FI - ELSE // negate all - IntRes2[i] := -1 XOR IntRes1[i] - FI - ELSE // don't negate - IntRes2[i] := IntRes1[i] +FOR j := 0 to 15 + i := j*16 + IF k[j] + dst[i+15:i] := a[i+15:i] + b[i+15:i] + ELSE + dst[i+15:i] := 0 FI ENDFOR - -// output -IF imm8[6] // byte / word mask - FOR i := 0 to UpperBound - j := i*size - IF IntRes2[i] - dst[j+size-1:j] := (imm8[0] ? 0xFF : 0xFFFF) - ELSE - dst[j+size-1:j] := 0 - FI - ENDFOR -ELSE // bit mask - dst[UpperBound:0] := IntRes[UpperBound:0] - dst[127:UpperBound+1] := 0 -FI +dst[MAX:256] := 0 - -
nmmintrin.h
+ +
immintrin.h
 - - SSE4.2 - String Compare - - - - - - Compare packed strings in "a" and "b" with lengths "la" and "lb" using the control in "imm8", and store the generated index in "dst". - [strcmp_note] - + + Integer + AVX512BW + Arithmetic + + + + Add packed 16-bit integers in "a" and "b", and store the results in "dst". -size := (imm8[0] ? 16 : 8) // 8 or 16-bit characters -UpperBound := (128 / size) - 1 - -// compare all characters -aInvalid := 0 -bInvalid := 0 -FOR i := 0 to UpperBound - m := i*size - FOR j := 0 to UpperBound - n := j*size - BoolRes[i][j] := (a[m+size-1:m] == b[n+size-1:n]) - - // invalidate characters after EOS - IF i == la - aInvalid := 1 - FI - IF j == lb - bInvalid := 1 - FI - - // override comparisons for invalid characters - CASE (imm8[3:2]) OF - 0: // equal any - IF (!aInvalid && bInvalid) - BoolRes[i][j] := 0 - ELSE IF (aInvalid && !bInvalid) - BoolRes[i][j] := 0 - ELSE If (aInvalid && bInvalid) - BoolRes[i][j] := 0 - FI - 1: // ranges - IF (!aInvalid && bInvalid) - BoolRes[i][j] := 0 - ELSE IF (aInvalid && !bInvalid) - BoolRes[i][j] := 0 - ELSE If (aInvalid && bInvalid) - BoolRes[i][j] := 0 - FI - 2: // equal each - IF (!aInvalid && bInvalid) - BoolRes[i][j] := 0 - ELSE IF (aInvalid && !bInvalid) - BoolRes[i][j] := 0 - ELSE If (aInvalid && bInvalid) - BoolRes[i][j] := 1 - FI - 3: // equal ordered - IF (!aInvalid && bInvalid) - BoolRes[i][j] := 0 - ELSE IF (aInvalid && !bInvalid) - BoolRes[i][j] := 1 - ELSE If (aInvalid && bInvalid) - BoolRes[i][j] := 1 - FI - ESAC - ENDFOR +FOR j := 0 to 31 + i := j*16 + dst[i+15:i] := a[i+15:i] + b[i+15:i] ENDFOR - -// aggregate results -CASE (imm8[3:2]) OF -0: // equal any - IntRes1 := 0 - FOR i := 0 to UpperBound - FOR j := 0 to UpperBound - IntRes1[i] := IntRes1[i] OR BoolRes[i][j] - ENDFOR - ENDFOR -1: // ranges - IntRes1 := 0 - FOR i := 0 to UpperBound - FOR j := 0 to UpperBound, j += 2 - IntRes1[i] := IntRes1[i] OR (BoolRes[i][j] AND BoolRes[i][j+1]) - ENDFOR - ENDFOR -2: // equal each - IntRes1 := 0 - FOR i := 0 to UpperBound - IntRes1[i] := BoolRes[i][i] - ENDFOR -3: // equal ordered - IntRes1 := (imm8[0] ? 0xFF : 0xFFFF) - FOR i := 0 to UpperBound - k := i - FOR j := 0 to UpperBound-i - IntRes1[i] := IntRes1[i] AND BoolRes[k][j] - k++ - ENDFOR - ENDFOR -ESAC - -// optionally negate results -FOR i := 0 to UpperBound - IF imm8[4] - IF imm8[5] // only negate valid - IF i >= lb // invalid, don't negate - IntRes2[i] := IntRes1[i] - ELSE // valid, negate - IntRes2[i] := -1 XOR IntRes1[i] - FI - ELSE // negate all - IntRes2[i] := -1 XOR IntRes1[i] - FI - ELSE // don't negate - IntRes2[i] := IntRes1[i] +dst[MAX:512] := 0 + + +
immintrin.h
+ + + Integer + AVX512BW + Arithmetic + + + + + + Add packed 16-bit integers in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +FOR j := 0 to 31 + i := j*16 + IF k[j] + dst[i+15:i] := a[i+15:i] + b[i+15:i] + ELSE + dst[i+15:i] := src[i+15:i] FI ENDFOR - -// output -IF imm8[6] // most significant bit - tmp := UpperBound - dst := tmp - DO WHILE ((tmp >= 0) AND a[tmp] = 0) - tmp := tmp - 1 - dst := tmp - OD -ELSE // least significant bit - tmp := 0 - dst := tmp - DO WHILE ((tmp <= UpperBound) AND a[tmp] = 0) - tmp := tmp + 1 - dst := tmp - OD -FI +dst[MAX:512] := 0 - -
nmmintrin.h
+ +
immintrin.h
 - - SSE4.2 - String Compare - - - - - - Compare packed strings in "a" and "b" with lengths "la" and "lb" using the control in "imm8", and returns 1 if any character in "b" was null, and 0 otherwise. - [strcmp_note] - + + Integer + AVX512BW + Arithmetic + + + + + Add packed 16-bit integers in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -size := (imm8[0] ? 16 : 8) // 8 or 16-bit characters -UpperBound := (128 / size) - 1 - -dst := (lb <= UpperBound) +FOR j := 0 to 31 + i := j*16 + IF k[j] + dst[i+15:i] := a[i+15:i] + b[i+15:i] + ELSE + dst[i+15:i] := 0 + FI +ENDFOR +dst[MAX:512] := 0 - -
nmmintrin.h
+ +
immintrin.h
 - - SSE4.2 - String Compare - - - - - - Compare packed strings in "a" and "b" with lengths "la" and "lb" using the control in "imm8", and returns 1 if the resulting mask was non-zero, and 0 otherwise. - [strcmp_note] - + + Integer + AVX512VL + AVX512BW + Arithmetic + + + + + + Add packed 16-bit integers in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -size := (imm8[0] ? 16 : 8) // 8 or 16-bit characters -UpperBound := (128 / size) - 1 - -// compare all characters -aInvalid := 0 -bInvalid := 0 -FOR i := 0 to UpperBound - m := i*size - FOR j := 0 to UpperBound - n := j*size - BoolRes[i][j] := (a[m+size-1:m] == b[n+size-1:n]) - - // invalidate characters after EOS - IF i == la - aInvalid := 1 - FI - IF j == lb - bInvalid := 1 - FI - - // override comparisons for invalid characters - CASE (imm8[3:2]) OF - 0: // equal any - IF (!aInvalid && bInvalid) - BoolRes[i][j] := 0 - ELSE IF (aInvalid && !bInvalid) - BoolRes[i][j] := 0 - ELSE If (aInvalid && bInvalid) - BoolRes[i][j] := 0 - FI - 1: // ranges - IF (!aInvalid && bInvalid) - BoolRes[i][j] := 0 - ELSE IF (aInvalid && !bInvalid) - BoolRes[i][j] := 0 - ELSE If (aInvalid && bInvalid) - BoolRes[i][j] := 0 - FI - 2: // equal each - IF (!aInvalid && bInvalid) - BoolRes[i][j] := 0 - ELSE IF (aInvalid && !bInvalid) - BoolRes[i][j] := 0 - ELSE If (aInvalid && bInvalid) - BoolRes[i][j] := 1 - FI - 3: // equal ordered - IF (!aInvalid && bInvalid) - BoolRes[i][j] := 0 - ELSE IF (aInvalid && !bInvalid) - BoolRes[i][j] := 1 - ELSE If (aInvalid && bInvalid) - BoolRes[i][j] := 1 - FI - ESAC - ENDFOR +FOR j := 0 to 7 + i := j*16 + IF k[j] + dst[i+15:i] := a[i+15:i] + b[i+15:i] + ELSE + dst[i+15:i] := src[i+15:i] + FI ENDFOR - -// aggregate results -CASE (imm8[3:2]) OF - 0: // equal any - IntRes1 := 0 - FOR i := 0 to UpperBound - FOR j := 0 to UpperBound - IntRes1[i] := IntRes1[i] OR BoolRes[i][j] - ENDFOR - ENDFOR - 1: // ranges - IntRes1 := 0 - FOR i := 0 to UpperBound - FOR j := 0 to UpperBound, j += 2 - IntRes1[i] := IntRes1[i] OR (BoolRes[i][j] AND BoolRes[i][j+1]) - ENDFOR - ENDFOR - 2: // equal each - IntRes1 := 0 - FOR i := 0 to UpperBound - IntRes1[i] := BoolRes[i][i] - ENDFOR - 3: // equal ordered - IntRes1 := (imm8[0] ? 0xFF : 0xFFFF) - FOR i := 0 to UpperBound - k := i - FOR j := 0 to UpperBound-i - IntRes1[i] := IntRes1[i] AND BoolRes[k][j] - k++ - ENDFOR - ENDFOR -ESAC - -// optionally negate results -FOR i := 0 to UpperBound - IF imm8[4] - IF imm8[5] // only negate valid - IF i >= lb // invalid, don't negate - IntRes2[i] := IntRes1[i] - ELSE // valid, negate - IntRes2[i] := -1 XOR IntRes1[i] - FI - ELSE // negate all - IntRes2[i] := -1 XOR IntRes1[i] - FI - ELSE // don't negate - IntRes2[i] := IntRes1[i] +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Integer + AVX512VL + AVX512BW + Arithmetic + + + + + Add packed 16-bit integers in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + +FOR j := 0 to 7 + i := j*16 + IF k[j] + dst[i+15:i] := a[i+15:i] + b[i+15:i] + ELSE + dst[i+15:i] := 0 FI ENDFOR - -// output -dst := (IntRes2 != 0) +dst[MAX:128] := 0 - -
nmmintrin.h
+ +
immintrin.h
 - - SSE4.2 - String Compare - - - - - - Compare packed strings in "a" and "b" with lengths "la" and "lb" using the control in "imm8", and returns 1 if any character in "a" was null, and 0 otherwise. - [strcmp_note] - + + Integer + AVX512VL + AVX512BW + Miscellaneous + + + + + + + Concatenate pairs of 16-byte blocks in "a" and "b" into a 32-byte temporary result, shift the result right by "imm8" bytes, and store the low 16 bytes in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -size := (imm8[0] ? 16 : 8) // 8 or 16-bit characters -UpperBound := (128 / size) - 1 - -dst := (la <= UpperBound) +FOR j := 0 to 1 + i := j*128 + tmp[255:0] := ((a[i+127:i] << 128)[255:0] OR b[i+127:i]) >> (imm8*8) + tmp_dst[i+127:i] := tmp[127:0] +ENDFOR +FOR j := 0 to 31 + i := j*8 + IF k[j] + dst[i+7:i] := tmp_dst[i+7:i] + ELSE + dst[i+7:i] := src[i+7:i] + FI +ENDFOR +dst[MAX:256] := 0 - -
nmmintrin.h
+ +
immintrin.h
 - - SSE4.2 - String Compare - - - - - - Compare packed strings in "a" and "b" with lengths "la" and "lb" using the control in "imm8", and returns bit 0 of the resulting bit mask. - [strcmp_note] - + + Integer + AVX512VL + AVX512BW + Miscellaneous + + + + + + Concatenate pairs of 16-byte blocks in "a" and "b" into a 32-byte temporary result, shift the result right by "imm8" bytes, and store the low 16 bytes in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -size := (imm8[0] ? 16 : 8) // 8 or 16-bit characters -UpperBound := (128 / size) - 1 - -// compare all characters -aInvalid := 0 -bInvalid := 0 -FOR i := 0 to UpperBound - m := i*size - FOR j := 0 to UpperBound - n := j*size - BoolRes[i][j] := (a[m+size-1:m] == b[n+size-1:n]) - - // invalidate characters after EOS - IF i == la - aInvalid := 1 - FI - IF j == lb - bInvalid := 1 - FI - - // override comparisons for invalid characters - CASE (imm8[3:2]) OF - 0: // equal any - IF (!aInvalid && bInvalid) - BoolRes[i][j] := 0 - ELSE IF (aInvalid && !bInvalid) - BoolRes[i][j] := 0 - ELSE If (aInvalid && bInvalid) - BoolRes[i][j] := 0 - FI - 1: // ranges - IF (!aInvalid && bInvalid) - BoolRes[i][j] := 0 - ELSE IF (aInvalid && !bInvalid) - BoolRes[i][j] := 0 - ELSE If (aInvalid && bInvalid) - BoolRes[i][j] := 0 - FI - 2: // equal each - IF (!aInvalid && bInvalid) - BoolRes[i][j] := 0 - ELSE IF (aInvalid && !bInvalid) - BoolRes[i][j] := 0 - ELSE If (aInvalid && bInvalid) - BoolRes[i][j] := 1 - FI - 3: // equal ordered - IF (!aInvalid && bInvalid) - BoolRes[i][j] := 0 - ELSE IF (aInvalid && !bInvalid) - BoolRes[i][j] := 1 - ELSE If (aInvalid && bInvalid) - BoolRes[i][j] := 1 - FI - ESAC - ENDFOR +FOR j := 0 to 1 + i := j*128 + tmp[255:0] := ((a[i+127:i] << 128)[255:0] OR b[i+127:i]) >> (imm8*8) + tmp_dst[i+127:i] := tmp[127:0] ENDFOR - -// aggregate results -CASE (imm8[3:2]) OF - 0: // equal any - IntRes1 := 0 - FOR i := 0 to UpperBound - FOR j := 0 to UpperBound - IntRes1[i] := IntRes1[i] OR BoolRes[i][j] - ENDFOR - ENDFOR - 1: // ranges - IntRes1 := 0 - FOR i := 0 to UpperBound - FOR j := 0 to UpperBound, j += 2 - IntRes1[i] := IntRes1[i] OR (BoolRes[i][j] AND BoolRes[i][j+1]) - ENDFOR - ENDFOR - 2: // equal each - IntRes1 := 0 - FOR i := 0 to UpperBound - IntRes1[i] := BoolRes[i][i] - ENDFOR - 3: // equal ordered - IntRes1 := (imm8[0] ? 0xFF : 0xFFFF) - FOR i := 0 to UpperBound - k := i - FOR j := 0 to UpperBound-i - IntRes1[i] := IntRes1[i] AND BoolRes[k][j] - k++ - ENDFOR - ENDFOR -ESAC - -// optionally negate results -FOR i := 0 to UpperBound - IF imm8[4] - IF imm8[5] // only negate valid - IF i >= lb // invalid, don't negate - IntRes2[i] := IntRes1[i] - ELSE // valid, negate - IntRes2[i] := -1 XOR IntRes1[i] - FI - ELSE // negate all - IntRes2[i] := -1 XOR IntRes1[i] - FI - ELSE // don't negate - IntRes2[i] := IntRes1[i] +FOR j := 0 to 31 + i := j*8 + IF k[j] + dst[i+7:i] := tmp_dst[i+7:i] + ELSE + dst[i+7:i] := 0 FI ENDFOR - -// output -dst := IntRes2[0 +dst[MAX:256] := 0 - -
nmmintrin.h
+ +
immintrin.h
 - - SSE4.2 - String Compare - - - - - - Compare packed strings in "a" and "b" with lengths "la" and "lb" using the control in "imm8", and returns 1 if "b" did not contain a null character and the resulting mask was zero, and 0 otherwise. - [strcmp_note] - + + Integer + AVX512BW + Miscellaneous + + + + + Concatenate pairs of 16-byte blocks in "a" and "b" into a 32-byte temporary result, shift the result right by "imm8" bytes, and store the low 16 bytes in "dst". -size := (imm8[0] ? 16 : 8) // 8 or 16-bit characters -UpperBound := (128 / size) - 1 - -// compare all characters -aInvalid := 0 -bInvalid := 0 -FOR i := 0 to UpperBound - m := i*size - FOR j := 0 to UpperBound - n := j*size - BoolRes[i][j] := (a[m+size-1:m] == b[n+size-1:n]) - - // invalidate characters after EOS - IF i == la - aInvalid := 1 - FI - IF j == lb - bInvalid := 1 - FI - - // override comparisons for invalid characters - CASE (imm8[3:2]) OF - 0: // equal any - IF (!aInvalid && bInvalid) - BoolRes[i][j] := 0 - ELSE IF (aInvalid && !bInvalid) - BoolRes[i][j] := 0 - ELSE If (aInvalid && bInvalid) - BoolRes[i][j] := 0 - FI - 1: // ranges - IF (!aInvalid && bInvalid) - BoolRes[i][j] := 0 - ELSE IF (aInvalid && !bInvalid) - BoolRes[i][j] := 0 - ELSE If (aInvalid && bInvalid) - BoolRes[i][j] := 0 - FI - 2: // equal each - IF (!aInvalid && bInvalid) - BoolRes[i][j] := 0 - ELSE IF (aInvalid && !bInvalid) - BoolRes[i][j] := 0 - ELSE If (aInvalid && bInvalid) - BoolRes[i][j] := 1 - FI - 3: // equal ordered - IF (!aInvalid && bInvalid) - BoolRes[i][j] := 0 - ELSE IF (aInvalid && !bInvalid) - BoolRes[i][j] := 1 - ELSE If (aInvalid && bInvalid) - BoolRes[i][j] := 1 - FI - ESAC - ENDFOR -ENDFOR - -// aggregate results -CASE (imm8[3:2]) OF - 0: // equal any - IntRes1 := 0 - FOR i := 0 to UpperBound - FOR j := 0 to UpperBound - IntRes1[i] := IntRes1[i] OR BoolRes[i][j] - ENDFOR - ENDFOR - 1: // ranges - IntRes1 := 0 - FOR i := 0 to UpperBound - FOR j := 0 to UpperBound, j += 2 - IntRes1[i] := IntRes1[i] OR (BoolRes[i][j] AND BoolRes[i][j+1]) - ENDFOR - ENDFOR - 2: // equal each - IntRes1 := 0 - FOR i := 0 to UpperBound - IntRes1[i] := BoolRes[i][i] - ENDFOR - 3: // equal ordered - IntRes1 := (imm8[0] ? 0xFF : 0xFFFF) - FOR i := 0 to UpperBound - k := i - FOR j := 0 to UpperBound-i - IntRes1[i] := IntRes1[i] AND BoolRes[k][j] - k++ - ENDFOR - ENDFOR -ESAC - -// optionally negate results -FOR i := 0 to UpperBound - IF imm8[4] - IF imm8[5] // only negate valid - IF i >= lb // invalid, don't negate - IntRes2[i] := IntRes1[i] - ELSE // valid, negate - IntRes2[i] := -1 XOR IntRes1[i] - FI - ELSE // negate all - IntRes2[i] := -1 XOR IntRes1[i] - FI - ELSE // don't negate - IntRes2[i] := IntRes1[i] - FI +FOR j := 0 to 3 + i := j*128 + tmp[255:0] := ((a[i+127:i] << 128)[255:0] OR b[i+127:i]) >> (imm8*8) + dst[i+127:i] := tmp[127:0] ENDFOR - -// output -dst := (IntRes2 == 0) AND (lb > UpperBound) +dst[MAX:512] := 0 - -
nmmintrin.h
+ +
immintrin.h
 - + Integer - SSE4.2 - Compare - - - Compare packed 64-bit integers in "a" and "b" for greater-than, and store the results in "dst". + AVX512BW + Miscellaneous + + + + + + + Concatenate pairs of 16-byte blocks in "a" and "b" into a 32-byte temporary result, shift the result right by "imm8" bytes, and store the low 16 bytes in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 1 - i := j*64 - dst[i+63:i] := ( a[i+63:i] > b[i+63:i] ) ? 0xFFFFFFFFFFFFFFFF : 0 +FOR j := 0 to 3 + i := j*128 + tmp[255:0] := ((a[i+127:i] << 128)[255:0] OR b[i+127:i]) >> (imm8*8) + tmp_dst[i+127:i] := tmp[127:0] ENDFOR +FOR j := 0 to 63 + i := j*8 + IF k[j] + dst[i+7:i] := tmp_dst[i+7:i] + ELSE + dst[i+7:i] := src[i+7:i] + FI +ENDFOR +dst[MAX:512] := 0 - -
nmmintrin.h
- - - Integer - SSE4.2 - Cryptography - - - Starting with the initial value in "crc", accumulates a CRC32 value for unsigned 8-bit integer "v", and stores the result in "dst". - -tmp1[7:0] := v[0:7] // bit reflection -tmp2[31:0] := crc[0:31] // bit reflection -tmp3[39:0] := tmp1[7:0] << 32 -tmp4[39:0] := tmp2[31:0] << 8 -tmp5[39:0] := tmp3[39:0] XOR tmp4[39:0] -tmp6[31:0] := tmp5[39:0] MOD2 0x11EDC6F41 -dst[31:0] := tmp6[0:31] // bit reflection - - -
nmmintrin.h
+ +
immintrin.h
 - + Integer - SSE4.2 - Cryptography - - - Starting with the initial value in "crc", accumulates a CRC32 value for unsigned 16-bit integer "v", and stores the result in "dst". + AVX512BW + Miscellaneous + + + + + + Concatenate pairs of 16-byte blocks in "a" and "b" into a 32-byte temporary result, shift the result right by "imm8" bytes, and store the low 16 bytes in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -tmp1[15:0] := v[0:15] // bit reflection -tmp2[31:0] := crc[0:31] // bit reflection -tmp3[47:0] := tmp1[15:0] << 32 -tmp4[47:0] := tmp2[31:0] << 16 -tmp5[47:0] := tmp3[47:0] XOR tmp4[47:0] -tmp6[31:0] := tmp5[47:0] MOD2 0x11EDC6F41 -dst[31:0] := tmp6[0:31] // bit reflection +FOR j := 0 to 3 + i := j*128 + tmp[255:0] := ((a[i+127:i] << 128)[255:0] OR b[i+127:i]) >> (imm8*8) + tmp_dst[i+127:i] := tmp[127:0] +ENDFOR +FOR j := 0 to 63 + i := j*8 + IF k[j] + dst[i+7:i] := tmp_dst[i+7:i] + ELSE + dst[i+7:i] := 0 + FI +ENDFOR +dst[MAX:512] := 0 - -
nmmintrin.h
+ +
immintrin.h
 - + Integer - SSE4.2 - Cryptography - - - Starting with the initial value in "crc", accumulates a CRC32 value for unsigned 32-bit integer "v", and stores the result in "dst". + AVX512VL + AVX512BW + Miscellaneous + + + + + + + Concatenate pairs of 16-byte blocks in "a" and "b" into a 32-byte temporary result, shift the result right by "imm8" bytes, and store the low 16 bytes in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -tmp1[31:0] := v[0:31] // bit reflection -tmp2[31:0] := crc[0:31] // bit reflection -tmp3[63:0] := tmp1[31:0] << 32 -tmp4[63:0] := tmp2[31:0] << 32 -tmp5[63:0] := tmp3[63:0] XOR tmp4[63:0] -tmp6[31:0] := tmp5[63:0] MOD2 0x11EDC6F41 -dst[31:0] := tmp6[0:31] // bit reflection +tmp_dst[255:0] := ((a[127:0] << 128)[255:0] OR b[127:0]) >> (imm8*8) +FOR j := 0 to 15 + i := j*8 + IF k[j] + dst[i+7:i] := tmp_dst[i+7:i] + ELSE + dst[i+7:i] := src[i+7:i] + FI +ENDFOR +dst[MAX:128] := 0 - -
nmmintrin.h
+ +
immintrin.h
 - + Integer - SSE4.2 - Cryptography - - - Starting with the initial value in "crc", accumulates a CRC32 value for unsigned 64-bit integer "v", and stores the result in "dst". + AVX512VL + AVX512BW + Miscellaneous + + + + + + Concatenate pairs of 16-byte blocks in "a" and "b" into a 32-byte temporary result, shift the result right by "imm8" bytes, and store the low 16 bytes in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -tmp1[63:0] := v[0:63] // bit reflection -tmp2[31:0] := crc[0:31] // bit reflection -tmp3[95:0] := tmp1[31:0] << 32 -tmp4[95:0] := tmp2[63:0] << 64 -tmp5[95:0] := tmp3[95:0] XOR tmp4[95:0] -tmp6[31:0] := tmp5[95:0] MOD2 0x11EDC6F41 -dst[31:0] := tmp6[0:31] // bit reflection +tmp_dst[255:0] := ((a[127:0] << 128)[255:0] OR b[127:0]) >> (imm8*8) +FOR j := 0 to 15 + i := j*8 + IF k[j] + dst[i+7:i] := tmp_dst[i+7:i] + ELSE + dst[i+7:i] := 0 + FI +ENDFOR +dst[MAX:128] := 0 - -
nmmintrin.h
+ +
immintrin.h
 - + Integer - POPCNT - Bit Manipulation - - - Count the number of bits set to 1 in unsigned 32-bit integer "a", and return that count in "dst". - + AVX512VL + AVX512BW + Arithmetic + + + + + + Average packed unsigned 8-bit integers in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -dst := 0 -FOR i := 0 to 31 - IF a[i] - dst := dst + 1 +FOR j := 0 to 31 + i := j*8 + IF k[j] + dst[i+7:i] := (a[i+7:i] + b[i+7:i] + 1) >> 1 + ELSE + dst[i+7:i] := src[i+7:i] FI ENDFOR +dst[MAX:256] := 0 - -
nmmintrin.h
+ +
immintrin.h
 - + Integer - POPCNT - Bit Manipulation - - - Count the number of bits set to 1 in unsigned 64-bit integer "a", and return that count in "dst". - + AVX512VL + AVX512BW + Arithmetic + + + + + Average packed unsigned 8-bit integers in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -dst := 0 -FOR i := 0 to 63 - IF a[i] - dst := dst + 1 +FOR j := 0 to 31 + i := j*8 + IF k[j] + dst[i+7:i] := (a[i+7:i] + b[i+7:i] + 1) >> 1 + ELSE + dst[i+7:i] := 0 FI ENDFOR +dst[MAX:256] := 0 - -
nmmintrin.h
- - - - Integer - AES - Cryptography - - - Perform one round of an AES encryption flow on data (state) in "a" using the round key in "RoundKey", and store the result in "dst"." - state := a -a[127:0] := ShiftRows(a[127:0]) -a[127:0] := SubBytes(a[127:0]) -a[127:0] := MixColumns(a[127:0]) -dst[127:0] := a[127:0] XOR RoundKey[127:0] - - -
wmmintrin.h
- - - Integer - AES - Cryptography - - - Perform the last round of an AES encryption flow on data (state) in "a" using the round key in "RoundKey", and store the result in "dst"." - state := a -a[127:0] := ShiftRows(a[127:0]) -a[127:0] := SubBytes(a[127:0]) -dst[127:0] := a[127:0] XOR RoundKey[127:0] - - -
wmmintrin.h
- - - Integer - AES - Cryptography - - - Perform one round of an AES decryption flow on data (state) in "a" using the round key in "RoundKey", and store the result in "dst"." - state := a -a[127:0] := InvShiftRows(a[127:0]) -a[127:0] := InvSubBytes(a[127:0]) -a[127:0] := InvMixColumns(a[127:0]) -dst[127:0] := a[127:0] XOR RoundKey[127:0] - - -
wmmintrin.h
- - - Integer - AES - Cryptography - - - Perform the last round of an AES decryption flow on data (state) in "a" using the round key in "RoundKey", and store the result in "dst"." - state := a -a[127:0] := InvShiftRows(a[127:0]) -a[127:0] := InvSubBytes(a[127:0]) -dst[127:0] := a[127:0] XOR RoundKey[127:0] - - -
wmmintrin.h
+ +
immintrin.h
 - + Integer - AES - Cryptography - - Perform the InvMixColumns transformation on "a" and store the result in "dst". + AVX512BW + Arithmetic + + + + Average packed unsigned 8-bit integers in "a" and "b", and store the results in "dst". -dst[127:0] := InvMixColumns(a[127:0]) +FOR j := 0 to 63 + i := j*8 + dst[i+7:i] := (a[i+7:i] + b[i+7:i] + 1) >> 1 +ENDFOR +dst[MAX:512] := 0 - -
wmmintrin.h
+ +
immintrin.h
 - + Integer - AES - Cryptography - - - Assist in expanding the AES cipher key by computing steps towards generating a round key for encryption cipher using data from "a" and an 8-bit round constant specified in "imm8", and store the result in "dst"." - + AVX512BW + Arithmetic + + + + + + Average packed unsigned 8-bit integers in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -X3[31:0] := a[127:96] -X2[31:0] := a[95:64] -X1[31:0] := a[63:32] -X0[31:0] := a[31:0] -RCON[31:0] := ZeroExtend(imm8[7:0]); -dst[31:0] := SubWord(X1) -dst[63:32] := (RotWord(SubWord(X1)) XOR RCON; -dst[95:64] := SubWord(X3) -dst[127:96] := RotWord(SubWord(X3)) XOR RCON; +FOR j := 0 to 63 + i := j*8 + IF k[j] + dst[i+7:i] := (a[i+7:i] + b[i+7:i] + 1) >> 1 + ELSE + dst[i+7:i] := src[i+7:i] + FI +ENDFOR +dst[MAX:512] := 0 - -
wmmintrin.h
+ +
immintrin.h
 - + Integer - PCLMULQDQ - Application-Targeted - - - - Perform a carry-less multiplication of two 64-bit integers, selected from "a" and "b" according to "imm8", and store the results in "dst". - - -IF (imm8[0] = 0) - TEMP1 := a[63:0]; -ELSE - TEMP1 := a[127:64]; -FI -IF (imm8[4] = 0) - TEMP2 := b[63:0]; -ELSE - TEMP2 := b[127:64]; -FI - -FOR i := 0 to 63 - TEMP[i] := (TEMP1[0] and TEMP2[i]); - FOR j := 1 to i - TEMP [i] := TEMP [i] XOR (TEMP1[j] AND TEMP2[i-j]) - ENDFOR - dst[i] := TEMP[i]; -ENDFOR -FOR i := 64 to 127 - TEMP [i] := 0; - FOR j := (i - 63) to 63 - TEMP [i] := TEMP [i] XOR (TEMP1[j] AND TEMP2[i-j]) - ENDFOR - dst[i] := TEMP[i]; + AVX512BW + Arithmetic + + + + + Average packed unsigned 8-bit integers in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + +FOR j := 0 to 63 + i := j*8 + IF k[j] + dst[i+7:i] := (a[i+7:i] + b[i+7:i] + 1) >> 1 + ELSE + dst[i+7:i] := 0 + FI ENDFOR -dst[127] := 0 +dst[MAX:512] := 0 - -
wmmintrin.h
+ +
immintrin.h
 - - - Floating Point - AVX + + Integer + AVX512VL + AVX512BW Arithmetic - - - Add packed double-precision (64-bit) floating-point elements in "a" and "b", and store the results in "dst". + + + + + + Average packed unsigned 8-bit integers in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*64 - dst[i+63:i] := a[i+63:i] + b[i+63:i] +FOR j := 0 to 15 + i := j*8 + IF k[j] + dst[i+7:i] := (a[i+7:i] + b[i+7:i] + 1) >> 1 + ELSE + dst[i+7:i] := src[i+7:i] + FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point - AVX + + Integer + AVX512VL + AVX512BW Arithmetic - - - Add packed single-precision (32-bit) floating-point elements in "a" and "b", and store the results in "dst". + + + + + Average packed unsigned 8-bit integers in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := j*32 - dst[i+31:i] := a[i+31:i] + b[i+31:i] +FOR j := 0 to 15 + i := j*8 + IF k[j] + dst[i+7:i] := (a[i+7:i] + b[i+7:i] + 1) >> 1 + ELSE + dst[i+7:i] := 0 + FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point - AVX + + Integer + AVX512VL + AVX512BW Arithmetic - - - Alternatively add and subtract packed double-precision (64-bit) floating-point elements in "a" to/from packed elements in "b", and store the results in "dst". + + + + + + Average packed unsigned 16-bit integers in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*64 - IF (j is even) - dst[i+63:i] := a[i+63:i] - b[i+63:i] +FOR j := 0 to 15 + i := j*16 + IF k[j] + dst[i+15:i] := (a[i+15:i] + b[i+15:i] + 1) >> 1 ELSE - dst[i+63:i] := a[i+63:i] + b[i+63:i] + dst[i+15:i] := src[i+15:i] FI ENDFOR dst[MAX:256] := 0 - +
immintrin.h
 - - Floating Point - AVX + + Integer + AVX512VL + AVX512BW Arithmetic - - - Alternatively add and subtract packed single-precision (32-bit) floating-point elements in "a" to/from packed elements in "b", and store the results in "dst". + + + + + Average packed unsigned 16-bit integers in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := j*32 - IF (j is even) - dst[i+31:i] := a[i+31:i] - b[i+31:i] +FOR j := 0 to 15 + i := j*16 + IF k[j] + dst[i+15:i] := (a[i+15:i] + b[i+15:i] + 1) >> 1 ELSE - dst[i+31:i] := a[i+31:i] + b[i+31:i] + dst[i+15:i] := 0 FI ENDFOR dst[MAX:256] := 0 - +
immintrin.h
 - - Floating Point - AVX - Logical - - - Compute the bitwise AND of packed double-precision (64-bit) floating-point elements in "a" and "b", and store the results in "dst". + + Integer + AVX512BW + Arithmetic + + + + Average packed unsigned 16-bit integers in "a" and "b", and store the results in "dst". -FOR j := 0 to 3 - i := j*64 - dst[i+63:i] := (a[i+63:i] AND b[i+63:i]) +FOR j := 0 to 31 + i := j*16 + dst[i+15:i] := (a[i+15:i] + b[i+15:i] + 1) >> 1 ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point - AVX - Logical - - - Compute the bitwise AND of packed single-precision (32-bit) floating-point elements in "a" and "b", and store the results in "dst". + + Integer + AVX512BW + Arithmetic + + + + + + Average packed unsigned 16-bit integers in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := j*32 - dst[i+31:i] := (a[i+31:i] AND b[i+31:i]) +FOR j := 0 to 31 + i := j*16 + IF k[j] + dst[i+15:i] := (a[i+15:i] + b[i+15:i] + 1) >> 1 + ELSE + dst[i+15:i] := src[i+15:i] + FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point - AVX - Logical - - - Compute the bitwise NOT of packed double-precision (64-bit) floating-point elements in "a" and then AND with "b", and store the results in "dst". + + Integer + AVX512BW + Arithmetic + + + + + Average packed unsigned 16-bit integers in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*64 - dst[i+63:i] := ((NOT a[i+63:i]) AND b[i+63:i]) +FOR j := 0 to 31 + i := j*16 + IF k[j] + dst[i+15:i] := (a[i+15:i] + b[i+15:i] + 1) >> 1 + ELSE + dst[i+15:i] := 0 + FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point - AVX - Logical - - - Compute the bitwise NOT of packed single-precision (32-bit) floating-point elements in "a" and then AND with "b", and store the results in "dst". + + Integer + AVX512VL + AVX512BW + Arithmetic + + + + + + Average packed unsigned 16-bit integers in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). FOR j := 0 to 7 - i := j*32 - dst[i+31:i] := ((NOT a[i+31:i]) AND b[i+31:i]) + i := j*16 + IF k[j] + dst[i+15:i] := (a[i+15:i] + b[i+15:i] + 1) >> 1 + ELSE + dst[i+15:i] := src[i+15:i] + FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point - AVX - Swizzle - - - - Blend packed double-precision (64-bit) floating-point elements from "a" and "b" using control mask "imm8", and store the results in "dst". + + Integer + AVX512VL + AVX512BW + Arithmetic + + + + + Average packed unsigned 16-bit integers in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*64 - IF imm8[j%8] - dst[i+63:i] := b[i+63:i] +FOR j := 0 to 7 + i := j*16 + IF k[j] + dst[i+15:i] := (a[i+15:i] + b[i+15:i] + 1) >> 1 ELSE - dst[i+63:i] := a[i+63:i] + dst[i+15:i] := 0 FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point - AVX - Swizzle - - - - Blend packed single-precision (32-bit) floating-point elements from "a" and "b" using control mask "imm8", and store the results in "dst". + + Integer + AVX512VL + AVX512BW + Miscellaneous + + + + + Blend packed 8-bit integers from "a" and "b" using control mask "k", and store the results in "dst". -FOR j := 0 to 7 - i := j*32 - IF imm8[j%8] - dst[i+31:i] := b[i+31:i] +FOR j := 0 to 31 + i := j*8 + IF k[j] + dst[i+7:i] := b[i+7:i] ELSE - dst[i+31:i] := a[i+31:i] + dst[i+7:i] := a[i+7:i] FI ENDFOR dst[MAX:256] := 0 - +
immintrin.h
 - - Floating Point - AVX - Swizzle - - - - Blend packed double-precision (64-bit) floating-point elements from "a" and "b" using "mask", and store the results in "dst". + + Integer + AVX512BW + Miscellaneous + + + + + Blend packed 8-bit integers from "a" and "b" using control mask "k", and store the results in "dst". -FOR j := 0 to 3 - i := j*64 - IF mask[i+63] - dst[i+63:i] := b[i+63:i] +FOR j := 0 to 63 + i := j*8 + IF k[j] + dst[i+7:i] := b[i+7:i] ELSE - dst[i+63:i] := a[i+63:i] + dst[i+7:i] := a[i+7:i] FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point - AVX - Swizzle - - - - Blend packed single-precision (32-bit) floating-point elements from "a" and "b" using "mask", and store the results in "dst". + + Integer + AVX512VL + AVX512BW + Miscellaneous + + + + + Blend packed 8-bit integers from "a" and "b" using control mask "k", and store the results in "dst". -FOR j := 0 to 7 - i := j*32 - IF mask[i+31] - dst[i+31:i] := b[i+31:i] +FOR j := 0 to 15 + i := j*8 + IF k[j] + dst[i+7:i] := b[i+7:i] ELSE - dst[i+31:i] := a[i+31:i] + dst[i+7:i] := a[i+7:i] FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point - AVX - Arithmetic - - - Divide packed double-precision (64-bit) floating-point elements in "a" by packed elements in "b", and store the results in "dst". - -FOR j := 0 to 3 - i := 64*j - dst[i+63:i] := a[i+63:i] / b[i+63:i] + + Integer + AVX512VL + AVX512BW + Miscellaneous + + + + + Blend packed 16-bit integers from "a" and "b" using control mask "k", and store the results in "dst". + +FOR j := 0 to 15 + i := j*16 + IF k[j] + dst[i+15:i] := b[i+15:i] + ELSE + dst[i+15:i] := a[i+15:i] + FI ENDFOR dst[MAX:256] := 0 - +
immintrin.h
 - - Floating Point - AVX - Arithmetic - - - Divide packed single-precision (32-bit) floating-point elements in "a" by packed elements in "b", and store the results in "dst". + + Integer + AVX512BW + Miscellaneous + + + + + Blend packed 16-bit integers from "a" and "b" using control mask "k", and store the results in "dst". -FOR j := 0 to 7 - i := 32*j - dst[i+31:i] := a[i+31:i] / b[i+31:i] +FOR j := 0 to 31 + i := j*16 + IF k[j] + dst[i+15:i] := b[i+15:i] + ELSE + dst[i+15:i] := a[i+15:i] + FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point - AVX - Arithmetic - - - - Conditionally multiply the packed single-precision (32-bit) floating-point elements in "a" and "b" using the high 4 bits in "imm8", sum the four products, and conditionally store the sum in "dst" using the low 4 bits of "imm8". + + Integer + AVX512VL + AVX512BW + Miscellaneous + + + + + Blend packed 16-bit integers from "a" and "b" using control mask "k", and store the results in "dst". -DP(a[127:0], b[127:0], imm8[7:0]) { - FOR j := 0 to 3 - i := j*32 - IF imm8[(4+j)%8] - temp[i+31:i] := a[i+31:i] * b[i+31:i] - ELSE - temp[i+31:i] := 0 - FI - ENDFOR - - sum[31:0] := (temp[127:96] + temp[95:64]) + (temp[63:32] + temp[31:0]) - - FOR j := 0 to 3 - i := j*32 - IF imm8[j%8] - tmpdst[i+31:i] := sum[31:0] - ELSE - tmpdst[i+31:i] := 0 - FI - ENDFOR - RETURN tmpdst[127:0] -} - -dst[127:0] := DP(a[127:0], b[127:0], imm8[7:0]) -dst[255:128] := DP(a[255:128], b[255:128], imm8[7:0]) -dst[MAX:256] := 0 +FOR j := 0 to 7 + i := j*16 + IF k[j] + dst[i+15:i] := b[i+15:i] + ELSE + dst[i+15:i] := a[i+15:i] + FI +ENDFOR +dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point - AVX - Arithmetic - - - Horizontally add adjacent pairs of double-precision (64-bit) floating-point elements in "a" and "b", and pack the results in "dst". + + Integer + AVX512VL + AVX512BW + Miscellaneous + + + + + Broadcast the low packed 8-bit integer from "a" to all elements of "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -dst[63:0] := a[127:64] + a[63:0] -dst[127:64] := b[127:64] + b[63:0] -dst[191:128] := a[255:192] + a[191:128] -dst[255:192] := b[255:192] + b[191:128] +FOR j := 0 to 31 + i := j*8 + IF k[j] + dst[i+7:i] := a[7:0] + ELSE + dst[i+7:i] := src[i+7:i] + FI +ENDFOR dst[MAX:256] := 0 - +
immintrin.h
 - - Floating Point - AVX - Arithmetic - - - Horizontally add adjacent pairs of single-precision (32-bit) floating-point elements in "a" and "b", and pack the results in "dst". + + Integer + AVX512VL + AVX512BW + Set + + + + + Broadcast 8-bit integer "a" to all elements of "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -dst[31:0] := a[63:32] + a[31:0] -dst[63:32] := a[127:96] + a[95:64] -dst[95:64] := b[63:32] + b[31:0] -dst[127:96] := b[127:96] + b[95:64] -dst[159:128] := a[191:160] + a[159:128] -dst[191:160] := a[255:224] + a[223:192] -dst[223:192] := b[191:160] + b[159:128] -dst[255:224] := b[255:224] + b[223:192] +FOR j := 0 to 31 + i := j*8 + IF k[j] + dst[i+7:i] := a[7:0] + ELSE + dst[i+7:i] := src[i+7:i] + FI +ENDFOR dst[MAX:256] := 0 - +
immintrin.h
 - - Floating Point - AVX - Arithmetic - - - Horizontally subtract adjacent pairs of double-precision (64-bit) floating-point elements in "a" and "b", and pack the results in "dst". + + Integer + AVX512VL + AVX512BW + Miscellaneous + + + + Broadcast the low packed 8-bit integer from "a" to all elements of "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -dst[63:0] := a[63:0] - a[127:64] -dst[127:64] := b[63:0] - b[127:64] -dst[191:128] := a[191:128] - a[255:192] -dst[255:192] := b[191:128] - b[255:192] +FOR j := 0 to 31 + i := j*8 + IF k[j] + dst[i+7:i] := a[7:0] + ELSE + dst[i+7:i] := 0 + FI +ENDFOR dst[MAX:256] := 0 - +
immintrin.h
 - - Floating Point - AVX - Arithmetic - - - Horizontally add adjacent pairs of single-precision (32-bit) floating-point elements in "a" and "b", and pack the results in "dst". + + Integer + AVX512VL + AVX512BW + Set + + + + Broadcast 8-bit integer "a" to all elements of "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -dst[31:0] := a[31:0] - a[63:32] -dst[63:32] := a[95:64] - a[127:96] -dst[95:64] := b[31:0] - b[63:32] -dst[127:96] := b[95:64] - b[127:96] -dst[159:128] := a[159:128] - a[191:160] -dst[191:160] := a[223:192] - a[255:224] -dst[223:192] := b[159:128] - b[191:160] -dst[255:224] := b[223:192] - b[255:224] +FOR j := 0 to 31 + i := j*8 + IF k[j] + dst[i+7:i] := a[7:0] + ELSE + dst[i+7:i] := 0 + FI +ENDFOR dst[MAX:256] := 0 - +
immintrin.h
 - - Floating Point - AVX - Special Math Functions - - - Compare packed double-precision (64-bit) floating-point elements in "a" and "b", and store packed maximum values in "dst". + + Integer + AVX512BW + Miscellaneous + + + Broadcast the low packed 8-bit integer from "a" to all elements of "dst". -FOR j := 0 to 3 - i := j*64 - dst[i+63:i] := MAX(a[i+63:i], b[i+63:i]) +FOR j := 0 to 63 + i := j*8 + dst[i+7:i] := a[7:0] ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point - AVX - Special Math Functions - - - Compare packed single-precision (32-bit) floating-point elements in "a" and "b", and store packed maximum values in "dst". + + Integer + AVX512BW + Miscellaneous + + + + + Broadcast the low packed 8-bit integer from "a" to all elements of "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := j*32 - dst[i+31:i] := MAX(a[i+31:i], b[i+31:i]) +FOR j := 0 to 63 + i := j*8 + IF k[j] + dst[i+7:i] := a[7:0] + ELSE + dst[i+7:i] := src[i+7:i] + FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point - AVX - Special Math Functions - - - Compare packed double-precision (64-bit) floating-point elements in "a" and "b", and store packed minimum values in "dst". - + + Integer + AVX512BW + Set + + + + + Broadcast 8-bit integer "a" to all elements of "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*64 - dst[i+63:i] := MIN(a[i+63:i], b[i+63:i]) +FOR j := 0 to 63 + i := j*8 + IF k[j] + dst[i+7:i] := a[7:0] + ELSE + dst[i+7:i] := src[i+7:i] + FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point - AVX - Special Math Functions - - - Compare packed single-precision (32-bit) floating-point elements in "a" and "b", and store packed minimum values in "dst". - + + Integer + AVX512BW + Miscellaneous + + + + Broadcast the low packed 8-bit integer from "a" to all elements of "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := j*32 - dst[i+31:i] := MIN(a[i+31:i], b[i+31:i]) +FOR j := 0 to 63 + i := j*8 + IF k[j] + dst[i+7:i] := a[7:0] + ELSE + dst[i+7:i] := 0 + FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point - AVX - Arithmetic - - - Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", and store the results in "dst". - + + Integer + AVX512BW + Set + + + + Broadcast 8-bit integer "a" to all elements of "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*64 - dst[i+63:i] := a[i+63:i] * b[i+63:i] +FOR j := 0 to 63 + i := j*8 + IF k[j] + dst[i+7:i] := a[7:0] + ELSE + dst[i+7:i] := 0 + FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point - AVX - Arithmetic - - - Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", and store the results in "dst". - + + Integer + AVX512VL + AVX512BW + Miscellaneous + + + + + Broadcast the low packed 8-bit integer from "a" to all elements of "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := j*32 - dst[i+31:i] := a[i+31:i] * b[i+31:i] +FOR j := 0 to 15 + i := j*8 + IF k[j] + dst[i+7:i] := a[7:0] + ELSE + dst[i+7:i] := src[i+7:i] + FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point - AVX - Logical - - - Compute the bitwise OR of packed double-precision (64-bit) floating-point elements in "a" and "b", and store the results in "dst". - + + Integer + AVX512VL + AVX512BW + Set + + + + + Broadcast 8-bit integer "a" to all elements of "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*64 - dst[i+63:i] := a[i+63:i] BITWISE OR b[i+63:i] +FOR j := 0 to 15 + i := j*8 + IF k[j] + dst[i+7:i] := a[7:0] + ELSE + dst[i+7:i] := src[i+7:i] + FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point - AVX - Logical - - - Compute the bitwise OR of packed single-precision (32-bit) floating-point elements in "a" and "b", and store the results in "dst". - + + Integer + AVX512VL + AVX512BW + Miscellaneous + + + + Broadcast the low packed 8-bit integer from "a" to all elements of "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := j*32 - dst[i+31:i] := a[i+31:i] BITWISE OR b[i+31:i] +FOR j := 0 to 15 + i := j*8 + IF k[j] + dst[i+7:i] := a[7:0] + ELSE + dst[i+7:i] := 0 + FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point - AVX - Swizzle - - - - Shuffle double-precision (64-bit) floating-point elements within 128-bit lanes using the control in "imm8", and store the results in "dst". + + Integer + AVX512VL + AVX512BW + Set + + + + Broadcast 8-bit integer "a" to all elements of "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -dst[63:0] := (imm8[0] == 0) ? a[63:0] : a[127:64] -dst[127:64] := (imm8[1] == 0) ? b[63:0] : b[127:64] -dst[191:128] := (imm8[2] == 0) ? a[191:128] : a[255:192] -dst[255:192] := (imm8[3] == 0) ? b[191:128] : b[255:192] -dst[MAX:256] := 0 +FOR j := 0 to 15 + i := j*8 + IF k[j] + dst[i+7:i] := a[7:0] + ELSE + dst[i+7:i] := 0 + FI +ENDFOR +dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point - AVX - Swizzle - - - - Shuffle single-precision (32-bit) floating-point elements in "a" within 128-bit lanes using the control in "imm8", and store the results in "dst". + + Integer + AVX512VL + AVX512BW + Miscellaneous + + + + + Broadcast the low packed 16-bit integer from "a" to all elements of "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -SELECT4(src, control){ - CASE(control[1:0]) - 0: tmp[31:0] := src[31:0] - 1: tmp[31:0] := src[63:32] - 2: tmp[31:0] := src[95:64] - 3: tmp[31:0] := src[127:96] - ESAC - RETURN tmp[31:0] -} - -dst[31:0] := SELECT4(a[127:0], imm8[1:0]) -dst[63:32] := SELECT4(a[127:0], imm8[3:2]) -dst[95:64] := SELECT4(b[127:0], imm8[5:4]) -dst[127:96] := SELECT4(b[127:0], imm8[7:6]) -dst[159:128] := SELECT4(a[255:128], imm8[1:0]) -dst[191:160] := SELECT4(a[255:128], imm8[3:2]) -dst[223:192] := SELECT4(b[255:128], imm8[5:4]) -dst[255:224] := SELECT4(b[255:128], imm8[7:6]) +FOR j := 0 to 15 + i := j*16 + IF k[j] + dst[i+15:i] := a[15:0] + ELSE + dst[i+15:i] := src[i+15:i] + FI +ENDFOR dst[MAX:256] := 0 - +
immintrin.h
 - - Floating Point - AVX - Arithmetic - - - Subtract packed double-precision (64-bit) floating-point elements in "b" from packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst". + + Integer + AVX512VL + AVX512BW + Set + + + + + Broadcast the low packed 16-bit integer from "a" to all elements of "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*64 - dst[i+63:i] := a[i+63:i] - b[i+63:i] +FOR j := 0 to 15 + i := j*16 + IF k[j] + dst[i+15:i] := a[15:0] + ELSE + dst[i+15:i] := src[i+15:i] + FI ENDFOR dst[MAX:256] := 0 - +
immintrin.h
 - - Floating Point - AVX - Arithmetic - - - Subtract packed single-precision (32-bit) floating-point elements in "b" from packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst". + + Integer + AVX512VL + AVX512BW + Miscellaneous + + + + Broadcast the low packed 16-bit integer from "a" to all elements of "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := j*32 - dst[i+31:i] := a[i+31:i] - b[i+31:i] +FOR j := 0 to 15 + i := j*16 + IF k[j] + dst[i+15:i] := a[15:0] + ELSE + dst[i+15:i] := 0 + FI ENDFOR dst[MAX:256] := 0 - +
immintrin.h
 - - Floating Point - AVX - Logical - - - Compute the bitwise XOR of packed double-precision (64-bit) floating-point elements in "a" and "b", and store the results in "dst". - + + Integer + AVX512VL + AVX512BW + Set + + + + Broadcast 16-bit integer "a" to all elements of "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*64 - dst[i+63:i] := a[i+63:i] XOR b[i+63:i] +FOR j := 0 to 15 + i := j*16 + IF k[j] + dst[i+15:i] := a[15:0] + ELSE + dst[i+15:i] := 0 + FI ENDFOR dst[MAX:256] := 0 - +
immintrin.h
 - - Floating Point - AVX - Logical - - - Compute the bitwise XOR of packed single-precision (32-bit) floating-point elements in "a" and "b", and store the results in "dst". - + + Integer + AVX512BW + Miscellaneous + + + Broadcast the low packed 16-bit integer from "a" to all elements of "dst". -FOR j := 0 to 7 - i := j*32 - dst[i+31:i] := a[i+31:i] XOR b[i+31:i] +FOR j := 0 to 31 + i := j*16 + dst[i+15:i] := a[15:0] ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point - AVX - Compare - - - - Compare packed double-precision (64-bit) floating-point elements in "a" and "b" based on the comparison operand specified by "imm8", and store the results in "dst". + + Integer + AVX512BW + Miscellaneous + + + + + Broadcast the low packed 16-bit integer from "a" to all elements of "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -CASE (imm8[7:0]) OF -0: OP := _CMP_EQ_OQ -1: OP := _CMP_LT_OS -2: OP := _CMP_LE_OS -3: OP := _CMP_UNORD_Q -4: OP := _CMP_NEQ_UQ -5: OP := _CMP_NLT_US -6: OP := _CMP_NLE_US -7: OP := _CMP_ORD_Q -8: OP := _CMP_EQ_UQ -9: OP := _CMP_NGE_US -10: OP := _CMP_NGT_US -11: OP := _CMP_FALSE_OQ -12: OP := _CMP_NEQ_OQ -13: OP := _CMP_GE_OS -14: OP := _CMP_GT_OS -15: OP := _CMP_TRUE_UQ -16: OP := _CMP_EQ_OS -17: OP := _CMP_LT_OQ -18: OP := _CMP_LE_OQ -19: OP := _CMP_UNORD_S -20: OP := _CMP_NEQ_US -21: OP := _CMP_NLT_UQ -22: OP := _CMP_NLE_UQ -23: OP := _CMP_ORD_S -24: OP := _CMP_EQ_US -25: OP := _CMP_NGE_UQ -26: OP := _CMP_NGT_UQ -27: OP := _CMP_FALSE_OS -28: OP := _CMP_NEQ_OS -29: OP := _CMP_GE_OQ -30: OP := _CMP_GT_OQ -31: OP := _CMP_TRUE_US -ESAC -FOR j := 0 to 1 - i := j*64 - dst[i+63:i] := ( a[i+63:i] OP b[i+63:i] ) ? 0xFFFFFFFFFFFFFFFF : 0 +FOR j := 0 to 31 + i := j*16 + IF k[j] + dst[i+15:i] := a[15:0] + ELSE + dst[i+15:i] := src[i+15:i] + FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point - AVX - Compare - - - - Compare packed double-precision (64-bit) floating-point elements in "a" and "b" based on the comparison operand specified by "imm8", and store the results in "dst". + + Integer + AVX512BW + Set + + + + + Broadcast 16-bit integer "a" to all elements of "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -CASE (imm8[7:0]) OF -0: OP := _CMP_EQ_OQ -1: OP := _CMP_LT_OS -2: OP := _CMP_LE_OS -3: OP := _CMP_UNORD_Q -4: OP := _CMP_NEQ_UQ -5: OP := _CMP_NLT_US -6: OP := _CMP_NLE_US -7: OP := _CMP_ORD_Q -8: OP := _CMP_EQ_UQ -9: OP := _CMP_NGE_US -10: OP := _CMP_NGT_US -11: OP := _CMP_FALSE_OQ -12: OP := _CMP_NEQ_OQ -13: OP := _CMP_GE_OS -14: OP := _CMP_GT_OS -15: OP := _CMP_TRUE_UQ -16: OP := _CMP_EQ_OS -17: OP := _CMP_LT_OQ -18: OP := _CMP_LE_OQ -19: OP := _CMP_UNORD_S -20: OP := _CMP_NEQ_US -21: OP := _CMP_NLT_UQ -22: OP := _CMP_NLE_UQ -23: OP := _CMP_ORD_S -24: OP := _CMP_EQ_US -25: OP := _CMP_NGE_UQ -26: OP := _CMP_NGT_UQ -27: OP := _CMP_FALSE_OS -28: OP := _CMP_NEQ_OS -29: OP := _CMP_GE_OQ -30: OP := _CMP_GT_OQ -31: OP := _CMP_TRUE_US -ESAC -FOR j := 0 to 3 - i := j*64 - dst[i+63:i] := ( a[i+63:i] OP b[i+63:i] ) ? 0xFFFFFFFFFFFFFFFF : 0 +FOR j := 0 to 31 + i := j*16 + IF k[j] + dst[i+15:i] := a[15:0] + ELSE + dst[i+15:i] := src[i+15:i] + FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point - AVX - Compare - - - - Compare packed single-precision (32-bit) floating-point elements in "a" and "b" based on the comparison operand specified by "imm8", and store the results in "dst". + + Integer + AVX512BW + Miscellaneous + + + + Broadcast the low packed 16-bit integer from "a" to all elements of "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -CASE (imm8[7:0]) OF -0: OP := _CMP_EQ_OQ -1: OP := _CMP_LT_OS -2: OP := _CMP_LE_OS -3: OP := _CMP_UNORD_Q -4: OP := _CMP_NEQ_UQ -5: OP := _CMP_NLT_US -6: OP := _CMP_NLE_US -7: OP := _CMP_ORD_Q -8: OP := _CMP_EQ_UQ -9: OP := _CMP_NGE_US -10: OP := _CMP_NGT_US -11: OP := _CMP_FALSE_OQ -12: OP := _CMP_NEQ_OQ -13: OP := _CMP_GE_OS -14: OP := _CMP_GT_OS -15: OP := _CMP_TRUE_UQ -16: OP := _CMP_EQ_OS -17: OP := _CMP_LT_OQ -18: OP := _CMP_LE_OQ -19: OP := _CMP_UNORD_S -20: OP := _CMP_NEQ_US -21: OP := _CMP_NLT_UQ -22: OP := _CMP_NLE_UQ -23: OP := _CMP_ORD_S -24: OP := _CMP_EQ_US -25: OP := _CMP_NGE_UQ -26: OP := _CMP_NGT_UQ -27: OP := _CMP_FALSE_OS -28: OP := _CMP_NEQ_OS -29: OP := _CMP_GE_OQ -30: OP := _CMP_GT_OQ -31: OP := _CMP_TRUE_US -ESAC -FOR j := 0 to 3 - i := j*32 - dst[i+31:i] := ( a[i+31:i] OP b[i+31:i] ) ? 0xFFFFFFFF : 0 +FOR j := 0 to 31 + i := j*16 + IF k[j] + dst[i+15:i] := a[15:0] + ELSE + dst[i+15:i] := 0 + FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point - AVX - Compare - - - - Compare packed single-precision (32-bit) floating-point elements in "a" and "b" based on the comparison operand specified by "imm8", and store the results in "dst". + + Integer + AVX512BW + Set + + + + Broadcast the low packed 16-bit integer from "a" to all elements of "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -CASE (imm8[7:0]) OF -0: OP := _CMP_EQ_OQ -1: OP := _CMP_LT_OS -2: OP := _CMP_LE_OS -3: OP := _CMP_UNORD_Q -4: OP := _CMP_NEQ_UQ -5: OP := _CMP_NLT_US -6: OP := _CMP_NLE_US -7: OP := _CMP_ORD_Q -8: OP := _CMP_EQ_UQ -9: OP := _CMP_NGE_US -10: OP := _CMP_NGT_US -11: OP := _CMP_FALSE_OQ -12: OP := _CMP_NEQ_OQ -13: OP := _CMP_GE_OS -14: OP := _CMP_GT_OS -15: OP := _CMP_TRUE_UQ -16: OP := _CMP_EQ_OS -17: OP := _CMP_LT_OQ -18: OP := _CMP_LE_OQ -19: OP := _CMP_UNORD_S -20: OP := _CMP_NEQ_US -21: OP := _CMP_NLT_UQ -22: OP := _CMP_NLE_UQ -23: OP := _CMP_ORD_S -24: OP := _CMP_EQ_US -25: OP := _CMP_NGE_UQ -26: OP := _CMP_NGT_UQ -27: OP := _CMP_FALSE_OS -28: OP := _CMP_NEQ_OS -29: OP := _CMP_GE_OQ -30: OP := _CMP_GT_OQ -31: OP := _CMP_TRUE_US -ESAC -FOR j := 0 to 7 - i := j*32 - dst[i+31:i] := ( a[i+31:i] OP b[i+31:i] ) ? 0xFFFFFFFF : 0 +FOR j := 0 to 31 + i := j*16 + IF k[j] + dst[i+15:i] := a[15:0] + ELSE + dst[i+15:i] := 0 + FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point - AVX - Compare - - - - Compare the lower double-precision (64-bit) floating-point element in "a" and "b" based on the comparison operand specified by "imm8", store the result in the lower element of "dst", and copy the upper element from "a" to the upper element of "dst". + + Integer + AVX512VL + AVX512BW + Miscellaneous + + + + + Broadcast the low packed 16-bit integer from "a" to all elements of "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -CASE (imm8[7:0]) OF -0: OP := _CMP_EQ_OQ -1: OP := _CMP_LT_OS -2: OP := _CMP_LE_OS -3: OP := _CMP_UNORD_Q -4: OP := _CMP_NEQ_UQ -5: OP := _CMP_NLT_US -6: OP := _CMP_NLE_US -7: OP := _CMP_ORD_Q -8: OP := _CMP_EQ_UQ -9: OP := _CMP_NGE_US -10: OP := _CMP_NGT_US -11: OP := _CMP_FALSE_OQ -12: OP := _CMP_NEQ_OQ -13: OP := _CMP_GE_OS -14: OP := _CMP_GT_OS -15: OP := _CMP_TRUE_UQ -16: OP := _CMP_EQ_OS -17: OP := _CMP_LT_OQ -18: OP := _CMP_LE_OQ -19: OP := _CMP_UNORD_S -20: OP := _CMP_NEQ_US -21: OP := _CMP_NLT_UQ -22: OP := _CMP_NLE_UQ -23: OP := _CMP_ORD_S -24: OP := _CMP_EQ_US -25: OP := _CMP_NGE_UQ -26: OP := _CMP_NGT_UQ -27: OP := _CMP_FALSE_OS -28: OP := _CMP_NEQ_OS -29: OP := _CMP_GE_OQ -30: OP := _CMP_GT_OQ -31: OP := _CMP_TRUE_US -ESAC - -dst[63:0] := ( a[63:0] OP b[63:0] ) ? 0xFFFFFFFFFFFFFFFF : 0 -dst[127:64] := a[127:64] +FOR j := 0 to 7 + i := j*16 + IF k[j] + dst[i+15:i] := a[15:0] + ELSE + dst[i+15:i] := src[i+15:i] + FI +ENDFOR dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point - AVX - Compare - - - - Compare the lower single-precision (32-bit) floating-point element in "a" and "b" based on the comparison operand specified by "imm8", store the result in the lower element of "dst", and copy the upper 3 packed elements from "a" to the upper elements of "dst". + + Integer + AVX512VL + AVX512BW + Set + + + + + Broadcast the low packed 16-bit integer from "a" to all elements of "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -CASE (imm8[7:0]) OF -0: OP := _CMP_EQ_OQ -1: OP := _CMP_LT_OS -2: OP := _CMP_LE_OS -3: OP := _CMP_UNORD_Q -4: OP := _CMP_NEQ_UQ -5: OP := _CMP_NLT_US -6: OP := _CMP_NLE_US -7: OP := _CMP_ORD_Q -8: OP := _CMP_EQ_UQ -9: OP := _CMP_NGE_US -10: OP := _CMP_NGT_US -11: OP := _CMP_FALSE_OQ -12: OP := _CMP_NEQ_OQ -13: OP := _CMP_GE_OS -14: OP := _CMP_GT_OS -15: OP := _CMP_TRUE_UQ -16: OP := _CMP_EQ_OS -17: OP := _CMP_LT_OQ -18: OP := _CMP_LE_OQ -19: OP := _CMP_UNORD_S -20: OP := _CMP_NEQ_US -21: OP := _CMP_NLT_UQ -22: OP := _CMP_NLE_UQ -23: OP := _CMP_ORD_S -24: OP := _CMP_EQ_US -25: OP := _CMP_NGE_UQ -26: OP := _CMP_NGT_UQ -27: OP := _CMP_FALSE_OS -28: OP := _CMP_NEQ_OS -29: OP := _CMP_GE_OQ -30: OP := _CMP_GT_OQ -31: OP := _CMP_TRUE_US -ESAC - -dst[31:0] := ( a[31:0] OP b[31:0] ) ? 0xFFFFFFFF : 0 -dst[127:32] := a[127:32] +FOR j := 0 to 7 + i := j*16 + IF k[j] + dst[i+15:i] := a[15:0] + ELSE + dst[i+15:i] := src[i+15:i] + FI +ENDFOR dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point + Integer - AVX - Convert - - Convert packed 32-bit integers in "a" to packed double-precision (64-bit) floating-point elements, and store the results in "dst". + AVX512VL + AVX512BW + Miscellaneous + + + + Broadcast the low packed 16-bit integer from "a" to all elements of "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*32 - m := j*64 - dst[m+63:m] := Convert_Int32_To_FP64(a[i+31:i]) +FOR j := 0 to 7 + i := j*16 + IF k[j] + dst[i+15:i] := a[15:0] + ELSE + dst[i+15:i] := 0 + FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point + Integer - AVX - Convert - - Convert packed 32-bit integers in "a" to packed single-precision (32-bit) floating-point elements, and store the results in "dst". + AVX512VL + AVX512BW + Set + + + + Broadcast the low packed 16-bit integer from "a" to all elements of "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). FOR j := 0 to 7 - i := 32*j - dst[i+31:i] := Convert_Int32_To_FP32(a[i+31:i]) + i := j*16 + IF k[j] + dst[i+15:i] := a[15:0] + ELSE + dst[i+15:i] := 0 + FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point - AVX - Convert - - Convert packed double-precision (64-bit) floating-point elements in "a" to packed single-precision (32-bit) floating-point elements, and store the results in "dst". - -FOR j := 0 to 3 - i := 32*j - k := 64*j - dst[i+31:i] := Convert_FP64_To_FP32(a[k+63:k]) + + Integer + Mask + AVX512VL + AVX512BW + Compare + + + + + Compare packed signed 8-bit integers in "a" and "b" based on the comparison operand specified by "imm8", and store the results in mask vector "k". + CASE (imm8[2:0]) OF +0: OP := _MM_CMPINT_EQ +1: OP := _MM_CMPINT_LT +2: OP := _MM_CMPINT_LE +3: OP := _MM_CMPINT_FALSE +4: OP := _MM_CMPINT_NE +5: OP := _MM_CMPINT_NLT +6: OP := _MM_CMPINT_NLE +7: OP := _MM_CMPINT_TRUE +ESAC +FOR j := 0 to 31 + i := j*8 + k[j] := ( a[i+7:i] OP b[i+7:i] ) ? 1 : 0 ENDFOR -dst[MAX:128] := 0 +k[MAX:32] := 0 - +
immintrin.h
 - - Floating Point + Integer - AVX - Convert - - Convert packed single-precision (32-bit) floating-point elements in "a" to packed 32-bit integers, and store the results in "dst". + Mask + AVX512VL + AVX512BW + Compare + + + + Compare packed signed 8-bit integers in "a" and "b" for equality, and store the results in mask vector "k". -FOR j := 0 to 7 - i := 32*j - dst[i+31:i] := Convert_FP32_To_Int32(a[i+31:i]) +FOR j := 0 to 31 + i := j*8 + k[j] := ( a[i+7:i] == b[i+7:i] ) ? 1 : 0 ENDFOR -dst[MAX:256] := 0 +k[MAX:32] := 0 - +
immintrin.h
 - - Floating Point - AVX - Convert - - Convert packed single-precision (32-bit) floating-point elements in "a" to packed double-precision (64-bit) floating-point elements, and store the results in "dst". + + Integer + Mask + AVX512VL + AVX512BW + Compare + + + + Compare packed signed 8-bit integers in "a" and "b" for greater-than-or-equal, and store the results in mask vector "k". -FOR j := 0 to 3 - i := 64*j - k := 32*j - dst[i+63:i] := Convert_FP32_To_FP64(a[k+31:k]) +FOR j := 0 to 31 + i := j*8 + k[j] := ( a[i+7:i] >= b[i+7:i] ) ? 1 : 0 ENDFOR -dst[MAX:256] := 0 +k[MAX:32] := 0 - +
immintrin.h
 - - Floating Point + Integer - AVX - Convert - - Convert packed double-precision (64-bit) floating-point elements in "a" to packed 32-bit integers with truncation, and store the results in "dst". + Mask + AVX512VL + AVX512BW + Compare + + + + Compare packed signed 8-bit integers in "a" and "b" for greater-than, and store the results in mask vector "k". -FOR j := 0 to 3 - i := 32*j - k := 64*j - dst[i+31:i] := Convert_FP64_To_Int32_Truncate(a[k+63:k]) +FOR j := 0 to 31 + i := j*8 + k[j] := ( a[i+7:i] > b[i+7:i] ) ? 1 : 0 ENDFOR -dst[MAX:128] := 0 +k[MAX:32] := 0 - +
immintrin.h
 - - Floating Point + Integer - AVX - Convert - - Convert packed double-precision (64-bit) floating-point elements in "a" to packed 32-bit integers, and store the results in "dst". + Mask + AVX512VL + AVX512BW + Compare + + + + Compare packed signed 8-bit integers in "a" and "b" for less-than-or-equal, and store the results in mask vector "k". -FOR j := 0 to 3 - i := 32*j - k := 64*j - dst[i+31:i] := Convert_FP64_To_Int32(a[k+63:k]) +FOR j := 0 to 31 + i := j*8 + k[j] := ( a[i+7:i] <= b[i+7:i] ) ? 1 : 0 ENDFOR -dst[MAX:128] := 0 +k[MAX:32] := 0 - +
immintrin.h
 - - Floating Point + Integer - AVX - Convert - - Convert packed single-precision (32-bit) floating-point elements in "a" to packed 32-bit integers with truncation, and store the results in "dst". + Mask + AVX512VL + AVX512BW + Compare + + + + Compare packed signed 8-bit integers in "a" and "b" for less-than, and store the results in mask vector "k". -FOR j := 0 to 7 - i := 32*j - dst[i+31:i] := Convert_FP32_To_Int32_Truncate(a[i+31:i]) +FOR j := 0 to 31 + i := j*8 + k[j] := ( a[i+7:i] < b[i+7:i] ) ? 1 : 0 ENDFOR -dst[MAX:256] := 0 +k[MAX:32] := 0 - +
immintrin.h
 - - Floating Point - AVX - Swizzle - - - Extract 128 bits (composed of 4 packed single-precision (32-bit) floating-point elements) from "a", selected with "imm8", and store the result in "dst". + + Integer + Mask + AVX512VL + AVX512BW + Compare + + + + Compare packed signed 8-bit integers in "a" and "b" for not-equal, and store the results in mask vector "k". -CASE imm8[7:0] of -0: dst[127:0] := a[127:0] -1: dst[127:0] := a[255:128] -ESAC -dst[MAX:128] := 0 +FOR j := 0 to 31 + i := j*8 + k[j] := ( a[i+7:i] != b[i+7:i] ) ? 1 : 0 +ENDFOR +k[MAX:32] := 0 - +
immintrin.h
 - - Floating Point - AVX - Swizzle - - - Extract 128 bits (composed of 2 packed double-precision (64-bit) floating-point elements) from "a", selected with "imm8", and store the result in "dst". - -CASE imm8[7:0] of -0: dst[127:0] := a[127:0] -1: dst[127:0] := a[255:128] + + Integer + Mask + AVX512VL + AVX512BW + Compare + + + + + + Compare packed signed 8-bit integers in "a" and "b" based on the comparison operand specified by "imm8", and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). + CASE (imm8[2:0]) OF +0: OP := _MM_CMPINT_EQ +1: OP := _MM_CMPINT_LT +2: OP := _MM_CMPINT_LE +3: OP := _MM_CMPINT_FALSE +4: OP := _MM_CMPINT_NE +5: OP := _MM_CMPINT_NLT +6: OP := _MM_CMPINT_NLE +7: OP := _MM_CMPINT_TRUE ESAC -dst[MAX:128] := 0 +FOR j := 0 to 31 + i := j*8 + IF k1[j] + k[j] := ( a[i+7:i] OP b[i+7:i] ) ? 1 : 0 + ELSE + k[j] := 0 + FI +ENDFOR +k[MAX:32] := 0 - +
immintrin.h
 - + Integer - AVX - Swizzle - - - Extract 128 bits (composed of integer data) from "a", selected with "imm8", and store the result in "dst". + Mask + AVX512VL + AVX512BW + Compare + + + + + Compare packed signed 8-bit integers in "a" and "b" for equality, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). -CASE imm8[7:0] of -0: dst[127:0] := a[127:0] -1: dst[127:0] := a[255:128] -ESAC -dst[MAX:128] := 0 +FOR j := 0 to 31 + i := j*8 + IF k1[j] + k[j] := ( a[i+7:i] == b[i+7:i] ) ? 1 : 0 + ELSE + k[j] := 0 + FI +ENDFOR +k[MAX:32] := 0 - +
immintrin.h
 - + Integer - AVX2 - Swizzle - - - Extract an 8-bit integer from "a", selected with "index", and store the result in "dst". + Mask + AVX512VL + AVX512BW + Compare + + + + + Compare packed signed 8-bit integers in "a" and "b" for greater-than-or-equal, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). -dst[7:0] := (a[255:0] >> (index * 8))[7:0] +FOR j := 0 to 31 + i := j*8 + IF k1[j] + k[j] := ( a[i+7:i] >= b[i+7:i] ) ? 1 : 0 + ELSE + k[j] := 0 + FI +ENDFOR +k[MAX:32] := 0 +
immintrin.h
 - + Integer - AVX2 - Swizzle - - - Extract a 16-bit integer from "a", selected with "index", and store the result in "dst". + Mask + AVX512VL + AVX512BW + Compare + + + + + Compare packed signed 8-bit integers in "a" and "b" for greater-than, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). -dst[15:0] := (a[255:0] >> (index * 16))[15:0] +FOR j := 0 to 31 + i := j*8 + IF k1[j] + k[j] := ( a[i+7:i] > b[i+7:i] ) ? 1 : 0 + ELSE + k[j] := 0 + FI +ENDFOR +k[MAX:32] := 0 +
immintrin.h
 - + Integer - AVX - Swizzle - - - Extract a 32-bit integer from "a", selected with "index", and store the result in "dst". + Mask + AVX512VL + AVX512BW + Compare + + + + + Compare packed signed 8-bit integers in "a" and "b" for less-than-or-equal, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). -dst[31:0] := (a[255:0] >> (index * 32))[31:0] +FOR j := 0 to 31 + i := j*8 + IF k1[j] + k[j] := ( a[i+7:i] <= b[i+7:i] ) ? 1 : 0 + ELSE + k[j] := 0 + FI +ENDFOR +k[MAX:32] := 0 +
immintrin.h
 - + Integer - AVX - Swizzle - - - Extract a 64-bit integer from "a", selected with "index", and store the result in "dst". + Mask + AVX512VL + AVX512BW + Compare + + + + + Compare packed signed 8-bit integers in "a" and "b" for less-than, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). -dst[63:0] := (a[255:0] >> (index * 64))[63:0] +FOR j := 0 to 31 + i := j*8 + IF k1[j] + k[j] := ( a[i+7:i] < b[i+7:i] ) ? 1 : 0 + ELSE + k[j] := 0 + FI +ENDFOR +k[MAX:32] := 0 +
immintrin.h
 - - AVX - General Support - - Zero the contents of all XMM or YMM registers. - -YMM0[MAX:0] := 0 -YMM1[MAX:0] := 0 -YMM2[MAX:0] := 0 -YMM3[MAX:0] := 0 -YMM4[MAX:0] := 0 -YMM5[MAX:0] := 0 -YMM6[MAX:0] := 0 -YMM7[MAX:0] := 0 -IF 64-bit mode - YMM8[MAX:0] := 0 - YMM9[MAX:0] := 0 - YMM10[MAX:0] := 0 - YMM11[MAX:0] := 0 - YMM12[MAX:0] := 0 - YMM13[MAX:0] := 0 - YMM14[MAX:0] := 0 - YMM15[MAX:0] := 0 -FI - - -
immintrin.h
- - - AVX - General Support - - Zero the upper 128 bits of all YMM registers; the lower 128-bits of the registers are unmodified. + + Integer + Mask + AVX512VL + AVX512BW + Compare + + + + + Compare packed signed 8-bit integers in "a" and "b" for not-equal, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). -YMM0[MAX:128] := 0 -YMM1[MAX:128] := 0 -YMM2[MAX:128] := 0 -YMM3[MAX:128] := 0 -YMM4[MAX:128] := 0 -YMM5[MAX:128] := 0 -YMM6[MAX:128] := 0 -YMM7[MAX:128] := 0 -IF 64-bit mode - YMM8[MAX:128] := 0 - YMM9[MAX:128] := 0 - YMM10[MAX:128] := 0 - YMM11[MAX:128] := 0 - YMM12[MAX:128] := 0 - YMM13[MAX:128] := 0 - YMM14[MAX:128] := 0 - YMM15[MAX:128] := 0 -FI - - +FOR j := 0 to 31 + i := j*8 + IF k1[j] + k[j] := ( a[i+7:i] != b[i+7:i] ) ? 1 : 0 + ELSE + k[j] := 0 + FI +ENDFOR +k[MAX:32] := 0 + +
immintrin.h
 - - Floating Point - AVX - Swizzle - - - Shuffle single-precision (32-bit) floating-point elements in "a" within 128-bit lanes using the control in "b", and store the results in "dst". - -SELECT4(src, control){ - CASE(control[1:0]) - 0: tmp[31:0] := src[31:0] - 1: tmp[31:0] := src[63:32] - 2: tmp[31:0] := src[95:64] - 3: tmp[31:0] := src[127:96] - ESAC - RETURN tmp[31:0] -} - -dst[31:0] := SELECT4(a[127:0], b[1:0]) -dst[63:32] := SELECT4(a[127:0], b[33:32]) -dst[95:64] := SELECT4(a[127:0], b[65:64]) -dst[127:96] := SELECT4(a[127:0], b[97:96]) -dst[159:128] := SELECT4(a[255:128], b[129:128]) -dst[191:160] := SELECT4(a[255:128], b[161:160]) -dst[223:192] := SELECT4(a[255:128], b[193:192]) -dst[255:224] := SELECT4(a[255:128], b[225:224]) -dst[MAX:256] := 0 + + Integer + Mask + AVX512BW + Compare + + + + + Compare packed signed 8-bit integers in "a" and "b" based on the comparison operand specified by "imm8", and store the results in mask vector "k". + CASE (imm8[2:0]) OF +0: OP := _MM_CMPINT_EQ +1: OP := _MM_CMPINT_LT +2: OP := _MM_CMPINT_LE +3: OP := _MM_CMPINT_FALSE +4: OP := _MM_CMPINT_NE +5: OP := _MM_CMPINT_NLT +6: OP := _MM_CMPINT_NLE +7: OP := _MM_CMPINT_TRUE +ESAC +FOR j := 0 to 63 + i := j*8 + k[j] := ( a[i+7:i] OP b[i+7:i] ) ? 1 : 0 +ENDFOR +k[MAX:64] := 0 - +
immintrin.h
 - - Floating Point - AVX - Swizzle - - - Shuffle single-precision (32-bit) floating-point elements in "a" using the control in "b", and store the results in "dst". + + Integer + Mask + AVX512BW + Compare + + + + Compare packed signed 8-bit integers in "a" and "b" for equality, and store the results in mask vector "k". -SELECT4(src, control){ - CASE(control[1:0]) - 0: tmp[31:0] := src[31:0] - 1: tmp[31:0] := src[63:32] - 2: tmp[31:0] := src[95:64] - 3: tmp[31:0] := src[127:96] - ESAC - RETURN tmp[31:0] -} - -dst[31:0] := SELECT4(a[127:0], b[1:0]) -dst[63:32] := SELECT4(a[127:0], b[33:32]) -dst[95:64] := SELECT4(a[127:0], b[65:64]) -dst[127:96] := SELECT4(a[127:0], b[97:96]) -dst[MAX:128] := 0 +FOR j := 0 to 63 + i := j*8 + k[j] := ( a[i+7:i] == b[i+7:i] ) ? 1 : 0 +ENDFOR +k[MAX:64] := 0 - +
immintrin.h
 - - Floating Point - AVX - Swizzle - - - Shuffle single-precision (32-bit) floating-point elements in "a" within 128-bit lanes using the control in "imm8", and store the results in "dst". + + Integer + Mask + AVX512BW + Compare + + + + Compare packed signed 8-bit integers in "a" and "b" for greater-than-or-equal, and store the results in mask vector "k". -SELECT4(src, control){ - CASE(control[1:0]) - 0: tmp[31:0] := src[31:0] - 1: tmp[31:0] := src[63:32] - 2: tmp[31:0] := src[95:64] - 3: tmp[31:0] := src[127:96] - ESAC - RETURN tmp[31:0] -} - -dst[31:0] := SELECT4(a[127:0], imm8[1:0]) -dst[63:32] := SELECT4(a[127:0], imm8[3:2]) -dst[95:64] := SELECT4(a[127:0], imm8[5:4]) -dst[127:96] := SELECT4(a[127:0], imm8[7:6]) -dst[159:128] := SELECT4(a[255:128], imm8[1:0]) -dst[191:160] := SELECT4(a[255:128], imm8[3:2]) -dst[223:192] := SELECT4(a[255:128], imm8[5:4]) -dst[255:224] := SELECT4(a[255:128], imm8[7:6]) -dst[MAX:256] := 0 +FOR j := 0 to 63 + i := j*8 + k[j] := ( a[i+7:i] >= b[i+7:i] ) ? 1 : 0 +ENDFOR +k[MAX:64] := 0 - +
immintrin.h
 - - Floating Point - AVX - Swizzle - - - Shuffle single-precision (32-bit) floating-point elements in "a" using the control in "imm8", and store the results in "dst". + + Integer + Mask + AVX512BW + Compare + + + + Compare packed signed 8-bit integers in "a" and "b" for greater-than, and store the results in mask vector "k". -SELECT4(src, control){ - CASE(control[1:0]) - 0: tmp[31:0] := src[31:0] - 1: tmp[31:0] := src[63:32] - 2: tmp[31:0] := src[95:64] - 3: tmp[31:0] := src[127:96] - ESAC - RETURN tmp[31:0] -} - -dst[31:0] := SELECT4(a[127:0], imm8[1:0]) -dst[63:32] := SELECT4(a[127:0], imm8[3:2]) -dst[95:64] := SELECT4(a[127:0], imm8[5:4]) -dst[127:96] := SELECT4(a[127:0], imm8[7:6]) -dst[MAX:128] := 0 +FOR j := 0 to 63 + i := j*8 + k[j] := ( a[i+7:i] > b[i+7:i] ) ? 1 : 0 +ENDFOR +k[MAX:64] := 0 - +
immintrin.h
 - - Floating Point - AVX - Swizzle - - - Shuffle double-precision (64-bit) floating-point elements in "a" within 128-bit lanes using the control in "b", and store the results in "dst". + + Integer + Mask + AVX512BW + Compare + + + + Compare packed signed 8-bit integers in "a" and "b" for less-than-or-equal, and store the results in mask vector "k". -IF (b[1] == 0) dst[63:0] := a[63:0] -IF (b[1] == 1) dst[63:0] := a[127:64] -IF (b[65] == 0) dst[127:64] := a[63:0] -IF (b[65] == 1) dst[127:64] := a[127:64] -IF (b[129] == 0) dst[191:128] := a[191:128] -IF (b[129] == 1) dst[191:128] := a[255:192] -IF (b[193] == 0) dst[255:192] := a[191:128] -IF (b[193] == 1) dst[255:192] := a[255:192] -dst[MAX:256] := 0 +FOR j := 0 to 63 + i := j*8 + k[j] := ( a[i+7:i] <= b[i+7:i] ) ? 1 : 0 +ENDFOR +k[MAX:64] := 0 - +
immintrin.h
 - - Floating Point - AVX - Swizzle - - - Shuffle double-precision (64-bit) floating-point elements in "a" using the control in "b", and store the results in "dst". + + Integer + Mask + AVX512BW + Compare + + + + Compare packed signed 8-bit integers in "a" and "b" for less-than, and store the results in mask vector "k". -IF (b[1] == 0) dst[63:0] := a[63:0] -IF (b[1] == 1) dst[63:0] := a[127:64] -IF (b[65] == 0) dst[127:64] := a[63:0] -IF (b[65] == 1) dst[127:64] := a[127:64] -dst[MAX:128] := 0 +FOR j := 0 to 63 + i := j*8 + k[j] := ( a[i+7:i] < b[i+7:i] ) ? 1 : 0 +ENDFOR +k[MAX:64] := 0 - +
immintrin.h
 - - Floating Point - AVX - Swizzle - - - Shuffle double-precision (64-bit) floating-point elements in "a" within 128-bit lanes using the control in "imm8", and store the results in "dst". + + Integer + Mask + AVX512BW + Compare + + + + Compare packed signed 8-bit integers in "a" and "b" for not-equal, and store the results in mask vector "k". -IF (imm8[0] == 0) dst[63:0] := a[63:0] -IF (imm8[0] == 1) dst[63:0] := a[127:64] -IF (imm8[1] == 0) dst[127:64] := a[63:0] -IF (imm8[1] == 1) dst[127:64] := a[127:64] -IF (imm8[2] == 0) dst[191:128] := a[191:128] -IF (imm8[2] == 1) dst[191:128] := a[255:192] -IF (imm8[3] == 0) dst[255:192] := a[191:128] -IF (imm8[3] == 1) dst[255:192] := a[255:192] -dst[MAX:256] := 0 +FOR j := 0 to 63 + i := j*8 + k[j] := ( a[i+7:i] != b[i+7:i] ) ? 1 : 0 +ENDFOR +k[MAX:64] := 0 - +
immintrin.h
 - - Floating Point - AVX - Swizzle - - - Shuffle double-precision (64-bit) floating-point elements in "a" using the control in "imm8", and store the results in "dst". - -IF (imm8[0] == 0) dst[63:0] := a[63:0] -IF (imm8[0] == 1) dst[63:0] := a[127:64] -IF (imm8[1] == 0) dst[127:64] := a[63:0] -IF (imm8[1] == 1) dst[127:64] := a[127:64] -dst[MAX:128] := 0 + + Integer + Mask + AVX512BW + Compare + + + + + + Compare packed signed 8-bit integers in "a" and "b" based on the comparison operand specified by "imm8", and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). + CASE (imm8[2:0]) OF +0: OP := _MM_CMPINT_EQ +1: OP := _MM_CMPINT_LT +2: OP := _MM_CMPINT_LE +3: OP := _MM_CMPINT_FALSE +4: OP := _MM_CMPINT_NE +5: OP := _MM_CMPINT_NLT +6: OP := _MM_CMPINT_NLE +7: OP := _MM_CMPINT_TRUE +ESAC +FOR j := 0 to 63 + i := j*8 + IF k1[j] + k[j] := ( a[i+7:i] OP b[i+7:i] ) ? 1 : 0 + ELSE + k[j] := 0 + FI +ENDFOR +k[MAX:64] := 0 - +
immintrin.h
 - - Floating Point - AVX - Swizzle - - - - Shuffle 128-bits (composed of 4 packed single-precision (32-bit) floating-point elements) selected by "imm8" from "a" and "b", and store the results in "dst". + + Integer + Mask + AVX512BW + Compare + + + + + Compare packed signed 8-bit integers in "a" and "b" for equality, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). -SELECT4(src1, src2, control){ - CASE(control[1:0]) - 0: tmp[127:0] := src1[127:0] - 1: tmp[127:0] := src1[255:128] - 2: tmp[127:0] := src2[127:0] - 3: tmp[127:0] := src2[255:128] - ESAC - IF control[3] - tmp[127:0] := 0 +FOR j := 0 to 63 + i := j*8 + IF k1[j] + k[j] := ( a[i+7:i] == b[i+7:i] ) ? 1 : 0 + ELSE + k[j] := 0 FI - RETURN tmp[127:0] -} - -dst[127:0] := SELECT4(a[255:0], b[255:0], imm8[3:0]) -dst[255:128] := SELECT4(a[255:0], b[255:0], imm8[7:4]) -dst[MAX:256] := 0 +ENDFOR +k[MAX:64] := 0 - +
immintrin.h
 - - Floating Point - AVX - Swizzle - - - - Shuffle 128-bits (composed of 2 packed double-precision (64-bit) floating-point elements) selected by "imm8" from "a" and "b", and store the results in "dst". + + Integer + Mask + AVX512BW + Compare + + + + + Compare packed signed 8-bit integers in "a" and "b" for greater-than-or-equal, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). -SELECT4(src1, src2, control){ - CASE(control[1:0]) - 0: tmp[127:0] := src1[127:0] - 1: tmp[127:0] := src1[255:128] - 2: tmp[127:0] := src2[127:0] - 3: tmp[127:0] := src2[255:128] - ESAC - IF control[3] - tmp[127:0] := 0 +FOR j := 0 to 63 + i := j*8 + IF k1[j] + k[j] := ( a[i+7:i] >= b[i+7:i] ) ? 1 : 0 + ELSE + k[j] := 0 FI - RETURN tmp[127:0] -} - -dst[127:0] := SELECT4(a[255:0], b[255:0], imm8[3:0]) -dst[255:128] := SELECT4(a[255:0], b[255:0], imm8[7:4]) -dst[MAX:256] := 0 +ENDFOR +k[MAX:64] := 0 - +
immintrin.h
 - + Integer - AVX - Swizzle - - - - Shuffle 128-bits (composed of integer data) selected by "imm8" from "a" and "b", and store the results in "dst". + Mask + AVX512BW + Compare + + + + + Compare packed signed 8-bit integers in "a" and "b" for greater-than, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). -SELECT4(src1, src2, control){ - CASE(control[1:0]) - 0: tmp[127:0] := src1[127:0] - 1: tmp[127:0] := src1[255:128] - 2: tmp[127:0] := src2[127:0] - 3: tmp[127:0] := src2[255:128] - ESAC - IF control[3] - tmp[127:0] := 0 +FOR j := 0 to 63 + i := j*8 + IF k1[j] + k[j] := ( a[i+7:i] > b[i+7:i] ) ? 1 : 0 + ELSE + k[j] := 0 FI - RETURN tmp[127:0] -} - -dst[127:0] := SELECT4(a[255:0], b[255:0], imm8[3:0]) -dst[255:128] := SELECT4(a[255:0], b[255:0], imm8[7:4]) -dst[MAX:256] := 0 +ENDFOR +k[MAX:64] := 0 - +
immintrin.h
 - - Floating Point - AVX - Load - - Broadcast a single-precision (32-bit) floating-point element from memory to all elements of "dst". + + Integer + Mask + AVX512BW + Compare + + + + + Compare packed signed 8-bit integers in "a" and "b" for less-than-or-equal, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). -tmp[31:0] = MEM[mem_addr+31:mem_addr] -FOR j := 0 to 7 - i := j*32 - dst[i+31:i] := tmp[31:0] +FOR j := 0 to 63 + i := j*8 + IF k1[j] + k[j] := ( a[i+7:i] <= b[i+7:i] ) ? 1 : 0 + ELSE + k[j] := 0 + FI ENDFOR -dst[MAX:256] := 0 +k[MAX:64] := 0 - +
immintrin.h
 - - Floating Point - AVX - Load - Swizzle - - Broadcast a single-precision (32-bit) floating-point element from memory to all elements of "dst". + + Integer + Mask + AVX512BW + Compare + + + + + Compare packed signed 8-bit integers in "a" and "b" for less-than, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). -tmp[31:0] = MEM[mem_addr+31:mem_addr] -FOR j := 0 to 3 - i := j*32 - dst[i+31:i] := tmp[31:0] +FOR j := 0 to 63 + i := j*8 + IF k1[j] + k[j] := ( a[i+7:i] < b[i+7:i] ) ? 1 : 0 + ELSE + k[j] := 0 + FI ENDFOR -dst[MAX:128] := 0 +k[MAX:64] := 0 - +
immintrin.h
 - - Floating Point - AVX - Load - Swizzle - - Broadcast a double-precision (64-bit) floating-point element from memory to all elements of "dst". + + Integer + Mask + AVX512BW + Compare + + + + + Compare packed signed 8-bit integers in "a" and "b" for not-equal, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). -tmp[63:0] = MEM[mem_addr+63:mem_addr] -FOR j := 0 to 3 - i := j*64 - dst[i+63:i] := tmp[63:0] +FOR j := 0 to 63 + i := j*8 + IF k1[j] + k[j] := ( a[i+7:i] != b[i+7:i] ) ? 1 : 0 + ELSE + k[j] := 0 + FI ENDFOR -dst[MAX:256] := 0 +k[MAX:64] := 0 - +
immintrin.h
 - - Floating Point - AVX - Load - Swizzle - - Broadcast 128 bits from memory (composed of 4 packed single-precision (32-bit) floating-point elements) to all elements of "dst". - -tmp[127:0] = MEM[mem_addr+127:mem_addr] -dst[127:0] := tmp[127:0] -dst[255:128] := tmp[127:0] -dst[MAX:256] := 0 + + Integer + Mask + AVX512VL + AVX512BW + Compare + + + + + Compare packed signed 8-bit integers in "a" and "b" based on the comparison operand specified by "imm8", and store the results in mask vector "k". + CASE (imm8[2:0]) OF +0: OP := _MM_CMPINT_EQ +1: OP := _MM_CMPINT_LT +2: OP := _MM_CMPINT_LE +3: OP := _MM_CMPINT_FALSE +4: OP := _MM_CMPINT_NE +5: OP := _MM_CMPINT_NLT +6: OP := _MM_CMPINT_NLE +7: OP := _MM_CMPINT_TRUE +ESAC +FOR j := 0 to 15 + i := j*8 + k[j] := ( a[i+7:i] OP b[i+7:i] ) ? 1 : 0 +ENDFOR +k[MAX:16] := 0 - +
immintrin.h
 - - Floating Point - AVX - Load - Swizzle - - Broadcast 128 bits from memory (composed of 2 packed double-precision (64-bit) floating-point elements) to all elements of "dst". + + Integer + Mask + AVX512VL + AVX512BW + Compare + + + + Compare packed signed 8-bit integers in "a" and "b" for equality, and store the results in mask vector "k". -tmp[127:0] = MEM[mem_addr+127:mem_addr] -dst[127:0] := tmp[127:0] -dst[255:128] := tmp[127:0] -dst[MAX:256] := 0 +FOR j := 0 to 15 + i := j*8 + k[j] := ( a[i+7:i] == b[i+7:i] ) ? 1 : 0 +ENDFOR +k[MAX:16] := 0 - +
immintrin.h
 - - Floating Point - AVX - Swizzle - - - - Copy "a" to "dst", then insert 128 bits (composed of 4 packed single-precision (32-bit) floating-point elements) from "b" into "dst" at the location specified by "imm8". + + Integer + Mask + AVX512VL + AVX512BW + Compare + + + + Compare packed signed 8-bit integers in "a" and "b" for greater-than-or-equal, and store the results in mask vector "k". -dst[255:0] := a[255:0] -CASE (imm8[1:0]) of -0: dst[127:0] := b[127:0] -1: dst[255:128] := b[127:0] -ESAC -dst[MAX:256] := 0 +FOR j := 0 to 15 + i := j*8 + k[j] := ( a[i+7:i] >= b[i+7:i] ) ? 1 : 0 +ENDFOR +k[MAX:16] := 0 - +
immintrin.h
 - - Floating Point - AVX - Swizzle - - - - Copy "a" to "dst", then insert 128 bits (composed of 2 packed double-precision (64-bit) floating-point elements) from "b" into "dst" at the location specified by "imm8". + + Integer + Mask + AVX512VL + AVX512BW + Compare + + + + Compare packed signed 8-bit integers in "a" and "b" for greater-than, and store the results in mask vector "k". -dst[255:0] := a[255:0] -CASE imm8[7:0] of -0: dst[127:0] := b[127:0] -1: dst[255:128] := b[127:0] -ESAC -dst[MAX:256] := 0 +FOR j := 0 to 15 + i := j*8 + k[j] := ( a[i+7:i] > b[i+7:i] ) ? 1 : 0 +ENDFOR +k[MAX:16] := 0 - +
immintrin.h
 - + Integer - AVX - Swizzle - - - - Copy "a" to "dst", then insert 128 bits from "b" into "dst" at the location specified by "imm8". + Mask + AVX512VL + AVX512BW + Compare + + + + Compare packed signed 8-bit integers in "a" and "b" for less-than-or-equal, and store the results in mask vector "k". -dst[255:0] := a[255:0] -CASE (imm8[1:0]) of -0: dst[127:0] := b[127:0] -1: dst[255:128] := b[127:0] -ESAC -dst[MAX:256] := 0 +FOR j := 0 to 15 + i := j*8 + k[j] := ( a[i+7:i] <= b[i+7:i] ) ? 1 : 0 +ENDFOR +k[MAX:16] := 0 - +
immintrin.h
 - + Integer - AVX - Swizzle - - - - Copy "a" to "dst", and insert the 8-bit integer "i" into "dst" at the location specified by "index". + Mask + AVX512VL + AVX512BW + Compare + + + + Compare packed signed 8-bit integers in "a" and "b" for less-than, and store the results in mask vector "k". -dst[255:0] := a[255:0] -sel := index*8 -dst[sel+7:sel] := i[7:0] +FOR j := 0 to 15 + i := j*8 + k[j] := ( a[i+7:i] < b[i+7:i] ) ? 1 : 0 +ENDFOR +k[MAX:16] := 0 +
immintrin.h
 - + Integer - AVX - Swizzle - - - - Copy "a" to "dst", and insert the 16-bit integer "i" into "dst" at the location specified by "index". + Mask + AVX512VL + AVX512BW + Compare + + + + Compare packed signed 8-bit integers in "a" and "b" for not-equal, and store the results in mask vector "k". -dst[255:0] := a[255:0] -sel := index*16 -dst[sel+15:sel] := i[15:0] +FOR j := 0 to 15 + i := j*8 + k[j] := ( a[i+7:i] != b[i+7:i] ) ? 1 : 0 +ENDFOR +k[MAX:16] := 0 +
immintrin.h
 - + Integer - AVX - Swizzle - - - - Copy "a" to "dst", and insert the 32-bit integer "i" into "dst" at the location specified by "index". - -dst[255:0] := a[255:0] -sel := index*32 -dst[sel+31:sel] := i[31:0] + Mask + AVX512VL + AVX512BW + Compare + + + + + + Compare packed signed 8-bit integers in "a" and "b" based on the comparison operand specified by "imm8", and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). + CASE (imm8[2:0]) OF +0: OP := _MM_CMPINT_EQ +1: OP := _MM_CMPINT_LT +2: OP := _MM_CMPINT_LE +3: OP := _MM_CMPINT_FALSE +4: OP := _MM_CMPINT_NE +5: OP := _MM_CMPINT_NLT +6: OP := _MM_CMPINT_NLE +7: OP := _MM_CMPINT_TRUE +ESAC +FOR j := 0 to 15 + i := j*8 + IF k1[j] + k[j] := ( a[i+7:i] OP b[i+7:i] ) ? 1 : 0 + ELSE + k[j] := 0 + FI +ENDFOR +k[MAX:16] := 0 +
immintrin.h
 - + Integer - AVX - Swizzle - - - - Copy "a" to "dst", and insert the 64-bit integer "i" into "dst" at the location specified by "index". + Mask + AVX512VL + AVX512BW + Compare + + + + + Compare packed signed 8-bit integers in "a" and "b" for equality, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). -dst[255:0] := a[255:0] -sel := index*64 -dst[sel+63:sel] := i[63:0] +FOR j := 0 to 15 + i := j*8 + IF k1[j] + k[j] := ( a[i+7:i] == b[i+7:i] ) ? 1 : 0 + ELSE + k[j] := 0 + FI +ENDFOR +k[MAX:16] := 0 +
immintrin.h
 - - Floating Point - AVX - Load - - Load 256-bits (composed of 4 packed double-precision (64-bit) floating-point elements) from memory into "dst". - "mem_addr" must be aligned on a 32-byte boundary or a general-protection exception may be generated. + + Integer + Mask + AVX512VL + AVX512BW + Compare + + + + + Compare packed signed 8-bit integers in "a" and "b" for greater-than-or-equal, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). -dst[255:0] := MEM[mem_addr+255:mem_addr] -dst[MAX:256] := 0 +FOR j := 0 to 15 + i := j*8 + IF k1[j] + k[j] := ( a[i+7:i] >= b[i+7:i] ) ? 1 : 0 + ELSE + k[j] := 0 + FI +ENDFOR +k[MAX:16] := 0 - +
immintrin.h
 - - Floating Point - AVX - Store - - - Store 256-bits (composed of 4 packed double-precision (64-bit) floating-point elements) from "a" into memory. - "mem_addr" must be aligned on a 32-byte boundary or a general-protection exception may be generated. + + Integer + Mask + AVX512VL + AVX512BW + Compare + + + + + Compare packed signed 8-bit integers in "a" and "b" for greater-than, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). -MEM[mem_addr+255:mem_addr] := a[255:0] +FOR j := 0 to 15 + i := j*8 + IF k1[j] + k[j] := ( a[i+7:i] > b[i+7:i] ) ? 1 : 0 + ELSE + k[j] := 0 + FI +ENDFOR +k[MAX:16] := 0 - +
immintrin.h
 - - Floating Point - AVX - Load - - Load 256-bits (composed of 8 packed single-precision (32-bit) floating-point elements) from memory into "dst". - "mem_addr" must be aligned on a 32-byte boundary or a general-protection exception may be generated. + + Integer + Mask + AVX512VL + AVX512BW + Compare + + + + + Compare packed signed 8-bit integers in "a" and "b" for less-than-or-equal, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). -dst[255:0] := MEM[mem_addr+255:mem_addr] -dst[MAX:256] := 0 +FOR j := 0 to 15 + i := j*8 + IF k1[j] + k[j] := ( a[i+7:i] <= b[i+7:i] ) ? 1 : 0 + ELSE + k[j] := 0 + FI +ENDFOR +k[MAX:16] := 0 - +
immintrin.h
 - - Floating Point - AVX - Store - - - Store 256-bits (composed of 8 packed single-precision (32-bit) floating-point elements) from "a" into memory. - "mem_addr" must be aligned on a 32-byte boundary or a general-protection exception may be generated. + + Integer + Mask + AVX512VL + AVX512BW + Compare + + + + + Compare packed signed 8-bit integers in "a" and "b" for less-than, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). -MEM[mem_addr+255:mem_addr] := a[255:0] +FOR j := 0 to 15 + i := j*8 + IF k1[j] + k[j] := ( a[i+7:i] < b[i+7:i] ) ? 1 : 0 + ELSE + k[j] := 0 + FI +ENDFOR +k[MAX:16] := 0 - +
immintrin.h
 - - Floating Point - AVX - Load - - Load 256-bits (composed of 4 packed double-precision (64-bit) floating-point elements) from memory into "dst". - "mem_addr" does not need to be aligned on any particular boundary. + + Integer + Mask + AVX512VL + AVX512BW + Compare + + + + + Compare packed signed 8-bit integers in "a" and "b" for not-equal, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). -dst[255:0] := MEM[mem_addr+255:mem_addr] -dst[MAX:256] := 0 +FOR j := 0 to 15 + i := j*8 + IF k1[j] + k[j] := ( a[i+7:i] != b[i+7:i] ) ? 1 : 0 + ELSE + k[j] := 0 + FI +ENDFOR +k[MAX:16] := 0 - +
immintrin.h
 - - Floating Point - AVX - Store - - - Store 256-bits (composed of 4 packed double-precision (64-bit) floating-point elements) from "a" into memory. - "mem_addr" does not need to be aligned on any particular boundary. - -MEM[mem_addr+255:mem_addr] := a[255:0] + + Integer + Mask + AVX512VL + AVX512BW + Compare + + + + + Compare packed unsigned 8-bit integers in "a" and "b" based on the comparison operand specified by "imm8", and store the results in mask vector "k". + CASE (imm8[2:0]) OF +0: OP := _MM_CMPINT_EQ +1: OP := _MM_CMPINT_LT +2: OP := _MM_CMPINT_LE +3: OP := _MM_CMPINT_FALSE +4: OP := _MM_CMPINT_NE +5: OP := _MM_CMPINT_NLT +6: OP := _MM_CMPINT_NLE +7: OP := _MM_CMPINT_TRUE +ESAC +FOR j := 0 to 31 + i := j*8 + k[j] := ( a[i+7:i] OP b[i+7:i] ) ? 1 : 0 +ENDFOR +k[MAX:32] := 0 - +
immintrin.h
 - - Floating Point - AVX - Load - - Load 256-bits (composed of 8 packed single-precision (32-bit) floating-point elements) from memory into "dst". - "mem_addr" does not need to be aligned on any particular boundary. + + Integer + Mask + AVX512VL + AVX512BW + Compare + + + + Compare packed unsigned 8-bit integers in "a" and "b" for equality, and store the results in mask vector "k". -dst[255:0] := MEM[mem_addr+255:mem_addr] -dst[MAX:256] := 0 +FOR j := 0 to 31 + i := j*8 + k[j] := ( a[i+7:i] == b[i+7:i] ) ? 1 : 0 +ENDFOR +k[MAX:32] := 0 - +
immintrin.h
 - - Floating Point - AVX - Store - - - Store 256-bits (composed of 8 packed single-precision (32-bit) floating-point elements) from "a" into memory. - "mem_addr" does not need to be aligned on any particular boundary. + + Integer + Mask + AVX512VL + AVX512BW + Compare + + + + Compare packed unsigned 8-bit integers in "a" and "b" for greater-than-or-equal, and store the results in mask vector "k". -MEM[mem_addr+255:mem_addr] := a[255:0] +FOR j := 0 to 31 + i := j*8 + k[j] := ( a[i+7:i] >= b[i+7:i] ) ? 1 : 0 +ENDFOR +k[MAX:32] := 0 - +
immintrin.h
 - + Integer - AVX - Load - - Load 256-bits of integer data from memory into "dst". - "mem_addr" must be aligned on a 32-byte boundary or a general-protection exception may be generated. + Mask + AVX512VL + AVX512BW + Compare + + + + Compare packed unsigned 8-bit integers in "a" and "b" for greater-than, and store the results in mask vector "k". -dst[255:0] := MEM[mem_addr+255:mem_addr] -dst[MAX:256] := 0 +FOR j := 0 to 31 + i := j*8 + k[j] := ( a[i+7:i] > b[i+7:i] ) ? 1 : 0 +ENDFOR +k[MAX:32] := 0 - +
immintrin.h
 - + Integer - AVX - Store - - - Store 256-bits of integer data from "a" into memory. - "mem_addr" must be aligned on a 32-byte boundary or a general-protection exception may be generated. + Mask + AVX512VL + AVX512BW + Compare + + + + Compare packed unsigned 8-bit integers in "a" and "b" for less-than-or-equal, and store the results in mask vector "k". -MEM[mem_addr+255:mem_addr] := a[255:0] +FOR j := 0 to 31 + i := j*8 + k[j] := ( a[i+7:i] <= b[i+7:i] ) ? 1 : 0 +ENDFOR +k[MAX:32] := 0 - +
immintrin.h
 - + Integer - AVX - Load - - Load 256-bits of integer data from memory into "dst". - "mem_addr" does not need to be aligned on any particular boundary. - + Mask + AVX512VL + AVX512BW + Compare + + + + Compare packed unsigned 8-bit integers in "a" and "b" for less-than, and store the results in mask vector "k". -dst[255:0] := MEM[mem_addr+255:mem_addr] -dst[MAX:256] := 0 +FOR j := 0 to 31 + i := j*8 + k[j] := ( a[i+7:i] < b[i+7:i] ) ? 1 : 0 +ENDFOR +k[MAX:32] := 0 - +
immintrin.h
 - + Integer - AVX - Store - - - Store 256-bits of integer data from "a" into memory. - "mem_addr" does not need to be aligned on any particular boundary. - + Mask + AVX512VL + AVX512BW + Compare + + + + Compare packed unsigned 8-bit integers in "a" and "b" for not-equal, and store the results in mask vector "k". -MEM[mem_addr+255:mem_addr] := a[255:0] +FOR j := 0 to 31 + i := j*8 + k[j] := ( a[i+7:i] != b[i+7:i] ) ? 1 : 0 +ENDFOR +k[MAX:32] := 0 - +
immintrin.h
 - - Floating Point - AVX - Load - - - Load packed double-precision (64-bit) floating-point elements from memory into "dst" using "mask" (elements are zeroed out when the high bit of the corresponding element is not set). - -FOR j := 0 to 3 - i := j*64 - IF mask[i+63] - dst[i+63:i] := MEM[mem_addr+i+63:mem_addr+i] + + Integer + Mask + AVX512VL + AVX512BW + Compare + + + + + + Compare packed unsigned 8-bit integers in "a" and "b" based on the comparison operand specified by "imm8", and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). + CASE (imm8[2:0]) OF +0: OP := _MM_CMPINT_EQ +1: OP := _MM_CMPINT_LT +2: OP := _MM_CMPINT_LE +3: OP := _MM_CMPINT_FALSE +4: OP := _MM_CMPINT_NE +5: OP := _MM_CMPINT_NLT +6: OP := _MM_CMPINT_NLE +7: OP := _MM_CMPINT_TRUE +ESAC +FOR j := 0 to 31 + i := j*8 + IF k1[j] + k[j] := ( a[i+7:i] OP b[i+7:i] ) ? 1 : 0 ELSE - dst[i+63:i] := 0 + k[j] := 0 FI ENDFOR -dst[MAX:256] := 0 +k[MAX:32] := 0 - +
immintrin.h
 - - Floating Point - AVX - Store - - - - Store packed double-precision (64-bit) floating-point elements from "a" into memory using "mask". + + Integer + Mask + AVX512VL + AVX512BW + Compare + + + + + Compare packed unsigned 8-bit integers in "a" and "b" for equality, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*64 - IF mask[i+63] - MEM[mem_addr+i+63:mem_addr+i] := a[i+63:i] +FOR j := 0 to 31 + i := j*8 + IF k1[j] + k[j] := ( a[i+7:i] == b[i+7:i] ) ? 1 : 0 + ELSE + k[j] := 0 FI ENDFOR +k[MAX:32] := 0 - +
immintrin.h
 - - Floating Point - AVX - Load - - - Load packed double-precision (64-bit) floating-point elements from memory into "dst" using "mask" (elements are zeroed out when the high bit of the corresponding element is not set). + + Integer + Mask + AVX512VL + AVX512BW + Compare + + + + + Compare packed unsigned 8-bit integers in "a" and "b" for greater-than-or-equal, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 1 - i := j*64 - IF mask[i+63] - dst[i+63:i] := MEM[mem_addr+i+63:mem_addr+i] +FOR j := 0 to 31 + i := j*8 + IF k1[j] + k[j] := ( a[i+7:i] >= b[i+7:i] ) ? 1 : 0 ELSE - dst[i+63:i] := 0 + k[j] := 0 FI ENDFOR -dst[MAX:128] := 0 +k[MAX:32] := 0 - +
immintrin.h
 - - Floating Point - AVX - Store - - - - Store packed double-precision (64-bit) floating-point elements from "a" into memory using "mask". + + Integer + Mask + AVX512VL + AVX512BW + Compare + + + + + Compare packed unsigned 8-bit integers in "a" and "b" for greater-than, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 1 - i := j*64 - IF mask[i+63] - MEM[mem_addr+i+63:mem_addr+i] := a[i+63:i] +FOR j := 0 to 31 + i := j*8 + IF k1[j] + k[j] := ( a[i+7:i] > b[i+7:i] ) ? 1 : 0 + ELSE + k[j] := 0 FI ENDFOR +k[MAX:32] := 0 - +
immintrin.h
 - - Floating Point - AVX - Load - - - Load packed single-precision (32-bit) floating-point elements from memory into "dst" using "mask" (elements are zeroed out when the high bit of the corresponding element is not set). + + Integer + Mask + AVX512VL + AVX512BW + Compare + + + + + Compare packed unsigned 8-bit integers in "a" and "b" for less-than-or-equal, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := j*32 - IF mask[i+31] - dst[i+31:i] := MEM[mem_addr+i+31:mem_addr+i] +FOR j := 0 to 31 + i := j*8 + IF k1[j] + k[j] := ( a[i+7:i] <= b[i+7:i] ) ? 1 : 0 ELSE - dst[i+31:i] := 0 + k[j] := 0 FI ENDFOR -dst[MAX:256] := 0 +k[MAX:32] := 0 - +
immintrin.h
 - - Floating Point - AVX - Store - - - - Store packed single-precision (32-bit) floating-point elements from "a" into memory using "mask". + + Integer + Mask + AVX512VL + AVX512BW + Compare + + + + + Compare packed unsigned 8-bit integers in "a" and "b" for less-than, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := j*32 - IF mask[i+31] - MEM[mem_addr+i+31:mem_addr+i] := a[i+31:i] +FOR j := 0 to 31 + i := j*8 + IF k1[j] + k[j] := ( a[i+7:i] < b[i+7:i] ) ? 1 : 0 + ELSE + k[j] := 0 FI ENDFOR +k[MAX:32] := 0 - +
immintrin.h
 - - Floating Point - AVX - Load - - - Load packed single-precision (32-bit) floating-point elements from memory into "dst" using "mask" (elements are zeroed out when the high bit of the corresponding element is not set). + + Integer + Mask + AVX512VL + AVX512BW + Compare + + + + + Compare packed unsigned 8-bit integers in "a" and "b" for not-equal, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*32 - IF mask[i+31] - dst[i+31:i] := MEM[mem_addr+i+31:mem_addr+i] +FOR j := 0 to 31 + i := j*8 + IF k1[j] + k[j] := ( a[i+7:i] != b[i+7:i] ) ? 1 : 0 ELSE - dst[i+31:i] := 0 + k[j] := 0 FI ENDFOR -dst[MAX:128] := 0 +k[MAX:32] := 0 - +
immintrin.h
 - - Floating Point - AVX - Store - - - - Store packed single-precision (32-bit) floating-point elements from "a" into memory using "mask". - -FOR j := 0 to 3 - i := j*32 - IF mask[i+31] - MEM[mem_addr+i+31:mem_addr+i] := a[i+31:i] - FI + + Integer + Mask + AVX512BW + Compare + + + + + Compare packed unsigned 8-bit integers in "a" and "b" based on the comparison operand specified by "imm8", and store the results in mask vector "k". + CASE (imm8[2:0]) OF +0: OP := _MM_CMPINT_EQ +1: OP := _MM_CMPINT_LT +2: OP := _MM_CMPINT_LE +3: OP := _MM_CMPINT_FALSE +4: OP := _MM_CMPINT_NE +5: OP := _MM_CMPINT_NLT +6: OP := _MM_CMPINT_NLE +7: OP := _MM_CMPINT_TRUE +ESAC +FOR j := 0 to 63 + i := j*8 + k[j] := ( a[i+7:i] OP b[i+7:i] ) ? 1 : 0 ENDFOR +k[MAX:64] := 0 - +
immintrin.h
 - - Floating Point - AVX - Move - - Duplicate odd-indexed single-precision (32-bit) floating-point elements from "a", and store the results in "dst". - + + Integer + Mask + AVX512BW + Compare + + + + Compare packed unsigned 8-bit integers in "a" and "b" for equality, and store the results in mask vector "k". -dst[31:0] := a[63:32] -dst[63:32] := a[63:32] -dst[95:64] := a[127:96] -dst[127:96] := a[127:96] -dst[159:128] := a[191:160] -dst[191:160] := a[191:160] -dst[223:192] := a[255:224] -dst[255:224] := a[255:224] -dst[MAX:256] := 0 +FOR j := 0 to 63 + i := j*8 + k[j] := ( a[i+7:i] == b[i+7:i] ) ? 1 : 0 +ENDFOR +k[MAX:64] := 0 - +
immintrin.h
 - - Floating Point - AVX - Move - - Duplicate even-indexed single-precision (32-bit) floating-point elements from "a", and store the results in "dst". - + + Integer + Mask + AVX512BW + Compare + + + + Compare packed unsigned 8-bit integers in "a" and "b" for greater-than-or-equal, and store the results in mask vector "k". -dst[31:0] := a[31:0] -dst[63:32] := a[31:0] -dst[95:64] := a[95:64] -dst[127:96] := a[95:64] -dst[159:128] := a[159:128] -dst[191:160] := a[159:128] -dst[223:192] := a[223:192] -dst[255:224] := a[223:192] -dst[MAX:256] := 0 - - -
immintrin.h
- - - Floating Point - AVX - Move - - Duplicate even-indexed double-precision (64-bit) floating-point elements from "a", and store the results in "dst". - - -dst[63:0] := a[63:0] -dst[127:64] := a[63:0] -dst[191:128] := a[191:128] -dst[255:192] := a[191:128] -dst[MAX:256] := 0 +FOR j := 0 to 63 + i := j*8 + k[j] := ( a[i+7:i] >= b[i+7:i] ) ? 1 : 0 +ENDFOR +k[MAX:64] := 0 - +
immintrin.h
 - + Integer - AVX - Load - - Load 256-bits of integer data from unaligned memory into "dst". This intrinsic may perform better than "_mm256_loadu_si256" when the data crosses a cache line boundary. + Mask + AVX512BW + Compare + + + + Compare packed unsigned 8-bit integers in "a" and "b" for greater-than, and store the results in mask vector "k". -dst[255:0] := MEM[mem_addr+255:mem_addr] -dst[MAX:256] := 0 +FOR j := 0 to 63 + i := j*8 + k[j] := ( a[i+7:i] > b[i+7:i] ) ? 1 : 0 +ENDFOR +k[MAX:64] := 0 - +
immintrin.h
 - + Integer - AVX - Store - - - Store 256-bits of integer data from "a" into memory using a non-temporal memory hint. - "mem_addr" must be aligned on a 32-byte boundary or a general-protection exception may be generated. + Mask + AVX512BW + Compare + + + + Compare packed unsigned 8-bit integers in "a" and "b" for less-than-or-equal, and store the results in mask vector "k". -MEM[mem_addr+255:mem_addr] := a[255:0] +FOR j := 0 to 63 + i := j*8 + k[j] := ( a[i+7:i] <= b[i+7:i] ) ? 1 : 0 +ENDFOR +k[MAX:64] := 0 - +
immintrin.h
 - - Floating Point - AVX - Store - - - Store 256-bits (composed of 4 packed double-precision (64-bit) floating-point elements) from "a" into memory using a non-temporal memory hint. - "mem_addr" must be aligned on a 32-byte boundary or a general-protection exception may be generated. + + Integer + Mask + AVX512BW + Compare + + + + Compare packed unsigned 8-bit integers in "a" and "b" for less-than, and store the results in mask vector "k". -MEM[mem_addr+255:mem_addr] := a[255:0] +FOR j := 0 to 63 + i := j*8 + k[j] := ( a[i+7:i] < b[i+7:i] ) ? 1 : 0 +ENDFOR +k[MAX:64] := 0 - +
immintrin.h
 - - Floating Point - AVX - Store - - - Store 256-bits (composed of 8 packed single-precision (32-bit) floating-point elements) from "a" into memory using a non-temporal memory hint. - "mem_addr" must be aligned on a 32-byte boundary or a general-protection exception may be generated. + + Integer + Mask + AVX512BW + Compare + + + + Compare packed unsigned 8-bit integers in "a" and "b" for not-equal, and store the results in mask vector "k". -MEM[mem_addr+255:mem_addr] := a[255:0] +FOR j := 0 to 63 + i := j*8 + k[j] := ( a[i+7:i] != b[i+7:i] ) ? 1 : 0 +ENDFOR +k[MAX:64] := 0 - +
immintrin.h
 - - Floating Point - AVX - Elementary Math Functions - - Compute the approximate reciprocal of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst". The maximum relative error for this approximation is less than 1.5*2^-12. - -FOR j := 0 to 7 - i := j*32 - dst[i+31:i] := APPROXIMATE(1.0/a[i+31:i]) + + Integer + Mask + AVX512BW + Compare + + + + + + Compare packed unsigned 8-bit integers in "a" and "b" based on the comparison operand specified by "imm8", and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). + CASE (imm8[2:0]) OF +0: OP := _MM_CMPINT_EQ +1: OP := _MM_CMPINT_LT +2: OP := _MM_CMPINT_LE +3: OP := _MM_CMPINT_FALSE +4: OP := _MM_CMPINT_NE +5: OP := _MM_CMPINT_NLT +6: OP := _MM_CMPINT_NLE +7: OP := _MM_CMPINT_TRUE +ESAC +FOR j := 0 to 63 + i := j*8 + IF k1[j] + k[j] := ( a[i+7:i] OP b[i+7:i] ) ? 1 : 0 + ELSE + k[j] := 0 + FI ENDFOR -dst[MAX:256] := 0 +k[MAX:64] := 0 - +
immintrin.h
 - - Floating Point - AVX - Elementary Math Functions - - Compute the approximate reciprocal square root of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst". The maximum relative error for this approximation is less than 1.5*2^-12. + + Integer + Mask + AVX512BW + Compare + + + + + Compare packed unsigned 8-bit integers in "a" and "b" for equality, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := j*32 - dst[i+31:i] := APPROXIMATE(1.0 / SQRT(a[i+31:i])) +FOR j := 0 to 63 + i := j*8 + IF k1[j] + k[j] := ( a[i+7:i] == b[i+7:i] ) ? 1 : 0 + ELSE + k[j] := 0 + FI ENDFOR -dst[MAX:256] := 0 +k[MAX:64] := 0 - +
immintrin.h
 - - Floating Point - AVX - Elementary Math Functions - - Compute the square root of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst". + + Integer + Mask + AVX512BW + Compare + + + + + Compare packed unsigned 8-bit integers in "a" and "b" for greater-than-or-equal, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*64 - dst[i+63:i] := SQRT(a[i+63:i]) +FOR j := 0 to 63 + i := j*8 + IF k1[j] + k[j] := ( a[i+7:i] >= b[i+7:i] ) ? 1 : 0 + ELSE + k[j] := 0 + FI ENDFOR -dst[MAX:256] := 0 +k[MAX:64] := 0 - +
immintrin.h
 - - Floating Point - AVX - Elementary Math Functions - - Compute the square root of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst". + + Integer + Mask + AVX512BW + Compare + + + + + Compare packed unsigned 8-bit integers in "a" and "b" for greater-than, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := j*32 - dst[i+31:i] := SQRT(a[i+31:i]) +FOR j := 0 to 63 + i := j*8 + IF k1[j] + k[j] := ( a[i+7:i] > b[i+7:i] ) ? 1 : 0 + ELSE + k[j] := 0 + FI ENDFOR -dst[MAX:256] := 0 +k[MAX:64] := 0 - +
immintrin.h
 - - Floating Point - AVX - Special Math Functions - - - Round the packed double-precision (64-bit) floating-point elements in "a" using the "rounding" parameter, and store the results as packed double-precision floating-point elements in "dst". - [round_note] - + + Integer + Mask + AVX512BW + Compare + + + + + Compare packed unsigned 8-bit integers in "a" and "b" for less-than-or-equal, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*64 - dst[i+63:i] := ROUND(a[i+63:i]) +FOR j := 0 to 63 + i := j*8 + IF k1[j] + k[j] := ( a[i+7:i] <= b[i+7:i] ) ? 1 : 0 + ELSE + k[j] := 0 + FI ENDFOR -dst[MAX:256] := 0 +k[MAX:64] := 0 - +
immintrin.h
 - - Floating Point - AVX - Special Math Functions - - - Round the packed single-precision (32-bit) floating-point elements in "a" using the "rounding" parameter, and store the results as packed single-precision floating-point elements in "dst". - [round_note] - + + Integer + Mask + AVX512BW + Compare + + + + + Compare packed unsigned 8-bit integers in "a" and "b" for less-than, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := j*32 - dst[i+31:i] := ROUND(a[i+31:i]) +FOR j := 0 to 63 + i := j*8 + IF k1[j] + k[j] := ( a[i+7:i] < b[i+7:i] ) ? 1 : 0 + ELSE + k[j] := 0 + FI ENDFOR -dst[MAX:256] := 0 +k[MAX:64] := 0 - +
immintrin.h
 - - Floating Point - AVX - Swizzle - - - Unpack and interleave double-precision (64-bit) floating-point elements from the high half of each 128-bit lane in "a" and "b", and store the results in "dst". + + Integer + Mask + AVX512BW + Compare + + + + + Compare packed unsigned 8-bit integers in "a" and "b" for not-equal, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). -INTERLEAVE_HIGH_QWORDS(src1[127:0], src2[127:0]){ - dst[63:0] := src1[127:64] - dst[127:64] := src2[127:64] - RETURN dst[127:0] -} - -dst[127:0] := INTERLEAVE_HIGH_QWORDS(a[127:0], b[127:0]) -dst[255:128] := INTERLEAVE_HIGH_QWORDS(a[255:128], b[255:128]) -dst[MAX:256] := 0 +FOR j := 0 to 63 + i := j*8 + IF k1[j] + k[j] := ( a[i+7:i] != b[i+7:i] ) ? 1 : 0 + ELSE + k[j] := 0 + FI +ENDFOR +k[MAX:64] := 0 - +
immintrin.h
 - - Floating Point - AVX - Swizzle - - - Unpack and interleave single-precision (32-bit) floating-point elements from the high half of each 128-bit lane in "a" and "b", and store the results in "dst". - -INTERLEAVE_HIGH_DWORDS(src1[127:0], src2[127:0]){ - dst[31:0] := src1[95:64] - dst[63:32] := src2[95:64] - dst[95:64] := src1[127:96] - dst[127:96] := src2[127:96] - RETURN dst[127:0] -} - -dst[127:0] := INTERLEAVE_HIGH_DWORDS(a[127:0], b[127:0]) -dst[255:128] := INTERLEAVE_HIGH_DWORDS(a[255:128], b[255:128]) -dst[MAX:256] := 0 + + Integer + Mask + AVX512VL + AVX512BW + Compare + + + + + Compare packed unsigned 8-bit integers in "a" and "b" based on the comparison operand specified by "imm8", and store the results in mask vector "k". + CASE (imm8[2:0]) OF +0: OP := _MM_CMPINT_EQ +1: OP := _MM_CMPINT_LT +2: OP := _MM_CMPINT_LE +3: OP := _MM_CMPINT_FALSE +4: OP := _MM_CMPINT_NE +5: OP := _MM_CMPINT_NLT +6: OP := _MM_CMPINT_NLE +7: OP := _MM_CMPINT_TRUE +ESAC +FOR j := 0 to 15 + i := j*8 + k[j] := ( a[i+7:i] OP b[i+7:i] ) ? 1 : 0 +ENDFOR +k[MAX:16] := 0 - +
immintrin.h
 - - Floating Point - AVX - Swizzle - - - Unpack and interleave double-precision (64-bit) floating-point elements from the low half of each 128-bit lane in "a" and "b", and store the results in "dst". + + Integer + Mask + AVX512VL + AVX512BW + Compare + + + + Compare packed unsigned 8-bit integers in "a" and "b" for equality, and store the results in mask vector "k". -INTERLEAVE_QWORDS(src1[127:0], src2[127:0]){ - dst[63:0] := src1[63:0] - dst[127:64] := src2[63:0] - RETURN dst[127:0] -} - -dst[127:0] := INTERLEAVE_QWORDS(a[127:0], b[127:0]) -dst[255:128] := INTERLEAVE_QWORDS(a[255:128], b[255:128]) -dst[MAX:256] := 0 +FOR j := 0 to 15 + i := j*8 + k[j] := ( a[i+7:i] == b[i+7:i] ) ? 1 : 0 +ENDFOR +k[MAX:16] := 0 - +
immintrin.h
 - - Floating Point - AVX - Swizzle - - - Unpack and interleave single-precision (32-bit) floating-point elements from the low half of each 128-bit lane in "a" and "b", and store the results in "dst". + + Integer + Mask + AVX512VL + AVX512BW + Compare + + + + Compare packed unsigned 8-bit integers in "a" and "b" for greater-than-or-equal, and store the results in mask vector "k". -INTERLEAVE_DWORDS(src1[127:0], src2[127:0]){ - dst[31:0] := src1[31:0] - dst[63:32] := src2[31:0] - dst[95:64] := src1[63:32] - dst[127:96] := src2[63:32] - RETURN dst[127:0] -} - -dst[127:0] := INTERLEAVE_DWORDS(a[127:0], b[127:0]) -dst[255:128] := INTERLEAVE_DWORDS(a[255:128], b[255:128]) -dst[MAX:256] := 0 +FOR j := 0 to 15 + i := j*8 + k[j] := ( a[i+7:i] >= b[i+7:i] ) ? 1 : 0 +ENDFOR +k[MAX:16] := 0 - +
immintrin.h
 - + Integer - AVX - Logical - - - Compute the bitwise AND of 256 bits (representing integer data) in "a" and "b", and set "ZF" to 1 if the result is zero, otherwise set "ZF" to 0. Compute the bitwise NOT of "a" and then AND with "b", and set "CF" to 1 if the result is zero, otherwise set "CF" to 0. Return the "ZF" value. + Mask + AVX512VL + AVX512BW + Compare + + + + Compare packed unsigned 8-bit integers in "a" and "b" for greater-than, and store the results in mask vector "k". -IF (a[255:0] AND b[255:0] == 0) - ZF := 1 -ELSE - ZF := 0 -FI -IF ((NOT a[255:0]) AND b[255:0] == 0) - CF := 1 -ELSE - CF := 0 -FI -RETURN ZF +FOR j := 0 to 15 + i := j*8 + k[j] := ( a[i+7:i] > b[i+7:i] ) ? 1 : 0 +ENDFOR +k[MAX:16] := 0 - +
immintrin.h
 - + Integer - AVX - Logical - - - Compute the bitwise AND of 256 bits (representing integer data) in "a" and "b", and set "ZF" to 1 if the result is zero, otherwise set "ZF" to 0. Compute the bitwise NOT of "a" and then AND with "b", and set "CF" to 1 if the result is zero, otherwise set "CF" to 0. Return the "CF" value. + Mask + AVX512VL + AVX512BW + Compare + + + + Compare packed unsigned 8-bit integers in "a" and "b" for less-than-or-equal, and store the results in mask vector "k". -IF (a[255:0] AND b[255:0] == 0) - ZF := 1 -ELSE - ZF := 0 -FI -IF ((NOT a[255:0]) AND b[255:0] == 0) - CF := 1 -ELSE - CF := 0 -FI -RETURN CF +FOR j := 0 to 15 + i := j*8 + k[j] := ( a[i+7:i] <= b[i+7:i] ) ? 1 : 0 +ENDFOR +k[MAX:16] := 0 - +
immintrin.h
 - + Integer - AVX - Logical - - - Compute the bitwise AND of 256 bits (representing integer data) in "a" and "b", and set "ZF" to 1 if the result is zero, otherwise set "ZF" to 0. Compute the bitwise NOT of "a" and then AND with "b", and set "CF" to 1 if the result is zero, otherwise set "CF" to 0. Return 1 if both the "ZF" and "CF" values are zero, otherwise return 0. + Mask + AVX512VL + AVX512BW + Compare + + + + Compare packed unsigned 8-bit integers in "a" and "b" for less-than, and store the results in mask vector "k". -IF (a[255:0] AND b[255:0] == 0) - ZF := 1 -ELSE - ZF := 0 -FI -IF ((NOT a[255:0]) AND b[255:0] == 0) - CF := 1 -ELSE - CF := 0 -FI -IF (ZF == 0 && CF == 0) - RETURN 1 -ELSE - RETURN 0 -FI +FOR j := 0 to 15 + i := j*8 + k[j] := ( a[i+7:i] < b[i+7:i] ) ? 1 : 0 +ENDFOR +k[MAX:16] := 0 - +
immintrin.h
 - - Floating Point - AVX - Logical - - - Compute the bitwise AND of 256 bits (representing double-precision (64-bit) floating-point elements) in "a" and "b", producing an intermediate 256-bit value, and set "ZF" to 1 if the sign bit of each 64-bit element in the intermediate value is zero, otherwise set "ZF" to 0. Compute the bitwise NOT of "a" and then AND with "b", producing an intermediate value, and set "CF" to 1 if the sign bit of each 64-bit element in the intermediate value is zero, otherwise set "CF" to 0. Return the "ZF" value. + + Integer + Mask + AVX512VL + AVX512BW + Compare + + + + Compare packed unsigned 8-bit integers in "a" and "b" for not-equal, and store the results in mask vector "k". -tmp[255:0] := a[255:0] AND b[255:0] -IF (tmp[63] == tmp[127] == tmp[191] == tmp[255] == 0) - ZF := 1 -ELSE - ZF := 0 -FI -tmp[255:0] := (NOT a[255:0]) AND b[255:0] -IF (tmp[63] == tmp[127] == tmp[191] == tmp[255] == 0) - CF := 1 -ELSE - CF := 0 -FI -RETURN ZF +FOR j := 0 to 15 + i := j*8 + k[j] := ( a[i+7:i] != b[i+7:i] ) ? 1 : 0 +ENDFOR +k[MAX:16] := 0 - +
immintrin.h
 - - Floating Point - AVX - Logical - - - Compute the bitwise AND of 256 bits (representing double-precision (64-bit) floating-point elements) in "a" and "b", producing an intermediate 256-bit value, and set "ZF" to 1 if the sign bit of each 64-bit element in the intermediate value is zero, otherwise set "ZF" to 0. Compute the bitwise NOT of "a" and then AND with "b", producing an intermediate value, and set "CF" to 1 if the sign bit of each 64-bit element in the intermediate value is zero, otherwise set "CF" to 0. Return the "CF" value. - -tmp[255:0] := a[255:0] AND b[255:0] -IF (tmp[63] == tmp[127] == tmp[191] == tmp[255] == 0) - ZF := 1 -ELSE - ZF := 0 -FI -tmp[255:0] := (NOT a[255:0]) AND b[255:0] -IF (tmp[63] == tmp[127] == tmp[191] == tmp[255] == 0) - CF := 1 -ELSE - CF := 0 -FI -RETURN CF + + Integer + Mask + AVX512VL + AVX512BW + Compare + + + + + + Compare packed unsigned 8-bit integers in "a" and "b" based on the comparison operand specified by "imm8", and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). + CASE (imm8[2:0]) OF +0: OP := _MM_CMPINT_EQ +1: OP := _MM_CMPINT_LT +2: OP := _MM_CMPINT_LE +3: OP := _MM_CMPINT_FALSE +4: OP := _MM_CMPINT_NE +5: OP := _MM_CMPINT_NLT +6: OP := _MM_CMPINT_NLE +7: OP := _MM_CMPINT_TRUE +ESAC +FOR j := 0 to 15 + i := j*8 + IF k1[j] + k[j] := ( a[i+7:i] OP b[i+7:i] ) ? 1 : 0 + ELSE + k[j] := 0 + FI +ENDFOR +k[MAX:16] := 0 - +
immintrin.h
 - - Floating Point - AVX - Logical - - - Compute the bitwise AND of 256 bits (representing double-precision (64-bit) floating-point elements) in "a" and "b", producing an intermediate 256-bit value, and set "ZF" to 1 if the sign bit of each 64-bit element in the intermediate value is zero, otherwise set "ZF" to 0. Compute the bitwise NOT of "a" and then AND with "b", producing an intermediate value, and set "CF" to 1 if the sign bit of each 64-bit element in the intermediate value is zero, otherwise set "CF" to 0. Return 1 if both the "ZF" and "CF" values are zero, otherwise return 0. + + Integer + Mask + AVX512VL + AVX512BW + Compare + + + + + Compare packed unsigned 8-bit integers in "a" and "b" for equality, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). -tmp[255:0] := a[255:0] AND b[255:0] -IF (tmp[63] == tmp[127] == tmp[191] == tmp[255] == 0) - ZF := 1 -ELSE - ZF := 0 -FI -tmp[255:0] := (NOT a[255:0]) AND b[255:0] -IF (tmp[63] == tmp[127] == tmp[191] == tmp[255] == 0) - CF := 1 -ELSE - CF := 0 -FI -IF (ZF == 0 && CF == 0) - RETURN 1 -ELSE - RETURN 0 -FI +FOR j := 0 to 15 + i := j*8 + IF k1[j] + k[j] := ( a[i+7:i] == b[i+7:i] ) ? 1 : 0 + ELSE + k[j] := 0 + FI +ENDFOR +k[MAX:16] := 0 - +
immintrin.h
 - - Floating Point - AVX - Logical - - - Compute the bitwise AND of 128 bits (representing double-precision (64-bit) floating-point elements) in "a" and "b", producing an intermediate 128-bit value, and set "ZF" to 1 if the sign bit of each 64-bit element in the intermediate value is zero, otherwise set "ZF" to 0. Compute the bitwise NOT of "a" and then AND with "b", producing an intermediate value, and set "CF" to 1 if the sign bit of each 64-bit element in the intermediate value is zero, otherwise set "CF" to 0. Return the "ZF" value. + + Integer + Mask + AVX512VL + AVX512BW + Compare + + + + + Compare packed unsigned 8-bit integers in "a" and "b" for greater-than-or-equal, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). -tmp[127:0] := a[127:0] AND b[127:0] -IF (tmp[63] == tmp[127] == 0) - ZF := 1 -ELSE - ZF := 0 -FI -tmp[127:0] := (NOT a[127:0]) AND b[127:0] -IF (tmp[63] == tmp[127] == 0) - CF := 1 -ELSE - CF := 0 -FI -RETURN ZF +FOR j := 0 to 15 + i := j*8 + IF k1[j] + k[j] := ( a[i+7:i] >= b[i+7:i] ) ? 1 : 0 + ELSE + k[j] := 0 + FI +ENDFOR +k[MAX:16] := 0 - +
immintrin.h
 - - Floating Point - AVX - Logical - - - Compute the bitwise AND of 128 bits (representing double-precision (64-bit) floating-point elements) in "a" and "b", producing an intermediate 128-bit value, and set "ZF" to 1 if the sign bit of each 64-bit element in the intermediate value is zero, otherwise set "ZF" to 0. Compute the bitwise NOT of "a" and then AND with "b", producing an intermediate value, and set "CF" to 1 if the sign bit of each 64-bit element in the intermediate value is zero, otherwise set "CF" to 0. Return the "CF" value. - -tmp[127:0] := a[127:0] AND b[127:0] -IF (tmp[63] == tmp[127] == 0) - ZF := 1 -ELSE - ZF := 0 -FI -tmp[127:0] := (NOT a[127:0]) AND b[127:0] -IF (tmp[63] == tmp[127] == 0) - CF := 1 -ELSE - CF := 0 -FI -RETURN CF - - -
immintrin.h
- - - Floating Point - AVX - Logical - - - Compute the bitwise AND of 128 bits (representing double-precision (64-bit) floating-point elements) in "a" and "b", producing an intermediate 128-bit value, and set "ZF" to 1 if the sign bit of each 64-bit element in the intermediate value is zero, otherwise set "ZF" to 0. Compute the bitwise NOT of "a" and then AND with "b", producing an intermediate value, and set "CF" to 1 if the sign bit of each 64-bit element in the intermediate value is zero, otherwise set "CF" to 0. Return 1 if both the "ZF" and "CF" values are zero, otherwise return 0. + + Integer + Mask + AVX512VL + AVX512BW + Compare + + + + + Compare packed unsigned 8-bit integers in "a" and "b" for greater-than, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). -tmp[127:0] := a[127:0] AND b[127:0] -IF (tmp[63] == tmp[127] == 0) - ZF := 1 -ELSE - ZF := 0 -FI -tmp[127:0] := (NOT a[127:0]) AND b[127:0] -IF (tmp[63] == tmp[127] == 0) - CF := 1 -ELSE - CF := 0 -FI -IF (ZF == 0 && CF == 0) - RETURN 1 -ELSE - RETURN 0 -FI +FOR j := 0 to 15 + i := j*8 + IF k1[j] + k[j] := ( a[i+7:i] > b[i+7:i] ) ? 1 : 0 + ELSE + k[j] := 0 + FI +ENDFOR +k[MAX:16] := 0 - +
immintrin.h
 - - Floating Point - AVX - Logical - - - Compute the bitwise AND of 256 bits (representing single-precision (32-bit) floating-point elements) in "a" and "b", producing an intermediate 256-bit value, and set "ZF" to 1 if the sign bit of each 32-bit element in the intermediate value is zero, otherwise set "ZF" to 0. Compute the bitwise NOT of "a" and then AND with "b", producing an intermediate value, and set "CF" to 1 if the sign bit of each 32-bit element in the intermediate value is zero, otherwise set "CF" to 0. Return the "ZF" value. + + Integer + Mask + AVX512VL + AVX512BW + Compare + + + + + Compare packed unsigned 8-bit integers in "a" and "b" for less-than-or-equal, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). -tmp[255:0] := a[255:0] AND b[255:0] -IF (tmp[31] == tmp[63] == tmp[95] == tmp[127] == tmp[159] == tmp[191] == tmp[223] == tmp[255] == 0) - ZF := 1 -ELSE - ZF := 0 -FI -tmp[255:0] := (NOT a[255:0]) AND b[255:0] -IF (tmp[31] == tmp[63] == tmp[95] == tmp[127] == tmp[159] == tmp[191] == tmp[223] == tmp[255] == 0) - CF := 1 -ELSE - CF := 0 -FI -RETURN ZF +FOR j := 0 to 15 + i := j*8 + IF k1[j] + k[j] := ( a[i+7:i] <= b[i+7:i] ) ? 1 : 0 + ELSE + k[j] := 0 + FI +ENDFOR +k[MAX:16] := 0 - +
immintrin.h
 - - Floating Point - AVX - Logical - - - Compute the bitwise AND of 256 bits (representing single-precision (32-bit) floating-point elements) in "a" and "b", producing an intermediate 256-bit value, and set "ZF" to 1 if the sign bit of each 32-bit element in the intermediate value is zero, otherwise set "ZF" to 0. Compute the bitwise NOT of "a" and then AND with "b", producing an intermediate value, and set "CF" to 1 if the sign bit of each 32-bit element in the intermediate value is zero, otherwise set "CF" to 0. Return the "CF" value. + + Integer + Mask + AVX512VL + AVX512BW + Compare + + + + + Compare packed unsigned 8-bit integers in "a" and "b" for less-than, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). -tmp[255:0] := a[255:0] AND b[255:0] -IF (tmp[31] == tmp[63] == tmp[95] == tmp[127] == tmp[159] == tmp[191] == tmp[223] == tmp[255] == 0) - ZF := 1 -ELSE - ZF := 0 -FI -tmp[255:0] := (NOT a[255:0]) AND b[255:0] -IF (tmp[31] == tmp[63] == tmp[95] == tmp[127] == tmp[159] == tmp[191] == tmp[223] == tmp[255] == 0) - CF := 1 -ELSE - CF := 0 -FI -RETURN CF +FOR j := 0 to 15 + i := j*8 + IF k1[j] + k[j] := ( a[i+7:i] < b[i+7:i] ) ? 1 : 0 + ELSE + k[j] := 0 + FI +ENDFOR +k[MAX:16] := 0 - +
immintrin.h
 - - Floating Point - AVX - Logical - - - Compute the bitwise AND of 256 bits (representing single-precision (32-bit) floating-point elements) in "a" and "b", producing an intermediate 256-bit value, and set "ZF" to 1 if the sign bit of each 32-bit element in the intermediate value is zero, otherwise set "ZF" to 0. Compute the bitwise NOT of "a" and then AND with "b", producing an intermediate value, and set "CF" to 1 if the sign bit of each 32-bit element in the intermediate value is zero, otherwise set "CF" to 0. Return 1 if both the "ZF" and "CF" values are zero, otherwise return 0. + + Integer + Mask + AVX512VL + AVX512BW + Compare + + + + + Compare packed unsigned 8-bit integers in "a" and "b" for not-equal, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). -tmp[255:0] := a[255:0] AND b[255:0] -IF (tmp[31] == tmp[63] == tmp[95] == tmp[127] == tmp[159] == tmp[191] == tmp[223] == tmp[255] == 0) - ZF := 1 -ELSE - ZF := 0 -FI -tmp[255:0] := (NOT a[255:0]) AND b[255:0] -IF (tmp[31] == tmp[63] == tmp[95] == tmp[127] == tmp[159] == tmp[191] == tmp[223] == tmp[255] == 0) - CF := 1 -ELSE - CF := 0 -FI -IF (ZF == 0 && CF == 0) - RETURN 1 -ELSE - RETURN 0 -FI +FOR j := 0 to 15 + i := j*8 + IF k1[j] + k[j] := ( a[i+7:i] != b[i+7:i] ) ? 1 : 0 + ELSE + k[j] := 0 + FI +ENDFOR +k[MAX:16] := 0 - +
immintrin.h
 - - Floating Point - AVX - Logical - - - Compute the bitwise AND of 128 bits (representing single-precision (32-bit) floating-point elements) in "a" and "b", producing an intermediate 128-bit value, and set "ZF" to 1 if the sign bit of each 32-bit element in the intermediate value is zero, otherwise set "ZF" to 0. Compute the bitwise NOT of "a" and then AND with "b", producing an intermediate value, and set "CF" to 1 if the sign bit of each 32-bit element in the intermediate value is zero, otherwise set "CF" to 0. Return the "ZF" value. - -tmp[127:0] := a[127:0] AND b[127:0] -IF (tmp[31] == tmp[63] == tmp[95] == tmp[127] == 0) - ZF := 1 -ELSE - ZF := 0 -FI -tmp[127:0] := (NOT a[127:0]) AND b[127:0] -IF (tmp[31] == tmp[63] == tmp[95] == tmp[127] == 0) - CF := 1 -ELSE - CF := 0 -FI -RETURN ZF + + Integer + Mask + AVX512VL + AVX512BW + Compare + + + + + Compare packed unsigned 16-bit integers in "a" and "b" based on the comparison operand specified by "imm8", and store the results in mask vector "k". + CASE (imm8[2:0]) OF +0: OP := _MM_CMPINT_EQ +1: OP := _MM_CMPINT_LT +2: OP := _MM_CMPINT_LE +3: OP := _MM_CMPINT_FALSE +4: OP := _MM_CMPINT_NE +5: OP := _MM_CMPINT_NLT +6: OP := _MM_CMPINT_NLE +7: OP := _MM_CMPINT_TRUE +ESAC +FOR j := 0 to 15 + i := j*16 + k[j] := ( a[i+15:i] OP b[i+15:i] ) ? 1 : 0 +ENDFOR +k[MAX:16] := 0 - +
immintrin.h
 - - Floating Point - AVX - Logical - - - Compute the bitwise AND of 128 bits (representing single-precision (32-bit) floating-point elements) in "a" and "b", producing an intermediate 128-bit value, and set "ZF" to 1 if the sign bit of each 32-bit element in the intermediate value is zero, otherwise set "ZF" to 0. Compute the bitwise NOT of "a" and then AND with "b", producing an intermediate value, and set "CF" to 1 if the sign bit of each 32-bit element in the intermediate value is zero, otherwise set "CF" to 0. Return the "CF" value. + + Integer + Mask + AVX512VL + AVX512BW + Compare + + + + Compare packed unsigned 16-bit integers in "a" and "b" for equality, and store the results in mask vector "k". -tmp[127:0] := a[127:0] AND b[127:0] -IF (tmp[31] == tmp[63] == tmp[95] == tmp[127] == 0) - ZF := 1 -ELSE - ZF := 0 -FI -tmp[127:0] := (NOT a[127:0]) AND b[127:0] -IF (tmp[31] == tmp[63] == tmp[95] == tmp[127] == 0) - CF := 1 -ELSE - CF := 0 -FI -RETURN CF +FOR j := 0 to 15 + i := j*16 + k[j] := ( a[i+15:i] == b[i+15:i] ) ? 1 : 0 +ENDFOR +k[MAX:16] := 0 - +
immintrin.h
 - - Floating Point - AVX - Logical - - - Compute the bitwise AND of 128 bits (representing single-precision (32-bit) floating-point elements) in "a" and "b", producing an intermediate 128-bit value, and set "ZF" to 1 if the sign bit of each 32-bit element in the intermediate value is zero, otherwise set "ZF" to 0. Compute the bitwise NOT of "a" and then AND with "b", producing an intermediate value, and set "CF" to 1 if the sign bit of each 32-bit element in the intermediate value is zero, otherwise set "CF" to 0. Return 1 if both the "ZF" and "CF" values are zero, otherwise return 0. + + Integer + Mask + AVX512VL + AVX512BW + Compare + + + + Compare packed unsigned 16-bit integers in "a" and "b" for greater-than-or-equal, and store the results in mask vector "k". -tmp[127:0] := a[127:0] AND b[127:0] -IF (tmp[31] == tmp[63] == tmp[95] == tmp[127] == 0) - ZF := 1 -ELSE - ZF := 0 -FI -tmp[127:0] := (NOT a[127:0]) AND b[127:0] -IF (tmp[31] == tmp[63] == tmp[95] == tmp[127] == 0) - CF := 1 -ELSE - CF := 0 -FI -IF (ZF == 0 && CF == 0) - RETURN 1 -ELSE - RETURN 0 -FI +FOR j := 0 to 15 + i := j*16 + k[j] := ( a[i+15:i] >= b[i+15:i] ) ? 1 : 0 +ENDFOR +k[MAX:16] := 0 - +
immintrin.h
 - - Floating Point - AVX - Miscellaneous - - Set each bit of mask "dst" based on the most significant bit of the corresponding packed double-precision (64-bit) floating-point element in "a". + + Integer + Mask + AVX512VL + AVX512BW + Compare + + + + Compare packed unsigned 16-bit integers in "a" and "b" for greater-than, and store the results in mask vector "k". -FOR j := 0 to 3 - i := j*64 - IF a[i+63] - dst[j] := 1 - ELSE - dst[j] := 0 - FI +FOR j := 0 to 15 + i := j*16 + k[j] := ( a[i+15:i] > b[i+15:i] ) ? 1 : 0 ENDFOR -dst[MAX:4] := 0 +k[MAX:16] := 0 - +
immintrin.h
 - - Floating Point - AVX - Miscellaneous - - Set each bit of mask "dst" based on the most significant bit of the corresponding packed single-precision (32-bit) floating-point element in "a". + + Integer + Mask + AVX512VL + AVX512BW + Compare + + + + Compare packed unsigned 16-bit integers in "a" and "b" for less-than-or-equal, and store the results in mask vector "k". -FOR j := 0 to 7 - i := j*32 - IF a[i+31] - dst[j] := 1 - ELSE - dst[j] := 0 - FI +FOR j := 0 to 15 + i := j*16 + k[j] := ( a[i+15:i] <= b[i+15:i] ) ? 1 : 0 ENDFOR -dst[MAX:8] := 0 +k[MAX:16] := 0 - +
immintrin.h
 - - Floating Point - AVX - Set - - Return vector of type __m256d with all elements set to zero. + + Integer + Mask + AVX512VL + AVX512BW + Compare + + + + Compare packed unsigned 16-bit integers in "a" and "b" for less-than, and store the results in mask vector "k". -dst[MAX:0] := 0 +FOR j := 0 to 15 + i := j*16 + k[j] := ( a[i+15:i] < b[i+15:i] ) ? 1 : 0 +ENDFOR +k[MAX:16] := 0 - +
immintrin.h
 - - Floating Point - AVX - Set - - Return vector of type __m256 with all elements set to zero. + + Integer + Mask + AVX512VL + AVX512BW + Compare + + + + Compare packed unsigned 16-bit integers in "a" and "b" for not-equal, and store the results in mask vector "k". -dst[MAX:0] := 0 +FOR j := 0 to 15 + i := j*16 + k[j] := ( a[i+15:i] != b[i+15:i] ) ? 1 : 0 +ENDFOR +k[MAX:16] := 0 - +
immintrin.h
 - + Integer - AVX - Set - - Return vector of type __m256i with all elements set to zero. - -dst[MAX:0] := 0 + Mask + AVX512VL + AVX512BW + Compare + + + + + + Compare packed unsigned 16-bit integers in "a" and "b" based on the comparison operand specified by "imm8", and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). + CASE (imm8[2:0]) OF +0: OP := _MM_CMPINT_EQ +1: OP := _MM_CMPINT_LT +2: OP := _MM_CMPINT_LE +3: OP := _MM_CMPINT_FALSE +4: OP := _MM_CMPINT_NE +5: OP := _MM_CMPINT_NLT +6: OP := _MM_CMPINT_NLE +7: OP := _MM_CMPINT_TRUE +ESAC +FOR j := 0 to 15 + i := j*16 + IF k1[j] + k[j] := ( a[i+15:i] OP b[i+15:i] ) ? 1 : 0 + ELSE + k[j] := 0 + FI +ENDFOR +k[MAX:16] := 0 - +
immintrin.h
 - - Floating Point - AVX - Set - - - - - Set packed double-precision (64-bit) floating-point elements in "dst" with the supplied values. + + Integer + Mask + AVX512VL + AVX512BW + Compare + + + + + Compare packed unsigned 16-bit integers in "a" and "b" for equality, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). -dst[63:0] := e0 -dst[127:64] := e1 -dst[191:128] := e2 -dst[255:192] := e3 -dst[MAX:256] := 0 +FOR j := 0 to 15 + i := j*16 + IF k1[j] + k[j] := ( a[i+15:i] == b[i+15:i] ) ? 1 : 0 + ELSE + k[j] := 0 + FI +ENDFOR +k[MAX:16] := 0 +
immintrin.h
 - - Floating Point - AVX - Set - - - - - - - - - Set packed single-precision (32-bit) floating-point elements in "dst" with the supplied values. + + Integer + Mask + AVX512VL + AVX512BW + Compare + + + + + Compare packed unsigned 16-bit integers in "a" and "b" for greater-than-or-equal, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). -dst[31:0] := e0 -dst[63:32] := e1 -dst[95:64] := e2 -dst[127:96] := e3 -dst[159:128] := e4 -dst[191:160] := e5 -dst[223:192] := e6 -dst[255:224] := e7 -dst[MAX:256] := 0 +FOR j := 0 to 15 + i := j*16 + IF k1[j] + k[j] := ( a[i+15:i] >= b[i+15:i] ) ? 1 : 0 + ELSE + k[j] := 0 + FI +ENDFOR +k[MAX:16] := 0 +
immintrin.h
 - + Integer - AVX - Set - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - Set packed 8-bit integers in "dst" with the supplied values in reverse order. + Mask + AVX512VL + AVX512BW + Compare + + + + + Compare packed unsigned 16-bit integers in "a" and "b" for greater-than, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). -dst[7:0] := e0 -dst[15:8] := e1 -dst[23:16] := e2 -dst[31:24] := e3 -dst[39:32] := e4 -dst[47:40] := e5 -dst[55:48] := e6 -dst[63:56] := e7 -dst[71:64] := e8 -dst[79:72] := e9 -dst[87:80] := e10 -dst[95:88] := e11 -dst[103:96] := e12 -dst[111:104] := e13 -dst[119:112] := e14 -dst[127:120] := e15 -dst[135:128] := e16 -dst[143:136] := e17 -dst[151:144] := e18 -dst[159:152] := e19 -dst[167:160] := e20 -dst[175:168] := e21 -dst[183:176] := e22 -dst[191:184] := e23 -dst[199:192] := e24 -dst[207:200] := e25 -dst[215:208] := e26 -dst[223:216] := e27 -dst[231:224] := e28 -dst[239:232] := e29 -dst[247:240] := e30 -dst[255:248] := e31 -dst[MAX:256] := 0 +FOR j := 0 to 15 + i := j*16 + IF k1[j] + k[j] := ( a[i+15:i] > b[i+15:i] ) ? 1 : 0 + ELSE + k[j] := 0 + FI +ENDFOR +k[MAX:16] := 0 +
immintrin.h
 - + Integer - AVX - Set - - - - - - - - - - - - - - - - - Set packed 16-bit integers in "dst" with the supplied values. + Mask + AVX512VL + AVX512BW + Compare + + + + + Compare packed unsigned 16-bit integers in "a" and "b" for less-than-or-equal, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). -dst[15:0] := e0 -dst[31:16] := e1 -dst[47:32] := e2 -dst[63:48] := e3 -dst[79:64] := e4 -dst[95:80] := e5 -dst[111:96] := e6 -dst[127:112] := e7 -dst[145:128] := e8 -dst[159:144] := e9 -dst[175:160] := e10 -dst[191:176] := e11 -dst[207:192] := e12 -dst[223:208] := e13 -dst[239:224] := e14 -dst[255:240] := e15 -dst[MAX:256] := 0 +FOR j := 0 to 15 + i := j*16 + IF k1[j] + k[j] := ( a[i+15:i] <= b[i+15:i] ) ? 1 : 0 + ELSE + k[j] := 0 + FI +ENDFOR +k[MAX:16] := 0 +
immintrin.h
 - + Integer - AVX - Set - - - - - - - - - Set packed 32-bit integers in "dst" with the supplied values. + Mask + AVX512VL + AVX512BW + Compare + + + + + Compare packed unsigned 16-bit integers in "a" and "b" for less-than, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). -dst[31:0] := e0 -dst[63:32] := e1 -dst[95:64] := e2 -dst[127:96] := e3 -dst[159:128] := e4 -dst[191:160] := e5 -dst[223:192] := e6 -dst[255:224] := e7 -dst[MAX:256] := 0 +FOR j := 0 to 15 + i := j*16 + IF k1[j] + k[j] := ( a[i+15:i] < b[i+15:i] ) ? 1 : 0 + ELSE + k[j] := 0 + FI +ENDFOR +k[MAX:16] := 0 +
immintrin.h
 - + Integer - AVX - Set - - - - - Set packed 64-bit integers in "dst" with the supplied values. + Mask + AVX512VL + AVX512BW + Compare + + + + + Compare packed unsigned 16-bit integers in "a" and "b" for not-equal, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). -dst[63:0] := e0 -dst[127:64] := e1 -dst[191:128] := e2 -dst[255:192] := e3 -dst[MAX:256] := 0 +FOR j := 0 to 15 + i := j*16 + IF k1[j] + k[j] := ( a[i+15:i] != b[i+15:i] ) ? 1 : 0 + ELSE + k[j] := 0 + FI +ENDFOR +k[MAX:16] := 0 +
immintrin.h
 - - Floating Point - AVX - Set - - - - - Set packed double-precision (64-bit) floating-point elements in "dst" with the supplied values in reverse order. - -dst[63:0] := e3 -dst[127:64] := e2 -dst[191:128] := e1 -dst[255:192] := e0 -dst[MAX:256] := 0 + + Integer + Mask + AVX512BW + Compare + + + + + Compare packed unsigned 16-bit integers in "a" and "b" based on the comparison operand specified by "imm8", and store the results in mask vector "k". + CASE (imm8[2:0]) OF +0: OP := _MM_CMPINT_EQ +1: OP := _MM_CMPINT_LT +2: OP := _MM_CMPINT_LE +3: OP := _MM_CMPINT_FALSE +4: OP := _MM_CMPINT_NE +5: OP := _MM_CMPINT_NLT +6: OP := _MM_CMPINT_NLE +7: OP := _MM_CMPINT_TRUE +ESAC +FOR j := 0 to 31 + i := j*16 + k[j] := ( a[i+15:i] OP b[i+15:i] ) ? 1 : 0 +ENDFOR +k[MAX:32] := 0 +
immintrin.h
 - - Floating Point - AVX - Set - - - - - - - - - Set packed single-precision (32-bit) floating-point elements in "dst" with the supplied values in reverse order. + + Integer + Mask + AVX512BW + Compare + + + + Compare packed unsigned 16-bit integers in "a" and "b" for equality, and store the results in mask vector "k". -dst[31:0] := e7 -dst[63:32] := e6 -dst[95:64] := e5 -dst[127:96] := e4 -dst[159:128] := e3 -dst[191:160] := e2 -dst[223:192] := e1 -dst[255:224] := e0 -dst[MAX:256] := 0 +FOR j := 0 to 31 + i := j*16 + k[j] := ( a[i+15:i] == b[i+15:i] ) ? 1 : 0 +ENDFOR +k[MAX:32] := 0 +
immintrin.h
 - + Integer - AVX - Set - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - Set packed 8-bit integers in "dst" with the supplied values in reverse order. + Mask + AVX512BW + Compare + + + + Compare packed unsigned 16-bit integers in "a" and "b" for greater-than-or-equal, and store the results in mask vector "k". -dst[7:0] := e31 -dst[15:8] := e30 -dst[23:16] := e29 -dst[31:24] := e28 -dst[39:32] := e27 -dst[47:40] := e26 -dst[55:48] := e25 -dst[63:56] := e24 -dst[71:64] := e23 -dst[79:72] := e22 -dst[87:80] := e21 -dst[95:88] := e20 -dst[103:96] := e19 -dst[111:104] := e18 -dst[119:112] := e17 -dst[127:120] := e16 -dst[135:128] := e15 -dst[143:136] := e14 -dst[151:144] := e13 -dst[159:152] := e12 -dst[167:160] := e11 -dst[175:168] := e10 -dst[183:176] := e9 -dst[191:184] := e8 -dst[199:192] := e7 -dst[207:200] := e6 -dst[215:208] := e5 -dst[223:216] := e4 -dst[231:224] := e3 -dst[239:232] := e2 -dst[247:240] := e1 -dst[255:248] := e0 -dst[MAX:256] := 0 +FOR j := 0 to 31 + i := j*16 + k[j] := ( a[i+15:i] >= b[i+15:i] ) ? 1 : 0 +ENDFOR +k[MAX:32] := 0 +
immintrin.h
 - + Integer - AVX - Set - - - - - - - - - - - - - - - - - Set packed 16-bit integers in "dst" with the supplied values in reverse order. + Mask + AVX512BW + Compare + + + + Compare packed unsigned 16-bit integers in "a" and "b" for greater-than, and store the results in mask vector "k". -dst[15:0] := e15 -dst[31:16] := e14 -dst[47:32] := e13 -dst[63:48] := e12 -dst[79:64] := e11 -dst[95:80] := e10 -dst[111:96] := e9 -dst[127:112] := e8 -dst[145:128] := e7 -dst[159:144] := e6 -dst[175:160] := e5 -dst[191:176] := e4 -dst[207:192] := e3 -dst[223:208] := e2 -dst[239:224] := e1 -dst[255:240] := e0 -dst[MAX:256] := 0 +FOR j := 0 to 31 + i := j*16 + k[j] := ( a[i+15:i] > b[i+15:i] ) ? 1 : 0 +ENDFOR +k[MAX:32] := 0 +
immintrin.h
 - + Integer - AVX - Set - - - - - - - - - Set packed 32-bit integers in "dst" with the supplied values in reverse order. + Mask + AVX512BW + Compare + + + + Compare packed unsigned 16-bit integers in "a" and "b" for less-than-or-equal, and store the results in mask vector "k". -dst[31:0] := e7 -dst[63:32] := e6 -dst[95:64] := e5 -dst[127:96] := e4 -dst[159:128] := e3 -dst[191:160] := e2 -dst[223:192] := e1 -dst[255:224] := e0 -dst[MAX:256] := 0 +FOR j := 0 to 31 + i := j*16 + k[j] := ( a[i+15:i] <= b[i+15:i] ) ? 1 : 0 +ENDFOR +k[MAX:32] := 0 +
immintrin.h
 - + Integer - AVX - Set - - - - - Set packed 64-bit integers in "dst" with the supplied values in reverse order. + Mask + AVX512BW + Compare + + + + Compare packed unsigned 16-bit integers in "a" and "b" for less-than, and store the results in mask vector "k". -dst[63:0] := e3 -dst[127:64] := e2 -dst[191:128] := e1 -dst[255:192] := e0 -dst[MAX:256] := 0 +FOR j := 0 to 31 + i := j*16 + k[j] := ( a[i+15:i] < b[i+15:i] ) ? 1 : 0 +ENDFOR +k[MAX:32] := 0 +
immintrin.h
 - - Floating Point - AVX - Set - - Broadcast double-precision (64-bit) floating-point value "a" to all elements of "dst". + + Integer + Mask + AVX512BW + Compare + + + + Compare packed unsigned 16-bit integers in "a" and "b" for not-equal, and store the results in mask vector "k". -FOR j := 0 to 3 - i := j*64 - dst[i+63:i] := a[63:0] +FOR j := 0 to 31 + i := j*16 + k[j] := ( a[i+15:i] != b[i+15:i] ) ? 1 : 0 ENDFOR -dst[MAX:256] := 0 +k[MAX:32] := 0 +
immintrin.h
 - - Floating Point - AVX - Set - - Broadcast single-precision (32-bit) floating-point value "a" to all elements of "dst". - -FOR j := 0 to 7 - i := j*32 - dst[i+31:i] := a[31:0] + + Integer + Mask + AVX512BW + Compare + + + + + + Compare packed unsigned 16-bit integers in "a" and "b" based on the comparison operand specified by "imm8", and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). + CASE (imm8[2:0]) OF +0: OP := _MM_CMPINT_EQ +1: OP := _MM_CMPINT_LT +2: OP := _MM_CMPINT_LE +3: OP := _MM_CMPINT_FALSE +4: OP := _MM_CMPINT_NE +5: OP := _MM_CMPINT_NLT +6: OP := _MM_CMPINT_NLE +7: OP := _MM_CMPINT_TRUE +ESAC +FOR j := 0 to 31 + i := j*16 + IF k1[j] + k[j] := ( a[i+15:i] OP b[i+15:i] ) ? 1 : 0 + ELSE + k[j] := 0 + FI ENDFOR -dst[MAX:256] := 0 +k[MAX:32] := 0 +
immintrin.h
 - + Integer - AVX - Set - - Broadcast 8-bit integer "a" to all elements of "dst". This intrinsic may generate the "vpbroadcastb". + Mask + AVX512BW + Compare + + + + + Compare packed unsigned 16-bit integers in "a" and "b" for equality, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). FOR j := 0 to 31 - i := j*8 - dst[i+7:i] := a[7:0] + i := j*16 + IF k1[j] + k[j] := ( a[i+15:i] == b[i+15:i] ) ? 1 : 0 + ELSE + k[j] := 0 + FI ENDFOR -dst[MAX:256] := 0 +k[MAX:32] := 0 +
immintrin.h
 - + Integer - AVX - Set - - Broadcast 16-bit integer "a" to all all elements of "dst". This intrinsic may generate the "vpbroadcastw". + Mask + AVX512BW + Compare + + + + + Compare packed unsigned 16-bit integers in "a" and "b" for greater-than-or-equal, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 15 +FOR j := 0 to 31 i := j*16 - dst[i+15:i] := a[15:0] + IF k1[j] + k[j] := ( a[i+15:i] >= b[i+15:i] ) ? 1 : 0 + ELSE + k[j] := 0 + FI ENDFOR -dst[MAX:256] := 0 +k[MAX:32] := 0 +
immintrin.h
 - + Integer - AVX - Set - - Broadcast 32-bit integer "a" to all elements of "dst". This intrinsic may generate the "vpbroadcastd". + Mask + AVX512BW + Compare + + + + + Compare packed unsigned 16-bit integers in "a" and "b" for greater-than, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := j*32 - dst[i+31:i] := a[31:0] +FOR j := 0 to 31 + i := j*16 + IF k1[j] + k[j] := ( a[i+15:i] > b[i+15:i] ) ? 1 : 0 + ELSE + k[j] := 0 + FI ENDFOR -dst[MAX:256] := 0 +k[MAX:32] := 0 +
immintrin.h
 - + Integer - AVX - Set - - Broadcast 64-bit integer "a" to all elements of "dst". This intrinsic may generate the "vpbroadcastq". + Mask + AVX512BW + Compare + + + + + Compare packed unsigned 16-bit integers in "a" and "b" for less-than-or-equal, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*64 - dst[i+63:i] := a[63:0] +FOR j := 0 to 31 + i := j*16 + IF k1[j] + k[j] := ( a[i+15:i] <= b[i+15:i] ) ? 1 : 0 + ELSE + k[j] := 0 + FI ENDFOR -dst[MAX:256] := 0 +k[MAX:32] := 0 +
immintrin.h
 - - Floating Point - AVX - Cast - - Cast vector of type __m256d to type __m256. - This intrinsic is only used for compilation and does not generate any instructions, thus it has zero latency. -
immintrin.h
- - - Floating Point - AVX - Cast - - Cast vector of type __m256 to type __m256d. - This intrinsic is only used for compilation and does not generate any instructions, thus it has zero latency. -
immintrin.h
- - - Floating Point + Integer - AVX - Cast - - Casts vector of type __m256 to type __m256i. This intrinsic is only used for compilation and does not generate any instructions, thus it has zero latency. - + Mask + AVX512BW + Compare + + + + + Compare packed unsigned 16-bit integers in "a" and "b" for less-than, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). + +FOR j := 0 to 31 + i := j*16 + IF k1[j] + k[j] := ( a[i+15:i] < b[i+15:i] ) ? 1 : 0 + ELSE + k[j] := 0 + FI +ENDFOR +k[MAX:32] := 0 + +
immintrin.h
 - - Floating Point + Integer - AVX - Cast - - Casts vector of type __m256d to type __m256i. This intrinsic is only used for compilation and does not generate any instructions, thus it has zero latency. - -
immintrin.h
- - - Floating Point - AVX - Cast - - Casts vector of type __m256i to type __m256. This intrinsic is only used for compilation and does not generate any instructions, thus it has zero latency. - -
immintrin.h
- - - Floating Point - AVX - Cast - - Casts vector of type __m256i to type __m256d. This intrinsic is only used for compilation and does not generate any instructions, thus it has zero latency. - -
immintrin.h
- - - Floating Point - AVX - Cast - - Casts vector of type __m256 to type __m128. This intrinsic is only used for compilation and does not generate any instructions, thus it has zero latency. - -
immintrin.h
- - - Floating Point - AVX - Cast - - Casts vector of type __m256d to type __m128d. This intrinsic is only used for compilation and does not generate any instructions, thus it has zero latency. - + Mask + AVX512BW + Compare + + + + + Compare packed unsigned 16-bit integers in "a" and "b" for not-equal, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). + +FOR j := 0 to 31 + i := j*16 + IF k1[j] + k[j] := ( a[i+15:i] != b[i+15:i] ) ? 1 : 0 + ELSE + k[j] := 0 + FI +ENDFOR +k[MAX:32] := 0 + +
immintrin.h
 - + Integer - AVX - Cast - - Casts vector of type __m256i to type __m128i. This intrinsic is only used for compilation and does not generate any instructions, thus it has zero latency. - -
immintrin.h
- - - Floating Point - AVX - Cast - - Casts vector of type __m128 to type __m256; the upper 128 bits of the result are undefined. This intrinsic is only used for compilation and does not generate any instructions, thus it has zero latency. - -
immintrin.h
- - - Floating Point - AVX - Cast - - Casts vector of type __m128d to type __m256d; the upper 128 bits of the result are undefined. This intrinsic is only used for compilation and does not generate any instructions, thus it has zero latency. - + Mask + AVX512VL + AVX512BW + Compare + + + + + Compare packed unsigned 16-bit integers in "a" and "b" based on the comparison operand specified by "imm8", and store the results in mask vector "k". + CASE (imm8[2:0]) OF +0: OP := _MM_CMPINT_EQ +1: OP := _MM_CMPINT_LT +2: OP := _MM_CMPINT_LE +3: OP := _MM_CMPINT_FALSE +4: OP := _MM_CMPINT_NE +5: OP := _MM_CMPINT_NLT +6: OP := _MM_CMPINT_NLE +7: OP := _MM_CMPINT_TRUE +ESAC +FOR j := 0 to 7 + i := j*16 + k[j] := ( a[i+15:i] OP b[i+15:i] ) ? 1 : 0 +ENDFOR +k[MAX:8] := 0 + +
immintrin.h
 - + Integer - AVX - Cast - - Casts vector of type __m128i to type __m256i; the upper 128 bits of the result are undefined. This intrinsic is only used for compilation and does not generate any instructions, thus it has zero latency. - -
immintrin.h
- - - Floating Point - AVX - Cast - - Casts vector of type __m128 to type __m256; the upper 128 bits of the result are zeroed. This intrinsic is only used for compilation and does not generate any instructions, thus it has zero latency. - -
immintrin.h
- - - Floating Point - AVX - Cast - - Casts vector of type __m128d to type __m256d; the upper 128 bits of the result are zeroed. This intrinsic is only used for compilation and does not generate any instructions, thus it has zero latency. - + Mask + AVX512VL + AVX512BW + Compare + + + + Compare packed unsigned 16-bit integers in "a" and "b" for equality, and store the results in mask vector "k". + +FOR j := 0 to 7 + i := j*16 + k[j] := ( a[i+15:i] == b[i+15:i] ) ? 1 : 0 +ENDFOR +k[MAX:8] := 0 + +
immintrin.h
 - + Integer - AVX - Cast - - Casts vector of type __m128i to type __m256i; the upper 128 bits of the result are zeroed. This intrinsic is only used for compilation and does not generate any instructions, thus it has zero latency. - -
immintrin.h
- - - Floating Point - AVX - Special Math Functions - - Round the packed single-precision (32-bit) floating-point elements in "a" down to an integer value, and store the results as packed single-precision floating-point elements in "dst". + Mask + AVX512VL + AVX512BW + Compare + + + + Compare packed unsigned 16-bit integers in "a" and "b" for greater-than-or-equal, and store the results in mask vector "k". FOR j := 0 to 7 - i := j*32 - dst[i+31:i] := FLOOR(a[i+31:i]) + i := j*16 + k[j] := ( a[i+15:i] >= b[i+15:i] ) ? 1 : 0 ENDFOR -dst[MAX:256] := 0 +k[MAX:8] := 0 - +
immintrin.h
 - - Floating Point - AVX - Special Math Functions - - Round the packed single-precision (32-bit) floating-point elements in "a" up to an integer value, and store the results as packed single-precision floating-point elements in "dst". + + Integer + Mask + AVX512VL + AVX512BW + Compare + + + + Compare packed unsigned 16-bit integers in "a" and "b" for greater-than, and store the results in mask vector "k". FOR j := 0 to 7 - i := j*32 - dst[i+31:i] := CEIL(a[i+31:i]) + i := j*16 + k[j] := ( a[i+15:i] > b[i+15:i] ) ? 1 : 0 ENDFOR -dst[MAX:256] := 0 +k[MAX:8] := 0 - +
immintrin.h
 - - Floating Point - AVX - Special Math Functions - - Round the packed double-precision (64-bit) floating-point elements in "a" down to an integer value, and store the results as packed double-precision floating-point elements in "dst". + + Integer + Mask + AVX512VL + AVX512BW + Compare + + + + Compare packed unsigned 16-bit integers in "a" and "b" for less-than-or-equal, and store the results in mask vector "k". -FOR j := 0 to 3 - i := j*64 - dst[i+63:i] := FLOOR(a[i+63:i]) +FOR j := 0 to 7 + i := j*16 + k[j] := ( a[i+15:i] <= b[i+15:i] ) ? 1 : 0 ENDFOR -dst[MAX:256] := 0 +k[MAX:8] := 0 - +
immintrin.h
 - - Floating Point - AVX - Special Math Functions - - Round the packed double-precision (64-bit) floating-point elements in "a" up to an integer value, and store the results as packed double-precision floating-point elements in "dst". + + Integer + Mask + AVX512VL + AVX512BW + Compare + + + + Compare packed unsigned 16-bit integers in "a" and "b" for less-than, and store the results in mask vector "k". -FOR j := 0 to 3 - i := j*64 - dst[i+63:i] := CEIL(a[i+63:i]) +FOR j := 0 to 7 + i := j*16 + k[j] := ( a[i+15:i] < b[i+15:i] ) ? 1 : 0 ENDFOR -dst[MAX:256] := 0 +k[MAX:8] := 0 - +
immintrin.h
 - - Floating Point - SSE - General Support - - Return vector of type __m128 with undefined elements. + + Integer + Mask + AVX512VL + AVX512BW + Compare + + + + Compare packed unsigned 16-bit integers in "a" and "b" for not-equal, and store the results in mask vector "k". + +FOR j := 0 to 7 + i := j*16 + k[j] := ( a[i+15:i] != b[i+15:i] ) ? 1 : 0 +ENDFOR +k[MAX:8] := 0 + +
immintrin.h
 - - Floating Point - SSE2 - General Support - - Return vector of type __m128d with undefined elements. + + Integer + Mask + AVX512VL + AVX512BW + Compare + + + + + + Compare packed unsigned 16-bit integers in "a" and "b" based on the comparison operand specified by "imm8", and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). + CASE (imm8[2:0]) OF +0: OP := _MM_CMPINT_EQ +1: OP := _MM_CMPINT_LT +2: OP := _MM_CMPINT_LE +3: OP := _MM_CMPINT_FALSE +4: OP := _MM_CMPINT_NE +5: OP := _MM_CMPINT_NLT +6: OP := _MM_CMPINT_NLE +7: OP := _MM_CMPINT_TRUE +ESAC +FOR j := 0 to 7 + i := j*16 + IF k1[j] + k[j] := ( a[i+15:i] OP b[i+15:i] ) ? 1 : 0 + ELSE + k[j] := 0 + FI +ENDFOR +k[MAX:8] := 0 + +
immintrin.h
 - + Integer - SSE2 - General Support - - Return vector of type __m128i with undefined elements. -
immintrin.h
- - - Floating Point - AVX - General Support - - Return vector of type __m256 with undefined elements. -
immintrin.h
- - - Floating Point - AVX - General Support - - Return vector of type __m256d with undefined elements. -
immintrin.h
- - - Integer - AVX - General Support - - Return vector of type __m256i with undefined elements. -
immintrin.h
- - - CLFLUSHOPT - General Support - - Invalidate and flush the cache line that contains "p" from all levels of the cache hierarchy. - -
immintrin.h
- - - CLWB - General Support - - Write back to memory the cache line that contains "p" from any level of the cache hierarchy in the cache coherence domain. - -
immintrin.h
- - - Floating Point - AVX - Set - - - Set packed __m256 vector "dst" with the supplied values. - -dst[127:0] := lo[127:0] -dst[255:128] := hi[127:0] -dst[MAX:256] := 0 - - -
immintrin.h
- - - Floating Point - AVX - Set - - - Set packed __m256d vector "dst" with the supplied values. + Mask + AVX512VL + AVX512BW + Compare + + + + + Compare packed unsigned 16-bit integers in "a" and "b" for equality, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). -dst[127:0] := lo[127:0] -dst[255:128] := hi[127:0] -dst[MAX:256] := 0 +FOR j := 0 to 7 + i := j*16 + IF k1[j] + k[j] := ( a[i+15:i] == b[i+15:i] ) ? 1 : 0 + ELSE + k[j] := 0 + FI +ENDFOR +k[MAX:8] := 0 - +
immintrin.h
 - + Integer - AVX - Set - - - Set packed __m256i vector "dst" with the supplied values. + Mask + AVX512VL + AVX512BW + Compare + + + + + Compare packed unsigned 16-bit integers in "a" and "b" for greater-than-or-equal, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). -dst[127:0] := lo[127:0] -dst[255:128] := hi[127:0] -dst[MAX:256] := 0 +FOR j := 0 to 7 + i := j*16 + IF k1[j] + k[j] := ( a[i+15:i] >= b[i+15:i] ) ? 1 : 0 + ELSE + k[j] := 0 + FI +ENDFOR +k[MAX:8] := 0 - +
immintrin.h
 - - Floating Point - AVX - Set - - - Set packed __m256 vector "dst" with the supplied values. + + Integer + Mask + AVX512VL + AVX512BW + Compare + + + + + Compare packed unsigned 16-bit integers in "a" and "b" for greater-than, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). -dst[127:0] := lo[127:0] -dst[255:128] := hi[127:0] -dst[MAX:256] := 0 +FOR j := 0 to 7 + i := j*16 + IF k1[j] + k[j] := ( a[i+15:i] > b[i+15:i] ) ? 1 : 0 + ELSE + k[j] := 0 + FI +ENDFOR +k[MAX:8] := 0 - +
immintrin.h
 - - Floating Point - AVX - Set - - - Set packed __m256d vector "dst" with the supplied values. + + Integer + Mask + AVX512VL + AVX512BW + Compare + + + + + Compare packed unsigned 16-bit integers in "a" and "b" for less-than-or-equal, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). -dst[127:0] := lo[127:0] -dst[255:128] := hi[127:0] -dst[MAX:256] := 0 +FOR j := 0 to 7 + i := j*16 + IF k1[j] + k[j] := ( a[i+15:i] <= b[i+15:i] ) ? 1 : 0 + ELSE + k[j] := 0 + FI +ENDFOR +k[MAX:8] := 0 - +
immintrin.h
 - + Integer - AVX - Set - - - Set packed __m256i vector "dst" with the supplied values. + Mask + AVX512VL + AVX512BW + Compare + + + + + Compare packed unsigned 16-bit integers in "a" and "b" for less-than, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). -dst[127:0] := lo[127:0] -dst[255:128] := hi[127:0] -dst[MAX:256] := 0 +FOR j := 0 to 7 + i := j*16 + IF k1[j] + k[j] := ( a[i+15:i] < b[i+15:i] ) ? 1 : 0 + ELSE + k[j] := 0 + FI +ENDFOR +k[MAX:8] := 0 - +
immintrin.h
 - - Floating Point - AVX - Load - - - Load two 128-bit values (composed of 4 packed single-precision (32-bit) floating-point elements) from memory, and combine them into a 256-bit value in "dst". - "hiaddr" and "loaddr" do not need to be aligned on any particular boundary. + + Integer + Mask + AVX512VL + AVX512BW + Compare + + + + + Compare packed unsigned 16-bit integers in "a" and "b" for not-equal, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). -dst[127:0] := MEM[loaddr+127:loaddr] -dst[255:128] := MEM[hiaddr+127:hiaddr] -dst[MAX:256] := 0 +FOR j := 0 to 7 + i := j*16 + IF k1[j] + k[j] := ( a[i+15:i] != b[i+15:i] ) ? 1 : 0 + ELSE + k[j] := 0 + FI +ENDFOR +k[MAX:8] := 0 +
immintrin.h
 - - Floating Point - AVX - Load - - - Load two 128-bit values (composed of 2 packed double-precision (64-bit) floating-point elements) from memory, and combine them into a 256-bit value in "dst". - "hiaddr" and "loaddr" do not need to be aligned on any particular boundary. - -dst[127:0] := MEM[loaddr+127:loaddr] -dst[255:128] := MEM[hiaddr+127:hiaddr] -dst[MAX:256] := 0 + + Integer + Mask + AVX512VL + AVX512BW + Compare + + + + + Compare packed signed 16-bit integers in "a" and "b" based on the comparison operand specified by "imm8", and store the results in mask vector "k". + CASE (imm8[2:0]) OF +0: OP := _MM_CMPINT_EQ +1: OP := _MM_CMPINT_LT +2: OP := _MM_CMPINT_LE +3: OP := _MM_CMPINT_FALSE +4: OP := _MM_CMPINT_NE +5: OP := _MM_CMPINT_NLT +6: OP := _MM_CMPINT_NLE +7: OP := _MM_CMPINT_TRUE +ESAC +FOR j := 0 to 15 + i := j*16 + k[j] := ( a[i+15:i] OP b[i+15:i] ) ? 1 : 0 +ENDFOR +k[MAX:16] := 0 +
immintrin.h
 - + Integer - AVX - Load - - - Load two 128-bit values (composed of integer data) from memory, and combine them into a 256-bit value in "dst". - "hiaddr" and "loaddr" do not need to be aligned on any particular boundary. + Mask + AVX512VL + AVX512BW + Compare + + + + Compare packed signed 16-bit integers in "a" and "b" for equality, and store the results in mask vector "k". -dst[127:0] := MEM[loaddr+127:loaddr] -dst[255:128] := MEM[hiaddr+127:hiaddr] -dst[MAX:256] := 0 +FOR j := 0 to 15 + i := j*16 + k[j] := ( a[i+15:i] == b[i+15:i] ) ? 1 : 0 +ENDFOR +k[MAX:16] := 0 +
immintrin.h
 - - Floating Point - AVX - Store - - - - Store the high and low 128-bit halves (each composed of 4 packed single-precision (32-bit) floating-point elements) from "a" into memory two different 128-bit locations. - "hiaddr" and "loaddr" do not need to be aligned on any particular boundary. + + Integer + Mask + AVX512VL + AVX512BW + Compare + + + + Compare packed signed 16-bit integers in "a" and "b" for greater-than-or-equal, and store the results in mask vector "k". -MEM[loaddr+127:loaddr] := a[127:0] -MEM[hiaddr+127:hiaddr] := a[255:128] +FOR j := 0 to 15 + i := j*16 + k[j] := ( a[i+15:i] >= b[i+15:i] ) ? 1 : 0 +ENDFOR +k[MAX:16] := 0 +
immintrin.h
 - - Floating Point - AVX - Store - - - - Store the high and low 128-bit halves (each composed of 2 packed double-precision (64-bit) floating-point elements) from "a" into memory two different 128-bit locations. - "hiaddr" and "loaddr" do not need to be aligned on any particular boundary. + + Integer + Mask + AVX512VL + AVX512BW + Compare + + + + Compare packed signed 16-bit integers in "a" and "b" for greater-than, and store the results in mask vector "k". -MEM[loaddr+127:loaddr] := a[127:0] -MEM[hiaddr+127:hiaddr] := a[255:128] +FOR j := 0 to 15 + i := j*16 + k[j] := ( a[i+15:i] > b[i+15:i] ) ? 1 : 0 +ENDFOR +k[MAX:16] := 0 +
immintrin.h
 - + Integer - AVX - Store - - - - Store the high and low 128-bit halves (each composed of integer data) from "a" into memory two different 128-bit locations. - "hiaddr" and "loaddr" do not need to be aligned on any particular boundary. + Mask + AVX512VL + AVX512BW + Compare + + + + Compare packed signed 16-bit integers in "a" and "b" for less-than-or-equal, and store the results in mask vector "k". -MEM[loaddr+127:loaddr] := a[127:0] -MEM[hiaddr+127:hiaddr] := a[255:128] +FOR j := 0 to 15 + i := j*16 + k[j] := ( a[i+15:i] <= b[i+15:i] ) ? 1 : 0 +ENDFOR +k[MAX:16] := 0 +
immintrin.h
 - + Integer - AVX2 - Special Math Functions - - Compute the absolute value of packed 8-bit integers in "a", and store the unsigned results in "dst". + Mask + AVX512VL + AVX512BW + Compare + + + + Compare packed signed 16-bit integers in "a" and "b" for less-than, and store the results in mask vector "k". -FOR j := 0 to 31 - i := j*8 - dst[i+7:i] := ABS(a[i+7:i]) +FOR j := 0 to 15 + i := j*16 + k[j] := ( a[i+15:i] < b[i+15:i] ) ? 1 : 0 ENDFOR -dst[MAX:256] := 0 +k[MAX:16] := 0 - +
immintrin.h
 - + Integer - AVX2 - Special Math Functions - - Compute the absolute value of packed 16-bit integers in "a", and store the unsigned results in "dst". + Mask + AVX512VL + AVX512BW + Compare + + + + Compare packed signed 16-bit integers in "a" and "b" for not-equal, and store the results in mask vector "k". FOR j := 0 to 15 i := j*16 - dst[i+15:i] := ABS(a[i+15:i]) + k[j] := ( a[i+15:i] != b[i+15:i] ) ? 1 : 0 ENDFOR -dst[MAX:256] := 0 +k[MAX:16] := 0 - +
immintrin.h
 - + Integer - AVX2 - Special Math Functions - - Compute the absolute value of packed 32-bit integers in "a", and store the unsigned results in "dst". - -FOR j := 0 to 7 - i := j*32 - dst[i+31:i] := ABS(a[i+31:i]) + Mask + AVX512VL + AVX512BW + Compare + + + + + + Compare packed signed 16-bit integers in "a" and "b" based on the comparison operand specified by "imm8", and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). + CASE (imm8[2:0]) OF +0: OP := _MM_CMPINT_EQ +1: OP := _MM_CMPINT_LT +2: OP := _MM_CMPINT_LE +3: OP := _MM_CMPINT_FALSE +4: OP := _MM_CMPINT_NE +5: OP := _MM_CMPINT_NLT +6: OP := _MM_CMPINT_NLE +7: OP := _MM_CMPINT_TRUE +ESAC +FOR j := 0 to 15 + i := j*16 + IF k1[j] + k[j] := ( a[i+15:i] OP b[i+15:i] ) ? 1 : 0 + ELSE + k[j] := 0 + FI ENDFOR -dst[MAX:256] := 0 +k[MAX:16] := 0 - +
immintrin.h
 - + Integer - AVX2 - Arithmetic - - - Add packed 8-bit integers in "a" and "b", and store the results in "dst". + Mask + AVX512VL + AVX512BW + Compare + + + + + Compare packed signed 16-bit integers in "a" and "b" for equality, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 31 - i := j*8 - dst[i+7:i] := a[i+7:i] + b[i+7:i] +FOR j := 0 to 15 + i := j*16 + IF k1[j] + k[j] := ( a[i+15:i] == b[i+15:i] ) ? 1 : 0 + ELSE + k[j] := 0 + FI ENDFOR -dst[MAX:256] := 0 +k[MAX:16] := 0 - +
immintrin.h
 - + Integer - AVX2 - Arithmetic - - - Add packed 16-bit integers in "a" and "b", and store the results in "dst". + Mask + AVX512VL + AVX512BW + Compare + + + + + Compare packed signed 16-bit integers in "a" and "b" for greater-than-or-equal, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). FOR j := 0 to 15 i := j*16 - dst[i+15:i] := a[i+15:i] + b[i+15:i] + IF k1[j] + k[j] := ( a[i+15:i] >= b[i+15:i] ) ? 1 : 0 + ELSE + k[j] := 0 + FI ENDFOR -dst[MAX:256] := 0 +k[MAX:16] := 0 - +
immintrin.h
 - + Integer - AVX2 - Arithmetic - - - Add packed 32-bit integers in "a" and "b", and store the results in "dst". + Mask + AVX512VL + AVX512BW + Compare + + + + + Compare packed signed 16-bit integers in "a" and "b" for greater-than, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := j*32 - dst[i+31:i] := a[i+31:i] + b[i+31:i] +FOR j := 0 to 15 + i := j*16 + IF k1[j] + k[j] := ( a[i+15:i] > b[i+15:i] ) ? 1 : 0 + ELSE + k[j] := 0 + FI ENDFOR -dst[MAX:256] := 0 +k[MAX:16] := 0 - +
immintrin.h
 - + Integer - AVX2 - Arithmetic - - - Add packed 64-bit integers in "a" and "b", and store the results in "dst". + Mask + AVX512VL + AVX512BW + Compare + + + + + Compare packed signed 16-bit integers in "a" and "b" for less-than-or-equal, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*64 - dst[i+63:i] := a[i+63:i] + b[i+63:i] +FOR j := 0 to 15 + i := j*16 + IF k1[j] + k[j] := ( a[i+15:i] <= b[i+15:i] ) ? 1 : 0 + ELSE + k[j] := 0 + FI ENDFOR -dst[MAX:256] := 0 +k[MAX:16] := 0 - +
immintrin.h
 - + Integer - AVX2 - Arithmetic - - - Add packed 8-bit integers in "a" and "b" using saturation, and store the results in "dst". + Mask + AVX512VL + AVX512BW + Compare + + + + + Compare packed signed 16-bit integers in "a" and "b" for less-than, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 31 - i := j*8 - dst[i+7:i] := Saturate_To_Int8( a[i+7:i] + b[i+7:i] ) +FOR j := 0 to 15 + i := j*16 + IF k1[j] + k[j] := ( a[i+15:i] < b[i+15:i] ) ? 1 : 0 + ELSE + k[j] := 0 + FI ENDFOR -dst[MAX:256] := 0 +k[MAX:16] := 0 - +
immintrin.h
 - + Integer - AVX2 - Arithmetic - - - Add packed 16-bit integers in "a" and "b" using saturation, and store the results in "dst". + Mask + AVX512VL + AVX512BW + Compare + + + + + Compare packed signed 16-bit integers in "a" and "b" for not-equal, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). FOR j := 0 to 15 i := j*16 - dst[i+15:i] := Saturate_To_Int16( a[i+15:i] + b[i+15:i] ) + IF k1[j] + k[j] := ( a[i+15:i] != b[i+15:i] ) ? 1 : 0 + ELSE + k[j] := 0 + FI ENDFOR -dst[MAX:256] := 0 +k[MAX:16] := 0 - +
immintrin.h
 - + Integer - AVX2 - Arithmetic - - - Add packed unsigned 8-bit integers in "a" and "b" using saturation, and store the results in "dst". - + Mask + AVX512BW + Compare + + + + + Compare packed signed 16-bit integers in "a" and "b" based on the comparison operand specified by "imm8", and store the results in mask vector "k". + CASE (imm8[2:0]) OF +0: OP := _MM_CMPINT_EQ +1: OP := _MM_CMPINT_LT +2: OP := _MM_CMPINT_LE +3: OP := _MM_CMPINT_FALSE +4: OP := _MM_CMPINT_NE +5: OP := _MM_CMPINT_NLT +6: OP := _MM_CMPINT_NLE +7: OP := _MM_CMPINT_TRUE +ESAC FOR j := 0 to 31 - i := j*8 - dst[i+7:i] := Saturate_To_UnsignedInt8( a[i+7:i] + b[i+7:i] ) + i := j*16 + k[j] := ( a[i+15:i] OP b[i+15:i] ) ? 1 : 0 ENDFOR -dst[MAX:256] := 0 +k[MAX:32] := 0 - +
immintrin.h
 - + Integer - AVX2 - Arithmetic - - - Add packed unsigned 16-bit integers in "a" and "b" using saturation, and store the results in "dst". + Mask + AVX512BW + Compare + + + + Compare packed signed 16-bit integers in "a" and "b" for equality, and store the results in mask vector "k". -FOR j := 0 to 15 +FOR j := 0 to 31 i := j*16 - dst[i+15:i] := Saturate_To_UnsignedInt16( a[i+15:i] + b[i+15:i] ) + k[j] := ( a[i+15:i] == b[i+15:i] ) ? 1 : 0 ENDFOR -dst[MAX:256] := 0 +k[MAX:32] := 0 - +
immintrin.h
 - + Integer - AVX2 - Miscellaneous - - - - Concatenate pairs of 16-byte blocks in "a" and "b" into a 32-byte temporary result, shift the result right by "count" bytes, and store the low 16 bytes in "dst". + Mask + AVX512BW + Compare + + + + Compare packed signed 16-bit integers in "a" and "b" for greater-than-or-equal, and store the results in mask vector "k". -FOR j := 0 to 1 - i := j*128 - tmp[255:0] := ((a[i+127:i] << 128) OR b[i+127:i]) >> (count[7:0]*8) - dst[i+127:i] := tmp[127:0] +FOR j := 0 to 31 + i := j*16 + k[j] := ( a[i+15:i] >= b[i+15:i] ) ? 1 : 0 ENDFOR -dst[MAX:256] := 0 +k[MAX:32] := 0 - +
immintrin.h
 - + Integer - AVX2 - Logical - - - Compute the bitwise AND of 256 bits (representing integer data) in "a" and "b", and store the result in "dst". + Mask + AVX512BW + Compare + + + + Compare packed signed 16-bit integers in "a" and "b" for greater-than, and store the results in mask vector "k". -dst[255:0] := (a[255:0] AND b[255:0]) -dst[MAX:256] := 0 +FOR j := 0 to 31 + i := j*16 + k[j] := ( a[i+15:i] > b[i+15:i] ) ? 1 : 0 +ENDFOR +k[MAX:32] := 0 - +
immintrin.h
 - + Integer - AVX2 - Logical - - - Compute the bitwise NOT of 256 bits (representing integer data) in "a" and then AND with "b", and store the result in "dst". + Mask + AVX512BW + Compare + + + + Compare packed signed 16-bit integers in "a" and "b" for less-than-or-equal, and store the results in mask vector "k". -dst[255:0] := ((NOT a[255:0]) AND b[255:0]) -dst[MAX:256] := 0 +FOR j := 0 to 31 + i := j*16 + k[j] := ( a[i+15:i] <= b[i+15:i] ) ? 1 : 0 +ENDFOR +k[MAX:32] := 0 - +
immintrin.h
 - + Integer - AVX2 - Probability/Statistics - - - Average packed unsigned 8-bit integers in "a" and "b", and store the results in "dst". + Mask + AVX512BW + Compare + + + + Compare packed signed 16-bit integers in "a" and "b" for less-than, and store the results in mask vector "k". FOR j := 0 to 31 - i := j*8 - dst[i+7:i] := (a[i+7:i] + b[i+7:i] + 1) >> 1 + i := j*16 + k[j] := ( a[i+15:i] < b[i+15:i] ) ? 1 : 0 ENDFOR -dst[MAX:256] := 0 +k[MAX:32] := 0 - +
immintrin.h
 - + Integer - AVX2 - Probability/Statistics - - - Average packed unsigned 16-bit integers in "a" and "b", and store the results in "dst". + Mask + AVX512BW + Compare + + + + Compare packed signed 16-bit integers in "a" and "b" for not-equal, and store the results in mask vector "k". -FOR j := 0 to 15 +FOR j := 0 to 31 i := j*16 - dst[i+15:i] := (a[i+15:i] + b[i+15:i] + 1) >> 1 + k[j] := ( a[i+15:i] != b[i+15:i] ) ? 1 : 0 ENDFOR -dst[MAX:256] := 0 +k[MAX:32] := 0 - +
immintrin.h
 - + Integer - AVX2 - Swizzle - - - - Blend packed 16-bit integers from "a" and "b" using control mask "imm8", and store the results in "dst". - -FOR j := 0 to 15 + Mask + AVX512BW + Compare + + + + + + Compare packed signed 16-bit integers in "a" and "b" based on the comparison operand specified by "imm8", and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). + CASE (imm8[2:0]) OF +0: OP := _MM_CMPINT_EQ +1: OP := _MM_CMPINT_LT +2: OP := _MM_CMPINT_LE +3: OP := _MM_CMPINT_FALSE +4: OP := _MM_CMPINT_NE +5: OP := _MM_CMPINT_NLT +6: OP := _MM_CMPINT_NLE +7: OP := _MM_CMPINT_TRUE +ESAC +FOR j := 0 to 31 i := j*16 - IF imm8[j%8] - dst[i+15:i] := b[i+15:i] - ELSE - dst[i+15:i] := a[i+15:i] + IF k1[j] + k[j] := ( a[i+15:i] OP b[i+15:i] ) ? 1 : 0 + ELSE + k[j] := 0 FI ENDFOR -dst[MAX:256] := 0 +k[MAX:32] := 0 - +
immintrin.h
 - + Integer - AVX2 - Swizzle - - - - Blend packed 32-bit integers from "a" and "b" using control mask "imm8", and store the results in "dst". + Mask + AVX512BW + Compare + + + + + Compare packed signed 16-bit integers in "a" and "b" for equality, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*32 - IF imm8[j%8] - dst[i+31:i] := b[i+31:i] - ELSE - dst[i+31:i] := a[i+31:i] +FOR j := 0 to 31 + i := j*16 + IF k1[j] + k[j] := ( a[i+15:i] == b[i+15:i] ) ? 1 : 0 + ELSE + k[j] := 0 FI ENDFOR -dst[MAX:128] := 0 +k[MAX:32] := 0 - +
immintrin.h
 - + Integer - AVX2 - Swizzle - - - - Blend packed 32-bit integers from "a" and "b" using control mask "imm8", and store the results in "dst". + Mask + AVX512BW + Compare + + + + + Compare packed signed 16-bit integers in "a" and "b" for greater-than-or-equal, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := j*32 - IF imm8[j%8] - dst[i+31:i] := b[i+31:i] - ELSE - dst[i+31:i] := a[i+31:i] +FOR j := 0 to 31 + i := j*16 + IF k1[j] + k[j] := ( a[i+15:i] >= b[i+15:i] ) ? 1 : 0 + ELSE + k[j] := 0 FI ENDFOR -dst[MAX:256] := 0 +k[MAX:32] := 0 - +
immintrin.h
 - + Integer - AVX2 - Swizzle - - - - Blend packed 8-bit integers from "a" and "b" using "mask", and store the results in "dst". + Mask + AVX512BW + Compare + + + + + Compare packed signed 16-bit integers in "a" and "b" for greater-than, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). FOR j := 0 to 31 - i := j*8 - IF mask[i+7] - dst[i+7:i] := b[i+7:i] - ELSE - dst[i+7:i] := a[i+7:i] + i := j*16 + IF k1[j] + k[j] := ( a[i+15:i] > b[i+15:i] ) ? 1 : 0 + ELSE + k[j] := 0 FI ENDFOR -dst[MAX:256] := 0 +k[MAX:32] := 0 - +
immintrin.h
 - + Integer - AVX2 - Swizzle - - Broadcast the low packed 8-bit integer from "a" to all elements of "dst". + Mask + AVX512BW + Compare + + + + + Compare packed signed 16-bit integers in "a" and "b" for less-than-or-equal, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*8 - dst[i+7:i] := a[7:0] +FOR j := 0 to 31 + i := j*16 + IF k1[j] + k[j] := ( a[i+15:i] <= b[i+15:i] ) ? 1 : 0 + ELSE + k[j] := 0 + FI ENDFOR -dst[MAX:128] := 0 +k[MAX:32] := 0 - +
immintrin.h
 - + Integer - AVX2 - Swizzle - - Broadcast the low packed 8-bit integer from "a" to all elements of "dst". + Mask + AVX512BW + Compare + + + + + Compare packed signed 16-bit integers in "a" and "b" for less-than, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). FOR j := 0 to 31 - i := j*8 - dst[i+7:i] := a[7:0] + i := j*16 + IF k1[j] + k[j] := ( a[i+15:i] < b[i+15:i] ) ? 1 : 0 + ELSE + k[j] := 0 + FI ENDFOR -dst[MAX:256] := 0 +k[MAX:32] := 0 - +
immintrin.h
 - + Integer - AVX2 - Swizzle - - Broadcast the low packed 32-bit integer from "a" to all elements of "dst". + Mask + AVX512BW + Compare + + + + + Compare packed signed 16-bit integers in "a" and "b" for not-equal, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*32 - dst[i+31:i] := a[31:0] +FOR j := 0 to 31 + i := j*16 + IF k1[j] + k[j] := ( a[i+15:i] != b[i+15:i] ) ? 1 : 0 + ELSE + k[j] := 0 + FI ENDFOR -dst[MAX:128] := 0 +k[MAX:32] := 0 - +
immintrin.h
 - + Integer - AVX2 - Swizzle - - Broadcast the low packed 32-bit integer from "a" to all elements of "dst". - + Mask + AVX512VL + AVX512BW + Compare + + + + + Compare packed signed 16-bit integers in "a" and "b" based on the comparison operand specified by "imm8", and store the results in mask vector "k". + CASE (imm8[2:0]) OF +0: OP := _MM_CMPINT_EQ +1: OP := _MM_CMPINT_LT +2: OP := _MM_CMPINT_LE +3: OP := _MM_CMPINT_FALSE +4: OP := _MM_CMPINT_NE +5: OP := _MM_CMPINT_NLT +6: OP := _MM_CMPINT_NLE +7: OP := _MM_CMPINT_TRUE +ESAC FOR j := 0 to 7 - i := j*32 - dst[i+31:i] := a[31:0] + i := j*16 + k[j] := ( a[i+15:i] OP b[i+15:i] ) ? 1 : 0 ENDFOR -dst[MAX:256] := 0 +k[MAX:8] := 0 - +
immintrin.h
 - + Integer - AVX2 - Swizzle - - Broadcast the low packed 64-bit integer from "a" to all elements of "dst". + Mask + AVX512VL + AVX512BW + Compare + + + + Compare packed signed 16-bit integers in "a" and "b" for equality, and store the results in mask vector "k". -FOR j := 0 to 1 - i := j*64 - dst[i+63:i] := a[63:0] +FOR j := 0 to 7 + i := j*16 + k[j] := ( a[i+15:i] == b[i+15:i] ) ? 1 : 0 ENDFOR -dst[MAX:128] := 0 +k[MAX:8] := 0 - +
immintrin.h
 - + Integer - AVX2 - Swizzle - - Broadcast the low packed 64-bit integer from "a" to all elements of "dst". + Mask + AVX512VL + AVX512BW + Compare + + + + Compare packed signed 16-bit integers in "a" and "b" for greater-than-or-equal, and store the results in mask vector "k". -FOR j := 0 to 3 - i := j*64 - dst[i+63:i] := a[63:0] +FOR j := 0 to 7 + i := j*16 + k[j] := ( a[i+15:i] >= b[i+15:i] ) ? 1 : 0 ENDFOR -dst[MAX:256] := 0 +k[MAX:8] := 0 - +
immintrin.h
 - - Floating Point - AVX2 - Swizzle - - Broadcast the low double-precision (64-bit) floating-point element from "a" to all elements of "dst". + + Integer + Mask + AVX512VL + AVX512BW + Compare + + + + Compare packed signed 16-bit integers in "a" and "b" for greater-than, and store the results in mask vector "k". -FOR j := 0 to 1 - i := j*64 - dst[i+63:i] := a[63:0] +FOR j := 0 to 7 + i := j*16 + k[j] := ( a[i+15:i] > b[i+15:i] ) ? 1 : 0 ENDFOR -dst[MAX:128] := 0 +k[MAX:8] := 0 - +
immintrin.h
 - - Floating Point - AVX2 - Swizzle - - Broadcast the low double-precision (64-bit) floating-point element from "a" to all elements of "dst". + + Integer + Mask + AVX512VL + AVX512BW + Compare + + + + Compare packed signed 16-bit integers in "a" and "b" for less-than-or-equal, and store the results in mask vector "k". -FOR j := 0 to 3 - i := j*64 - dst[i+63:i] := a[63:0] +FOR j := 0 to 7 + i := j*16 + k[j] := ( a[i+15:i] <= b[i+15:i] ) ? 1 : 0 ENDFOR -dst[MAX:256] := 0 +k[MAX:8] := 0 - +
immintrin.h
 - + Integer - AVX2 - Swizzle - - Broadcast 128 bits of integer data from "a" to all 128-bit lanes in "dst". - + Mask + AVX512VL + AVX512BW + Compare + + + + Compare packed signed 16-bit integers in "a" and "b" for less-than, and store the results in mask vector "k". -dst[127:0] := a[127:0] -dst[255:128] := a[127:0] -dst[MAX:256] := 0 +FOR j := 0 to 7 + i := j*16 + k[j] := ( a[i+15:i] < b[i+15:i] ) ? 1 : 0 +ENDFOR +k[MAX:8] := 0 - +
immintrin.h
 - + Integer - AVX2 - Swizzle - - Broadcast 128 bits of integer data from "a" to all 128-bit lanes in "dst". - + Mask + AVX512VL + AVX512BW + Compare + + + + Compare packed signed 16-bit integers in "a" and "b" for not-equal, and store the results in mask vector "k". -dst[127:0] := a[127:0] -dst[255:128] := a[127:0] -dst[MAX:256] := 0 +FOR j := 0 to 7 + i := j*16 + k[j] := ( a[i+15:i] != b[i+15:i] ) ? 1 : 0 +ENDFOR +k[MAX:8] := 0 - +
immintrin.h
 - - Floating Point - AVX2 - Swizzle - - Broadcast the low single-precision (32-bit) floating-point element from "a" to all elements of "dst". - -FOR j := 0 to 3 - i := j*32 - dst[i+31:i] := a[31:0] + + Integer + Mask + AVX512VL + AVX512BW + Compare + + + + + + Compare packed signed 16-bit integers in "a" and "b" based on the comparison operand specified by "imm8", and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). + CASE (imm8[2:0]) OF +0: OP := _MM_CMPINT_EQ +1: OP := _MM_CMPINT_LT +2: OP := _MM_CMPINT_LE +3: OP := _MM_CMPINT_FALSE +4: OP := _MM_CMPINT_NE +5: OP := _MM_CMPINT_NLT +6: OP := _MM_CMPINT_NLE +7: OP := _MM_CMPINT_TRUE +ESAC +FOR j := 0 to 7 + i := j*16 + IF k1[j] + k[j] := ( a[i+15:i] OP b[i+15:i] ) ? 1 : 0 + ELSE + k[j] := 0 + FI ENDFOR -dst[MAX:128] := 0 +k[MAX:8] := 0 - +
immintrin.h
 - - Floating Point - AVX2 - Swizzle - - Broadcast the low single-precision (32-bit) floating-point element from "a" to all elements of "dst". + + Integer + Mask + AVX512VL + AVX512BW + Compare + + + + + Compare packed signed 16-bit integers in "a" and "b" for equality, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). FOR j := 0 to 7 - i := j*32 - dst[i+31:i] := a[31:0] + i := j*16 + IF k1[j] + k[j] := ( a[i+15:i] == b[i+15:i] ) ? 1 : 0 + ELSE + k[j] := 0 + FI ENDFOR -dst[MAX:256] := 0 +k[MAX:8] := 0 - +
immintrin.h
 - + Integer - AVX2 - Swizzle - - Broadcast the low packed 16-bit integer from "a" to all elements of "dst". + Mask + AVX512VL + AVX512BW + Compare + + + + + Compare packed signed 16-bit integers in "a" and "b" for greater-than-or-equal, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). FOR j := 0 to 7 i := j*16 - dst[i+15:i] := a[15:0] + IF k1[j] + k[j] := ( a[i+15:i] >= b[i+15:i] ) ? 1 : 0 + ELSE + k[j] := 0 + FI ENDFOR -dst[MAX:128] := 0 +k[MAX:8] := 0 - +
immintrin.h
 - + Integer - AVX2 - Swizzle - - Broadcast the low packed 16-bit integer from "a" to all elements of "dst". + Mask + AVX512VL + AVX512BW + Compare + + + + + Compare packed signed 16-bit integers in "a" and "b" for greater-than, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 15 +FOR j := 0 to 7 i := j*16 - dst[i+15:i] := a[15:0] + IF k1[j] + k[j] := ( a[i+15:i] > b[i+15:i] ) ? 1 : 0 + ELSE + k[j] := 0 + FI ENDFOR -dst[MAX:256] := 0 +k[MAX:8] := 0 - +
immintrin.h
 - + Integer - AVX2 + Mask + AVX512VL + AVX512BW Compare - - - Compare packed 8-bit integers in "a" and "b" for equality, and store the results in "dst". + + + + + Compare packed signed 16-bit integers in "a" and "b" for less-than-or-equal, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 31 - i := j*8 - dst[i+7:i] := ( a[i+7:i] == b[i+7:i] ) ? 0xFF : 0 +FOR j := 0 to 7 + i := j*16 + IF k1[j] + k[j] := ( a[i+15:i] <= b[i+15:i] ) ? 1 : 0 + ELSE + k[j] := 0 + FI ENDFOR -dst[MAX:256] := 0 +k[MAX:8] := 0 - +
immintrin.h
 - + Integer - AVX2 + Mask + AVX512VL + AVX512BW Compare - - - Compare packed 16-bit integers in "a" and "b" for equality, and store the results in "dst". + + + + + Compare packed signed 16-bit integers in "a" and "b" for less-than, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 15 +FOR j := 0 to 7 i := j*16 - dst[i+15:i] := ( a[i+15:i] == b[i+15:i] ) ? 0xFFFF : 0 + IF k1[j] + k[j] := ( a[i+15:i] < b[i+15:i] ) ? 1 : 0 + ELSE + k[j] := 0 + FI ENDFOR -dst[MAX:256] := 0 +k[MAX:8] := 0 - +
immintrin.h
 - + Integer - AVX2 + Mask + AVX512VL + AVX512BW Compare - - - Compare packed 32-bit integers in "a" and "b" for equality, and store the results in "dst". + + + + + Compare packed signed 16-bit integers in "a" and "b" for not-equal, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). FOR j := 0 to 7 - i := j*32 - dst[i+31:i] := ( a[i+31:i] == b[i+31:i] ) ? 0xFFFFFFFF : 0 + i := j*16 + IF k1[j] + k[j] := ( a[i+15:i] != b[i+15:i] ) ? 1 : 0 + ELSE + k[j] := 0 + FI ENDFOR -dst[MAX:256] := 0 +k[MAX:8] := 0 - +
immintrin.h
 - + Integer - AVX2 - Compare - - - Compare packed 64-bit integers in "a" and "b" for equality, and store the results in "dst". + AVX512VL + AVX512BW + Miscellaneous + + + + + + Shuffle 16-bit integers in "a" and "b" across lanes using the corresponding selector and index in "idx", and store the results in "dst" using writemask "k" (elements are copied from "idx" when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*64 - dst[i+63:i] := ( a[i+63:i] == b[i+63:i] ) ? 0xFFFFFFFFFFFFFFFF : 0 +FOR j := 0 to 15 + i := j*16 + IF k[j] + off := 16*idx[i+3:i] + dst[i+15:i] := idx[i+4] ? b[off+15:off] : a[off+15:off] + ELSE + dst[i+15:i] := idx[i+15:i] + FI ENDFOR dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - AVX2 - Compare - - - Compare packed 8-bit integers in "a" and "b" for greater-than, and store the results in "dst". + AVX512VL + AVX512BW + Miscellaneous + + + + + + Shuffle 16-bit integers in "a" and "b" across lanes using the corresponding selector and index in "idx", and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). -FOR j := 0 to 31 - i := j*8 - dst[i+7:i] := ( a[i+7:i] > b[i+7:i] ) ? 0xFF : 0 +FOR j := 0 to 15 + i := j*16 + IF k[j] + off := 16*idx[i+3:i] + dst[i+15:i] := idx[i+4] ? b[off+15:off] : a[off+15:off] + ELSE + dst[i+15:i] := a[i+15:i] + FI ENDFOR dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - AVX2 - Compare - - - Compare packed 16-bit integers in "a" and "b" for greater-than, and store the results in "dst". + AVX512VL + AVX512BW + Miscellaneous + + + + + + Shuffle 16-bit integers in "a" and "b" across lanes using the corresponding selector and index in "idx", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). FOR j := 0 to 15 i := j*16 - dst[i+15:i] := ( a[i+15:i] > b[i+15:i] ) ? 0xFFFF : 0 + IF k[j] + off := 16*idx[i+3:i] + dst[i+15:i] := idx[i+4] ? b[off+15:off] : a[off+15:off] + ELSE + dst[i+15:i] := 0 + FI ENDFOR dst[MAX:256] := 0 - + +
immintrin.h
 - + Integer - AVX2 - Compare - - - Compare packed 32-bit integers in "a" and "b" for greater-than, and store the results in "dst". + AVX512VL + AVX512BW + Miscellaneous + + + + + Shuffle 16-bit integers in "a" and "b" across lanes using the corresponding selector and index in "idx", and store the results in "dst". -FOR j := 0 to 7 - i := j*32 - dst[i+31:i] := ( a[i+31:i] > b[i+31:i] ) ? 0xFFFFFFFF : 0 +FOR j := 0 to 15 + i := j*16 + off := 16*idx[i+3:i] + dst[i+15:i] := idx[i+4] ? b[off+15:off] : a[off+15:off] ENDFOR dst[MAX:256] := 0 - + +
immintrin.h
 - + Integer - AVX2 - Compare - - - Compare packed 64-bit integers in "a" and "b" for greater-than, and store the results in "dst". + AVX512BW + Miscellaneous + + + + + + Shuffle 16-bit integers in "a" and "b" across lanes using the corresponding selector and index in "idx", and store the results in "dst" using writemask "k" (elements are copied from "idx" when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*64 - dst[i+63:i] := ( a[i+63:i] > b[i+63:i] ) ? 0xFFFFFFFFFFFFFFFF : 0 +FOR j := 0 to 31 + i := j*16 + IF k[j] + off := 16*idx[i+4:i] + dst[i+15:i] := idx[i+5] ? b[off+15:off] : a[off+15:off] + ELSE + dst[i+15:i] := idx[i+15:i] + FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX2 - Convert - - Sign extend packed 16-bit integers in "a" to packed 32-bit integers, and store the results in "dst". + AVX512BW + Miscellaneous + + + + + + Shuffle 16-bit integers in "a" and "b" across lanes using the corresponding selector and index in "idx", and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). -FOR j:= 0 to 7 - i := 32*j - k := 16*j - dst[i+31:i] := SignExtend(a[k+15:k]) +FOR j := 0 to 31 + i := j*16 + IF k[j] + off := 16*idx[i+4:i] + dst[i+15:i] := idx[i+5] ? b[off+15:off] : a[off+15:off] + ELSE + dst[i+15:i] := a[i+15:i] + FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX2 - Convert - - Sign extend packed 16-bit integers in "a" to packed 64-bit integers, and store the results in "dst". + AVX512BW + Miscellaneous + + + + + + Shuffle 16-bit integers in "a" and "b" across lanes using the corresponding selector and index in "idx", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j:= 0 to 3 - i := 64*j - k := 16*j - dst[i+63:i] := SignExtend(a[k+15:k]) +FOR j := 0 to 31 + i := j*16 + IF k[j] + off := 16*idx[i+4:i] + dst[i+15:i] := idx[i+5] ? b[off+15:off] : a[off+15:off] + ELSE + dst[i+15:i] := 0 + FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - + +
immintrin.h
 - + Integer - AVX2 - Convert - - Sign extend packed 32-bit integers in "a" to packed 64-bit integers, and store the results in "dst". + AVX512BW + Miscellaneous + + + + + Shuffle 16-bit integers in "a" and "b" across lanes using the corresponding selector and index in "idx", and store the results in "dst". -FOR j:= 0 to 3 - i := 64*j - k := 32*j - dst[i+63:i] := SignExtend(a[k+31:k]) +FOR j := 0 to 31 + i := j*16 + off := 16*idx[i+4:i] + dst[i+15:i] := idx[i+5] ? b[off+15:off] : a[off+15:off] ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - + +
immintrin.h
 - + Integer - AVX2 - Convert - - Sign extend packed 8-bit integers in "a" to packed 16-bit integers, and store the results in "dst". - -FOR j := 0 to 15 - i := j*8 - l := j*16 - dst[l+15:l] := SignExtend(a[i+7:i]) + AVX512VL + AVX512BW + Miscellaneous + + + + + + Shuffle 16-bit integers in "a" and "b" using the corresponding selector and index in "idx", and store the results in "dst" using writemask "k" (elements are copied from "idx" when the corresponding mask bit is not set). + +FOR j := 0 to 7 + i := j*16 + IF k[j] + off := 16*idx[i+2:i] + dst[i+15:i] := idx[i+3] ? b[off+15:off] : a[off+15:off] + ELSE + dst[i+15:i] := idx[i+15:i] + FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - AVX2 - Convert - - Sign extend packed 8-bit integers in "a" to packed 32-bit integers, and store the results in "dst". + AVX512VL + AVX512BW + Miscellaneous + + + + + + Shuffle 16-bit integers in "a" and "b" using the corresponding selector and index in "idx", and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). FOR j := 0 to 7 - i := 32*j - k := 8*j - dst[i+31:i] := SignExtend(a[k+7:k]) + i := j*16 + IF k[j] + off := 16*idx[i+2:i] + dst[i+15:i] := idx[i+3] ? b[off+15:off] : a[off+15:off] + ELSE + dst[i+15:i] := a[i+15:i] + FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - AVX2 - Convert - - Sign extend packed 8-bit integers in the low 8 bytes of "a" to packed 64-bit integers, and store the results in "dst". + AVX512VL + AVX512BW + Miscellaneous + + + + + + Shuffle 16-bit integers in "a" and "b" using the corresponding selector and index in "idx", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := 64*j - k := 8*j - dst[i+63:i] := SignExtend(a[k+7:k]) +FOR j := 0 to 7 + i := j*16 + IF k[j] + off := 16*idx[i+2:i] + dst[i+15:i] := idx[i+3] ? b[off+15:off] : a[off+15:off] + ELSE + dst[i+15:i] := 0 + FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:128] := 0 - + +
immintrin.h
 - + Integer - AVX2 - Convert - - Zero extend packed unsigned 16-bit integers in "a" to packed 32-bit integers, and store the results in "dst". + AVX512VL + AVX512BW + Miscellaneous + + + + + Shuffle 16-bit integers in "a" and "b" using the corresponding selector and index in "idx", and store the results in "dst". FOR j := 0 to 7 - i := 32*j - k := 16*j - dst[i+31:i] := ZeroExtend(a[k+15:k]) + i := j*16 + off := 16*idx[i+2:i] + dst[i+15:i] := idx[i+3] ? b[off+15:off] : a[off+15:off] ENDFOR -dst[MAX:256] := 0 +dst[MAX:128] := 0 - + +
immintrin.h
 - + Integer - AVX2 - Convert - - Zero extend packed unsigned 16-bit integers in "a" to packed 64-bit integers, and store the results in "dst". + AVX512VL + AVX512BW + Miscellaneous + + + + + + Shuffle 16-bit integers in "a" across lanes using the corresponding index in "idx", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j:= 0 to 3 - i := 64*j - k := 16*j - dst[i+63:i] := ZeroExtend(a[k+15:k]) +FOR j := 0 to 15 + i := j*16 + id := idx[i+3:i]*16 + IF k[j] + dst[i+15:i] := a[id+15:id] + ELSE + dst[i+15:i] := src[i+15:i] + FI ENDFOR dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - AVX2 - Convert - - Zero extend packed unsigned 32-bit integers in "a" to packed 64-bit integers, and store the results in "dst". + AVX512VL + AVX512BW + Miscellaneous + + + + + Shuffle 16-bit integers in "a" across lanes using the corresponding index in "idx", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j:= 0 to 3 - i := 64*j - k := 32*j - dst[i+63:i] := ZeroExtend(a[k+31:k]) +FOR j := 0 to 15 + i := j*16 + id := idx[i+3:i]*16 + IF k[j] + dst[i+15:i] := a[id+15:id] + ELSE + dst[i+15:i] := 0 + FI ENDFOR dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - AVX2 - Convert - - Zero extend packed unsigned 8-bit integers in "a" to packed 16-bit integers, and store the results in "dst". + AVX512VL + AVX512BW + Miscellaneous + + + + Shuffle 16-bit integers in "a" across lanes using the corresponding index in "idx", and store the results in "dst". FOR j := 0 to 15 - i := j*8 - l := j*16 - dst[l+15:l] := ZeroExtend(a[i+7:i]) + i := j*16 + id := idx[i+3:i]*16 + dst[i+15:i] := a[id+15:id] ENDFOR dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - AVX2 - Convert - - Zero extend packed unsigned 8-bit integers in "a" to packed 32-bit integers, and store the results in "dst". + AVX512BW + Miscellaneous + + + + + + Shuffle 16-bit integers in "a" across lanes using the corresponding index in "idx", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := 32*j - k := 8*j - dst[i+31:i] := ZeroExtend(a[k+7:k]) +FOR j := 0 to 31 + i := j*16 + id := idx[i+4:i]*16 + IF k[j] + dst[i+15:i] := a[id+15:id] + ELSE + dst[i+15:i] := src[i+15:i] + FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX2 - Convert - - Zero extend packed unsigned 8-bit integers in the low 8 byte sof "a" to packed 64-bit integers, and store the results in "dst". + AVX512BW + Miscellaneous + + + + + Shuffle 16-bit integers in "a" across lanes using the corresponding index in "idx", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := 64*j - k := 8*j - dst[i+63:i] := ZeroExtend(a[k+7:k]) +FOR j := 0 to 31 + i := j*16 + id := idx[i+4:i]*16 + IF k[j] + dst[i+15:i] := a[id+15:id] + ELSE + dst[i+15:i] := 0 + FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX2 - Swizzle - - - Extract 128 bits (composed of integer data) from "a", selected with "imm8", and store the result in "dst". + AVX512BW + Miscellaneous + + + + Shuffle 16-bit integers in "a" across lanes using the corresponding index in "idx", and store the results in "dst". -CASE imm8[7:0] of -0: dst[127:0] := a[127:0] -1: dst[127:0] := a[255:128] -ESAC -dst[MAX:128] := 0 +FOR j := 0 to 31 + i := j*16 + id := idx[i+4:i]*16 + dst[i+15:i] := a[id+15:id] +ENDFOR +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX2 - Arithmetic - - - Horizontally add adjacent pairs of 16-bit integers in "a" and "b", and pack the signed 16-bit results in "dst". + AVX512VL + AVX512BW + Miscellaneous + + + + + + Shuffle 16-bit integers in "a" using the corresponding index in "idx", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -dst[15:0] := a[31:16] + a[15:0] -dst[31:16] := a[63:48] + a[47:32] -dst[47:32] := a[95:80] + a[79:64] -dst[63:48] := a[127:112] + a[111:96] -dst[79:64] := b[31:16] + b[15:0] -dst[95:80] := b[63:48] + b[47:32] -dst[111:96] := b[95:80] + b[79:64] -dst[127:112] := b[127:112] + b[111:96] -dst[143:128] := a[159:144] + a[143:128] -dst[159:144] := a[191:176] + a[175:160] -dst[175:160] := a[223:208] + a[207:192] -dst[191:176] := a[255:240] + a[239:224] -dst[207:192] := b[127:112] + b[143:128] -dst[223:208] := b[159:144] + b[175:160] -dst[239:224] := b[191:176] + b[207:192] -dst[255:240] := b[223:208] + b[239:224] -dst[MAX:256] := 0 +FOR j := 0 to 7 + i := j*16 + id := idx[i+2:i]*16 + IF k[j] + dst[i+15:i] := a[id+15:id] + ELSE + dst[i+15:i] := src[i+15:i] + FI +ENDFOR +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - AVX2 - Arithmetic - - - Horizontally add adjacent pairs of 32-bit integers in "a" and "b", and pack the signed 32-bit results in "dst". + AVX512VL + AVX512BW + Miscellaneous + + + + + Shuffle 16-bit integers in "a" using the corresponding index in "idx", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -dst[31:0] := a[63:32] + a[31:0] -dst[63:32] := a[127:96] + a[95:64] -dst[95:64] := b[63:32] + b[31:0] -dst[127:96] := b[127:96] + b[95:64] -dst[159:128] := a[191:160] + a[159:128] -dst[191:160] := a[255:224] + a[223:192] -dst[223:192] := b[191:160] + b[159:128] -dst[255:224] := b[255:224] + b[223:192] -dst[MAX:256] := 0 +FOR j := 0 to 7 + i := j*16 + id := idx[i+2:i]*16 + IF k[j] + dst[i+15:i] := a[id+15:id] + ELSE + dst[i+15:i] := 0 + FI +ENDFOR +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - AVX2 - Arithmetic - - - Horizontally add adjacent pairs of 16-bit integers in "a" and "b" using saturation, and pack the signed 16-bit results in "dst". + AVX512VL + AVX512BW + Miscellaneous + + + + Shuffle 16-bit integers in "a" using the corresponding index in "idx", and store the results in "dst". -dst[15:0]= Saturate_To_Int16(a[31:16] + a[15:0]) -dst[31:16] = Saturate_To_Int16(a[63:48] + a[47:32]) -dst[47:32] = Saturate_To_Int16(a[95:80] + a[79:64]) -dst[63:48] = Saturate_To_Int16(a[127:112] + a[111:96]) -dst[79:64] = Saturate_To_Int16(b[31:16] + b[15:0]) -dst[95:80] = Saturate_To_Int16(b[63:48] + b[47:32]) -dst[111:96] = Saturate_To_Int16(b[95:80] + b[79:64]) -dst[127:112] = Saturate_To_Int16(b[127:112] + b[111:96]) -dst[143:128] = Saturate_To_Int16(a[159:144] + a[143:128]) -dst[159:144] = Saturate_To_Int16(a[191:176] + a[175:160]) -dst[175:160] = Saturate_To_Int16( a[223:208] + a[207:192]) -dst[191:176] = Saturate_To_Int16(a[255:240] + a[239:224]) -dst[207:192] = Saturate_To_Int16(b[127:112] + b[143:128]) -dst[223:208] = Saturate_To_Int16(b[159:144] + b[175:160]) -dst[239:224] = Saturate_To_Int16(b[191-160] + b[159-128]) -dst[255:240] = Saturate_To_Int16(b[255:240] + b[239:224]) -dst[MAX:256] := 0 +FOR j := 0 to 7 + i := j*16 + id := idx[i+2:i]*16 + dst[i+15:i] := a[id+15:id] +ENDFOR +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - AVX2 + AVX512VL + AVX512BW Arithmetic - - - Horizontally subtract adjacent pairs of 16-bit integers in "a" and "b", and pack the signed 16-bit results in "dst". + + + + + + Multiply packed unsigned 8-bit integers in "a" by packed signed 8-bit integers in "b", producing intermediate signed 16-bit integers. Horizontally add adjacent pairs of intermediate signed 16-bit integers, and pack the saturated results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -dst[15:0] := a[15:0] - a[31:16] -dst[31:16] := a[47:32] - a[63:48] -dst[47:32] := a[79:64] - a[95:80] -dst[63:48] := a[111:96] - a[127:112] -dst[79:64] := b[15:0] - b[31:16] -dst[95:80] := b[47:32] - b[63:48] -dst[111:96] := b[79:64] - b[95:80] -dst[127:112] := b[111:96] - b[127:112] -dst[143:128] := a[143:128] - a[159:144] -dst[159:144] := a[175:160] - a[191:176] -dst[175:160] := a[207:192] - a[223:208] -dst[191:176] := a[239:224] - a[255:240] -dst[207:192] := b[143:128] - b[159:144] -dst[223:208] := b[175:160] - b[191:176] -dst[239:224] := b[207:192] - b[223:208] -dst[255:240] := b[239:224] - b[255:240] +FOR j := 0 to 15 + i := j*16 + IF k[j] + dst[i+15:i] := Saturate16( a[i+15:i+8]*b[i+15:i+8] + a[i+7:i]*b[i+7:i] ) + ELSE + dst[i+15:i] := src[i+15:i] + FI +ENDFOR dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - AVX2 + AVX512VL + AVX512BW Arithmetic - - - Horizontally subtract adjacent pairs of 32-bit integers in "a" and "b", and pack the signed 32-bit results in "dst". + + + + + Multiply packed unsigned 8-bit integers in "a" by packed signed 8-bit integers in "b", producing intermediate signed 16-bit integers. Horizontally add adjacent pairs of intermediate signed 16-bit integers, and pack the saturated results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -dst[31:0] := a[31:0] - a[63:32] -dst[63:32] := a[95:64] - a[127:96] -dst[95:64] := b[31:0] - b[63:32] -dst[127:96] := b[95:64] - b[127:96] -dst[159:128] := a[159:128] - a[191:160] -dst[191:160] := a[223:192] - a[255:224] -dst[223:192] := b[159:128] - b[191:160] -dst[255:224] := b[223:192] - b[255:224] +FOR j := 0 to 15 + i := j*16 + IF k[j] + dst[i+15:i] := Saturate16( a[i+15:i+8]*b[i+15:i+8] + a[i+7:i]*b[i+7:i] ) + ELSE + dst[i+15:i] := 0 + FI +ENDFOR dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - AVX2 + AVX512BW Arithmetic - - - Horizontally subtract adjacent pairs of 16-bit integers in "a" and "b" using saturation, and pack the signed 16-bit results in "dst". - -dst[15:0]= Saturate_To_Int16(a[15:0] - a[31:16]) -dst[31:16] = Saturate_To_Int16(a[47:32] - a[63:48]) -dst[47:32] = Saturate_To_Int16(a[79:64] - a[95:80]) -dst[63:48] = Saturate_To_Int16(a[111:96] - a[127:112]) -dst[79:64] = Saturate_To_Int16(b[15:0] - b[31:16]) -dst[95:80] = Saturate_To_Int16(b[47:32] - b[63:48]) -dst[111:96] = Saturate_To_Int16(b[79:64] - b[95:80]) -dst[127:112] = Saturate_To_Int16(b[111:96] - b[127:112]) -dst[143:128]= Saturate_To_Int16(a[143:128] - a[159:144]) -dst[159:144] = Saturate_To_Int16(a[175:160] - a[191:176]) -dst[175:160] = Saturate_To_Int16(a[207:192] - a[223:208]) -dst[191:176] = Saturate_To_Int16(a[239:224] - a[255:240]) -dst[207:192] = Saturate_To_Int16(b[143:128] - b[159:144]) -dst[223:208] = Saturate_To_Int16(b[175:160] - b[191:176]) -dst[239:224] = Saturate_To_Int16(b[207:192] - b[223:208]) -dst[255:240] = Saturate_To_Int16(b[239:224] - b[255:240]) -dst[MAX:256] := 0 - - -
immintrin.h
- - - Floating Point - AVX2 - Load - - - - - Gather double-precision (64-bit) floating-point elements from memory using 32-bit indices. 64-bit elements are loaded from addresses starting at "base_addr" and offset by each 32-bit element in "vindex" (each index is scaled by the factor in "scale"). Gathered elements are merged into "dst". "scale" should be 1, 2, 4 or 8. - + + + + Vertically multiply each unsigned 8-bit integer from "a" with the corresponding signed 8-bit integer from "b", producing intermediate signed 16-bit integers. Horizontally add adjacent pairs of intermediate signed 16-bit integers, and pack the saturated results in "dst". -FOR j := 0 to 1 - i := j*64 - m := j*32 - dst[i+63:i] := MEM[base_addr + SignExtend(vindex[m+31:m])*scale] +FOR j := 0 to 31 + i := j*16 + dst[i+15:i] := Saturate16( a[i+15:i+8]*b[i+15:i+8] + a[i+7:i]*b[i+7:i] ) ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point - AVX2 - Load - - - - - Gather double-precision (64-bit) floating-point elements from memory using 32-bit indices. 64-bit elements are loaded from addresses starting at "base_addr" and offset by each 32-bit element in "vindex" (each index is scaled by the factor in "scale"). Gathered elements are merged into "dst". "scale" should be 1, 2, 4 or 8. - + + Integer + AVX512BW + Arithmetic + + + + + + Multiply packed unsigned 8-bit integers in "a" by packed signed 8-bit integers in "b", producing intermediate signed 16-bit integers. Horizontally add adjacent pairs of intermediate signed 16-bit integers, and pack the saturated results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*64 - m := j*32 - dst[i+63:i] := MEM[base_addr + SignExtend(vindex[m+31:m])*scale] +FOR j := 0 to 31 + i := j*16 + IF k[j] + dst[i+15:i] := Saturate16( a[i+15:i+8]*b[i+15:i+8] + a[i+7:i]*b[i+7:i] ) + ELSE + dst[i+15:i] := src[i+15:i] + FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point - AVX2 - Load - - - - Gather single-precision (32-bit) floating-point elements from memory using 32-bit indices. 32-bit elements are loaded from addresses starting at "base_addr" and offset by each 32-bit element in "vindex" (each index is scaled by the factor in "scale"). Gathered elements are merged into "dst". "scale" should be 1, 2, 4 or 8. + + Integer + AVX512BW + Arithmetic + + + + + Multiply packed unsigned 8-bit integers in "a" by packed signed 8-bit integers in "b", producing intermediate signed 16-bit integers. Horizontally add adjacent pairs of intermediate signed 16-bit integers, and pack the saturated results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*32 - dst[i+31:i] := MEM[base_addr + SignExtend(vindex[i+31:i])*scale] +FOR j := 0 to 31 + i := j*16 + IF k[j] + dst[i+15:i] := Saturate16( a[i+15:i+8]*b[i+15:i+8] + a[i+7:i]*b[i+7:i] ) + ELSE + dst[i+15:i] := 0 + FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point - AVX2 - Load - - - - Gather single-precision (32-bit) floating-point elements from memory using 32-bit indices. 32-bit elements are loaded from addresses starting at "base_addr" and offset by each 32-bit element in "vindex" (each index is scaled by the factor in "scale"). Gathered elements are merged into "dst". "scale" should be 1, 2, 4 or 8. + + Integer + AVX512VL + AVX512BW + Arithmetic + + + + + + Multiply packed unsigned 8-bit integers in "a" by packed signed 8-bit integers in "b", producing intermediate signed 16-bit integers. Horizontally add adjacent pairs of intermediate signed 16-bit integers, and pack the saturated results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). FOR j := 0 to 7 - i := j*32 - dst[i+31:i] := MEM[base_addr + SignExtend(vindex[i+31:i])*scale] + i := j*16 + IF k[j] + dst[i+15:i] := Saturate16( a[i+15:i+8]*b[i+15:i+8] + a[i+7:i]*b[i+7:i] ) + ELSE + dst[i+15:i] := src[i+15:i] + FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - AVX2 - Load - - - - - Gather 32-bit integers from memory using 32-bit indices. 32-bit elements are loaded from addresses starting at "base_addr" and offset by each 32-bit element in "vindex" (each index is scaled by the factor in "scale"). Gathered elements are merged into "dst". "scale" should be 1, 2, 4 or 8. - + AVX512VL + AVX512BW + Arithmetic + + + + + Multiply packed unsigned 8-bit integers in "a" by packed signed 8-bit integers in "b", producing intermediate signed 16-bit integers. Horizontally add adjacent pairs of intermediate signed 16-bit integers, and pack the saturated results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*32 - dst[i+31:i] := MEM[base_addr + SignExtend(vindex[i+31:i])*scale] +FOR j := 0 to 7 + i := j*16 + IF k[j] + dst[i+15:i] := Saturate16( a[i+15:i+8]*b[i+15:i+8] + a[i+7:i]*b[i+7:i] ) + ELSE + dst[i+15:i] := 0 + FI ENDFOR dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - AVX2 - Load - - - - - Gather 32-bit integers from memory using 32-bit indices. 32-bit elements are loaded from addresses starting at "base_addr" and offset by each 32-bit element in "vindex" (each index is scaled by the factor in "scale"). Gathered elements are merged into "dst". "scale" should be 1, 2, 4 or 8. - + AVX512VL + AVX512BW + Arithmetic + + + + + + Multiply packed signed 16-bit integers in "a" and "b", producing intermediate signed 32-bit integers. Horizontally add adjacent pairs of intermediate 32-bit integers, and pack the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). FOR j := 0 to 7 i := j*32 - dst[i+31:i] := MEM[base_addr + SignExtend(vindex[i+31:i])*scale] + IF k[j] + dst[i+31:i] := SignExtend32(a[i+31:i+16]*b[i+31:i+16]) + SignExtend32(a[i+15:i]*b[i+15:i]) + ELSE + dst[i+31:i] := src[i+31:i] + FI ENDFOR dst[MAX:256] := 0 - -
immintrin.h
- - - Integer - AVX2 - Load - - - - Gather 64-bit integers from memory using 32-bit indices. 64-bit elements are loaded from addresses starting at "base_addr" and offset by each 32-bit element in "vindex" (each index is scaled by the factor in "scale"). Gathered elements are merged into "dst". "scale" should be 1, 2, 4 or 8. - - -FOR j := 0 to 1 - i := j*64 - m := j*32 - dst[i+63:i] := MEM[base_addr + SignExtend(vindex[m+31:m])*scale] -ENDFOR -dst[MAX:128] := 0 - - +
immintrin.h
 - + Integer - AVX2 - Load - - - - Gather 64-bit integers from memory using 32-bit indices. 64-bit elements are loaded from addresses starting at "base_addr" and offset by each 32-bit element in "vindex" (each index is scaled by the factor in "scale"). Gathered elements are merged into "dst". "scale" should be 1, 2, 4 or 8. - + AVX512VL + AVX512BW + Arithmetic + + + + + Multiply packed signed 16-bit integers in "a" and "b", producing intermediate signed 32-bit integers. Horizontally add adjacent pairs of intermediate 32-bit integers, and pack the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*64 - m := j*32 - dst[i+63:i] := MEM[base_addr + SignExtend(vindex[m+31:m])*scale] +FOR j := 0 to 7 + i := j*32 + IF k[j] + dst[i+31:i] := SignExtend32(a[i+31:i+16]*b[i+31:i+16]) + SignExtend32(a[i+15:i]*b[i+15:i]) + ELSE + dst[i+31:i] := 0 + FI ENDFOR dst[MAX:256] := 0 - +
immintrin.h
 - - Floating Point - AVX2 - Load - - - - - Gather double-precision (64-bit) floating-point elements from memory using 64-bit indices. 64-bit elements are loaded from addresses starting at "base_addr" and offset by each 64-bit element in "vindex" (each index is scaled by the factor in "scale"). Gathered elements are merged into "dst". "scale" should be 1, 2, 4 or 8. - + + Integer + AVX512BW + Arithmetic + + + + Multiply packed signed 16-bit integers in "a" and "b", producing intermediate signed 32-bit integers. Horizontally add adjacent pairs of intermediate 32-bit integers, and pack the results in "dst". -FOR j := 0 to 1 - i := j*64 - dst[i+63:i] := MEM[base_addr + SignExtend(vindex[m+63:m])*scale] +FOR j := 0 to 15 + i := j*32 + dst[i+31:i] := SignExtend32(a[i+31:i+16]*b[i+31:i+16]) + SignExtend32(a[i+15:i]*b[i+15:i]) ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point - AVX2 - Load - - - - - Gather double-precision (64-bit) floating-point elements from memory using 64-bit indices. 64-bit elements are loaded from addresses starting at "base_addr" and offset by each 64-bit element in "vindex" (each index is scaled by the factor in "scale"). Gathered elements are merged into "dst". "scale" should be 1, 2, 4 or 8. - + + Integer + AVX512BW + Arithmetic + + + + + + Multiply packed signed 16-bit integers in "a" and "b", producing intermediate signed 32-bit integers. Horizontally add adjacent pairs of intermediate 32-bit integers, and pack the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*64 - dst[i+63:i] := MEM[base_addr + SignExtend(vindex[m+63:m])*scale] +FOR j := 0 to 15 + i := j*32 + IF k[j] + dst[i+31:i] := SignExtend32(a[i+31:i+16]*b[i+31:i+16]) + SignExtend32(a[i+15:i]*b[i+15:i]) + ELSE + dst[i+31:i] := src[i+31:i] + FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point - AVX2 - Load - - - - - Gather single-precision (32-bit) floating-point elements from memory using 64-bit indices. 32-bit elements are loaded from addresses starting at "base_addr" and offset by each 64-bit element in "vindex" (each index is scaled by the factor in "scale"). Gathered elements are merged into "dst". "scale" should be 1, 2, 4 or 8. - + + Integer + AVX512BW + Arithmetic + + + + + Multiply packed signed 16-bit integers in "a" and "b", producing intermediate signed 32-bit integers. Horizontally add adjacent pairs of intermediate 32-bit integers, and pack the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 1 +FOR j := 0 to 15 i := j*32 - m := j*64 - dst[i+31:i] := MEM[base_addr + SignExtend(vindex[m+63:m])*scale] + IF k[j] + dst[i+31:i] := SignExtend32(a[i+31:i+16]*b[i+31:i+16]) + SignExtend32(a[i+15:i]*b[i+15:i]) + ELSE + dst[i+31:i] := 0 + FI ENDFOR -dst[MAX:64] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point - AVX2 - Load - - - - - Gather single-precision (32-bit) floating-point elements from memory using 64-bit indices. 32-bit elements are loaded from addresses starting at "base_addr" and offset by each 64-bit element in "vindex" (each index is scaled by the factor in "scale"). Gathered elements are merged into "dst". "scale" should be 1, 2, 4 or 8. - + + Integer + AVX512VL + AVX512BW + Arithmetic + + + + + + Multiply packed signed 16-bit integers in "a" and "b", producing intermediate signed 32-bit integers. Horizontally add adjacent pairs of intermediate 32-bit integers, and pack the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). FOR j := 0 to 3 i := j*32 - m := j*64 - dst[i+31:i] := MEM[base_addr + SignExtend(vindex[m+63:m])*scale] + IF k[j] + dst[i+31:i] := SignExtend32(a[i+31:i+16]*b[i+31:i+16]) + SignExtend32(a[i+15:i]*b[i+15:i]) + ELSE + dst[i+31:i] := src[i+31:i] + FI ENDFOR dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - AVX2 - Load - - - - Gather 32-bit integers from memory using 64-bit indices. 32-bit elements are loaded from addresses starting at "base_addr" and offset by each 64-bit element in "vindex" (each index is scaled by the factor in "scale"). Gathered elements are merged into "dst". "scale" should be 1, 2, 4 or 8. - + AVX512VL + AVX512BW + Arithmetic + + + + + Multiply packed signed 16-bit integers in "a" and "b", producing intermediate signed 32-bit integers. Horizontally add adjacent pairs of intermediate 32-bit integers, and pack the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 1 +FOR j := 0 to 3 i := j*32 - m := j*64 - dst[i+31:i] := MEM[base_addr + SignExtend(vindex[m+63:m])*scale] + IF k[j] + dst[i+31:i] := SignExtend32(a[i+31:i+16]*b[i+31:i+16]) + SignExtend32(a[i+15:i]*b[i+15:i]) + ELSE + dst[i+31:i] := 0 + FI ENDFOR -dst[MAX:64] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - AVX2 - Load - - - - Gather 32-bit integers from memory using 64-bit indices. 32-bit elements are loaded from addresses starting at "base_addr" and offset by each 64-bit element in "vindex" (each index is scaled by the factor in "scale"). Gathered elements are merged into "dst". "scale" should be 1, 2, 4 or 8. - + AVX512VL + AVX512BW + Arithmetic + + + + + + Compare packed signed 8-bit integers in "a" and "b", and store packed maximum values in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*32 - m := j*64 - dst[i+31:i] := MEM[base_addr + SignExtend(vindex[m+63:m])*scale] +FOR j := 0 to 31 + i := j*8 + IF k[j] + dst[i+7:i] := MAX(a[i+7:i], b[i+7:i]) + ELSE + dst[i+7:i] := src[i+7:i] + FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - AVX2 - Load - - - - Gather 64-bit integers from memory using 64-bit indices. 64-bit elements are loaded from addresses starting at "base_addr" and offset by each 64-bit element in "vindex" (each index is scaled by the factor in "scale"). Gathered elements are merged into "dst". "scale" should be 1, 2, 4 or 8. - + AVX512VL + AVX512BW + Arithmetic + + + + + Compare packed signed 8-bit integers in "a" and "b", and store packed maximum values in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 1 - i := j*64 - dst[i+63:i] := MEM[base_addr + SignExtend(vindex[m+63:m])*scale] +FOR j := 0 to 31 + i := j*8 + IF k[j] + dst[i+7:i] := MAX(a[i+7:i], b[i+7:i]) + ELSE + dst[i+7:i] := 0 + FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - AVX2 - Load - - - - Gather 64-bit integers from memory using 64-bit indices. 64-bit elements are loaded from addresses starting at "base_addr" and offset by each 64-bit element in "vindex" (each index is scaled by the factor in "scale"). Gathered elements are merged into "dst". "scale" should be 1, 2, 4 or 8. - + AVX512BW + Arithmetic + + + + + + Compare packed signed 8-bit integers in "a" and "b", and store packed maximum values in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*64 - dst[i+63:i] := MEM[base_addr + SignExtend(vindex[m+63:m])*scale] +FOR j := 0 to 63 + i := j*8 + IF k[j] + dst[i+7:i] := MAX(a[i+7:i], b[i+7:i]) + ELSE + dst[i+7:i] := src[i+7:i] + FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX2 - Swizzle - - - - Copy "a" to "dst", then insert 128 bits (composed of integer data) from "b" into "dst" at the location specified by "imm8". + AVX512BW + Arithmetic + + + + + Compare packed signed 8-bit integers in "a" and "b", and store packed maximum values in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -dst[255:0] := a[255:0] -CASE (imm8[1:0]) of -0: dst[127:0] := b[127:0] -1: dst[255:128] := b[127:0] -ESAC -dst[MAX:256] := 0 +FOR j := 0 to 63 + i := j*8 + IF k[j] + dst[i+7:i] := MAX(a[i+7:i], b[i+7:i]) + ELSE + dst[i+7:i] := 0 + FI +ENDFOR +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX2 + AVX512BW Arithmetic - - - Multiply packed signed 16-bit integers in "a" and "b", producing intermediate signed 32-bit integers. Horizontally add adjacent pairs of intermediate 32-bit integers, and pack the results in "dst". - + + + + Compare packed signed 8-bit integers in "a" and "b", and store packed maximum values in "dst". -FOR j := 0 to 7 - i := j*32 - dst[i+31:i] := a[i+31:i+16]*b[i+31:i+16] + a[i+15:i]*b[i+15:i] +FOR j := 0 to 63 + i := j*8 + dst[i+7:i] := MAX(a[i+7:i], b[i+7:i]) ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX2 + AVX512VL + AVX512BW Arithmetic - - - Vertically multiply each unsigned 8-bit integer from "a" with the corresponding signed 8-bit integer from "b", producing intermediate signed 16-bit integers. Horizontally add adjacent pairs of intermediate signed 16-bit integers, and pack the saturated results in "dst". - + + + + + + Compare packed signed 8-bit integers in "a" and "b", and store packed maximum values in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). FOR j := 0 to 15 - i := j*16 - dst[i+15:i] := Saturate_To_Int16( a[i+15:i+8]*b[i+15:i+8] + a[i+7:i]*b[i+7:i] ) -ENDFOR -dst[MAX:256] := 0 - - -
immintrin.h
- - - Floating Point - AVX2 - Load - - - - - - - Gather double-precision (64-bit) floating-point elements from memory using 32-bit indices. 64-bit elements are loaded from addresses starting at "base_addr" and offset by each 32-bit element in "vindex" (each index is scaled by the factor in "scale"). Gathered elements are merged into "dst" using "mask" (elements are copied from "src" when the highest bit is not set in the corresponding element). "scale" should be 1, 2, 4 or 8. - - -FOR j := 0 to 1 - i := j*64 - m := j*32 - IF mask[i+63] - dst[i+63:i] := MEM[base_addr + SignExtend(vindex[m+31:m])*scale] - mask[i+63] := 0 + i := j*8 + IF k[j] + dst[i+7:i] := MAX(a[i+7:i], b[i+7:i]) ELSE - dst[i+63:i] := src[i+63:i] + dst[i+7:i] := src[i+7:i] FI ENDFOR -mask[MAX:128] := 0 dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point - AVX2 - Load - - - - - - - Gather double-precision (64-bit) floating-point elements from memory using 32-bit indices. 64-bit elements are loaded from addresses starting at "base_addr" and offset by each 32-bit element in "vindex" (each index is scaled by the factor in "scale"). Gathered elements are merged into "dst" using "mask" (elements are copied from "src" when the highest bit is not set in the corresponding element). "scale" should be 1, 2, 4 or 8. - + + Integer + AVX512VL + AVX512BW + Arithmetic + + + + + Compare packed signed 8-bit integers in "a" and "b", and store packed maximum values in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*64 - m := j*32 - IF mask[i+63] - dst[i+63:i] := MEM[base_addr + SignExtend(vindex[m+31:m])*scale] - mask[i+63] := 0 +FOR j := 0 to 15 + i := j*8 + IF k[j] + dst[i+7:i] := MAX(a[i+7:i], b[i+7:i]) ELSE - dst[i+63:i] := src[i+63:i] + dst[i+7:i] := 0 FI ENDFOR -mask[MAX:256] := 0 -dst[MAX:256] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point - AVX2 - Load - - - - - - - Gather single-precision (32-bit) floating-point elements from memory using 32-bit indices. 32-bit elements are loaded from addresses starting at "base_addr" and offset by each 32-bit element in "vindex" (each index is scaled by the factor in "scale"). Gathered elements are merged into "dst" using "mask" (elements are copied from "src" when the highest bit is not set in the corresponding element). "scale" should be 1, 2, 4 or 8. - + + Integer + AVX512VL + AVX512BW + Arithmetic + + + + + + Compare packed signed 16-bit integers in "a" and "b", and store packed maximum values in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*32 - IF mask[i+31] - dst[i+31:i] := MEM[base_addr + SignExtend(vindex[i+31:i])*scale] - mask[i+31] := 0 +FOR j := 0 to 15 + i := j*16 + IF k[j] + dst[i+15:i] := MAX(a[i+15:i], b[i+15:i]) ELSE - dst[i+31:i] := src[i+31:i] + dst[i+15:i] := src[i+15:i] FI ENDFOR -mask[MAX:128] := 0 -dst[MAX:128] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - - Floating Point - AVX2 - Load - - - - - - - Gather single-precision (32-bit) floating-point elements from memory using 32-bit indices. 32-bit elements are loaded from addresses starting at "base_addr" and offset by each 32-bit element in "vindex" (each index is scaled by the factor in "scale"). Gathered elements are merged into "dst" using "mask" (elements are copied from "src" when the highest bit is not set in the corresponding element). "scale" should be 1, 2, 4 or 8. - + + Integer + AVX512VL + AVX512BW + Arithmetic + + + + + Compare packed signed 16-bit integers in "a" and "b", and store packed maximum values in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := j*32 - IF mask[i+31] - dst[i+31:i] := MEM[base_addr + SignExtend(vindex[i+31:i])*scale] - mask[i+31] := 0 +FOR j := 0 to 15 + i := j*16 + IF k[j] + dst[i+15:i] := MAX(a[i+15:i], b[i+15:i]) ELSE - dst[i+31:i] := src[i+31:i] + dst[i+15:i] := 0 FI ENDFOR -mask[MAX:256] := 0 dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - AVX2 - Load - - - - - - - Gather 32-bit integers from memory using 32-bit indices. 32-bit elements are loaded from addresses starting at "base_addr" and offset by each 32-bit element in "vindex" (each index is scaled by the factor in "scale"). Gathered elements are merged into "dst" using "mask" (elements are copied from "src" when the highest bit is not set in the corresponding element). "scale" should be 1, 2, 4 or 8. - + AVX512BW + Arithmetic + + + + + + Compare packed signed 16-bit integers in "a" and "b", and store packed maximum values in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*32 - IF mask[i+31] - dst[i+31:i] := MEM[base_addr + SignExtend(vindex[i+31:i])*scale] - mask[i+31] := 0 +FOR j := 0 to 31 + i := j*16 + IF k[j] + dst[i+15:i] := MAX(a[i+15:i], b[i+15:i]) ELSE - dst[i+31:i] := src[i+31:i] + dst[i+15:i] := src[i+15:i] FI ENDFOR -mask[MAX:128] := 0 -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX2 - Load - - - - - - - Gather 32-bit integers from memory using 32-bit indices. 32-bit elements are loaded from addresses starting at "base_addr" and offset by each 32-bit element in "vindex" (each index is scaled by the factor in "scale"). Gathered elements are merged into "dst" using "mask" (elements are copied from "src" when the highest bit is not set in the corresponding element). "scale" should be 1, 2, 4 or 8. - + AVX512BW + Arithmetic + + + + + Compare packed signed 16-bit integers in "a" and "b", and store packed maximum values in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := j*32 - IF mask[i+31] - dst[i+31:i] := MEM[base_addr + SignExtend(vindex[i+31:i])*scale] - mask[i+31] := 0 +FOR j := 0 to 31 + i := j*16 + IF k[j] + dst[i+15:i] := MAX(a[i+15:i], b[i+15:i]) ELSE - dst[i+31:i] := src[i+31:i] + dst[i+15:i] := 0 FI ENDFOR -mask[MAX:256] := 0 -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX2 - Load - - - - - - - Gather 64-bit integers from memory using 32-bit indices. 64-bit elements are loaded from addresses starting at "base_addr" and offset by each 32-bit element in "vindex" (each index is scaled by the factor in "scale"). Gathered elements are merged into "dst" using "mask" (elements are copied from "src" when the highest bit is not set in the corresponding element). "scale" should be 1, 2, 4 or 8. - + AVX512BW + Arithmetic + + + + Compare packed signed 16-bit integers in "a" and "b", and store packed maximum values in "dst". -FOR j := 0 to 1 - i := j*64 - m := j*32 - IF mask[i+63] - dst[i+63:i] := MEM[base_addr + SignExtend(vindex[m+31:m])*scale] - mask[i+63] := 0 - ELSE - dst[i+63:i] := src[i+63:i] - FI +FOR j := 0 to 31 + i := j*16 + dst[i+15:i] := MAX(a[i+15:i], b[i+15:i]) ENDFOR -mask[MAX:128] := 0 -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX2 - Load - - - - - - - Gather 64-bit integers from memory using 32-bit indices. 64-bit elements are loaded from addresses starting at "base_addr" and offset by each 32-bit element in "vindex" (each index is scaled by the factor in "scale"). Gathered elements are merged into "dst" using "mask" (elements are copied from "src" when the highest bit is not set in the corresponding element). "scale" should be 1, 2, 4 or 8. - + AVX512VL + AVX512BW + Arithmetic + + + + + + Compare packed signed 16-bit integers in "a" and "b", and store packed maximum values in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*64 - m := j*32 - IF mask[i+63] - dst[i+63:i] := MEM[base_addr + SignExtend(vindex[m+31:m])*scale] - mask[i+63] := 0 +FOR j := 0 to 7 + i := j*16 + IF k[j] + dst[i+15:i] := MAX(a[i+15:i], b[i+15:i]) ELSE - dst[i+63:i] := src[i+63:i] + dst[i+15:i] := src[i+15:i] FI ENDFOR -mask[MAX:256] := 0 -dst[MAX:256] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point - AVX2 - Load - - - - - - - Gather double-precision (64-bit) floating-point elements from memory using 64-bit indices. 64-bit elements are loaded from addresses starting at "base_addr" and offset by each 64-bit element in "vindex" (each index is scaled by the factor in "scale"). Gathered elements are merged into "dst" using "mask" (elements are copied from "src" when the highest bit is not set in the corresponding element). "scale" should be 1, 2, 4 or 8. - + + Integer + AVX512VL + AVX512BW + Arithmetic + + + + + Compare packed signed 16-bit integers in "a" and "b", and store packed maximum values in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 1 - i := j*64 - IF mask[i+63] - dst[i+63:i] := MEM[base_addr + SignExtend(vindex[m+63:m])*scale] - mask[i+63] := 0 +FOR j := 0 to 7 + i := j*16 + IF k[j] + dst[i+15:i] := MAX(a[i+15:i], b[i+15:i]) ELSE - dst[i+63:i] := src[i+63:i] + dst[i+15:i] := 0 FI ENDFOR -mask[MAX:128] := 0 dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point - AVX2 - Load - - - - - - - Gather double-precision (64-bit) floating-point elements from memory using 64-bit indices. 64-bit elements are loaded from addresses starting at "base_addr" and offset by each 64-bit element in "vindex" (each index is scaled by the factor in "scale"). Gathered elements are merged into "dst" using "mask" (elements are copied from "src" when the highest bit is not set in the corresponding element). "scale" should be 1, 2, 4 or 8. - + + Integer + AVX512VL + AVX512BW + Arithmetic + + + + + + Compare packed unsigned 8-bit integers in "a" and "b", and store packed maximum values in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*64 - IF mask[i+63] - dst[i+63:i] := MEM[base_addr + SignExtend(vindex[m+63:m])*scale] - mask[i+63] := 0 +FOR j := 0 to 31 + i := j*8 + IF k[j] + dst[i+7:i] := MAX(a[i+7:i], b[i+7:i]) ELSE - dst[i+63:i] := src[i+63:i] + dst[i+7:i] := src[i+7:i] FI ENDFOR -mask[MAX:256] := 0 dst[MAX:256] := 0 - +
immintrin.h
 - - Floating Point - AVX2 - Load - - - - - - - Gather single-precision (32-bit) floating-point elements from memory using 64-bit indices. 32-bit elements are loaded from addresses starting at "base_addr" and offset by each 64-bit element in "vindex" (each index is scaled by the factor in "scale"). Gathered elements are merged into "dst" using "mask" (elements are copied from "src" when the highest bit is not set in the corresponding element). "scale" should be 1, 2, 4 or 8. - + + Integer + AVX512VL + AVX512BW + Arithmetic + + + + + Compare packed unsigned 8-bit integers in "a" and "b", and store packed maximum values in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 1 - i := j*32 - m := j*64 - IF mask[i+31] - dst[i+31:i] := MEM[base_addr + SignExtend(vindex[m+63:m])*scale] - mask[i+31] := 0 +FOR j := 0 to 31 + i := j*8 + IF k[j] + dst[i+7:i] := MAX(a[i+7:i], b[i+7:i]) ELSE - dst[i+31:i] := src[i+31:i] + dst[i+7:i] := 0 FI ENDFOR -mask[MAX:64] := 0 -dst[MAX:64] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - - Floating Point - AVX2 - Load - - - - - - - Gather single-precision (32-bit) floating-point elements from memory using 64-bit indices. 32-bit elements are loaded from addresses starting at "base_addr" and offset by each 64-bit element in "vindex" (each index is scaled by the factor in "scale"). Gathered elements are merged into "dst" using "mask" (elements are copied from "src" when the highest bit is not set in the corresponding element). "scale" should be 1, 2, 4 or 8. - + + Integer + AVX512BW + Arithmetic + + + + + + Compare packed unsigned 8-bit integers in "a" and "b", and store packed maximum values in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*32 - m := j*64 - IF mask[i+31] - dst[i+31:i] := MEM[base_addr + SignExtend(vindex[m+63:m])*scale] - mask[i+31] := 0 +FOR j := 0 to 63 + i := j*8 + IF k[j] + dst[i+7:i] := MAX(a[i+7:i], b[i+7:i]) ELSE - dst[i+31:i] := src[i+31:i] + dst[i+7:i] := src[i+7:i] FI ENDFOR -mask[MAX:128] := 0 -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX2 - Load - - - - - - - Gather 32-bit integers from memory using 64-bit indices. 32-bit elements are loaded from addresses starting at "base_addr" and offset by each 64-bit element in "vindex" (each index is scaled by the factor in "scale"). Gathered elements are merged into "dst" using "mask" (elements are copied from "src" when the highest bit is not set in the corresponding element). "scale" should be 1, 2, 4 or 8. - + AVX512BW + Arithmetic + + + + + Compare packed unsigned 8-bit integers in "a" and "b", and store packed maximum values in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 1 - i := j*32 - m := j*64 - IF mask[i+31] - dst[i+31:i] := MEM[base_addr + SignExtend(vindex[m+63:m])*scale] - mask[i+31] := 0 +FOR j := 0 to 63 + i := j*8 + IF k[j] + dst[i+7:i] := MAX(a[i+7:i], b[i+7:i]) ELSE - dst[i+31:i] := src[i+31:i] + dst[i+7:i] := 0 FI ENDFOR -mask[MAX:64] := 0 -dst[MAX:64] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX2 - Load - - - - - - - Gather 32-bit integers from memory using 64-bit indices. 32-bit elements are loaded from addresses starting at "base_addr" and offset by each 64-bit element in "vindex" (each index is scaled by the factor in "scale"). Gathered elements are merged into "dst" using "mask" (elements are copied from "src" when the highest bit is not set in the corresponding element). "scale" should be 1, 2, 4 or 8. - + AVX512BW + Arithmetic + + + + Compare packed unsigned 8-bit integers in "a" and "b", and store packed maximum values in "dst". -FOR j := 0 to 3 - i := j*32 - m := j*64 - IF mask[i+31] - dst[i+31:i] := MEM[base_addr + SignExtend(vindex[m+63:m])*scale] - mask[i+31] := 0 - ELSE - dst[i+31:i] := src[i+31:i] - FI +FOR j := 0 to 63 + i := j*8 + dst[i+7:i] := MAX(a[i+7:i], b[i+7:i]) ENDFOR -mask[MAX:128] := 0 -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX2 - Load - - - - - - - Gather 64-bit integers from memory using 64-bit indices. 64-bit elements are loaded from addresses starting at "base_addr" and offset by each 64-bit element in "vindex" (each index is scaled by the factor in "scale"). Gathered elements are merged into "dst" using "mask" (elements are copied from "src" when the highest bit is not set in the corresponding element). "scale" should be 1, 2, 4 or 8. - + AVX512VL + AVX512BW + Arithmetic + + + + + + Compare packed unsigned 8-bit integers in "a" and "b", and store packed maximum values in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 1 - i := j*64 - IF mask[i+63] - dst[i+63:i] := MEM[base_addr + SignExtend(vindex[m+63:m])*scale] - mask[i+63] := 0 +FOR j := 0 to 15 + i := j*8 + IF k[j] + dst[i+7:i] := MAX(a[i+7:i], b[i+7:i]) ELSE - dst[i+63:i] := src[i+63:i] + dst[i+7:i] := src[i+7:i] FI ENDFOR -mask[MAX:128] := 0 dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - AVX2 - Load - - - - - - - Gather 64-bit integers from memory using 64-bit indices. 64-bit elements are loaded from addresses starting at "base_addr" and offset by each 64-bit element in "vindex" (each index is scaled by the factor in "scale"). Gathered elements are merged into "dst" using "mask" (elements are copied from "src" when the highest bit is not set in the corresponding element). "scale" should be 1, 2, 4 or 8. - + AVX512VL + AVX512BW + Arithmetic + + + + + Compare packed unsigned 8-bit integers in "a" and "b", and store packed maximum values in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*64 - IF mask[i+63] - dst[i+63:i] := MEM[base_addr + SignExtend(vindex[m+63:m])*scale] - mask[i+63] := 0 +FOR j := 0 to 15 + i := j*8 + IF k[j] + dst[i+7:i] := MAX(a[i+7:i], b[i+7:i]) ELSE - dst[i+63:i] := src[i+63:i] + dst[i+7:i] := 0 FI ENDFOR -mask[MAX:256] := 0 -dst[MAX:256] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - AVX2 - Load - - - Load packed 32-bit integers from memory into "dst" using "mask" (elements are zeroed out when the highest bit is not set in the corresponding element). + AVX512VL + AVX512BW + Arithmetic + + + + + + Compare packed unsigned 16-bit integers in "a" and "b", and store packed maximum values in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*32 - IF mask[i+31] - dst[i+31:i] := MEM[mem_addr+i+31:mem_addr+i] +FOR j := 0 to 15 + i := j*16 + IF k[j] + dst[i+15:i] := MAX(a[i+15:i], b[i+15:i]) ELSE - dst[i+31:i] := 0 + dst[i+15:i] := src[i+15:i] FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - AVX2 - Load - - - Load packed 32-bit integers from memory into "dst" using "mask" (elements are zeroed out when the highest bit is not set in the corresponding element). + AVX512VL + AVX512BW + Arithmetic + + + + + Compare packed unsigned 16-bit integers in "a" and "b", and store packed maximum values in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := j*32 - IF mask[i+31] - dst[i+31:i] := MEM[mem_addr+i+31:mem_addr+i] +FOR j := 0 to 15 + i := j*16 + IF k[j] + dst[i+15:i] := MAX(a[i+15:i], b[i+15:i]) ELSE - dst[i+31:i] := 0 + dst[i+15:i] := 0 FI ENDFOR dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - AVX2 - Load - - - Load packed 64-bit integers from memory into "dst" using "mask" (elements are zeroed out when the highest bit is not set in the corresponding element). + AVX512BW + Arithmetic + + + + + + Compare packed unsigned 16-bit integers in "a" and "b", and store packed maximum values in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 1 - i := j*64 - IF mask[i+63] - dst[i+63:i] := MEM[mem_addr+i+63:mem_addr+i] +FOR j := 0 to 31 + i := j*16 + IF k[j] + dst[i+15:i] := MAX(a[i+15:i], b[i+15:i]) ELSE - dst[i+63:i] := 0 + dst[i+15:i] := src[i+15:i] FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX2 - Load - - - Load packed 64-bit integers from memory into "dst" using "mask" (elements are zeroed out when the highest bit is not set in the corresponding element). + AVX512BW + Arithmetic + + + + + Compare packed unsigned 16-bit integers in "a" and "b", and store packed maximum values in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*64 - IF mask[i+63] - dst[i+63:i] := MEM[mem_addr+i+63:mem_addr+i] +FOR j := 0 to 31 + i := j*16 + IF k[j] + dst[i+15:i] := MAX(a[i+15:i], b[i+15:i]) ELSE - dst[i+63:i] := 0 + dst[i+15:i] := 0 FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX2 - Store - - - - Store packed 32-bit integers from "a" into memory using "mask" (elements are not stored when the highest bit is not set in the corresponding element). - + AVX512BW + Arithmetic + + + + Compare packed unsigned 16-bit integers in "a" and "b", and store packed maximum values in "dst". -FOR j := 0 to 3 - i := j*32 - IF mask[i+31] - MEM[mem_addr+i+31:mem_addr+i] := a[i+31:i] - FI +FOR j := 0 to 31 + i := j*16 + dst[i+15:i] := MAX(a[i+15:i], b[i+15:i]) ENDFOR +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX2 - Store - - - - Store packed 32-bit integers from "a" into memory using "mask" (elements are not stored when the highest bit is not set in the corresponding element). - + AVX512VL + AVX512BW + Arithmetic + + + + + + Compare packed unsigned 16-bit integers in "a" and "b", and store packed maximum values in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). FOR j := 0 to 7 - i := j*32 - IF mask[i+31] - MEM[mem_addr+i+31:mem_addr+i] := a[i+31:i] + i := j*16 + IF k[j] + dst[i+15:i] := MAX(a[i+15:i], b[i+15:i]) + ELSE + dst[i+15:i] := src[i+15:i] FI ENDFOR +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - AVX2 - Store - - - - Store packed 64-bit integers from "a" into memory using "mask" (elements are not stored when the highest bit is not set in the corresponding element). - + AVX512VL + AVX512BW + Arithmetic + + + + + Compare packed unsigned 16-bit integers in "a" and "b", and store packed maximum values in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 1 - i := j*64 - IF mask[i+63] - MEM[mem_addr+i+63:mem_addr+i] := a[i+63:i] +FOR j := 0 to 7 + i := j*16 + IF k[j] + dst[i+15:i] := MAX(a[i+15:i], b[i+15:i]) + ELSE + dst[i+15:i] := 0 FI ENDFOR +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - AVX2 - Store - - - - Store packed 64-bit integers from "a" into memory using "mask" (elements are not stored when the highest bit is not set in the corresponding element). - + AVX512VL + AVX512BW + Arithmetic + + + + + + Compare packed signed 8-bit integers in "a" and "b", and store packed minimum values in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*64 - IF mask[i+63] - MEM[mem_addr+i+63:mem_addr+i] := a[i+63:i] +FOR j := 0 to 31 + i := j*8 + IF k[j] + dst[i+7:i] := MIN(a[i+7:i], b[i+7:i]) + ELSE + dst[i+7:i] := src[i+7:i] FI ENDFOR +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - AVX2 - Special Math Functions - - - Compare packed 8-bit integers in "a" and "b", and store packed maximum values in "dst". + AVX512VL + AVX512BW + Arithmetic + + + + + Compare packed signed 8-bit integers in "a" and "b", and store packed minimum values in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). FOR j := 0 to 31 i := j*8 - IF a[i+7:i] > b[i+7:i] - dst[i+7:i] := a[i+7:i] + IF k[j] + dst[i+7:i] := MIN(a[i+7:i], b[i+7:i]) ELSE - dst[i+7:i] := b[i+7:i] + dst[i+7:i] := 0 FI ENDFOR dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - AVX2 - Special Math Functions - - - Compare packed 16-bit integers in "a" and "b", and store packed maximum values in "dst". - + AVX512BW + Arithmetic + + + + + + Compare packed signed 8-bit integers in "a" and "b", and store packed minimum values in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*16 - IF a[i+15:i] > b[i+15:i] - dst[i+15:i] := a[i+15:i] +FOR j := 0 to 63 + i := j*8 + IF k[j] + dst[i+7:i] := MIN(a[i+7:i], b[i+7:i]) ELSE - dst[i+15:i] := b[i+15:i] + dst[i+7:i] := src[i+7:i] FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX2 - Special Math Functions - - - Compare packed 32-bit integers in "a" and "b", and store packed maximum values in "dst". + AVX512BW + Arithmetic + + + + + Compare packed signed 8-bit integers in "a" and "b", and store packed minimum values in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := j*32 - IF a[i+31:i] > b[i+31:i] - dst[i+31:i] := a[i+31:i] +FOR j := 0 to 63 + i := j*8 + IF k[j] + dst[i+7:i] := MIN(a[i+7:i], b[i+7:i]) ELSE - dst[i+31:i] := b[i+31:i] + dst[i+7:i] := 0 FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX2 - Special Math Functions - - - Compare packed unsigned 8-bit integers in "a" and "b", and store packed maximum values in "dst". - + AVX512BW + Arithmetic + + + + Compare packed signed 8-bit integers in "a" and "b", and store packed minimum values in "dst". -FOR j := 0 to 31 +FOR j := 0 to 63 i := j*8 - IF a[i+7:i] > b[i+7:i] - dst[i+7:i] := a[i+7:i] - ELSE - dst[i+7:i] := b[i+7:i] - FI + dst[i+7:i] := MIN(a[i+7:i], b[i+7:i]) ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX2 - Special Math Functions - - - Compare packed unsigned 16-bit integers in "a" and "b", and store packed maximum values in "dst". - + AVX512VL + AVX512BW + Arithmetic + + + + + + Compare packed signed 8-bit integers in "a" and "b", and store packed minimum values in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). FOR j := 0 to 15 - i := j*16 - IF a[i+15:i] > b[i+15:i] - dst[i+15:i] := a[i+15:i] + i := j*8 + IF k[j] + dst[i+7:i] := MIN(a[i+7:i], b[i+7:i]) ELSE - dst[i+15:i] := b[i+15:i] + dst[i+7:i] := src[i+7:i] FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - AVX2 - Special Math Functions - - - Compare packed unsigned 32-bit integers in "a" and "b", and store packed maximum values in "dst". + AVX512VL + AVX512BW + Arithmetic + + + + + Compare packed signed 8-bit integers in "a" and "b", and store packed minimum values in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := j*32 - IF a[i+31:i] > b[i+31:i] - dst[i+31:i] := a[i+31:i] +FOR j := 0 to 15 + i := j*8 + IF k[j] + dst[i+7:i] := MIN(a[i+7:i], b[i+7:i]) ELSE - dst[i+31:i] := b[i+31:i] + dst[i+7:i] := 0 FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - AVX2 - Special Math Functions - - - Compare packed 8-bit integers in "a" and "b", and store packed minimum values in "dst". - + AVX512VL + AVX512BW + Arithmetic + + + + + + Compare packed signed 16-bit integers in "a" and "b", and store packed minimum values in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 31 - i := j*8 - IF a[i+7:i] < b[i+7:i] - dst[i+7:i] := a[i+7:i] +FOR j := 0 to 15 + i := j*16 + IF k[j] + dst[i+15:i] := MIN(a[i+15:i], b[i+15:i]) ELSE - dst[i+7:i] := b[i+7:i] + dst[i+15:i] := src[i+15:i] FI ENDFOR dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - AVX2 - Special Math Functions - - - Compare packed 16-bit integers in "a" and "b", and store packed minimum values in "dst". - + AVX512VL + AVX512BW + Arithmetic + + + + + Compare packed signed 16-bit integers in "a" and "b", and store packed minimum values in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). FOR j := 0 to 15 i := j*16 - IF a[i+15:i] < b[i+15:i] - dst[i+15:i] := a[i+15:i] + IF k[j] + dst[i+15:i] := MIN(a[i+15:i], b[i+15:i]) ELSE - dst[i+15:i] := b[i+15:i] + dst[i+15:i] := 0 FI ENDFOR dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - AVX2 - Special Math Functions - - - Compare packed 32-bit integers in "a" and "b", and store packed minimum values in "dst". - + AVX512BW + Arithmetic + + + + + + Compare packed signed 16-bit integers in "a" and "b", and store packed minimum values in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := j*32 - IF a[i+31:i] < b[i+31:i] - dst[i+31:i] := a[i+31:i] +FOR j := 0 to 31 + i := j*16 + IF k[j] + dst[i+15:i] := MIN(a[i+15:i], b[i+15:i]) ELSE - dst[i+31:i] := b[i+31:i] + dst[i+15:i] := src[i+15:i] FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX2 - Special Math Functions - - - Compare packed unsigned 8-bit integers in "a" and "b", and store packed minimum values in "dst". - + AVX512BW + Arithmetic + + + + + Compare packed signed 16-bit integers in "a" and "b", and store packed minimum values in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). FOR j := 0 to 31 - i := j*8 - IF a[i+7:i] < b[i+7:i] - dst[i+7:i] := a[i+7:i] + i := j*16 + IF k[j] + dst[i+15:i] := MIN(a[i+15:i], b[i+15:i]) ELSE - dst[i+7:i] := b[i+7:i] + dst[i+15:i] := 0 FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX2 - Special Math Functions - - - Compare packed unsigned 16-bit integers in "a" and "b", and store packed minimum values in "dst". - + AVX512BW + Arithmetic + + + + Compare packed signed 16-bit integers in "a" and "b", and store packed minimum values in "dst". -FOR j := 0 to 15 +FOR j := 0 to 31 i := j*16 - IF a[i+15:i] < b[i+15:i] - dst[i+15:i] := a[i+15:i] + dst[i+15:i] := MIN(a[i+15:i], b[i+15:i]) +ENDFOR +dst[MAX:512] := 0 + + +
immintrin.h
+ + + Integer + AVX512VL + AVX512BW + Arithmetic + + + + + + Compare packed signed 16-bit integers in "a" and "b", and store packed minimum values in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +FOR j := 0 to 7 + i := j*16 + IF k[j] + dst[i+15:i] := MIN(a[i+15:i], b[i+15:i]) ELSE - dst[i+15:i] := b[i+15:i] + dst[i+15:i] := src[i+15:i] FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - AVX2 - Special Math Functions - - - Compare packed unsigned 32-bit integers in "a" and "b", and store packed minimum values in "dst". - + AVX512VL + AVX512BW + Arithmetic + + + + + Compare packed signed 16-bit integers in "a" and "b", and store packed minimum values in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). FOR j := 0 to 7 - i := j*32 - IF a[i+31:i] < b[i+31:i] - dst[i+31:i] := a[i+31:i] + i := j*16 + IF k[j] + dst[i+15:i] := MIN(a[i+15:i], b[i+15:i]) ELSE - dst[i+31:i] := b[i+31:i] + dst[i+15:i] := 0 FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - AVX2 - Miscellaneous - - -Create mask from the most significant bit of each 8-bit element in "a", and store the result in "dst". - + AVX512VL + AVX512BW + Arithmetic + + + + + + Compare packed unsigned 8-bit integers in "a" and "b", and store packed minimum values in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). FOR j := 0 to 31 i := j*8 - dst[j] := a[i+7] + IF k[j] + dst[i+7:i] := MIN(a[i+7:i], b[i+7:i]) + ELSE + dst[i+7:i] := src[i+7:i] + FI ENDFOR +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - AVX2 - Miscellaneous - - - - Compute the sum of absolute differences (SADs) of quadruplets of unsigned 8-bit integers in "a" compared to those in "b", and store the 16-bit results in "dst". - Eight SADs are performed for each 128-bit lane using one quadruplet from "b" and eight quadruplets from "a". One quadruplet is selected from "b" starting at on the offset specified in "imm8". Eight quadruplets are formed from sequential 8-bit integers selected from "a" starting at the offset specified in "imm8". + AVX512VL + AVX512BW + Arithmetic + + + + + Compare packed unsigned 8-bit integers in "a" and "b", and store packed minimum values in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -MPSADBW(a[127:0], b[127:0], imm8[2:0]) { - a_offset := imm8[2]*32 - b_offset := imm8[1:0]*32 - FOR j := 0 to 7 - i := j*8 - k := a_offset+i - l := b_offset - tmp[i+15:i] := ABS(a[k+7:k] - b[l+7:l]) + ABS(a[k+15:k+8] - b[l+15:l+8]) + ABS(a[k+23:k+16] - b[l+23:l+16]) + ABS(a[k+31:k+24] - b[l+31:l+24]) - ENDFOR - RETURN tmp[127:0] -} - -dst[127:0] := MPSADBW(a[127:0], b[127:0], imm8[2:0]) -dst[255:128] := MPSADBW(a[255:128], b[255:128], imm8[5:3]) +FOR j := 0 to 31 + i := j*8 + IF k[j] + dst[i+7:i] := MIN(a[i+7:i], b[i+7:i]) + ELSE + dst[i+7:i] := 0 + FI +ENDFOR dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - AVX2 + AVX512BW Arithmetic - - - Multiply the low 32-bit integers from each packed 64-bit element in "a" and "b", and store the signed 64-bit results in "dst". + + + + + + Compare packed unsigned 8-bit integers in "a" and "b", and store packed minimum values in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*64 - dst[i+63:i] := a[i+31:i] * b[i+31:i] +FOR j := 0 to 63 + i := j*8 + IF k[j] + dst[i+7:i] := MIN(a[i+7:i], b[i+7:i]) + ELSE + dst[i+7:i] := src[i+7:i] + FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX2 + AVX512BW Arithmetic - - - Multiply the low unsigned 32-bit integers from each packed 64-bit element in "a" and "b", and store the unsigned 64-bit results in "dst". + + + + + Compare packed unsigned 8-bit integers in "a" and "b", and store packed minimum values in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*64 - dst[i+63:i] := a[i+31:i] * b[i+31:i] +FOR j := 0 to 63 + i := j*8 + IF k[j] + dst[i+7:i] := MIN(a[i+7:i], b[i+7:i]) + ELSE + dst[i+7:i] := 0 + FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX2 + AVX512BW Arithmetic - - - Multiply the packed 16-bit integers in "a" and "b", producing intermediate 32-bit integers, and store the high 16 bits of the intermediate integers in "dst". + + + + Compare packed unsigned 8-bit integers in "a" and "b", and store packed minimum values in "dst". -FOR j := 0 to 15 - i := j*16 - tmp[31:0] := a[i+15:i] * b[i+15:i] - dst[i+15:i] := tmp[31:16] +FOR j := 0 to 63 + i := j*8 + dst[i+7:i] := MIN(a[i+7:i], b[i+7:i]) ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX2 + AVX512VL + AVX512BW Arithmetic - - - Multiply the packed unsigned 16-bit integers in "a" and "b", producing intermediate 32-bit integers, and store the high 16 bits of the intermediate integers in "dst". + + + + + + Compare packed unsigned 8-bit integers in "a" and "b", and store packed minimum values in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). FOR j := 0 to 15 - i := j*16 - tmp[31:0] := a[i+15:i] * b[i+15:i] - dst[i+15:i] := tmp[31:16] + i := j*8 + IF k[j] + dst[i+7:i] := MIN(a[i+7:i], b[i+7:i]) + ELSE + dst[i+7:i] := src[i+7:i] + FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - AVX2 + AVX512VL + AVX512BW Arithmetic - - - Multiply packed 16-bit integers in "a" and "b", producing intermediate signed 32-bit integers. Truncate each intermediate integer to the 18 most significant bits, round by adding 1, and store bits [16:1] to "dst". + + + + + Compare packed unsigned 8-bit integers in "a" and "b", and store packed minimum values in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). FOR j := 0 to 15 - i := j*16 - tmp[31:0] := ((a[i+15:i] * b[i+15:i]) >> 14) + 1 - dst[i+15:i] := tmp[16:1] + i := j*8 + IF k[j] + dst[i+7:i] := MIN(a[i+7:i], b[i+7:i]) + ELSE + dst[i+7:i] := 0 + FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - AVX2 + AVX512VL + AVX512BW Arithmetic - - - Multiply the packed 16-bit integers in "a" and "b", producing intermediate 32-bit integers, and store the low 16 bits of the intermediate integers in "dst". + + + + + + Compare packed unsigned 16-bit integers in "a" and "b", and store packed minimum values in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). FOR j := 0 to 15 i := j*16 - tmp[31:0] := a[i+15:i] * b[i+15:i] - dst[i+15:i] := tmp[15:0] + IF k[j] + dst[i+15:i] := MIN(a[i+15:i], b[i+15:i]) + ELSE + dst[i+15:i] := src[i+15:i] + FI ENDFOR dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - AVX2 + AVX512VL + AVX512BW Arithmetic - - - Multiply the packed 32-bit integers in "a" and "b", producing intermediate 64-bit integers, and store the low 32 bits of the intermediate integers in "dst". + + + + + Compare packed unsigned 16-bit integers in "a" and "b", and store packed minimum values in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := j*32 - tmp[63:0] := a[i+31:i] * b[i+31:i] - dst[i+31:i] := tmp[31:0] +FOR j := 0 to 15 + i := j*16 + IF k[j] + dst[i+15:i] := MIN(a[i+15:i], b[i+15:i]) + ELSE + dst[i+15:i] := 0 + FI ENDFOR dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - AVX2 - Logical - - - Compute the bitwise OR of 256 bits (representing integer data) in "a" and "b", and store the result in "dst". + AVX512BW + Arithmetic + + + + + + Compare packed unsigned 16-bit integers in "a" and "b", and store packed minimum values in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -dst[255:0] := (a[255:0] OR b[255:0]) -dst[MAX:256] := 0 +FOR j := 0 to 31 + i := j*16 + IF k[j] + dst[i+15:i] := MIN(a[i+15:i], b[i+15:i]) + ELSE + dst[i+15:i] := src[i+15:i] + FI +ENDFOR +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - AVX2 - Miscellaneous - - - Convert packed 16-bit integers from "a" and "b" to packed 8-bit integers using signed saturation, and store the results in "dst". - - -dst[7:0] := Saturate_Int16_To_Int8 (a[15:0]) -dst[15:8] := Saturate_Int16_To_Int8 (a[31:16]) -dst[23:16] := Saturate_Int16_To_Int8 (a[47:32]) -dst[31:24] := Saturate_Int16_To_Int8 (a[63:48]) -dst[39:32] := Saturate_Int16_To_Int8 (a[79:64]) -dst[47:40] := Saturate_Int16_To_Int8 (a[95:80]) -dst[55:48] := Saturate_Int16_To_Int8 (a[111:96]) -dst[63:56] := Saturate_Int16_To_Int8 (a[127:112]) -dst[71:64] := Saturate_Int16_To_Int8 (b[15:0]) -dst[79:72] := Saturate_Int16_To_Int8 (b[31:16]) -dst[87:80] := Saturate_Int16_To_Int8 (b[47:32]) -dst[95:88] := Saturate_Int16_To_Int8 (b[63:48]) -dst[103:96] := Saturate_Int16_To_Int8 (b[79:64]) -dst[111:104] := Saturate_Int16_To_Int8 (b[95:80]) -dst[119:112] := Saturate_Int16_To_Int8 (b[111:96]) -dst[127:120] := Saturate_Int16_To_Int8 (b[127:112]) -dst[135:128] := Saturate_Int16_To_Int8 (a[143:128]) -dst[143:136] := Saturate_Int16_To_Int8 (a[159:144]) -dst[151:144] := Saturate_Int16_To_Int8 (a[175:160]) -dst[159:152] := Saturate_Int16_To_Int8 (a[191:176]) -dst[167:160] := Saturate_Int16_To_Int8 (a[207:192]) -dst[175:168] := Saturate_Int16_To_Int8 (a[223:208]) -dst[183:176] := Saturate_Int16_To_Int8 (a[239:224]) -dst[191:184] := Saturate_Int16_To_Int8 (a[255:240]) -dst[199:192] := Saturate_Int16_To_Int8 (b[143:128]) -dst[207:200] := Saturate_Int16_To_Int8 (b[159:144]) -dst[215:208] := Saturate_Int16_To_Int8 (b[175:160]) -dst[223:216] := Saturate_Int16_To_Int8 (b[191:176]) -dst[231:224] := Saturate_Int16_To_Int8 (b[207:192]) -dst[239:232] := Saturate_Int16_To_Int8 (b[223:208]) -dst[247:240] := Saturate_Int16_To_Int8 (b[239:224]) -dst[255:248] := Saturate_Int16_To_Int8 (b[255:240]) -dst[MAX:256] := 0 - - -
immintrin.h
- - + Integer - AVX2 - Miscellaneous - - - Convert packed 32-bit integers from "a" and "b" to packed 16-bit integers using signed saturation, and store the results in "dst". + AVX512BW + Arithmetic + + + + + Compare packed unsigned 16-bit integers in "a" and "b", and store packed minimum values in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -dst[15:0] := Saturate_Int32_To_Int16 (a[31:0]) -dst[31:16] := Saturate_Int32_To_Int16 (a[63:32]) -dst[47:32] := Saturate_Int32_To_Int16 (a[95:64]) -dst[63:48] := Saturate_Int32_To_Int16 (a[127:96]) -dst[79:64] := Saturate_Int32_To_Int16 (b[31:0]) -dst[95:80] := Saturate_Int32_To_Int16 (b[63:32]) -dst[111:96] := Saturate_Int32_To_Int16 (b[95:64]) -dst[127:112] := Saturate_Int32_To_Int16 (b[127:96]) -dst[143:128] := Saturate_Int32_To_Int16 (a[159:128]) -dst[159:144] := Saturate_Int32_To_Int16 (a[191:160]) -dst[175:160] := Saturate_Int32_To_Int16 (a[223:192]) -dst[191:176] := Saturate_Int32_To_Int16 (a[255:224]) -dst[207:192] := Saturate_Int32_To_Int16 (b[159:128]) -dst[223:208] := Saturate_Int32_To_Int16 (b[191:160]) -dst[239:224] := Saturate_Int32_To_Int16 (b[223:192]) -dst[255:240] := Saturate_Int32_To_Int16 (b[255:224]) -dst[MAX:256] := 0 +FOR j := 0 to 31 + i := j*16 + IF k[j] + dst[i+15:i] := MIN(a[i+15:i], b[i+15:i]) + ELSE + dst[i+15:i] := 0 + FI +ENDFOR +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - AVX2 - Miscellaneous - - - Convert packed 16-bit integers from "a" and "b" to packed 8-bit integers using unsigned saturation, and store the results in "dst". - -dst[7:0] := Saturate_Int16_To_UnsignedInt8 (a[15:0]) -dst[15:8] := Saturate_Int16_To_UnsignedInt8 (a[31:16]) -dst[23:16] := Saturate_Int16_To_UnsignedInt8 (a[47:32]) -dst[31:24] := Saturate_Int16_To_UnsignedInt8 (a[63:48]) -dst[39:32] := Saturate_Int16_To_UnsignedInt8 (a[79:64]) -dst[47:40] := Saturate_Int16_To_UnsignedInt8 (a[95:80]) -dst[55:48] := Saturate_Int16_To_UnsignedInt8 (a[111:96]) -dst[63:56] := Saturate_Int16_To_UnsignedInt8 (a[127:112]) -dst[71:64] := Saturate_Int16_To_UnsignedInt8 (b[15:0]) -dst[79:72] := Saturate_Int16_To_UnsignedInt8 (b[31:16]) -dst[87:80] := Saturate_Int16_To_UnsignedInt8 (b[47:32]) -dst[95:88] := Saturate_Int16_To_UnsignedInt8 (b[63:48]) -dst[103:96] := Saturate_Int16_To_UnsignedInt8 (b[79:64]) -dst[111:104] := Saturate_Int16_To_UnsignedInt8 (b[95:80]) -dst[119:112] := Saturate_Int16_To_UnsignedInt8 (b[111:96]) -dst[127:120] := Saturate_Int16_To_UnsignedInt8 (b[127:112]) -dst[135:128] := Saturate_Int16_To_UnsignedInt8 (a[143:128]) -dst[143:136] := Saturate_Int16_To_UnsignedInt8 (a[159:144]) -dst[151:144] := Saturate_Int16_To_UnsignedInt8 (a[175:160]) -dst[159:152] := Saturate_Int16_To_UnsignedInt8 (a[191:176]) -dst[167:160] := Saturate_Int16_To_UnsignedInt8 (a[207:192]) -dst[175:168] := Saturate_Int16_To_UnsignedInt8 (a[223:208]) -dst[183:176] := Saturate_Int16_To_UnsignedInt8 (a[239:224]) -dst[191:184] := Saturate_Int16_To_UnsignedInt8 (a[255:240]) -dst[199:192] := Saturate_Int16_To_UnsignedInt8 (b[143:128]) -dst[207:200] := Saturate_Int16_To_UnsignedInt8 (b[159:144]) -dst[215:208] := Saturate_Int16_To_UnsignedInt8 (b[175:160]) -dst[223:216] := Saturate_Int16_To_UnsignedInt8 (b[191:176]) -dst[231:224] := Saturate_Int16_To_UnsignedInt8 (b[207:192]) -dst[239:232] := Saturate_Int16_To_UnsignedInt8 (b[223:208]) -dst[247:240] := Saturate_Int16_To_UnsignedInt8 (b[239:224]) -dst[255:248] := Saturate_Int16_To_UnsignedInt8 (b[255:240]) -dst[MAX:256] := 0 - - -
immintrin.h
- - + Integer - AVX2 - Miscellaneous - - - Convert packed 32-bit integers from "a" and "b" to packed 16-bit integers using unsigned saturation, and store the results in "dst". + AVX512BW + Arithmetic + + + + Compare packed unsigned 16-bit integers in "a" and "b", and store packed minimum values in "dst". -dst[15:0] := Saturate_Int32_To_UnsignedInt16 (a[31:0]) -dst[31:16] := Saturate_Int32_To_UnsignedInt16 (a[63:32]) -dst[47:32] := Saturate_Int32_To_UnsignedInt16 (a[95:64]) -dst[63:48] := Saturate_Int32_To_UnsignedInt16 (a[127:96]) -dst[79:64] := Saturate_Int32_To_UnsignedInt16 (b[31:0]) -dst[95:80] := Saturate_Int32_To_UnsignedInt16 (b[63:32]) -dst[111:96] := Saturate_Int32_To_UnsignedInt16 (b[95:64]) -dst[127:112] := Saturate_Int32_To_UnsignedInt16 (b[127:96]) -dst[143:128] := Saturate_Int32_To_UnsignedInt16 (a[159:128]) -dst[159:144] := Saturate_Int32_To_UnsignedInt16 (a[191:160]) -dst[175:160] := Saturate_Int32_To_UnsignedInt16 (a[223:192]) -dst[191:176] := Saturate_Int32_To_UnsignedInt16 (a[255:224]) -dst[207:192] := Saturate_Int32_To_UnsignedInt16 (b[159:128]) -dst[223:208] := Saturate_Int32_To_UnsignedInt16 (b[191:160]) -dst[239:224] := Saturate_Int32_To_UnsignedInt16 (b[223:192]) -dst[255:240] := Saturate_Int32_To_UnsignedInt16 (b[255:224]) -dst[MAX:256] := 0 +FOR j := 0 to 31 + i := j*16 + dst[i+15:i] := MIN(a[i+15:i], b[i+15:i]) +ENDFOR +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX2 - Swizzle - - - - Shuffle 128-bits (composed of integer data) selected by "imm8" from "a" and "b", and store the results in "dst". + AVX512VL + AVX512BW + Arithmetic + + + + + + Compare packed unsigned 16-bit integers in "a" and "b", and store packed minimum values in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -SELECT4(src1, src2, control){ - CASE(control[1:0]) - 0: tmp[127:0] := src1[127:0] - 1: tmp[127:0] := src1[255:128] - 2: tmp[127:0] := src2[127:0] - 3: tmp[127:0] := src2[255:128] - ESAC - IF control[3] - tmp[127:0] := 0 +FOR j := 0 to 7 + i := j*16 + IF k[j] + dst[i+15:i] := MIN(a[i+15:i], b[i+15:i]) + ELSE + dst[i+15:i] := src[i+15:i] FI - RETURN tmp[127:0] -} - -dst[127:0] := SELECT4(a[255:0], b[255:0], imm8[3:0]) -dst[255:128] := SELECT4(a[255:0], b[255:0], imm8[7:4]) -dst[MAX:256] := 0 +ENDFOR +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - AVX2 - Swizzle - - - Shuffle 64-bit integers in "a" across lanes using the control in "imm8", and store the results in "dst". + AVX512VL + AVX512BW + Arithmetic + + + + + Compare packed unsigned 16-bit integers in "a" and "b", and store packed minimum values in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -SELECT4(src, control){ - CASE(control[1:0]) - 0: tmp[63:0] := src[63:0] - 1: tmp[63:0] := src[127:64] - 2: tmp[63:0] := src[191:128] - 3: tmp[63:0] := src[255:192] - ESAC - RETURN tmp[63:0] -} - -dst[63:0] := SELECT4(a[255:0], imm8[1:0]) -dst[127:64] := SELECT4(a[255:0], imm8[3:2]) -dst[191:128] := SELECT4(a[255:0], imm8[5:4]) -dst[255:192] := SELECT4(a[255:0], imm8[7:6]) -dst[MAX:256] := 0 +FOR j := 0 to 7 + i := j*16 + IF k[j] + dst[i+15:i] := MIN(a[i+15:i], b[i+15:i]) + ELSE + dst[i+15:i] := 0 + FI +ENDFOR +dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point - AVX2 - Swizzle - - - Shuffle double-precision (64-bit) floating-point elements in "a" across lanes using the control in "imm8", and store the results in "dst". + + Integer + Mask + AVX512VL + AVX512BW + Miscellaneous + + + Set each bit of mask register "k" based on the most significant bit of the corresponding packed 8-bit integer in "a". -SELECT4(src, control){ - CASE(control[1:0]) - 0: tmp[63:0] := src[63:0] - 1: tmp[63:0] := src[127:64] - 2: tmp[63:0] := src[191:128] - 3: tmp[63:0] := src[255:192] - ESAC - RETURN tmp[63:0] -} - -dst[63:0] := SELECT4(a[255:0], imm8[1:0]) -dst[127:64] := SELECT4(a[255:0], imm8[3:2]) -dst[191:128] := SELECT4(a[255:0], imm8[5:4]) -dst[255:192] := SELECT4(a[255:0], imm8[7:6]) -dst[MAX:256] := 0 +FOR j := 0 to 31 + i := j*8 + IF a[i+7] + k[j] := 1 + ELSE + k[j] := 0 + FI +ENDFOR +k[MAX:32] := 0 - +
immintrin.h
 - + Integer - AVX2 - Swizzle - - - Shuffle 32-bit integers in "a" across lanes using the corresponding index in "idx", and store the results in "dst". - + Mask + AVX512BW + Miscellaneous + + + Set each bit of mask register "k" based on the most significant bit of the corresponding packed 8-bit integer in "a". -FOR j := 0 to 7 - i := j*32 - id := idx[i+2:i]*32 - dst[i+31:i] := a[id+31:id] +FOR j := 0 to 63 + i := j*8 + IF a[i+7] + k[j] := 1 + ELSE + k[j] := 0 + FI ENDFOR -dst[MAX:256] := 0 +k[MAX:64] := 0 - +
immintrin.h
 - - Floating Point - AVX2 - Swizzle - - - Shuffle single-precision (32-bit) floating-point elements in "a" across lanes using the corresponding index in "idx". + + Integer + Mask + AVX512VL + AVX512BW + Miscellaneous + + + Set each bit of mask register "k" based on the most significant bit of the corresponding packed 8-bit integer in "a". -FOR j := 0 to 7 - i := j*32 - id := idx[i+2:i]*32 - dst[i+31:i] := a[id+31:id] +FOR j := 0 to 15 + i := j*8 + IF a[i+7] + k[j] := 1 + ELSE + k[j] := 0 + FI ENDFOR -dst[MAX:256] := 0 +k[MAX:16] := 0 - +
immintrin.h
 - + Integer - AVX2 - Arithmetic - - - Compute the absolute differences of packed unsigned 8-bit integers in "a" and "b", then horizontally sum each consecutive 8 differences to produce four unsigned 16-bit integers, and pack these unsigned 16-bit integers in the low 16 bits of 64-bit elements in "dst". + AVX512VL + AVX512BW + Miscellaneous + + + Set each packed 8-bit integer in "dst" to all ones or all zeros based on the value of the corresponding bit in "k". FOR j := 0 to 31 i := j*8 - tmp[i+7:i] := ABS(a[i+7:i] - b[i+7:i]) -ENDFOR -FOR j := 0 to 4 - i := j*64 - dst[i+15:i] := tmp[i+7:i] + tmp[i+15:i+8] + tmp[i+23:i+16] + tmp[i+31:i+24] + tmp[i+39:i+32] + tmp[i+47:i+40] + tmp[i+55:i+48] + tmp[i+63:i+56] - dst[i+63:i+16] := 0 + IF k[j] + dst[i+7:i] := 0xFF + ELSE + dst[i+7:i] := 0 + FI ENDFOR dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - AVX2 - Swizzle - - - Shuffle 32-bit integers in "a" within 128-bit lanes using the control in "imm8", and store the results in "dst". + AVX512BW + Miscellaneous + + + Set each packed 8-bit integer in "dst" to all ones or all zeros based on the value of the corresponding bit in "k". -SELECT4(src, control){ - CASE(control[1:0]) - 0: tmp[31:0] := src[31:0] - 1: tmp[31:0] := src[63:32] - 2: tmp[31:0] := src[95:64] - 3: tmp[31:0] := src[127:96] - ESAC - RETURN tmp[31:0] -} - -dst[31:0] := SELECT4(a[127:0], imm8[1:0]) -dst[63:32] := SELECT4(a[127:0], imm8[3:2]) -dst[95:64] := SELECT4(a[127:0], imm8[5:4]) -dst[127:96] := SELECT4(a[127:0], imm8[7:6]) -dst[159:128] := SELECT4(a[255:128], imm8[1:0]) -dst[191:160] := SELECT4(a[255:128], imm8[3:2]) -dst[223:192] := SELECT4(a[255:128], imm8[5:4]) -dst[255:224] := SELECT4(a[255:128], imm8[7:6]) -dst[MAX:256] := 0 +FOR j := 0 to 63 + i := j*8 + IF k[j] + dst[i+7:i] := 0xFF + ELSE + dst[i+7:i] := 0 + FI +ENDFOR +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX2 - Swizzle - - - Shuffle 8-bit integers in "a" within 128-bit lanes according to shuffle control mask in the corresponding 8-bit element of "b", and store the results in "dst". + AVX512BW + AVX512VL + Miscellaneous + + + Set each packed 8-bit integer in "dst" to all ones or all zeros based on the value of the corresponding bit in "k". FOR j := 0 to 15 i := j*8 - IF b[i+7] == 1 - dst[i+7:i] := 0 - ELSE - index[3:0] := b[i+3:i] - dst[i+7:i] := a[index*8+7:index*8] - FI - IF b[128+i+7] == 1 - dst[128+i+7:128+i] := 0 + IF k[j] + dst[i+7:i] := 0xFF ELSE - index[3:0] := b[128+i+3:128+i] - dst[128+i+7:128+i] := a[128+index*8+7:128+index*8] + dst[i+7:i] := 0 FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - AVX2 - Swizzle - - - Shuffle 16-bit integers in the high 64 bits of 128-bit lanes of "a" using the control in "imm8". Store the results in the high 64 bits of 128-bit lanes of "dst", with the low 64 bits of 128-bit lanes being copied from from "a" to "dst". + AVX512VL + AVX512BW + Miscellaneous + + + Set each packed 16-bit integer in "dst" to all ones or all zeros based on the value of the corresponding bit in "k". -dst[63:0] := a[63:0] -dst[79:64] := (a >> (imm8[1:0] * 16))[79:64] -dst[95:80] := (a >> (imm8[3:2] * 16))[79:64] -dst[111:96] := (a >> (imm8[5:4] * 16))[79:64] -dst[127:112] := (a >> (imm8[7:6] * 16))[79:64] -dst[191:128] := a[191:128] -dst[207:192] := (a >> (imm8[1:0] * 16))[207:192] -dst[223:208] := (a >> (imm8[3:2] * 16))[207:192] -dst[239:224] := (a >> (imm8[5:4] * 16))[207:192] -dst[255:240] := (a >> (imm8[7:6] * 16))[207:192] +FOR j := 0 to 15 + i := j*16 + IF k[j] + dst[i+15:i] := 0xFFFF + ELSE + dst[i+15:i] := 0 + FI +ENDFOR dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - AVX2 - Swizzle - - - Shuffle 16-bit integers in the low 64 bits of 128-bit lanes of "a" using the control in "imm8". Store the results in the low 64 bits of 128-bit lanes of "dst", with the high 64 bits of 128-bit lanes being copied from from "a" to "dst". + AVX512BW + Miscellaneous + + + Set each packed 16-bit integer in "dst" to all ones or all zeros based on the value of the corresponding bit in "k". -dst[15:0] := (a >> (imm8[1:0] * 16))[15:0] -dst[31:16] := (a >> (imm8[3:2] * 16))[15:0] -dst[47:32] := (a >> (imm8[5:4] * 16))[15:0] -dst[63:48] := (a >> (imm8[7:6] * 16))[15:0] -dst[127:64] := a[127:64] -dst[143:128] := (a >> (imm8[1:0] * 16))[143:128] -dst[159:144] := (a >> (imm8[3:2] * 16))[143:128] -dst[175:160] := (a >> (imm8[5:4] * 16))[143:128] -dst[191:176] := (a >> (imm8[7:6] * 16))[143:128] -dst[255:192] := a[255:192] -dst[MAX:256] := 0 +FOR j := 0 to 31 + i := j*16 + IF k[j] + dst[i+15:i] := 0xFFFF + ELSE + dst[i+15:i] := 0 + FI +ENDFOR +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX2 - Arithmetic - - - Negate packed 8-bit integers in "a" when the corresponding signed 8-bit integer in "b" is negative, and store the results in "dst". Element in "dst" are zeroed out when the corresponding element in "b" is zero. + AVX512VL + AVX512BW + Miscellaneous + + + Set each packed 16-bit integer in "dst" to all ones or all zeros based on the value of the corresponding bit in "k". -FOR j := 0 to 31 - i := j*8 - IF b[i+7:i] < 0 - dst[i+7:i] := NEG(a[i+7:i]) - ELSE IF b[i+7:i] = 0 - dst[i+7:i] := 0 +FOR j := 0 to 7 + i := j*16 + IF k[j] + dst[i+15:i] := 0xFFFF ELSE - dst[i+7:i] := a[i+7:i] + dst[i+15:i] := 0 FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - AVX2 - Arithmetic - - - Negate packed 16-bit integers in "a" when the corresponding signed 16-bit integer in "b" is negative, and store the results in "dst". Element in "dst" are zeroed out when the corresponding element in "b" is zero. + AVX512VL + AVX512BW + Convert + + + Convert packed signed 16-bit integers in "a" to packed 8-bit integers with signed saturation, and store the results in "dst". FOR j := 0 to 15 - i := j*16 - IF b[i+15:i] < 0 - dst[i+15:i] := NEG(a[i+15:i]) - ELSE IF b[i+15:i] = 0 - dst[i+15:i] := 0 - ELSE - dst[i+15:i] := a[i+15:i] - FI + i := 16*j + l := 8*j + dst[l+7:l] := Saturate8(a[i+15:i]) ENDFOR -dst[MAX:256] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - AVX2 - Arithmetic - - - Negate packed 32-bit integers in "a" when the corresponding signed 32-bit integer in "b" is negative, and store the results in "dst". Element in "dst" are zeroed out when the corresponding element in "b" is zero. + AVX512VL + AVX512BW + Convert + + + + + Convert packed signed 16-bit integers in "a" to packed 8-bit integers with signed saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := j*32 - IF b[i+31:i] < 0 - dst[i+31:i] := NEG(a[i+31:i]) - ELSE IF b[i+31:i] = 0 - dst[i+31:i] := 0 +FOR j := 0 to 15 + i := 16*j + l := 8*j + IF k[j] + dst[l+7:l] := Saturate8(a[i+15:i]) ELSE - dst[i+31:i] := a[i+31:i] + dst[l+7:l] := src[l+7:l] FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - AVX2 - Shift - - - Shift 128-bit lanes in "a" left by "imm8" bytes while shifting in zeros, and store the results in "dst". - -tmp := imm8[7:0] -IF tmp > 15 - tmp := 16 -FI -dst[127:0] := a[127:0] << (tmp*8) -dst[255:128] := a[255:128] << (tmp*8) -dst[MAX:256] := 0 - - -
immintrin.h
- - - Integer - AVX2 - Shift - - - Shift 128-bit lanes in "a" left by "imm8" bytes while shifting in zeros, and store the results in "dst". - -tmp := imm8[7:0] -IF tmp > 15 - tmp := 16 -FI -dst[127:0] := a[127:0] << (tmp*8) -dst[255:128] := a[255:128] << (tmp*8) -dst[MAX:256] := 0 - - -
immintrin.h
- - - Integer - AVX2 - Shift - - - Shift packed 16-bit integers in "a" left by "count" while shifting in zeros, and store the results in "dst". + AVX512VL + AVX512BW + Convert + Store + + + + + Convert packed signed 16-bit integers in "a" to packed 8-bit integers with signed saturation, and store the active results (those with their respective bit set in writemask "k") to unaligned memory at "base_addr". FOR j := 0 to 15 - i := j*16 - IF count[63:0] > 15 - dst[i+15:i] := 0 - ELSE - dst[i+15:i] := ZeroExtend(a[i+15:i] << count[63:0]) + i := 16*j + l := 8*j + IF k[j] + MEM[base_addr+l+7:base_addr+l] := Saturate8(a[i+15:i]) FI ENDFOR -dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - AVX2 - Shift - - - Shift packed 16-bit integers in "a" left by "imm8" while shifting in zeros, and store the results in "dst". + AVX512VL + AVX512BW + Convert + + + + Convert packed signed 16-bit integers in "a" to packed 8-bit integers with signed saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). FOR j := 0 to 15 - i := j*16 - IF imm8[7:0] > 15 - dst[i+15:i] := 0 + i := 16*j + l := 8*j + IF k[j] + dst[l+7:l] := Saturate8(a[i+15:i]) ELSE - dst[i+15:i] := ZeroExtend(a[i+15:i] << imm8[7:0]) + dst[l+7:l] := 0 FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - AVX2 - Shift - - - Shift packed 32-bit integers in "a" left by "count" while shifting in zeros, and store the results in "dst". + AVX512BW + Convert + + + Convert packed signed 16-bit integers in "a" to packed 8-bit integers with signed saturation, and store the results in "dst". -FOR j := 0 to 7 - i := j*32 - IF count[63:0] > 31 - dst[i+31:i] := 0 - ELSE - dst[i+31:i] := ZeroExtend(a[i+31:i] << count[63:0]) - FI +FOR j := 0 to 31 + i := 16*j + l := 8*j + dst[l+7:l] := Saturate8(a[i+15:i]) ENDFOR dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - AVX2 - Shift - - - Shift packed 32-bit integers in "a" left by "imm8" while shifting in zeros, and store the results in "dst". + AVX512BW + Convert + + + + + Convert packed signed 16-bit integers in "a" to packed 8-bit integers with signed saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := j*32 - IF imm8[7:0] > 31 - dst[i+31:i] := 0 +FOR j := 0 to 31 + i := 16*j + l := 8*j + IF k[j] + dst[l+7:l] := Saturate8(a[i+15:i]) ELSE - dst[i+31:i] := ZeroExtend(a[i+31:i] << imm8[7:0]) + dst[l+7:l] := src[l+7:l] FI ENDFOR dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - AVX2 - Shift - - - Shift packed 64-bit integers in "a" left by "count" while shifting in zeros, and store the results in "dst". + AVX512BW + Convert + Store + + + + + Convert packed signed 16-bit integers in "a" to packed 8-bit integers with signed saturation, and store the active results (those with their respective bit set in writemask "k") to unaligned memory at "base_addr". -FOR j := 0 to 3 - i := j*64 - IF count[63:0] > 63 - dst[i+63:i] := 0 - ELSE - dst[i+63:i] := ZeroExtend(a[i+63:i] << count[63:0]) +FOR j := 0 to 31 + i := 16*j + l := 8*j + IF k[j] + MEM[base_addr+l+7:base_addr+l] := Saturate8(a[i+15:i]) FI ENDFOR -dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - AVX2 - Shift - - - Shift packed 64-bit integers in "a" left by "imm8" while shifting in zeros, and store the results in "dst". + AVX512BW + Convert + + + + Convert packed signed 16-bit integers in "a" to packed 8-bit integers with signed saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*64 - IF imm8[7:0] > 63 - dst[i+63:i] := 0 +FOR j := 0 to 31 + i := 16*j + l := 8*j + IF k[j] + dst[l+7:l] := Saturate8(a[i+15:i]) ELSE - dst[i+63:i] := ZeroExtend(a[i+63:i] << imm8[7:0]) + dst[l+7:l] := 0 FI ENDFOR dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - AVX2 - Shift - - - Shift packed 32-bit integers in "a" left by the amount specified by the corresponding element in "count" while shifting in zeros, and store the results in "dst". + AVX512VL + AVX512BW + Convert + + + Convert packed signed 16-bit integers in "a" to packed 8-bit integers with signed saturation, and store the results in "dst". -FOR j := 0 to 3 - i := j*32 - dst[i+31:i] := ZeroExtend(a[i+31:i] << count[i+31:i]) +FOR j := 0 to 7 + i := 16*j + l := 8*j + dst[l+7:l] := Saturate8(a[i+15:i]) ENDFOR -dst[MAX:128] := 0 +dst[MAX:64] := 0 - +
immintrin.h
 - + Integer - AVX2 - Shift - - - Shift packed 32-bit integers in "a" left by the amount specified by the corresponding element in "count" while shifting in zeros, and store the results in "dst". + AVX512VL + AVX512BW + Convert + + + + + Convert packed signed 16-bit integers in "a" to packed 8-bit integers with signed saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). FOR j := 0 to 7 - i := j*32 - dst[i+31:i] := ZeroExtend(a[i+31:i] << count[i+31:i]) + i := 16*j + l := 8*j + IF k[j] + dst[l+7:l] := Saturate8(a[i+15:i]) + ELSE + dst[l+7:l] := src[l+7:l] + FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:64] := 0 - +
immintrin.h
 - + Integer - AVX2 - Shift - - - Shift packed 64-bit integers in "a" left by the amount specified by the corresponding element in "count" while shifting in zeros, and store the results in "dst". + AVX512VL + AVX512BW + Convert + Store + + + + + Convert packed signed 16-bit integers in "a" to packed 8-bit integers with signed saturation, and store the active results (those with their respective bit set in writemask "k") to unaligned memory at "base_addr". -FOR j := 0 to 1 - i := j*64 - dst[i+63:i] := ZeroExtend(a[i+63:i] << count[i+63:i]) +FOR j := 0 to 7 + i := 16*j + l := 8*j + IF k[j] + MEM[base_addr+l+7:base_addr+l] := Saturate8(a[i+15:i]) + FI ENDFOR -dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - AVX2 - Shift - - - Shift packed 64-bit integers in "a" left by the amount specified by the corresponding element in "count" while shifting in zeros, and store the results in "dst". + AVX512VL + AVX512BW + Convert + + + + Convert packed signed 16-bit integers in "a" to packed 8-bit integers with signed saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*64 - dst[i+63:i] := ZeroExtend(a[i+63:i] << count[i+63:i]) +FOR j := 0 to 7 + i := 16*j + l := 8*j + IF k[j] + dst[l+7:l] := Saturate8(a[i+15:i]) + ELSE + dst[l+7:l] := 0 + FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:64] := 0 - +
immintrin.h
 - + Integer - AVX2 - Shift - - - Shift packed 16-bit integers in "a" right by "count" while shifting in sign bits, and store the results in "dst". + AVX512VL + AVX512BW + Convert + + + + + Sign extend packed 8-bit integers in "a" to packed 16-bit integers, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). FOR j := 0 to 15 - i := j*16 - IF count[63:0] > 15 - dst[i+15:i] := SignBit + i := j*8 + l := j*16 + IF k[j] + dst[l+15:l] := SignExtend16(a[i+7:i]) ELSE - dst[i+15:i] := SignExtend(a[i+15:i] >> count[63:0]) + dst[l+15:l] := src[l+15:l] FI ENDFOR dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - AVX2 - Shift - - - Shift packed 16-bit integers in "a" right by "imm8" while shifting in sign bits, and store the results in "dst". + AVX512VL + AVX512BW + Convert + + + + Sign extend packed 8-bit integers in "a" to packed 16-bit integers, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). FOR j := 0 to 15 - i := j*16 - IF imm8[7:0] > 15 - dst[i+15:i] := SignBit + i := j*8 + l := j*16 + IF k[j] + dst[l+15:l] := SignExtend16(a[i+7:i]) ELSE - dst[i+15:i] := SignExtend(a[i+15:i] >> imm8[7:0]) + dst[l+15:l] := 0 FI ENDFOR dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - AVX2 - Shift - - - Shift packed 32-bit integers in "a" right by "count" while shifting in sign bits, and store the results in "dst". + AVX512BW + Convert + + + Sign extend packed 8-bit integers in "a" to packed 16-bit integers, and store the results in "dst". -FOR j := 0 to 7 - i := j*32 - IF count[63:0] > 31 - dst[i+31:i] := SignBit +FOR j := 0 to 31 + i := j*8 + l := j*16 + dst[l+15:l] := SignExtend16(a[i+7:i]) +ENDFOR +dst[MAX:512] := 0 + + +
immintrin.h
+ + + Integer + AVX512BW + Convert + + + + + Sign extend packed 8-bit integers in "a" to packed 16-bit integers, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +FOR j := 0 to 31 + i := j*8 + l := j*16 + IF k[j] + dst[l+15:l] := SignExtend16(a[i+7:i]) ELSE - dst[i+31:i] := SignExtend(a[i+31:i] >> count[63:0]) + dst[l+15:l] := src[l+15:l] FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX2 - Shift - - - Shift packed 32-bit integers in "a" right by "imm8" while shifting in sign bits, and store the results in "dst". + AVX512BW + Convert + + + + Sign extend packed 8-bit integers in "a" to packed 16-bit integers, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := j*32 - IF imm8[7:0] > 31 - dst[i+31:i] := SignBit +FOR j := 0 to 31 + i := j*8 + l := j*16 + IF k[j] + dst[l+15:l] := SignExtend16(a[i+7:i]) ELSE - dst[i+31:i] := SignExtend(a[i+31:i] >> imm8[7:0]) + dst[l+15:l] := 0 FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX2 - Shift - - - Shift packed 32-bit integers in "a" right by the amount specified by the corresponding element in "count" while shifting in sign bits, and store the results in "dst". + AVX512VL + AVX512BW + Convert + + + + + Sign extend packed 8-bit integers in "a" to packed 16-bit integers, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*32 - dst[i+31:i] := SignExtend(a[i+31:i] >> count[i+31:i]) +FOR j := 0 to 7 + i := j*8 + l := j*16 + IF k[j] + dst[l+15:l] := SignExtend16(a[i+7:i]) + ELSE + dst[l+15:l] := src[l+15:l] + FI ENDFOR dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - AVX2 - Shift - - - Shift packed 32-bit integers in "a" right by the amount specified by the corresponding element in "count" while shifting in sign bits, and store the results in "dst". + AVX512VL + AVX512BW + Convert + + + + Sign extend packed 8-bit integers in "a" to packed 16-bit integers, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). FOR j := 0 to 7 - i := j*32 - dst[i+31:i] := SignExtend(a[i+31:i] >> count[i+31:i]) + i := j*8 + l := j*16 + IF k[j] + dst[l+15:l] := SignExtend16(a[i+7:i]) + ELSE + dst[l+15:l] := 0 + FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - AVX2 - Shift - - - Shift 128-bit lanes in "a" right by "imm8" bytes while shifting in zeros, and store the results in "dst". + AVX512VL + AVX512BW + Convert + + + Convert packed unsigned 16-bit integers in "a" to packed unsigned 8-bit integers with unsigned saturation, and store the results in "dst". -tmp := imm8[7:0] -IF tmp > 15 - tmp := 16 -FI -dst[127:0] := a[127:0] >> (tmp*8) -dst[255:128] := a[255:128] >> (tmp*8) -dst[MAX:256] := 0 +FOR j := 0 to 15 + i := 16*j + l := 8*j + dst[l+7:l] := SaturateU8(a[i+15:i]) +ENDFOR +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - AVX2 - Shift - - - Shift 128-bit lanes in "a" right by "imm8" bytes while shifting in zeros, and store the results in "dst". + AVX512VL + AVX512BW + Convert + + + + + Convert packed unsigned 16-bit integers in "a" to packed unsigned 8-bit integers with unsigned saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -tmp := imm8[7:0] -IF tmp > 15 - tmp := 16 -FI -dst[127:0] := a[127:0] >> (tmp*8) -dst[255:128] := a[255:128] >> (tmp*8) -dst[MAX:256] := 0 +FOR j := 0 to 15 + i := 16*j + l := 8*j + IF k[j] + dst[l+7:l] := SaturateU8(a[i+15:i]) + ELSE + dst[l+7:l] := src[l+7:l] + FI +ENDFOR +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - AVX2 - Shift - - - Shift packed 16-bit integers in "a" right by "count" while shifting in zeros, and store the results in "dst". + AVX512VL + AVX512BW + Convert + Store + + + + + Convert packed unsigned 16-bit integers in "a" to packed unsigned 8-bit integers with unsigned saturation, and store the active results (those with their respective bit set in writemask "k") to unaligned memory at "base_addr". FOR j := 0 to 15 - i := j*16 - IF count[63:0] > 15 - dst[i+15:i] := 0 - ELSE - dst[i+15:i] := ZeroExtend(a[i+15:i] >> count[63:0]) + i := 16*j + l := 8*j + IF k[j] + MEM[base_addr+l+7:base_addr+l] := SaturateU8(a[i+15:i]) FI ENDFOR -dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - AVX2 - Shift - - - Shift packed 16-bit integers in "a" right by "imm8" while shifting in zeros, and store the results in "dst". + AVX512VL + AVX512BW + Convert + + + + Convert packed unsigned 16-bit integers in "a" to packed unsigned 8-bit integers with unsigned saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). FOR j := 0 to 15 - i := j*16 - IF imm8[7:0] > 15 - dst[i+15:i] := 0 + i := 16*j + l := 8*j + IF k[j] + dst[l+7:l] := SaturateU8(a[i+15:i]) ELSE - dst[i+15:i] := ZeroExtend(a[i+15:i] >> imm8[7:0]) + dst[l+7:l] := 0 FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - AVX2 - Shift - - - Shift packed 32-bit integers in "a" right by "count" while shifting in zeros, and store the results in "dst". + AVX512BW + Convert + + + Convert packed unsigned 16-bit integers in "a" to packed unsigned 8-bit integers with unsigned saturation, and store the results in "dst". -FOR j := 0 to 7 - i := j*32 - IF count[63:0] > 31 - dst[i+31:i] := 0 - ELSE - dst[i+31:i] := ZeroExtend(a[i+31:i] >> count[63:0]) - FI +FOR j := 0 to 31 + i := 16*j + l := 8*j + dst[l+7:l] := SaturateU8(a[i+15:i]) ENDFOR dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - AVX2 - Shift - - - Shift packed 32-bit integers in "a" right by "imm8" while shifting in zeros, and store the results in "dst". + AVX512BW + Convert + + + + + Convert packed unsigned 16-bit integers in "a" to packed unsigned 8-bit integers with unsigned saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := j*32 - IF imm8[7:0] > 31 - dst[i+31:i] := 0 +FOR j := 0 to 31 + i := 16*j + l := 8*j + IF k[j] + dst[l+7:l] := SaturateU8(a[i+15:i]) ELSE - dst[i+31:i] := ZeroExtend(a[i+31:i] >> imm8[7:0]) + dst[l+7:l] := src[l+7:l] FI ENDFOR dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - AVX2 - Shift - - - Shift packed 64-bit integers in "a" right by "count" while shifting in zeros, and store the results in "dst". + AVX512BW + Convert + Store + + + + + Convert packed unsigned 16-bit integers in "a" to packed unsigned 8-bit integers with unsigned saturation, and store the active results (those with their respective bit set in writemask "k") to unaligned memory at "base_addr". -FOR j := 0 to 3 - i := j*64 - IF count[63:0] > 63 - dst[i+63:i] := 0 - ELSE - dst[i+63:i] := ZeroExtend(a[i+63:i] >> count[63:0]) +FOR j := 0 to 31 + i := 16*j + l := 8*j + IF k[j] + MEM[base_addr+l+7:base_addr+l] := SaturateU8(a[i+15:i]) FI ENDFOR -dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - AVX2 - Shift - - - Shift packed 64-bit integers in "a" right by "imm8" while shifting in zeros, and store the results in "dst". + AVX512BW + Convert + + + + Convert packed unsigned 16-bit integers in "a" to packed unsigned 8-bit integers with unsigned saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*64 - IF imm8[7:0] > 63 - dst[i+63:i] := 0 +FOR j := 0 to 31 + i := 16*j + l := 8*j + IF k[j] + dst[l+7:l] := SaturateU8(a[i+15:i]) ELSE - dst[i+63:i] := ZeroExtend(a[i+63:i] >> imm8[7:0]) + dst[l+7:l] := 0 FI ENDFOR dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - AVX2 - Shift - - - Shift packed 32-bit integers in "a" right by the amount specified by the corresponding element in "count" while shifting in zeros, and store the results in "dst". + AVX512VL + AVX512BW + Convert + + + Convert packed unsigned 16-bit integers in "a" to packed unsigned 8-bit integers with unsigned saturation, and store the results in "dst". -FOR j := 0 to 3 - i := j*32 - dst[i+31:i] := ZeroExtend(a[i+31:i] >> count[i+31:i]) +FOR j := 0 to 7 + i := 16*j + l := 8*j + dst[l+7:l] := SaturateU8(a[i+15:i]) ENDFOR -dst[MAX:128] := 0 +dst[MAX:64] := 0 - +
immintrin.h
 - + Integer - AVX2 - Shift - - - Shift packed 32-bit integers in "a" right by the amount specified by the corresponding element in "count" while shifting in zeros, and store the results in "dst". + AVX512VL + AVX512BW + Convert + + + + + Convert packed unsigned 16-bit integers in "a" to packed unsigned 8-bit integers with unsigned saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). FOR j := 0 to 7 - i := j*32 - dst[i+31:i] := ZeroExtend(a[i+31:i] >> count[i+31:i]) + i := 16*j + l := 8*j + IF k[j] + dst[l+7:l] := SaturateU8(a[i+15:i]) + ELSE + dst[l+7:l] := src[l+7:l] + FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:64] := 0 - +
immintrin.h
 - + Integer - AVX2 - Shift - - - Shift packed 64-bit integers in "a" right by the amount specified by the corresponding element in "count" while shifting in zeros, and store the results in "dst". + AVX512VL + AVX512BW + Convert + Store + + + + + Convert packed unsigned 16-bit integers in "a" to packed unsigned 8-bit integers with unsigned saturation, and store the active results (those with their respective bit set in writemask "k") to unaligned memory at "base_addr". -FOR j := 0 to 1 - i := j*64 - dst[i+63:i] := ZeroExtend(a[i+63:i] >> count[i+63:i]) +FOR j := 0 to 7 + i := 16*j + l := 8*j + IF k[j] + MEM[base_addr+l+7:base_addr+l] := SaturateU8(a[i+15:i]) + FI ENDFOR -dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - AVX2 - Shift - - - Shift packed 64-bit integers in "a" right by the amount specified by the corresponding element in "count" while shifting in zeros, and store the results in "dst". + AVX512VL + AVX512BW + Convert + + + + Convert packed unsigned 16-bit integers in "a" to packed unsigned 8-bit integers with unsigned saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*64 - dst[i+63:i] := ZeroExtend(a[i+63:i] >> count[i+63:i]) +FOR j := 0 to 7 + i := 16*j + l := 8*j + IF k[j] + dst[l+7:l] := SaturateU8(a[i+15:i]) + ELSE + dst[l+7:l] := 0 + FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:64] := 0 - +
immintrin.h
 - + Integer - AVX2 - Load - - Load 256-bits of integer data from memory into "dst" using a non-temporal memory hint. - "mem_addr" must be aligned on a 32-byte boundary or a general-protection exception may be generated. + Mask + AVX512VL + AVX512BW + Miscellaneous + + + Set each bit of mask register "k" based on the most significant bit of the corresponding packed 16-bit integer in "a". -dst[255:0] := MEM[mem_addr+255:mem_addr] -dst[MAX:256] := 0 +FOR j := 0 to 15 + i := j*16 + IF a[i+15] + k[j] := 1 + ELSE + k[j] := 0 + FI +ENDFOR +k[MAX:16] := 0 - +
immintrin.h
 - + Integer - AVX2 - Arithmetic - - - Subtract packed 8-bit integers in "b" from packed 8-bit integers in "a", and store the results in "dst". + Mask + AVX512BW + Miscellaneous + + + Set each bit of mask register "k" based on the most significant bit of the corresponding packed 16-bit integer in "a". FOR j := 0 to 31 - i := j*8 - dst[i+7:i] := a[i+7:i] - b[i+7:i] + i := j*16 + IF a[i+15] + k[j] := 1 + ELSE + k[j] := 0 + FI ENDFOR -dst[MAX:256] := 0 +k[MAX:32] := 0 - +
immintrin.h
 - + Integer - AVX2 - Arithmetic - - - Subtract packed 16-bit integers in "b" from packed 16-bit integers in "a", and store the results in "dst". + Mask + AVX512VL + AVX512BW + Miscellaneous + + + Set each bit of mask register "k" based on the most significant bit of the corresponding packed 16-bit integer in "a". -FOR j := 0 to 15 +FOR j := 0 to 7 i := j*16 - dst[i+15:i] := a[i+15:i] - b[i+15:i] + IF a[i+15] + k[j] := 1 + ELSE + k[j] := 0 + FI ENDFOR -dst[MAX:256] := 0 +k[MAX:8] := 0 - +
immintrin.h
 - + Integer - AVX2 - Arithmetic - - - Subtract packed 32-bit integers in "b" from packed 32-bit integers in "a", and store the results in "dst". + AVX512VL + AVX512BW + Convert + + + Convert packed 16-bit integers in "a" to packed 8-bit integers with truncation, and store the results in "dst". -FOR j := 0 to 7 - i := j*32 - dst[i+31:i] := a[i+31:i] - b[i+31:i] +FOR j := 0 to 15 + i := 16*j + l := 8*j + dst[l+7:l] := Truncate8(a[i+15:i]) ENDFOR -dst[MAX:256] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - AVX2 - Arithmetic - - - Subtract packed 64-bit integers in "b" from packed 64-bit integers in "a", and store the results in "dst". + AVX512VL + AVX512BW + Convert + + + + + Convert packed 16-bit integers in "a" to packed 8-bit integers with truncation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*64 - dst[i+63:i] := a[i+63:i] - b[i+63:i] +FOR j := 0 to 15 + i := 16*j + l := 8*j + IF k[j] + dst[l+7:l] := Truncate8(a[i+15:i]) + ELSE + dst[l+7:l] := src[l+7:l] + FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - AVX2 - Arithmetic - - - Subtract packed 8-bit integers in "b" from packed 8-bit integers in "a" using saturation, and store the results in "dst". + AVX512VL + AVX512BW + Convert + Store + + + + + Convert packed 16-bit integers in "a" to packed 8-bit integers with truncation, and store the active results (those with their respective bit set in writemask "k") to unaligned memory at "base_addr". -FOR j := 0 to 31 - i := j*8 - dst[i+7:i] := Saturate_To_Int8(a[i+7:i] - b[i+7:i]) +FOR j := 0 to 15 + i := 16*j + l := 8*j + IF k[j] + MEM[base_addr+l+7:base_addr+l] := Truncate8(a[i+15:i]) + FI ENDFOR -dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - AVX2 - Arithmetic - - - Subtract packed 16-bit integers in "b" from packed 16-bit integers in "a" using saturation, and store the results in "dst". + AVX512VL + AVX512BW + Convert + + + + Convert packed 16-bit integers in "a" to packed 8-bit integers with truncation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). FOR j := 0 to 15 - i := j*16 - dst[i+15:i] := Saturate_To_Int16(a[i+15:i] - b[i+15:i]) + i := 16*j + l := 8*j + IF k[j] + dst[l+7:l] := Truncate8(a[i+15:i]) + ELSE + dst[l+7:l] := 0 + FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - AVX2 - Arithmetic - - - Subtract packed unsigned 8-bit integers in "b" from packed unsigned 8-bit integers in "a" using saturation, and store the results in "dst". + AVX512BW + Convert + + + Convert packed 16-bit integers in "a" to packed 8-bit integers with truncation, and store the results in "dst". FOR j := 0 to 31 - i := j*8 - dst[i+7:i] := Saturate_To_UnsignedInt8(a[i+7:i] - b[i+7:i]) + i := 16*j + l := 8*j + dst[l+7:l] := Truncate8(a[i+15:i]) ENDFOR dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - AVX2 - Arithmetic - - - Subtract packed unsigned 16-bit integers in "b" from packed unsigned 16-bit integers in "a" using saturation, and store the results in "dst". + AVX512BW + Convert + + + + + Convert packed 16-bit integers in "a" to packed 8-bit integers with truncation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*16 - dst[i+15:i] := Saturate_To_UnsignedInt16(a[i+15:i] - b[i+15:i]) +FOR j := 0 to 31 + i := 16*j + l := 8*j + IF k[j] + dst[l+7:l] := Truncate8(a[i+15:i]) + ELSE + dst[l+7:l] := src[l+7:l] + FI ENDFOR dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - AVX2 - Logical - - - Compute the bitwise XOR of 256 bits (representing integer data) in "a" and "b", and store the result in "dst". + AVX512BW + Convert + Store + + + + + Convert packed 16-bit integers in "a" to packed 8-bit integers with truncation, and store the active results (those with their respective bit set in writemask "k") to unaligned memory at "base_addr". -dst[255:0] := (a[255:0] XOR b[255:0]) -dst[MAX:256] := 0 +FOR j := 0 to 31 + i := 16*j + l := 8*j + IF k[j] + MEM[base_addr+l+7:base_addr+l] := Truncate8(a[i+15:i]) + FI +ENDFOR - +
immintrin.h
 - + Integer - AVX2 - Swizzle - - - Unpack and interleave 8-bit integers from the high half of each 128-bit lane in "a" and "b", and store the results in "dst". + AVX512BW + Convert + + + + Convert packed 16-bit integers in "a" to packed 8-bit integers with truncation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -INTERLEAVE_HIGH_BYTES(src1[127:0], src2[127:0]){ - dst[7:0] := src1[71:64] - dst[15:8] := src2[71:64] - dst[23:16] := src1[79:72] - dst[31:24] := src2[79:72] - dst[39:32] := src1[87:80] - dst[47:40] := src2[87:80] - dst[55:48] := src1[95:88] - dst[63:56] := src2[95:88] - dst[71:64] := src1[103:96] - dst[79:72] := src2[103:96] - dst[87:80] := src1[111:104] - dst[95:88] := src2[111:104] - dst[103:96] := src1[119:112] - dst[111:104] := src2[119:112] - dst[119:112] := src1[127:120] - dst[127:120] := src2[127:120] - RETURN dst[127:0] -} - -dst[127:0] := INTERLEAVE_HIGH_BYTES(a[127:0], b[127:0]) -dst[255:128] := INTERLEAVE_HIGH_BYTES(a[255:128], b[255:128]) +FOR j := 0 to 31 + i := 16*j + l := 8*j + IF k[j] + dst[l+7:l] := Truncate8(a[i+15:i]) + ELSE + dst[l+7:l] := 0 + FI +ENDFOR dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - AVX2 - Swizzle - - - Unpack and interleave 16-bit integers from the high half of each 128-bit lane in "a" and "b", and store the results in "dst". + AVX512VL + AVX512BW + Convert + + + Convert packed 16-bit integers in "a" to packed 8-bit integers with truncation, and store the results in "dst". -INTERLEAVE_HIGH_WORDS(src1[127:0], src2[127:0]){ - dst[15:0] := src1[79:64] - dst[31:16] := src2[79:64] - dst[47:32] := src1[95:80] - dst[63:48] := src2[95:80] - dst[79:64] := src1[111:96] - dst[95:80] := src2[111:96] - dst[111:96] := src1[127:112] - dst[127:112] := src2[127:112] - RETURN dst[127:0] -} - -dst[127:0] := INTERLEAVE_HIGH_WORDS(a[127:0], b[127:0]) -dst[255:128] := INTERLEAVE_HIGH_WORDS(a[255:128], b[255:128]) -dst[MAX:256] := 0 +FOR j := 0 to 7 + i := 16*j + l := 8*j + dst[l+7:l] := Truncate8(a[i+15:i]) +ENDFOR +dst[MAX:64] := 0 - +
immintrin.h
 - + Integer - AVX2 - Swizzle - - - Unpack and interleave 32-bit integers from the high half of each 128-bit lane in "a" and "b", and store the results in "dst". + AVX512VL + AVX512BW + Convert + + + + + Convert packed 16-bit integers in "a" to packed 8-bit integers with truncation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -INTERLEAVE_HIGH_DWORDS(src1[127:0], src2[127:0]){ - dst[31:0] := src1[95:64] - dst[63:32] := src2[95:64] - dst[95:64] := src1[127:96] - dst[127:96] := src2[127:96] - RETURN dst[127:0] -} - -dst[127:0] := INTERLEAVE_HIGH_DWORDS(a[127:0], b[127:0]) -dst[255:128] := INTERLEAVE_HIGH_DWORDS(a[255:128], b[255:128]) -dst[MAX:256] := 0 +FOR j := 0 to 7 + i := 16*j + l := 8*j + IF k[j] + dst[l+7:l] := Truncate8(a[i+15:i]) + ELSE + dst[l+7:l] := src[l+7:l] + FI +ENDFOR +dst[MAX:64] := 0 - +
immintrin.h
 - + Integer - AVX2 - Swizzle - - - Unpack and interleave 64-bit integers from the high half of each 128-bit lane in "a" and "b", and store the results in "dst". + AVX512VL + AVX512BW + Convert + Store + + + + + Convert packed 16-bit integers in "a" to packed 8-bit integers with truncation, and store the active results (those with their respective bit set in writemask "k") to unaligned memory at "base_addr". -INTERLEAVE_HIGH_QWORDS(src1[127:0], src2[127:0]){ - dst[63:0] := src1[127:64] - dst[127:64] := src2[127:64] - RETURN dst[127:0] -} - -dst[127:0] := INTERLEAVE_HIGH_QWORDS(a[127:0], b[127:0]) -dst[255:128] := INTERLEAVE_HIGH_QWORDS(a[255:128], b[255:128]) -dst[MAX:256] := 0 +FOR j := 0 to 7 + i := 16*j + l := 8*j + IF k[j] + MEM[base_addr+l+7:base_addr+l] := Truncate8(a[i+15:i]) + FI +ENDFOR - +
immintrin.h
 - + Integer - AVX2 - Swizzle - - - Unpack and interleave 8-bit integers from the low half of each 128-bit lane in "a" and "b", and store the results in "dst". + AVX512VL + AVX512BW + Convert + + + + Convert packed 16-bit integers in "a" to packed 8-bit integers with truncation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -INTERLEAVE_BYTES(src1[127:0], src2[127:0]){ - dst[7:0] := src1[7:0] - dst[15:8] := src2[7:0] - dst[23:16] := src1[15:8] - dst[31:24] := src2[15:8] - dst[39:32] := src1[23:16] - dst[47:40] := src2[23:16] - dst[55:48] := src1[31:24] - dst[63:56] := src2[31:24] - dst[71:64] := src1[39:32] - dst[79:72] := src2[39:32] - dst[87:80] := src1[47:40] - dst[95:88] := src2[47:40] - dst[103:96] := src1[55:48] - dst[111:104] := src2[55:48] - dst[119:112] := src1[63:56] - dst[127:120] := src2[63:56] - RETURN dst[127:0] -} - -dst[127:0] := INTERLEAVE_BYTES(a[127:0], b[127:0]) -dst[255:128] := INTERLEAVE_BYTES(a[255:128], b[255:128]) -dst[MAX:256] := 0 +FOR j := 0 to 7 + i := 16*j + l := 8*j + IF k[j] + dst[l+7:l] := Truncate8(a[i+15:i]) + ELSE + dst[l+7:l] := 0 + FI +ENDFOR +dst[MAX:64] := 0 - +
immintrin.h
 - + Integer - AVX2 - Swizzle - - - Unpack and interleave 16-bit integers from the low half of each 128-bit lane in "a" and "b", and store the results in "dst". + AVX512VL + AVX512BW + Convert + + + + + Zero extend packed unsigned 8-bit integers in "a" to packed 16-bit integers, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -INTERLEAVE_WORDS(src1[127:0], src2[127:0]){ - dst[15:0] := src1[15:0] - dst[31:16] := src2[15:0] - dst[47:32] := src1[31:16] - dst[63:48] := src2[31:16] - dst[79:64] := src1[47:32] - dst[95:80] := src2[47:32] - dst[111:96] := src1[63:48] - dst[127:112] := src2[63:48] - RETURN dst[127:0] -} - -dst[127:0] := INTERLEAVE_WORDS(a[127:0], b[127:0]) -dst[255:128] := INTERLEAVE_WORDS(a[255:128], b[255:128]) +FOR j := 0 to 15 + i := j*8 + l := j*16 + IF k[j] + dst[l+15:l] := ZeroExtend16(a[i+7:i]) + ELSE + dst[l+15:l] := src[l+15:l] + FI +ENDFOR dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - AVX2 - Swizzle - - - Unpack and interleave 32-bit integers from the low half of each 128-bit lane in "a" and "b", and store the results in "dst". + AVX512VL + AVX512BW + Convert + + + + Zero extend packed unsigned 8-bit integers in "a" to packed 16-bit integers, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -INTERLEAVE_DWORDS(src1[127:0], src2[127:0]){ - dst[31:0] := src1[31:0] - dst[63:32] := src2[31:0] - dst[95:64] := src1[63:32] - dst[127:96] := src2[63:32] - RETURN dst[127:0] -} - -dst[127:0] := INTERLEAVE_DWORDS(a[127:0], b[127:0]) -dst[255:128] := INTERLEAVE_DWORDS(a[255:128], b[255:128]) +FOR j := 0 to 15 + i := j*8 + l := j*16 + IF k[j] + dst[l+15:l] := ZeroExtend16(a[i+7:i]) + ELSE + dst[l+15:l] := 0 + FI +ENDFOR dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - AVX2 - Swizzle - - - Unpack and interleave 64-bit integers from the low half of each 128-bit lane in "a" and "b", and store the results in "dst". - -INTERLEAVE_QWORDS(src1[127:0], src2[127:0]){ - dst[63:0] := src1[63:0] - dst[127:64] := src2[63:0] - RETURN dst[127:0] -} - -dst[127:0] := INTERLEAVE_QWORDS(a[127:0], b[127:0]) -dst[255:128] := INTERLEAVE_QWORDS(a[255:128], b[255:128]) -dst[MAX:256] := 0 - - -
immintrin.h
- - - Floating Point - FMA - Arithmetic - - - - Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", add the intermediate result to packed elements in "c", and store the results in "dst". - + AVX512BW + Convert + + + Zero extend packed unsigned 8-bit integers in "a" to packed 16-bit integers, and store the results in "dst". -FOR j := 0 to 1 - i := j*64 - dst[i+63:i] := (a[i+63:i] * b[i+63:i]) + c[i+63:i] +FOR j := 0 to 31 + i := j*8 + l := j*16 + dst[l+15:l] := ZeroExtend16(a[i+7:i]) ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - - - +
immintrin.h
 - - Floating Point - FMA - Arithmetic - - - - Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", add the intermediate result to packed elements in "c", and store the results in "dst". - + + Integer + AVX512BW + Convert + + + + + Zero extend packed unsigned 8-bit integers in "a" to packed 16-bit integers, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*64 - dst[i+63:i] := (a[i+63:i] * b[i+63:i]) + c[i+63:i] +FOR j := 0 to 31 + i := j*8 + l := j*16 + IF k[j] + dst[l+15:l] := ZeroExtend16(a[i+7:i]) + ELSE + dst[l+15:l] := src[l+15:l] + FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - - - +
immintrin.h
 - - Floating Point - FMA - Arithmetic - - - - Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", add the intermediate result to packed elements in "c", and store the results in "dst". - + + Integer + AVX512BW + Convert + + + + Zero extend packed unsigned 8-bit integers in "a" to packed 16-bit integers, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*32 - dst[i+31:i] := (a[i+31:i] * b[i+31:i]) + c[i+31:i] +FOR j := 0 to 31 + i := j*8 + l := j*16 + IF k[j] + dst[l+15:l] := ZeroExtend16(a[i+7:i]) + ELSE + dst[l+15:l] := 0 + FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - - - +
immintrin.h
 - - Floating Point - FMA - Arithmetic - - - - Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", add the intermediate result to packed elements in "c", and store the results in "dst". - + + Integer + AVX512VL + AVX512BW + Convert + + + + + Zero extend packed unsigned 8-bit integers in "a" to packed 16-bit integers, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). FOR j := 0 to 7 - i := j*32 - dst[i+31:i] := (a[i+31:i] * b[i+31:i]) + c[i+31:i] + i := j*8 + l := j*16 + IF k[j] + dst[l+15:l] := ZeroExtend16(a[i+7:i]) + ELSE + dst[l+15:l] := src[l+15:l] + FI ENDFOR -dst[MAX:256] := 0 - - - - -
immintrin.h
- - - Floating Point - FMA - Arithmetic - - - - Multiply the lower double-precision (64-bit) floating-point elements in "a" and "b", and add the intermediate result to the lower element in "c". Store the result in the lower element of "dst", and copy the upper element from "a" to the upper element of "dst". - - -dst[63:0] := (a[63:0] * b[63:0]) + c[63:0] -dst[127:64] := a[127:64] dst[MAX:128] := 0 - - - +
immintrin.h
 - - Floating Point - FMA - Arithmetic - - - - Multiply the lower single-precision (32-bit) floating-point elements in "a" and "b", and add the intermediate result to the lower element in "c". Store the result in the lower element of "dst", and copy the upper 3 packed elements from "a" to the upper elements of "dst". - + + Integer + AVX512VL + AVX512BW + Convert + + + + Zero extend packed unsigned 8-bit integers in "a" to packed 16-bit integers, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -dst[31:0] := (a[31:0] * b[31:0]) + c[31:0] -dst[127:32] := a[127:32] +FOR j := 0 to 7 + i := j*8 + l := j*16 + IF k[j] + dst[l+15:l] := ZeroExtend16(a[i+7:i]) + ELSE + dst[l+15:l] := 0 + FI +ENDFOR dst[MAX:128] := 0 - - - +
immintrin.h
 - - Floating Point - FMA + + Integer + AVX512VL + AVX512BW Arithmetic - - - - Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", alternatively add and subtract packed elements in "c" to/from the intermediate result, and store the results in "dst". - + + + + + + Multiply packed signed 16-bit integers in "a" and "b", producing intermediate signed 32-bit integers. Truncate each intermediate integer to the 18 most significant bits, round by adding 1, and store bits [16:1] to "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 1 - i := j*64 - IF (j is even) - dst[i+63:i] := (a[i+63:i] * b[i+63:i]) - c[i+63:i] +FOR j := 0 to 15 + i := j*16 + IF k[j] + tmp[31:0] := ((SignExtend32(a[i+15:i]) * SignExtend32(b[i+15:i])) >> 14) + 1 + dst[i+15:i] := tmp[16:1] ELSE - dst[i+63:i] := (a[i+63:i] * b[i+63:i]) + c[i+63:i] + dst[i+15:i] := src[i+15:i] FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:256] := 0 - - - +
immintrin.h
 - - Floating Point - FMA + + Integer + AVX512VL + AVX512BW Arithmetic - - - - Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", alternatively add and subtract packed elements in "c" to/from the intermediate result, and store the results in "dst". - + + + + + Multiply packed signed 16-bit integers in "a" and "b", producing intermediate signed 32-bit integers. Truncate each intermediate integer to the 18 most significant bits, round by adding 1, and store bits [16:1] to "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*64 - IF (j is even) - dst[i+63:i] := (a[i+63:i] * b[i+63:i]) - c[i+63:i] +FOR j := 0 to 15 + i := j*16 + IF k[j] + tmp[31:0] := ((SignExtend32(a[i+15:i]) * SignExtend32(b[i+15:i])) >> 14) + 1 + dst[i+15:i] := tmp[16:1] ELSE - dst[i+63:i] := (a[i+63:i] * b[i+63:i]) + c[i+63:i] + dst[i+15:i] := 0 FI ENDFOR dst[MAX:256] := 0 - - - +
immintrin.h
 - - Floating Point - FMA + + Integer + AVX512BW Arithmetic - - - - Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", alternatively add and subtract packed elements in "c" to/from the intermediate result, and store the results in "dst". - + + + + + + Multiply packed signed 16-bit integers in "a" and "b", producing intermediate signed 32-bit integers. Truncate each intermediate integer to the 18 most significant bits, round by adding 1, and store bits [16:1] to "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*32 - IF (j is even) - dst[i+31:i] := (a[i+31:i] * b[i+31:i]) - c[i+31:i] +FOR j := 0 to 31 + i := j*16 + IF k[j] + tmp[31:0] := ((SignExtend32(a[i+15:i]) * SignExtend32(b[i+15:i])) >> 14) + 1 + dst[i+15:i] := tmp[16:1] ELSE - dst[i+31:i] := (a[i+31:i] * b[i+31:i]) + c[i+31:i] + dst[i+15:i] := src[i+15:i] FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - - - +
immintrin.h
 - - Floating Point - FMA + + Integer + AVX512BW Arithmetic - - - - Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", alternatively add and subtract packed elements in "c" to/from the intermediate result, and store the results in "dst". - + + + + + Multiply packed signed 16-bit integers in "a" and "b", producing intermediate signed 32-bit integers. Truncate each intermediate integer to the 18 most significant bits, round by adding 1, and store bits [16:1] to "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := j*32 - IF (j is even) - dst[i+31:i] := (a[i+31:i] * b[i+31:i]) - c[i+31:i] +FOR j := 0 to 31 + i := j*16 + IF k[j] + tmp[31:0] := ((SignExtend32(a[i+15:i]) * SignExtend32(b[i+15:i])) >> 14) + 1 + dst[i+15:i] := tmp[16:1] ELSE - dst[i+31:i] := (a[i+31:i] * b[i+31:i]) + c[i+31:i] + dst[i+15:i] := 0 FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - - - +
immintrin.h
 - - Floating Point - FMA + + Integer + AVX512BW Arithmetic - - - - Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", subtract packed elements in "c" from the intermediate result, and store the results in "dst". + + + + Multiply packed signed 16-bit integers in "a" and "b", producing intermediate signed 32-bit integers. Truncate each intermediate integer to the 18 most significant bits, round by adding 1, and store bits [16:1] to "dst". -FOR j := 0 to 1 - i := j*64 - dst[i+63:i] := (a[i+63:i] * b[i+63:i]) - c[i+63:i] +FOR j := 0 to 31 + i := j*16 + tmp[31:0] := ((SignExtend32(a[i+15:i]) * SignExtend32(b[i+15:i])) >> 14) + 1 + dst[i+15:i] := tmp[16:1] ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - - - +
immintrin.h
 - - Floating Point - FMA + + Integer + AVX512VL + AVX512BW Arithmetic - - - - Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", subtract packed elements in "c" from the intermediate result, and store the results in "dst". + + + + + + Multiply packed signed 16-bit integers in "a" and "b", producing intermediate signed 32-bit integers. Truncate each intermediate integer to the 18 most significant bits, round by adding 1, and store bits [16:1] to "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*64 - dst[i+63:i] := (a[i+63:i] * b[i+63:i]) - c[i+63:i] +FOR j := 0 to 7 + i := j*16 + IF k[j] + tmp[31:0] := ((SignExtend32(a[i+15:i]) * SignExtend32(b[i+15:i])) >> 14) + 1 + dst[i+15:i] := tmp[16:1] + ELSE + dst[i+15:i] := src[i+15:i] + FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:128] := 0 - - - +
immintrin.h
 - - Floating Point - FMA + + Integer + AVX512VL + AVX512BW Arithmetic - - - - Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", subtract packed elements in "c" from the intermediate result, and store the results in "dst". - + + + + + Multiply packed signed 16-bit integers in "a" and "b", producing intermediate signed 32-bit integers. Truncate each intermediate integer to the 18 most significant bits, round by adding 1, and store bits [16:1] to "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*32 - dst[i+31:i] := (a[i+31:i] * b[i+31:i]) - c[i+31:i] +FOR j := 0 to 7 + i := j*16 + IF k[j] + tmp[31:0] := ((SignExtend32(a[i+15:i]) * SignExtend32(b[i+15:i])) >> 14) + 1 + dst[i+15:i] := tmp[16:1] + ELSE + dst[i+15:i] := 0 + FI ENDFOR dst[MAX:128] := 0 - - - +
immintrin.h
 - - Floating Point - FMA + + Integer + AVX512VL + AVX512BW Arithmetic - - - - Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", subtract packed elements in "c" from the intermediate result, and store the results in "dst". - + + + + + + Multiply the packed unsigned 16-bit integers in "a" and "b", producing intermediate 32-bit integers, and store the high 16 bits of the intermediate integers in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := j*32 - dst[i+31:i] := (a[i+31:i] * b[i+31:i]) - c[i+31:i] +FOR j := 0 to 15 + i := j*16 + IF k[j] + tmp[31:0] := a[i+15:i] * b[i+15:i] + dst[i+15:i] := tmp[31:16] + ELSE + dst[i+15:i] := src[i+15:i] + FI ENDFOR dst[MAX:256] := 0 - - - +
immintrin.h
 - - Floating Point - FMA + + Integer + AVX512VL + AVX512BW Arithmetic - - - - Multiply the lower double-precision (64-bit) floating-point elements in "a" and "b", and subtract the lower element in "c" from the intermediate result. Store the result in the lower element of "dst", and copy the upper element from "a" to the upper element of "dst". + + + + + Multiply the packed unsigned 16-bit integers in "a" and "b", producing intermediate 32-bit integers, and store the high 16 bits of the intermediate integers in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -dst[63:0] := (a[63:0] * b[63:0]) - c[63:0] -dst[127:64] := a[127:64] -dst[MAX:128] := 0 +FOR j := 0 to 15 + i := j*16 + IF k[j] + tmp[31:0] := a[i+15:i] * b[i+15:i] + dst[i+15:i] := tmp[31:16] + ELSE + dst[i+15:i] := 0 + FI +ENDFOR +dst[MAX:256] := 0 - - - +
immintrin.h
 - - Floating Point - FMA + + Integer + AVX512BW Arithmetic - - - - Multiply the lower single-precision (32-bit) floating-point elements in "a" and "b", and subtract the lower element in "c" from the intermediate result. Store the result in the lower element of "dst", and copy the upper 3 packed elements from "a" to the upper elements of "dst". + + + + + + Multiply the packed unsigned 16-bit integers in "a" and "b", producing intermediate 32-bit integers, and store the high 16 bits of the intermediate integers in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -dst[31:0] := (a[31:0] * b[31:0]) - c[31:0] -dst[127:32] := a[127:32] -dst[MAX:128] := 0 +FOR j := 0 to 31 + i := j*16 + IF k[j] + tmp[31:0] := a[i+15:i] * b[i+15:i] + dst[i+15:i] := tmp[31:16] + ELSE + dst[i+15:i] := src[i+15:i] + FI +ENDFOR +dst[MAX:512] := 0 - - - +
immintrin.h
 - - Floating Point - FMA + + Integer + AVX512BW Arithmetic - - - - Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", alternatively subtract and add packed elements in "c" from/to the intermediate result, and store the results in "dst". - + + + + + Multiply the packed unsigned 16-bit integers in "a" and "b", producing intermediate 32-bit integers, and store the high 16 bits of the intermediate integers in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 1 - i := j*64 - IF (j is even) - dst[i+63:i] := (a[i+63:i] * b[i+63:i]) + c[i+63:i] +FOR j := 0 to 31 + i := j*16 + IF k[j] + tmp[31:0] := a[i+15:i] * b[i+15:i] + dst[i+15:i] := tmp[31:16] ELSE - dst[i+63:i] := (a[i+63:i] * b[i+63:i]) - c[i+63:i] + dst[i+15:i] := 0 FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - - - +
immintrin.h
 - - Floating Point - FMA + + Integer + AVX512BW Arithmetic - - - - Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", alternatively subtract and add packed elements in "c" from/to the intermediate result, and store the results in "dst". - + + + + Multiply the packed unsigned 16-bit integers in "a" and "b", producing intermediate 32-bit integers, and store the high 16 bits of the intermediate integers in "dst". -FOR j := 0 to 3 - i := j*64 - IF (j is even) - dst[i+63:i] := (a[i+63:i] * b[i+63:i]) + c[i+63:i] - ELSE - dst[i+63:i] := (a[i+63:i] * b[i+63:i]) - c[i+63:i] - FI +FOR j := 0 to 31 + i := j*16 + tmp[31:0] := a[i+15:i] * b[i+15:i] + dst[i+15:i] := tmp[31:16] ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - - - +
immintrin.h
 - - Floating Point - FMA + + Integer + AVX512VL + AVX512BW Arithmetic - - - - Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", alternatively subtract and add packed elements in "c" from/to the intermediate result, and store the results in "dst". - + + + + + + Multiply the packed unsigned 16-bit integers in "a" and "b", producing intermediate 32-bit integers, and store the high 16 bits of the intermediate integers in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*32 - IF (j is even) - dst[i+31:i] := (a[i+31:i] * b[i+31:i]) + c[i+31:i] +FOR j := 0 to 7 + i := j*16 + IF k[j] + tmp[31:0] := a[i+15:i] * b[i+15:i] + dst[i+15:i] := tmp[31:16] ELSE - dst[i+31:i] := (a[i+31:i] * b[i+31:i]) - c[i+31:i] + dst[i+15:i] := src[i+15:i] FI ENDFOR dst[MAX:128] := 0 - - - +
immintrin.h
 - - Floating Point - FMA + + Integer + AVX512VL + AVX512BW Arithmetic - - - - Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", alternatively subtract and add packed elements in "c" from/to the intermediate result, and store the results in "dst". - + + + + + Multiply the packed unsigned 16-bit integers in "a" and "b", producing intermediate 32-bit integers, and store the high 16 bits of the intermediate integers in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). FOR j := 0 to 7 - i := j*32 - IF (j is even) - dst[i+31:i] := (a[i+31:i] * b[i+31:i]) + c[i+31:i] + i := j*16 + IF k[j] + tmp[31:0] := a[i+15:i] * b[i+15:i] + dst[i+15:i] := tmp[31:16] ELSE - dst[i+31:i] := (a[i+31:i] * b[i+31:i]) - c[i+31:i] + dst[i+15:i] := 0 FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:128] := 0 - - - +
immintrin.h
 - - Floating Point - FMA + + Integer + AVX512VL + AVX512BW Arithmetic - - - - Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", add the negated intermediate result to packed elements in "c", and store the results in "dst". - + + + + + + Multiply the packed signed 16-bit integers in "a" and "b", producing intermediate 32-bit integers, and store the high 16 bits of the intermediate integers in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 1 - i := j*64 - dst[i+63:i] := -(a[i+63:i] * b[i+63:i]) + c[i+63:i] -ENDFOR -dst[MAX:128] := 0 +FOR j := 0 to 15 + i := j*16 + IF k[j] + tmp[31:0] := SignExtend32(a[i+15:i]) * SignExtend32(b[i+15:i]) + dst[i+15:i] := tmp[31:16] + ELSE + dst[i+15:i] := src[i+15:i] + FI +ENDFOR +dst[MAX:256] := 0 - - - +
immintrin.h
 - - Floating Point - FMA + + Integer + AVX512VL + AVX512BW Arithmetic - - - - Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", add the negated intermediate result to packed elements in "c", and store the results in "dst". - + + + + + Multiply the packed signed 16-bit integers in "a" and "b", producing intermediate 32-bit integers, and store the high 16 bits of the intermediate integers in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*64 - dst[i+63:i] := -(a[i+63:i] * b[i+63:i]) + c[i+63:i] -ENDFOR +FOR j := 0 to 15 + i := j*16 + IF k[j] + tmp[31:0] := SignExtend32(a[i+15:i]) * SignExtend32(b[i+15:i]) + dst[i+15:i] := tmp[31:16] + ELSE + dst[i+15:i] := 0 + FI +ENDFOR dst[MAX:256] := 0 - - - +
immintrin.h
 - - Floating Point - FMA + + Integer + AVX512BW Arithmetic - - - - Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", add the negated intermediate result to packed elements in "c", and store the results in "dst". - + + + + + + Multiply the packed signed 16-bit integers in "a" and "b", producing intermediate 32-bit integers, and store the high 16 bits of the intermediate integers in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*32 - a[i+31:i] := -(a[i+31:i] * b[i+31:i]) + c[i+31:i] -ENDFOR -dst[MAX:128] := 0 +FOR j := 0 to 31 + i := j*16 + IF k[j] + tmp[31:0] := SignExtend32(a[i+15:i]) * SignExtend32(b[i+15:i]) + dst[i+15:i] := tmp[31:16] + ELSE + dst[i+15:i] := src[i+15:i] + FI +ENDFOR +dst[MAX:512] := 0 - - - +
immintrin.h
 - - Floating Point - FMA + + Integer + AVX512BW Arithmetic - - - - Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", add the negated intermediate result to packed elements in "c", and store the results in "dst". - + + + + + Multiply the packed signed 16-bit integers in "a" and "b", producing intermediate 32-bit integers, and store the high 16 bits of the intermediate integers in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := j*32 - a[i+31:i] := -(a[i+31:i] * b[i+31:i]) + c[i+31:i] -ENDFOR -dst[MAX:256] := 0 +FOR j := 0 to 31 + i := j*16 + IF k[j] + tmp[31:0] := SignExtend32(a[i+15:i]) * SignExtend32(b[i+15:i]) + dst[i+15:i] := tmp[31:16] + ELSE + dst[i+15:i] := 0 + FI +ENDFOR +dst[MAX:512] := 0 - - - +
immintrin.h
 - - Floating Point - FMA + + Integer + AVX512BW Arithmetic - - - - Multiply the lower double-precision (64-bit) floating-point elements in "a" and "b", and add the negated intermediate result to the lower element in "c". Store the result in the lower element of "dst", and copy the upper element from "a" to the upper element of "dst". + + + + Multiply the packed signed 16-bit integers in "a" and "b", producing intermediate 32-bit integers, and store the high 16 bits of the intermediate integers in "dst". -dst[63:0] := -(a[63:0] * b[63:0]) + c[63:0] -dst[127:64] := a[127:64] -dst[MAX:128] := 0 +FOR j := 0 to 31 + i := j*16 + tmp[31:0] := SignExtend32(a[i+15:i]) * SignExtend32(b[i+15:i]) + dst[i+15:i] := tmp[31:16] +ENDFOR +dst[MAX:512] := 0 - - - +
immintrin.h
 - - Floating Point - FMA + + Integer + AVX512VL + AVX512BW Arithmetic - - - - Multiply the lower single-precision (32-bit) floating-point elements in "a" and "b", and add the negated intermediate result to the lower element in "c". Store the result in the lower element of "dst", and copy the upper 3 packed elements from "a" to the upper elements of "dst". + + + + + + Multiply the packed signed 16-bit integers in "a" and "b", producing intermediate 32-bit integers, and store the high 16 bits of the intermediate integers in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -dst[31:0] := -(a[31:0] * b[31:0]) + c[31:0] -dst[127:32] := a[127:32] +FOR j := 0 to 7 + i := j*16 + IF k[j] + tmp[31:0] := SignExtend32(a[i+15:i]) * SignExtend32(b[i+15:i]) + dst[i+15:i] := tmp[31:16] + ELSE + dst[i+15:i] := src[i+15:i] + FI +ENDFOR dst[MAX:128] := 0 - - - +
immintrin.h
 - - Floating Point - FMA + + Integer + AVX512VL + AVX512BW Arithmetic - - - - Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", subtract packed elements in "c" from the negated intermediate result, and store the results in "dst". + + + + + Multiply the packed signed 16-bit integers in "a" and "b", producing intermediate 32-bit integers, and store the high 16 bits of the intermediate integers in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 1 - i := j*64 - dst[i+63:i] := -(a[i+63:i] * b[i+63:i]) - c[i+63:i] -ENDFOR +FOR j := 0 to 7 + i := j*16 + IF k[j] + tmp[31:0] := SignExtend32(a[i+15:i]) * SignExtend32(b[i+15:i]) + dst[i+15:i] := tmp[31:16] + ELSE + dst[i+15:i] := 0 + FI +ENDFOR dst[MAX:128] := 0 - - - +
immintrin.h
 - - Floating Point - FMA + + Integer + AVX512VL + AVX512BW Arithmetic - - - - Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", subtract packed elements in "c" from the negated intermediate result, and store the results in "dst". + + + + + + Multiply the packed 16-bit integers in "a" and "b", producing intermediate 32-bit integers, and store the low 16 bits of the intermediate integers in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*64 - dst[i+63:i] := -(a[i+63:i] * b[i+63:i]) - c[i+63:i] -ENDFOR +FOR j := 0 to 15 + i := j*16 + IF k[j] + tmp[31:0] := SignExtend32(a[i+15:i]) * SignExtend32(b[i+15:i]) + dst[i+15:i] := tmp[15:0] + ELSE + dst[i+15:i] := src[i+15:i] + FI +ENDFOR dst[MAX:256] := 0 - - - +
immintrin.h
 - - Floating Point - FMA + + Integer + AVX512VL + AVX512BW Arithmetic - - - - Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", subtract packed elements in "c" from the negated intermediate result, and store the results in "dst". + + + + + Multiply the packed 16-bit integers in "a" and "b", producing intermediate 32-bit integers, and store the low 16 bits of the intermediate integers in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*32 - dst[i+31:i] := -(a[i+31:i] * b[i+31:i]) - c[i+31:i] -ENDFOR -dst[MAX:128] := 0 +FOR j := 0 to 15 + i := j*16 + IF k[j] + tmp[31:0] := SignExtend32(a[i+15:i]) * SignExtend32(b[i+15:i]) + dst[i+15:i] := tmp[15:0] + ELSE + dst[i+15:i] := 0 + FI +ENDFOR +dst[MAX:256] := 0 - - - +
immintrin.h
 - - Floating Point - FMA + + Integer + AVX512BW Arithmetic - - - - Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", subtract packed elements in "c" from the negated intermediate result, and store the results in "dst". + + + + + + Multiply the packed 16-bit integers in "a" and "b", producing intermediate 32-bit integers, and store the low 16 bits of the intermediate integers in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := j*32 - dst[i+31:i] := -(a[i+31:i] * b[i+31:i]) - c[i+31:i] -ENDFOR -dst[MAX:256] := 0 +FOR j := 0 to 31 + i := j*16 + IF k[j] + tmp[31:0] := SignExtend32(a[i+15:i]) * SignExtend32(b[i+15:i]) + dst[i+15:i] := tmp[15:0] + ELSE + dst[i+15:i] := src[i+15:i] + FI +ENDFOR +dst[MAX:512] := 0 - - - +
immintrin.h
 - - Floating Point - FMA + + Integer + AVX512BW Arithmetic - - - - Multiply the lower double-precision (64-bit) floating-point elements in "a" and "b", and subtract the lower element in "c" from the negated intermediate result. Store the result in the lower element of "dst", and copy the upper element from "a" to the upper element of "dst". + + + + + Multiply the packed 16-bit integers in "a" and "b", producing intermediate 32-bit integers, and store the low 16 bits of the intermediate integers in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -dst[63:0] := -(a[63:0] * b[63:0]) - c[63:0] -dst[127:64] := a[127:64] -dst[MAX:128] := 0 +FOR j := 0 to 31 + i := j*16 + IF k[j] + tmp[31:0] := SignExtend32(a[i+15:i]) * SignExtend32(b[i+15:i]) + dst[i+15:i] := tmp[15:0] + ELSE + dst[i+15:i] := 0 + FI +ENDFOR +dst[MAX:512] := 0 - - - - b +
immintrin.h
 - - Floating Point - FMA + + Integer + AVX512BW Arithmetic - - - - Multiply the lower single-precision (32-bit) floating-point elements in "a" and "b", and subtract the lower element in "c" from the negated intermediate result. Store the result in the lower element of "dst", and copy the upper 3 packed elements from "a" to the upper elements of "dst". + + + + Multiply the packed 16-bit integers in "a" and "b", producing intermediate 32-bit integers, and store the low 16 bits of the intermediate integers in "dst". -dst[31:0] := -(a[31:0] * b[31:0]) - c[31:0] -dst[127:32] := a[127:32] -dst[MAX:128] := 0 +FOR j := 0 to 31 + i := j*16 + tmp[31:0] := SignExtend32(a[i+15:i]) * SignExtend32(b[i+15:i]) + dst[i+15:i] := tmp[15:0] +ENDFOR +dst[MAX:512] := 0 - - - +
immintrin.h
 - + Integer - BMI1 - Bit Manipulation - - - - Extract contiguous bits from unsigned 32-bit integer "a", and store the result in "dst". Extract the number of bits specified by "len", starting at the bit specified by "start". + AVX512VL + AVX512BW + Arithmetic + + + + + + Multiply the packed 16-bit integers in "a" and "b", producing intermediate 32-bit integers, and store the low 16 bits of the intermediate integers in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -tmp := ZERO_EXTEND_TO_512(a) -dst := ZERO_EXTEND(tmp[start+len-1:start]) +FOR j := 0 to 7 + i := j*16 + IF k[j] + tmp[31:0] := SignExtend32(a[i+15:i]) * SignExtend32(b[i+15:i]) + dst[i+15:i] := tmp[15:0] + ELSE + dst[i+15:i] := src[i+15:i] + FI +ENDFOR +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - BMI1 - Bit Manipulation - - - Extract contiguous bits from unsigned 32-bit integer "a", and store the result in "dst". Extract the number of bits specified by bits 15:8 of "control", starting at the bit specified by bits 0:7 of "control". + AVX512VL + AVX512BW + Arithmetic + + + + + Multiply the packed 16-bit integers in "a" and "b", producing intermediate 32-bit integers, and store the low 16 bits of the intermediate integers in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -start := control[7:0] -len := control[15:8] -tmp := ZERO_EXTEND_TO_512(a) -dst := ZERO_EXTEND(tmp[start+len-1:start]) +FOR j := 0 to 7 + i := j*16 + IF k[j] + tmp[31:0] := SignExtend32(a[i+15:i]) * SignExtend32(b[i+15:i]) + dst[i+15:i] := tmp[15:0] + ELSE + dst[i+15:i] := 0 + FI +ENDFOR +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - BMI1 - Bit Manipulation - - - - Extract contiguous bits from unsigned 64-bit integer "a", and store the result in "dst". Extract the number of bits specified by "len", starting at the bit specified by "start". + AVX512BW + Miscellaneous + + + + Compute the absolute differences of packed unsigned 8-bit integers in "a" and "b", then horizontally sum each consecutive 8 differences to produce eight unsigned 16-bit integers, and pack these unsigned 16-bit integers in the low 16 bits of 64-bit elements in "dst". -tmp := ZERO_EXTEND_TO_512(a) -dst := ZERO_EXTEND(tmp[start+len-1:start]) +FOR j := 0 to 63 + i := j*8 + tmp[i+7:i] := ABS(a[i+7:i] - b[i+7:i]) +ENDFOR +FOR j := 0 to 7 + i := j*64 + dst[i+15:i] := tmp[i+7:i] + tmp[i+15:i+8] + tmp[i+23:i+16] + tmp[i+31:i+24] + \ + tmp[i+39:i+32] + tmp[i+47:i+40] + tmp[i+55:i+48] + tmp[i+63:i+56] + dst[i+63:i+16] := 0 +ENDFOR +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - BMI1 - Bit Manipulation - - - Extract contiguous bits from unsigned 64-bit integer "a", and store the result in "dst". Extract the number of bits specified by bits 15:8 of "control", starting at the bit specified by bits 0:7 of "control".. + AVX512VL + AVX512BW + Miscellaneous + + + + + + Shuffle packed 8-bit integers in "a" according to shuffle control mask in the corresponding 8-bit element of "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -start := control[7:0] -len := control[15:8] -tmp := ZERO_EXTEND_TO_512(a) -dst := ZERO_EXTEND(tmp[start+len-1:start]) +FOR j := 0 to 31 + i := j*8 + IF k[j] + IF b[i+7] == 1 + dst[i+7:i] := 0 + ELSE + index[4:0] := b[i+3:i] + (j & 0x10) + dst[i+7:i] := a[index*8+7:index*8] + FI + ELSE + dst[i+7:i] := src[i+7:i] + FI +ENDFOR +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - BMI1 - Bit Manipulation - - Extract the lowest set bit from unsigned 32-bit integer "a" and set the corresponding bit in "dst". All other bits in "dst" are zeroed, and all bits are zeroed if no bits are set in "a". + AVX512VL + AVX512BW + Miscellaneous + + + + + Shuffle packed 8-bit integers in "a" according to shuffle control mask in the corresponding 8-bit element of "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -dst := (-a) BITWISE AND a +FOR j := 0 to 31 + i := j*8 + IF k[j] + IF b[i+7] == 1 + dst[i+7:i] := 0 + ELSE + index[4:0] := b[i+3:i] + (j & 0x10) + dst[i+7:i] := a[index*8+7:index*8] + FI + ELSE + dst[i+7:i] := 0 + FI +ENDFOR +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - BMI1 - Bit Manipulation - - Extract the lowest set bit from unsigned 64-bit integer "a" and set the corresponding bit in "dst". All other bits in "dst" are zeroed, and all bits are zeroed if no bits are set in "a". + AVX512BW + Miscellaneous + + + + + + Shuffle 8-bit integers in "a" within 128-bit lanes using the control in the corresponding 8-bit element of "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -dst := (-a) BITWISE AND a +FOR j := 0 to 63 + i := j*8 + IF k[j] + IF b[i+7] == 1 + dst[i+7:i] := 0 + ELSE + index[5:0] := b[i+3:i] + (j & 0x30) + dst[i+7:i] := a[index*8+7:index*8] + FI + ELSE + dst[i+7:i] := src[i+7:i] + FI +ENDFOR +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - BMI1 - Bit Manipulation - - Set all the lower bits of "dst" up to and including the lowest set bit in unsigned 32-bit integer "a". + AVX512BW + Miscellaneous + + + + + Shuffle packed 8-bit integers in "a" according to shuffle control mask in the corresponding 8-bit element of "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -dst := (a - 1) XOR a +FOR j := 0 to 63 + i := j*8 + IF k[j] + IF b[i+7] == 1 + dst[i+7:i] := 0 + ELSE + index[5:0] := b[i+3:i] + (j & 0x30) + dst[i+7:i] := a[index*8+7:index*8] + FI + ELSE + dst[i+7:i] := 0 + FI +ENDFOR +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - BMI1 - Bit Manipulation - - Set all the lower bits of "dst" up to and including the lowest set bit in unsigned 64-bit integer "a". + AVX512BW + Miscellaneous + + + + Shuffle packed 8-bit integers in "a" according to shuffle control mask in the corresponding 8-bit element of "b", and store the results in "dst". -dst := (a - 1) XOR a +FOR j := 0 to 63 + i := j*8 + IF b[i+7] == 1 + dst[i+7:i] := 0 + ELSE + index[5:0] := b[i+3:i] + (j & 0x30) + dst[i+7:i] := a[index*8+7:index*8] + FI +ENDFOR +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - BMI1 - Bit Manipulation - - Copy all bits from unsigned 32-bit integer "a" to "dst", and reset (set to 0) the bit in "dst" that corresponds to the lowest set bit in "a". + AVX512VL + AVX512BW + Miscellaneous + + + + + + Shuffle packed 8-bit integers in "a" according to shuffle control mask in the corresponding 8-bit element of "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -dst := (a - 1) BITWISE AND a +FOR j := 0 to 15 + i := j*8 + IF k[j] + IF b[i+7] == 1 + dst[i+7:i] := 0 + ELSE + index[3:0] := b[i+3:i] + dst[i+7:i] := a[index*8+7:index*8] + FI + ELSE + dst[i+7:i] := src[i+7:i] + FI +ENDFOR +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - BMI1 - Bit Manipulation - - Copy all bits from unsigned 64-bit integer "a" to "dst", and reset (set to 0) the bit in "dst" that corresponds to the lowest set bit in "a". + AVX512VL + AVX512BW + Miscellaneous + + + + + Shuffle packed 8-bit integers in "a" according to shuffle control mask in the corresponding 8-bit element of "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -dst := (a - 1) BITWISE AND a +FOR j := 0 to 15 + i := j*8 + IF k[j] + IF b[i+7] == 1 + dst[i+7:i] := 0 + ELSE + index[3:0] := b[i+3:i] + dst[i+7:i] := a[index*8+7:index*8] + FI + ELSE + dst[i+7:i] := 0 + FI +ENDFOR +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - BMI2 - Bit Manipulation - - - Copy all bits from unsigned 32-bit integer "a" to "dst", and reset (set to 0) the high bits in "dst" starting at "index". + AVX512VL + AVX512BW + Miscellaneous + + + + + + Shuffle 16-bit integers in the high 64 bits of 128-bit lanes of "a" using the control in "imm8". Store the results in the high 64 bits of 128-bit lanes of "dst", with the low 64 bits of 128-bit lanes being copied from from "a" to "dst", using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -n := index[7:0] -dst := a -IF (n < 32) - dst[31:n] := 0 -FI +tmp_dst[63:0] := a[63:0] +tmp_dst[79:64] := (a >> (imm8[1:0] * 16))[79:64] +tmp_dst[95:80] := (a >> (imm8[3:2] * 16))[79:64] +tmp_dst[111:96] := (a >> (imm8[5:4] * 16))[79:64] +tmp_dst[127:112] := (a >> (imm8[7:6] * 16))[79:64] +tmp_dst[191:128] := a[191:128] +tmp_dst[207:192] := (a >> (imm8[1:0] * 16))[207:192] +tmp_dst[223:208] := (a >> (imm8[3:2] * 16))[207:192] +tmp_dst[239:224] := (a >> (imm8[5:4] * 16))[207:192] +tmp_dst[255:240] := (a >> (imm8[7:6] * 16))[207:192] +FOR j := 0 to 15 + i := j*16 + IF k[j] + dst[i+15:i] := tmp_dst[i+15:i] + ELSE + dst[i+15:i] := src[i+15:i] + FI +ENDFOR +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - BMI2 - Bit Manipulation - - - Copy all bits from unsigned 64-bit integer "a" to "dst", and reset (set to 0) the high bits in "dst" starting at "index". - -n := index[7:0] -dst := a -IF (n < 64) - dst[63:n] := 0 -FI + AVX512VL + AVX512BW + Miscellaneous + + + + + Shuffle 16-bit integers in the high 64 bits of 128-bit lanes of "a" using the control in "imm8". Store the results in the high 64 bits of 128-bit lanes of "dst", with the low 64 bits of 128-bit lanes being copied from from "a" to "dst", using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + +tmp_dst[63:0] := a[63:0] +tmp_dst[79:64] := (a >> (imm8[1:0] * 16))[79:64] +tmp_dst[95:80] := (a >> (imm8[3:2] * 16))[79:64] +tmp_dst[111:96] := (a >> (imm8[5:4] * 16))[79:64] +tmp_dst[127:112] := (a >> (imm8[7:6] * 16))[79:64] +tmp_dst[191:128] := a[191:128] +tmp_dst[207:192] := (a >> (imm8[1:0] * 16))[207:192] +tmp_dst[223:208] := (a >> (imm8[3:2] * 16))[207:192] +tmp_dst[239:224] := (a >> (imm8[5:4] * 16))[207:192] +tmp_dst[255:240] := (a >> (imm8[7:6] * 16))[207:192] +FOR j := 0 to 15 + i := j*16 + IF k[j] + dst[i+15:i] := tmp_dst[i+15:i] + ELSE + dst[i+15:i] := 0 + FI +ENDFOR +dst[MAX:256] := 0 - +
immintrin.h
 - - INVPCID - OS-Targeted - - - - Invalidate mappings in the Translation Lookaside Buffers (TLBs) and paging-structure caches for the processor context identifier (PCID) specified by "descriptor" based on the invalidation type specified in "type". - The PCID "descriptor" is specified as a 16-byte memory operand (with no alignment restrictions) where bits [11:0] specify the PCID, and bits [127:64] specify the linear address; bits [63:12] are reserved. - The types supported are: - 0) Individual-address invalidation: If "type" is 0, the logical processor invalidates mappings for a single linear address and tagged with the PCID specified in "descriptor", except global translations. The instruction may also invalidate global translations, mappings for other linear addresses, or mappings tagged with other PCIDs. - 1) Single-context invalidation: If "type" is 1, the logical processor invalidates all mappings tagged with the PCID specified in "descriptor" except global translations. In some cases, it may invalidate mappings for other PCIDs as well. - 2) All-context invalidation: If "type" is 2, the logical processor invalidates all mappings tagged with any PCID. - 3) All-context invalidation, retaining global translations: If "type" is 3, the logical processor invalidates all mappings tagged with any PCID except global translations, ignoring "descriptor". The instruction may also invalidate global translations as well. - + + Integer + AVX512BW + Miscellaneous + + + + + + Shuffle 16-bit integers in the high 64 bits of 128-bit lanes of "a" using the control in "imm8". Store the results in the high 64 bits of 128-bit lanes of "dst", with the low 64 bits of 128-bit lanes being copied from from "a" to "dst", using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -CASE type OF -0: // individual-address invalidation retaining global translations - OP_PCID := descriptor[11:0] - ADDR := descriptor[127:64] - BREAK -1: // single PCID invalidation retaining globals - OP_PCID := descriptor[11:0] - // invalidate all mappings tagged with OP_PCID except global translations - BREAK -2: // all PCID invalidation - // invalidate all mappings tagged with any PCID - BREAK -3: // all PCID invalidation retaining global translations - // invalidate all mappings tagged with any PCID except global translations - BREAK -ESAC +tmp_dst[63:0] := a[63:0] +tmp_dst[79:64] := (a >> (imm8[1:0] * 16))[79:64] +tmp_dst[95:80] := (a >> (imm8[3:2] * 16))[79:64] +tmp_dst[111:96] := (a >> (imm8[5:4] * 16))[79:64] +tmp_dst[127:112] := (a >> (imm8[7:6] * 16))[79:64] +tmp_dst[191:128] := a[191:128] +tmp_dst[207:192] := (a >> (imm8[1:0] * 16))[207:192] +tmp_dst[223:208] := (a >> (imm8[3:2] * 16))[207:192] +tmp_dst[239:224] := (a >> (imm8[5:4] * 16))[207:192] +tmp_dst[255:240] := (a >> (imm8[7:6] * 16))[207:192] +tmp_dst[319:256] := a[319:256] +tmp_dst[335:320] := (a >> (imm8[1:0] * 16))[335:320] +tmp_dst[351:336] := (a >> (imm8[3:2] * 16))[335:320] +tmp_dst[367:352] := (a >> (imm8[5:4] * 16))[335:320] +tmp_dst[383:368] := (a >> (imm8[7:6] * 16))[335:320] +tmp_dst[447:384] := a[447:384] +tmp_dst[463:448] := (a >> (imm8[1:0] * 16))[463:448] +tmp_dst[479:464] := (a >> (imm8[3:2] * 16))[463:448] +tmp_dst[495:480] := (a >> (imm8[5:4] * 16))[463:448] +tmp_dst[511:496] := (a >> (imm8[7:6] * 16))[463:448] +FOR j := 0 to 31 + i := j*16 + IF k[j] + dst[i+15:i] := tmp_dst[i+15:i] + ELSE + dst[i+15:i] := src[i+15:i] + FI +ENDFOR +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - LZCNT - Bit Manipulation - - Count the number of leading zero bits in unsigned 32-bit integer "a", and return that count in "dst". + AVX512BW + Miscellaneous + + + + + Shuffle 16-bit integers in the high 64 bits of 128-bit lanes of "a" using the control in "imm8". Store the results in the high 64 bits of 128-bit lanes of "dst", with the low 64 bits of 128-bit lanes being copied from from "a" to "dst", using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -tmp := 31 -dst := 0 -DO WHILE (tmp >= 0 AND a[tmp] = 0) - tmp := tmp - 1 - dst := dst + 1 -OD +tmp_dst[63:0] := a[63:0] +tmp_dst[79:64] := (a >> (imm8[1:0] * 16))[79:64] +tmp_dst[95:80] := (a >> (imm8[3:2] * 16))[79:64] +tmp_dst[111:96] := (a >> (imm8[5:4] * 16))[79:64] +tmp_dst[127:112] := (a >> (imm8[7:6] * 16))[79:64] +tmp_dst[191:128] := a[191:128] +tmp_dst[207:192] := (a >> (imm8[1:0] * 16))[207:192] +tmp_dst[223:208] := (a >> (imm8[3:2] * 16))[207:192] +tmp_dst[239:224] := (a >> (imm8[5:4] * 16))[207:192] +tmp_dst[255:240] := (a >> (imm8[7:6] * 16))[207:192] +tmp_dst[319:256] := a[319:256] +tmp_dst[335:320] := (a >> (imm8[1:0] * 16))[335:320] +tmp_dst[351:336] := (a >> (imm8[3:2] * 16))[335:320] +tmp_dst[367:352] := (a >> (imm8[5:4] * 16))[335:320] +tmp_dst[383:368] := (a >> (imm8[7:6] * 16))[335:320] +tmp_dst[447:384] := a[447:384] +tmp_dst[463:448] := (a >> (imm8[1:0] * 16))[463:448] +tmp_dst[479:464] := (a >> (imm8[3:2] * 16))[463:448] +tmp_dst[495:480] := (a >> (imm8[5:4] * 16))[463:448] +tmp_dst[511:496] := (a >> (imm8[7:6] * 16))[463:448] +FOR j := 0 to 31 + i := j*16 + IF k[j] + dst[i+15:i] := tmp_dst[i+15:i] + ELSE + dst[i+15:i] := 0 + FI +ENDFOR +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - LZCNT - Bit Manipulation - - Count the number of leading zero bits in unsigned 64-bit integer "a", and return that count in "dst". + AVX512BW + Miscellaneous + + + + Shuffle 16-bit integers in the high 64 bits of 128-bit lanes of "a" using the control in "imm8". Store the results in the high 64 bits of 128-bit lanes of "dst", with the low 64 bits of 128-bit lanes being copied from from "a" to "dst". -tmp := 63 -dst := 0 -DO WHILE (tmp >= 0 AND a[tmp] = 0) - tmp := tmp - 1 - dst := dst + 1 -OD +dst[63:0] := a[63:0] +dst[79:64] := (a >> (imm8[1:0] * 16))[79:64] +dst[95:80] := (a >> (imm8[3:2] * 16))[79:64] +dst[111:96] := (a >> (imm8[5:4] * 16))[79:64] +dst[127:112] := (a >> (imm8[7:6] * 16))[79:64] +dst[191:128] := a[191:128] +dst[207:192] := (a >> (imm8[1:0] * 16))[207:192] +dst[223:208] := (a >> (imm8[3:2] * 16))[207:192] +dst[239:224] := (a >> (imm8[5:4] * 16))[207:192] +dst[255:240] := (a >> (imm8[7:6] * 16))[207:192] +dst[319:256] := a[319:256] +dst[335:320] := (a >> (imm8[1:0] * 16))[335:320] +dst[351:336] := (a >> (imm8[3:2] * 16))[335:320] +dst[367:352] := (a >> (imm8[5:4] * 16))[335:320] +dst[383:368] := (a >> (imm8[7:6] * 16))[335:320] +dst[447:384] := a[447:384] +dst[463:448] := (a >> (imm8[1:0] * 16))[463:448] +dst[479:464] := (a >> (imm8[3:2] * 16))[463:448] +dst[495:480] := (a >> (imm8[5:4] * 16))[463:448] +dst[511:496] := (a >> (imm8[7:6] * 16))[463:448] +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - BMI2 - Bit Manipulation - - - Deposit contiguous low bits from unsigned 32-bit integer "a" to "dst" at the corresponding bit locations specified by "mask"; all other bits in "dst" are set to zero. + AVX512VL + AVX512BW + Miscellaneous + + + + + + Shuffle 16-bit integers in the high 64 bits of "a" using the control in "imm8". Store the results in the high 64 bits of "dst", with the low 64 bits being copied from from "a" to "dst", using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -tmp := a -dst := 0 -m := 0 -k := 0 -DO WHILE m < 32 - IF mask[m] = 1 - dst[m] := tmp[k] - k := k + 1 +tmp_dst[63:0] := a[63:0] +tmp_dst[79:64] := (a >> (imm8[1:0] * 16))[79:64] +tmp_dst[95:80] := (a >> (imm8[3:2] * 16))[79:64] +tmp_dst[111:96] := (a >> (imm8[5:4] * 16))[79:64] +tmp_dst[127:112] := (a >> (imm8[7:6] * 16))[79:64] +FOR j := 0 to 7 + i := j*16 + IF k[j] + dst[i+15:i] := tmp_dst[i+15:i] + ELSE + dst[i+15:i] := src[i+15:i] FI - m := m + 1 -OD +ENDFOR +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - BMI2 - Bit Manipulation - - - Deposit contiguous low bits from unsigned 64-bit integer "a" to "dst" at the corresponding bit locations specified by "mask"; all other bits in "dst" are set to zero. + AVX512VL + AVX512BW + Miscellaneous + + + + + Shuffle 16-bit integers in the high 64 bits of "a" using the control in "imm8". Store the results in the high 64 bits of "dst", with the low 64 bits being copied from from "a" to "dst", using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -tmp := a -dst := 0 -m := 0 -k := 0 -DO WHILE m < 64 - IF mask[m] = 1 - dst[m] := tmp[k] - k := k + 1 +tmp_dst[63:0] := a[63:0] +tmp_dst[79:64] := (a >> (imm8[1:0] * 16))[79:64] +tmp_dst[95:80] := (a >> (imm8[3:2] * 16))[79:64] +tmp_dst[111:96] := (a >> (imm8[5:4] * 16))[79:64] +tmp_dst[127:112] := (a >> (imm8[7:6] * 16))[79:64] +FOR j := 0 to 7 + i := j*16 + IF k[j] + dst[i+15:i] := tmp_dst[i+15:i] + ELSE + dst[i+15:i] := 0 FI - m := m + 1 -OD +ENDFOR +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - BMI2 - Bit Manipulation - - - Extract bits from unsigned 32-bit integer "a" at the corresponding bit locations specified by "mask" to contiguous low bits in "dst"; the remaining upper bits in "dst" are set to zero. + AVX512VL + AVX512BW + Miscellaneous + + + + + + Shuffle 16-bit integers in the low 64 bits of 128-bit lanes of "a" using the control in "imm8". Store the results in the low 64 bits of 128-bit lanes of "dst", with the high 64 bits of 128-bit lanes being copied from from "a" to "dst", using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -tmp := a -dst := 0 -m := 0 -k := 0 -DO WHILE m < 32 - IF mask[m] = 1 - dst[k] := tmp[m] - k := k + 1 +tmp_dst[15:0] := (a >> (imm8[1:0] * 16))[15:0] +tmp_dst[31:16] := (a >> (imm8[3:2] * 16))[15:0] +tmp_dst[47:32] := (a >> (imm8[5:4] * 16))[15:0] +tmp_dst[63:48] := (a >> (imm8[7:6] * 16))[15:0] +tmp_dst[127:64] := a[127:64] +tmp_dst[143:128] := (a >> (imm8[1:0] * 16))[143:128] +tmp_dst[159:144] := (a >> (imm8[3:2] * 16))[143:128] +tmp_dst[175:160] := (a >> (imm8[5:4] * 16))[143:128] +tmp_dst[191:176] := (a >> (imm8[7:6] * 16))[143:128] +tmp_dst[255:192] := a[255:192] +FOR j := 0 to 15 + i := j*16 + IF k[j] + dst[i+15:i] := tmp_dst[i+15:i] + ELSE + dst[i+15:i] := src[i+15:i] FI - m := m + 1 -OD +ENDFOR +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - BMI2 - Bit Manipulation - - - Extract bits from unsigned 64-bit integer "a" at the corresponding bit locations specified by "mask" to contiguous low bits in "dst"; the remaining upper bits in "dst" are set to zero. + AVX512VL + AVX512BW + Miscellaneous + + + + + Shuffle 16-bit integers in the low 64 bits of 128-bit lanes of "a" using the control in "imm8". Store the results in the low 64 bits of 128-bit lanes of "dst", with the high 64 bits of 128-bit lanes being copied from from "a" to "dst", using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -tmp := a -dst := 0 -m := 0 -k := 0 -DO WHILE m < 64 - IF mask[m] = 1 - dst[k] := tmp[m] - k := k + 1 +tmp_dst[15:0] := (a >> (imm8[1:0] * 16))[15:0] +tmp_dst[31:16] := (a >> (imm8[3:2] * 16))[15:0] +tmp_dst[47:32] := (a >> (imm8[5:4] * 16))[15:0] +tmp_dst[63:48] := (a >> (imm8[7:6] * 16))[15:0] +tmp_dst[127:64] := a[127:64] +tmp_dst[143:128] := (a >> (imm8[1:0] * 16))[143:128] +tmp_dst[159:144] := (a >> (imm8[3:2] * 16))[143:128] +tmp_dst[175:160] := (a >> (imm8[5:4] * 16))[143:128] +tmp_dst[191:176] := (a >> (imm8[7:6] * 16))[143:128] +tmp_dst[255:192] := a[255:192] +FOR j := 0 to 15 + i := j*16 + IF k[j] + dst[i+15:i] := tmp_dst[i+15:i] + ELSE + dst[i+15:i] := 0 FI - m := m + 1 -OD +ENDFOR +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - BMI1 - Bit Manipulation - - - Compute the bitwise NOT of 32-bit integer "a" and then AND with b, and store the results in dst. + AVX512BW + Miscellaneous + + + + + + Shuffle 16-bit integers in the low 64 bits of 128-bit lanes of "a" using the control in "imm8". Store the results in the low 64 bits of 128-bit lanes of "dst", with the high 64 bits of 128-bit lanes being copied from from "a" to "dst", using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -dst[31:0] := ((NOT a[31:0]) AND b[31:0]) +tmp_dst[15:0] := (a >> (imm8[1:0] * 16))[15:0] +tmp_dst[31:16] := (a >> (imm8[3:2] * 16))[15:0] +tmp_dst[47:32] := (a >> (imm8[5:4] * 16))[15:0] +tmp_dst[63:48] := (a >> (imm8[7:6] * 16))[15:0] +tmp_dst[127:64] := a[127:64] +tmp_dst[143:128] := (a >> (imm8[1:0] * 16))[143:128] +tmp_dst[159:144] := (a >> (imm8[3:2] * 16))[143:128] +tmp_dst[175:160] := (a >> (imm8[5:4] * 16))[143:128] +tmp_dst[191:176] := (a >> (imm8[7:6] * 16))[143:128] +tmp_dst[255:192] := a[255:192] +tmp_dst[271:256] := (a >> (imm8[1:0] * 16))[271:256] +tmp_dst[287:272] := (a >> (imm8[3:2] * 16))[271:256] +tmp_dst[303:288] := (a >> (imm8[5:4] * 16))[271:256] +tmp_dst[319:304] := (a >> (imm8[7:6] * 16))[271:256] +tmp_dst[383:320] := a[383:320] +tmp_dst[399:384] := (a >> (imm8[1:0] * 16))[399:384] +tmp_dst[415:400] := (a >> (imm8[3:2] * 16))[399:384] +tmp_dst[431:416] := (a >> (imm8[5:4] * 16))[399:384] +tmp_dst[447:432] := (a >> (imm8[7:6] * 16))[399:384] +tmp_dst[511:448] := a[511:448] +FOR j := 0 to 31 + i := j*16 + IF k[j] + dst[i+15:i] := tmp_dst[i+15:i] + ELSE + dst[i+15:i] := src[i+15:i] + FI +ENDFOR +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - BMI1 - Bit Manipulation - - - Compute the bitwise NOT of 64-bit integer "a" and then AND with b, and store the results in dst. + AVX512BW + Miscellaneous + + + + + Shuffle 16-bit integers in the low 64 bits of 128-bit lanes of "a" using the control in "imm8". Store the results in the low 64 bits of 128-bit lanes of "dst", with the high 64 bits of 128-bit lanes being copied from from "a" to "dst", using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -dst[63:0] := ((NOT a[63:0]) AND b[63:0]) +tmp_dst[15:0] := (a >> (imm8[1:0] * 16))[15:0] +tmp_dst[31:16] := (a >> (imm8[3:2] * 16))[15:0] +tmp_dst[47:32] := (a >> (imm8[5:4] * 16))[15:0] +tmp_dst[63:48] := (a >> (imm8[7:6] * 16))[15:0] +tmp_dst[127:64] := a[127:64] +tmp_dst[143:128] := (a >> (imm8[1:0] * 16))[143:128] +tmp_dst[159:144] := (a >> (imm8[3:2] * 16))[143:128] +tmp_dst[175:160] := (a >> (imm8[5:4] * 16))[143:128] +tmp_dst[191:176] := (a >> (imm8[7:6] * 16))[143:128] +tmp_dst[255:192] := a[255:192] +tmp_dst[271:256] := (a >> (imm8[1:0] * 16))[271:256] +tmp_dst[287:272] := (a >> (imm8[3:2] * 16))[271:256] +tmp_dst[303:288] := (a >> (imm8[5:4] * 16))[271:256] +tmp_dst[319:304] := (a >> (imm8[7:6] * 16))[271:256] +tmp_dst[383:320] := a[383:320] +tmp_dst[399:384] := (a >> (imm8[1:0] * 16))[399:384] +tmp_dst[415:400] := (a >> (imm8[3:2] * 16))[399:384] +tmp_dst[431:416] := (a >> (imm8[5:4] * 16))[399:384] +tmp_dst[447:432] := (a >> (imm8[7:6] * 16))[399:384] +tmp_dst[511:448] := a[511:448] +FOR j := 0 to 31 + i := j*16 + IF k[j] + dst[i+15:i] := tmp_dst[i+15:i] + ELSE + dst[i+15:i] := 0 + FI +ENDFOR +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - BMI2 - Arithmetic - - - - Multiply unsigned 32-bit integers "a" and "b", store the low 32-bits of the result in "dst", and store the high 32-bits in "hi". This does not read or write arithmetic flags. + AVX512BW + Miscellaneous + + + + Shuffle 16-bit integers in the low 64 bits of 128-bit lanes of "a" using the control in "imm8". Store the results in the low 64 bits of 128-bit lanes of "dst", with the high 64 bits of 128-bit lanes being copied from from "a" to "dst". -dst[31:0] := (a * b)[31:0] -hi[31:0] := (a * b)[63:32] +dst[15:0] := (a >> (imm8[1:0] * 16))[15:0] +dst[31:16] := (a >> (imm8[3:2] * 16))[15:0] +dst[47:32] := (a >> (imm8[5:4] * 16))[15:0] +dst[63:48] := (a >> (imm8[7:6] * 16))[15:0] +dst[127:64] := a[127:64] +dst[143:128] := (a >> (imm8[1:0] * 16))[143:128] +dst[159:144] := (a >> (imm8[3:2] * 16))[143:128] +dst[175:160] := (a >> (imm8[5:4] * 16))[143:128] +dst[191:176] := (a >> (imm8[7:6] * 16))[143:128] +dst[255:192] := a[255:192] +dst[271:256] := (a >> (imm8[1:0] * 16))[271:256] +dst[287:272] := (a >> (imm8[3:2] * 16))[271:256] +dst[303:288] := (a >> (imm8[5:4] * 16))[271:256] +dst[319:304] := (a >> (imm8[7:6] * 16))[271:256] +dst[383:320] := a[383:320] +dst[399:384] := (a >> (imm8[1:0] * 16))[399:384] +dst[415:400] := (a >> (imm8[3:2] * 16))[399:384] +dst[431:416] := (a >> (imm8[5:4] * 16))[399:384] +dst[447:432] := (a >> (imm8[7:6] * 16))[399:384] +dst[511:448] := a[511:448] +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - BMI2 - Arithmetic - - - - Multiply unsigned 64-bit integers "a" and "b", store the low 64-bits of the result in "dst", and store the high 64-bits in "hi". This does not read or write arithmetic flags. + AVX512VL + AVX512BW + Miscellaneous + + + + + + Shuffle 16-bit integers in the low 64 bits of "a" using the control in "imm8". Store the results in the low 64 bits of "dst", with the high 64 bits being copied from from "a" to "dst", using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -dst[63:0] := (a * b)[63:0] -hi[63:0] := (a * b)[127:64] +tmp_dst[15:0] := (a >> (imm8[1:0] * 16))[15:0] +tmp_dst[31:16] := (a >> (imm8[3:2] * 16))[15:0] +tmp_dst[47:32] := (a >> (imm8[5:4] * 16))[15:0] +tmp_dst[63:48] := (a >> (imm8[7:6] * 16))[15:0] +tmp_dst[127:64] := a[127:64] +FOR j := 0 to 7 + i := j*16 + IF k[j] + dst[i+15:i] := tmp_dst[i+15:i] + ELSE + dst[i+15:i] := src[i+15:i] + FI +ENDFOR +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - BMI1 - Bit Manipulation - - Count the number of trailing zero bits in unsigned 32-bit integer "a", and return that count in "dst". + AVX512VL + AVX512BW + Miscellaneous + + + + + Shuffle 16-bit integers in the low 64 bits of "a" using the control in "imm8". Store the results in the low 64 bits of "dst", with the high 64 bits being copied from from "a" to "dst", using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -tmp := 0 -dst := 0 -DO WHILE ((tmp < 32) AND a[tmp] = 0) - tmp := tmp + 1 - dst := dst + 1 -OD +tmp_dst[15:0] := (a >> (imm8[1:0] * 16))[15:0] +tmp_dst[31:16] := (a >> (imm8[3:2] * 16))[15:0] +tmp_dst[47:32] := (a >> (imm8[5:4] * 16))[15:0] +tmp_dst[63:48] := (a >> (imm8[7:6] * 16))[15:0] +tmp_dst[127:64] := a[127:64] +FOR j := 0 to 7 + i := j*16 + IF k[j] + dst[i+15:i] := tmp_dst[i+15:i] + ELSE + dst[i+15:i] := 0 + FI +ENDFOR +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - BMI1 - Bit Manipulation - - Count the number of trailing zero bits in unsigned 64-bit integer "a", and return that count in "dst". + AVX512BW + Shift + + + + Shift 128-bit lanes in "a" left by "imm8" bytes while shifting in zeros, and store the results in "dst". -tmp := 0 -dst := 0 -DO WHILE ((tmp < 64) AND a[tmp] = 0) - tmp := tmp + 1 - dst := dst + 1 -OD +tmp := imm8[7:0] +IF tmp > 15 + tmp := 16 +FI +dst[127:0] := a[127:0] << (tmp*8) +dst[255:128] := a[255:128] << (tmp*8) +dst[383:256] := a[383:256] << (tmp*8) +dst[511:384] := a[511:384] << (tmp*8) +dst[MAX:512] := 0 - +
immintrin.h
 - - RTM - General Support - - - Force an RTM abort. The EAX register is updated to reflect an XABORT instruction caused the abort, and the "imm8" parameter will be provided in bits [31:24] of EAX. - Following an RTM abort, the logical processor resumes execution at the fallback address computed through the outermost XBEGIN instruction. - + + Integer + AVX512VL + AVX512BW + Shift + + + + + + Shift packed 16-bit integers in "a" left by the amount specified by the corresponding element in "count" while shifting in zeros, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -IF RTM_ACTIVE = 0 - // nop -ELSE - // restore architectural register state - // discard memory updates performed in transaction - // update EAX with status and imm8 value - RTM_NEST_COUNT := 0 - RTM_ACTIVE := 0 - IF 64-bit Mode - RIP := fallbackRIP +FOR j := 0 to 15 + i := j*16 + IF k[j] + IF count[i+15:i] < 16 + dst[i+15:i] := ZeroExtend16(a[i+15:i] << count[i+15:i]) + ELSE + dst[i+15:i] := 0 + FI ELSE - EIP := fallbackEIP - FI -FI + dst[i+15:i] := src[i+15:i] + FI +ENDFOR +dst[MAX:256] := 0 - +
immintrin.h
 - - RTM - General Support - - - Specify the start of an RTM code region. - If the logical processor was not already in transactional execution, then this call causes the logical processor to transition into transactional execution. - On an RTM abort, the logical processor discards all architectural register and memory updates performed during the RTM execution, restores architectural state, and starts execution beginning at the fallback address computed from the outermost XBEGIN instruction. Return status of ~0 (0xFFFF) if continuing inside transaction; all other codes are aborts. - + + Integer + AVX512VL + AVX512BW + Shift + + + + + Shift packed 16-bit integers in "a" left by the amount specified by the corresponding element in "count" while shifting in zeros, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -IF RTM_NEST_COUNT < MAX_RTM_NEST_COUNT - RTM_NEST_COUNT := RTM_NEST_COUNT + 1 - IF RTM_NEST_COUNT = 1 - IF 64-bit Mode - fallbackRIP := RIP + SignExtend(IMM) - ELSE IF 32-bit Mode - fallbackEIP := EIP + SignExtend(IMM) - ELSE // 16-bit Mode - fallbackEIP := (EIP + SignExtend(IMM)) AND 0x0000FFFF +FOR j := 0 to 15 + i := j*16 + IF k[j] + IF count[i+15:i] < 16 + dst[i+15:i] := ZeroExtend16(a[i+15:i] << count[i+15:i]) + ELSE + dst[i+15:i] := 0 FI - - RTM_ACTIVE := 1 - // enter RTM execution, record register state, start tracking memory state - FI -ELSE - // RTM abort (see _xabort) -FI + ELSE + dst[i+15:i] := 0 + FI +ENDFOR +dst[MAX:256] := 0 - +
immintrin.h
 - - RTM - General Support - - - Specify the end of an RTM code region. - If this corresponds to the outermost scope, the logical processor will attempt to commit the logical processor state atomically. - If the commit fails, the logical processor will perform an RTM abort. - + + Integer + AVX512VL + AVX512BW + Shift + + + + Shift packed 16-bit integers in "a" left by the amount specified by the corresponding element in "count" while shifting in zeros, and store the results in "dst". -IF RTM_ACTIVE = 1 - RTM_NEST_COUNT := RTM_NEST_COUNT - 1 - IF RTM_NEST_COUNT = 0 - // try to commit transaction - IF fail to commit transaction - // RTM abort (see _xabort) - ELSE - RTM_ACTIVE = 0 - FI +FOR j := 0 to 15 + i := j*16 + IF count[i+15:i] < 16 + dst[i+15:i] := ZeroExtend16(a[i+15:i] << count[i+15:i]) + ELSE + dst[i+15:i] := 0 FI -FI +ENDFOR +dst[MAX:256] := 0 - +
immintrin.h
 - - RTM - General Support - - Query the transactional execution status, return 0 if inside a transactionally executing RTM or HLE region, and return 1 otherwise. + + Integer + AVX512BW + Shift + + + + + + Shift packed 16-bit integers in "a" left by the amount specified by the corresponding element in "count" while shifting in zeros, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -IF (RTM_ACTIVE = 1 OR HLE_ACTIVE = 1) - dst := 0 -ELSE - dst := 1 -FI +FOR j := 0 to 31 + i := j*16 + IF k[j] + IF count[i+15:i] < 16 + dst[i+15:i] := ZeroExtend16(a[i+15:i] << count[i+15:i]) + ELSE + dst[i+15:i] := 0 + FI + ELSE + dst[i+15:i] := src[i+15:i] + FI +ENDFOR +dst[MAX:512] := 0 - +
immintrin.h
 - - RDTSCP - General Support - - Copy the current 64-bit value of the processor's time-stamp counter into "dst", and store the IA32_TSC_AUX MSR (signature value) into memory at "mem_addr". + + Integer + AVX512BW + Shift + + + + + Shift packed 16-bit integers in "a" left by the amount specified by the corresponding element in "count" while shifting in zeros, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -dst[63:0] := TimeStampCounter -MEM[mem_addr+31:mem_addr] := IA32_TSC_AUX[31:0] +FOR j := 0 to 31 + i := j*16 + IF k[j] + IF count[i+15:i] < 16 + dst[i+15:i] := ZeroExtend16(a[i+15:i] << count[i+15:i]) + ELSE + dst[i+15:i] := 0 + FI + ELSE + dst[i+15:i] := 0 + FI +ENDFOR +dst[MAX:512] := 0 - +
immintrin.h
 - - RDPID - General Support - - Copy the IA32_TSC_AUX MSR (signature value) into "dst". + + Integer + AVX512BW + Shift + + + + Shift packed 16-bit integers in "a" left by the amount specified by the corresponding element in "count" while shifting in zeros, and store the results in "dst". -dst[31:0] := IA32_TSC_AUX[31:0] +FOR j := 0 to 31 + i := j*16 + IF count[i+15:i] < 16 + dst[i+15:i] := ZeroExtend16(a[i+15:i] << count[i+15:i]) + ELSE + dst[i+15:i] := 0 + FI +ENDFOR +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - Bit Manipulation - - Set "dst" to the index of the lowest set bit in 32-bit integer "a". If no bits are set in "a" then "dst" is undefined. + AVX512VL + AVX512BW + Shift + + + + + + Shift packed 16-bit integers in "a" left by the amount specified by the corresponding element in "count" while shifting in zeros, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -tmp := 0 -IF a = 0 - dst := undefined -ELSE - DO WHILE ((tmp < 32) AND a[tmp] = 0) - tmp := tmp + 1 - dst := tmp - OD -FI +FOR j := 0 to 7 + i := j*16 + IF k[j] + IF count[i+15:i] < 16 + dst[i+15:i] := ZeroExtend16(a[i+15:i] << count[i+15:i]) + ELSE + dst[i+15:i] := 0 + FI + ELSE + dst[i+15:i] := src[i+15:i] + FI +ENDFOR +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - Bit Manipulation - - Set "dst" to the index of the highest set bit in 32-bit integer "a". If no bits are set in "a" then "dst" is undefined. + AVX512VL + AVX512BW + Shift + + + + + Shift packed 16-bit integers in "a" left by the amount specified by the corresponding element in "count" while shifting in zeros, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -tmp := 31 -IF a = 0 - dst := undefined -ELSE - DO WHILE ((tmp > 0) AND a[tmp] = 0) - tmp := tmp - 1 - dst := tmp - OD -FI +FOR j := 0 to 7 + i := j*16 + IF k[j] + IF count[i+15:i] < 16 + dst[i+15:i] := ZeroExtend16(a[i+15:i] << count[i+15:i]) + ELSE + dst[i+15:i] := 0 + FI + ELSE + dst[i+15:i] := 0 + FI +ENDFOR +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - Bit Manipulation - - - Set "index" to the index of the lowest set bit in 32-bit integer "mask". If no bits are set in "mask", then set "dst" to 0, otherwise set "dst" to 1. + AVX512VL + AVX512BW + Shift + + + + Shift packed 16-bit integers in "a" left by the amount specified by the corresponding element in "count" while shifting in zeros, and store the results in "dst". -tmp := 0 -IF mask = 0 - dst := 0 -ELSE - DO WHILE ((tmp < 32) AND mask[tmp] = 0) - tmp := tmp + 1 - index := tmp - dst := 1 - OD -FI +FOR j := 0 to 7 + i := j*16 + IF count[i+15:i] < 16 + dst[i+15:i] := ZeroExtend16(a[i+15:i] << count[i+15:i]) + ELSE + dst[i+15:i] := 0 + FI +ENDFOR +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - Bit Manipulation - - - Set "index" to the index of the highest set bit in 32-bit integer "mask". If no bits are set in "mask", then set "dst" to 0, otherwise set "dst" to 1. + AVX512VL + AVX512BW + Shift + + + + + + Shift packed 16-bit integers in "a" left by "count" while shifting in zeros, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -tmp := 31 -IF mask = 0 - dst := 0 -ELSE - DO WHILE ((tmp > 0) AND mask[tmp] = 0) - tmp := tmp - 1 - index := tmp - dst := 1 - OD -FI +FOR j := 0 to 15 + i := j*16 + IF k[j] + IF count[63:0] > 15 + dst[i+15:i] := 0 + ELSE + dst[i+15:i] := ZeroExtend16(a[i+15:i] << count[63:0]) + FI + ELSE + dst[i+15:i] := src[i+15:i] + FI +ENDFOR +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - Bit Manipulation - - - Set "index" to the index of the lowest set bit in 64-bit integer "mask". If no bits are set in "mask", then set "dst" to 0, otherwise set "dst" to 1. + AVX512VL + AVX512BW + Shift + + + + + + Shift packed 16-bit integers in "a" left by "imm8" while shifting in zeros, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -tmp := 0 -IF mask = 0 - dst := 0 -ELSE - DO WHILE ((tmp < 64) AND mask[tmp] = 0) - tmp := tmp + 1 - index := tmp - dst := 1 - OD -FI +FOR j := 0 to 15 + i := j*16 + IF k[j] + IF imm8[7:0] > 15 + dst[i+15:i] := 0 + ELSE + dst[i+15:i] := ZeroExtend16(a[i+15:i] << imm8[7:0]) + FI + ELSE + dst[i+15:i] := src[i+15:i] + FI +ENDFOR +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - Bit Manipulation - - - Set "index" to the index of the highest set bit in 64-bit integer "mask". If no bits are set in "mask", then set "dst" to 0, otherwise set "dst" to 1. + AVX512VL + AVX512BW + Shift + + + + + Shift packed 16-bit integers in "a" left by "count" while shifting in zeros, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -tmp := 31 -IF mask = 0 - dst := 0 -ELSE - DO WHILE ((tmp > 0) AND mask[tmp] = 0) - tmp := tmp - 1 - index := tmp - dst := 1 - OD -FI +FOR j := 0 to 15 + i := j*16 + IF k[j] + IF count[63:0] > 15 + dst[i+15:i] := 0 + ELSE + dst[i+15:i] := ZeroExtend16(a[i+15:i] << count[63:0]) + FI + ELSE + dst[i+15:i] := 0 + FI +ENDFOR +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - Bit Manipulation - - - Return the bit at index "b" of 32-bit integer "a". + AVX512VL + AVX512BW + Shift + + + + + Shift packed 16-bit integers in "a" left by "imm8" while shifting in zeros, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -dst := a[b] +FOR j := 0 to 15 + i := j*16 + IF k[j] + IF imm8[7:0] > 15 + dst[i+15:i] := 0 + ELSE + dst[i+15:i] := ZeroExtend16(a[i+15:i] << imm8[7:0]) + FI + ELSE + dst[i+15:i] := 0 + FI +ENDFOR +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - Bit Manipulation - - - Return the bit at index "b" of 32-bit integer "a", and set that bit to its complement. + AVX512BW + Shift + + + + + + Shift packed 16-bit integers in "a" left by "count" while shifting in zeros, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -dst := a[b] -a[b] := ~a[b] +FOR j := 0 to 31 + i := j*16 + IF k[j] + IF count[63:0] > 15 + dst[i+15:i] := 0 + ELSE + dst[i+15:i] := ZeroExtend16(a[i+15:i] << count[63:0]) + FI + ELSE + dst[i+15:i] := src[i+15:i] + FI +ENDFOR +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - Bit Manipulation - - - Return the bit at index "b" of 32-bit integer "a", and set that bit to zero. + AVX512BW + Shift + + + + + + Shift packed 16-bit integers in "a" left by "imm8" while shifting in zeros, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -dst := a[b] -a[b] := 0 +FOR j := 0 to 31 + i := j*16 + IF k[j] + IF imm8[7:0] > 15 + dst[i+15:i] := 0 + ELSE + dst[i+15:i] := ZeroExtend16(a[i+15:i] << imm8[7:0]) + FI + ELSE + dst[i+15:i] := src[i+15:i] + FI +ENDFOR +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - Bit Manipulation - - - Return the bit at index "b" of 32-bit integer "a", and set that bit to one. + AVX512BW + Shift + + + + + Shift packed 16-bit integers in "a" left by "count" while shifting in zeros, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -dst := a[b] -a[b] := 1 +FOR j := 0 to 31 + i := j*16 + IF k[j] + IF count[63:0] > 15 + dst[i+15:i] := 0 + ELSE + dst[i+15:i] := ZeroExtend16(a[i+15:i] << count[63:0]) + FI + ELSE + dst[i+15:i] := 0 + FI +ENDFOR +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - Bit Manipulation - - - Return the bit at index "b" of 64-bit integer "a". + AVX512BW + Shift + + + + + Shift packed 16-bit integers in "a" left by "imm8" while shifting in zeros, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -dst := a[b] +FOR j := 0 to 31 + i := j*16 + IF k[j] + IF imm8[7:0] > 15 + dst[i+15:i] := 0 + ELSE + dst[i+15:i] := ZeroExtend16(a[i+15:i] << imm8[7:0]) + FI + ELSE + dst[i+15:i] := 0 + FI +ENDFOR +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - Bit Manipulation - - - Return the bit at index "b" of 64-bit integer "a", and set that bit to its complement. + AVX512BW + Shift + + + + Shift packed 16-bit integers in "a" left by "count" while shifting in zeros, and store the results in "dst". -dst := a[b] -a[b] := ~a[b] +FOR j := 0 to 31 + i := j*16 + IF count[63:0] > 15 + dst[i+15:i] := 0 + ELSE + dst[i+15:i] := ZeroExtend16(a[i+15:i] << count[63:0]) + FI +ENDFOR +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - Bit Manipulation - - - Return the bit at index "b" of 64-bit integer "a", and set that bit to zero. + AVX512BW + Shift + + + + Shift packed 16-bit integers in "a" left by "imm8" while shifting in zeros, and store the results in "dst". -dst := a[b] -a[b] := 0 +FOR j := 0 to 31 + i := j*16 + IF imm8[7:0] > 15 + dst[i+15:i] := 0 + ELSE + dst[i+15:i] := ZeroExtend16(a[i+15:i] << imm8[7:0]) + FI +ENDFOR +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - Bit Manipulation - - - Return the bit at index "b" of 64-bit integer "a", and set that bit to one. + AVX512VL + AVX512BW + Shift + + + + + + Shift packed 16-bit integers in "a" left by "count" while shifting in zeros, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -dst := a[b] -a[b] := 1 +FOR j := 0 to 7 + i := j*16 + IF k[j] + IF count[63:0] > 15 + dst[i+15:i] := 0 + ELSE + dst[i+15:i] := ZeroExtend16(a[i+15:i] << count[63:0]) + FI + ELSE + dst[i+15:i] := src[i+15:i] + FI +ENDFOR +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - Bit Manipulation - - Reverse the byte order of 32-bit integer "a", and store the result in "dst". This intrinsic is provided for conversion between little and big endian values. + AVX512VL + AVX512BW + Shift + + + + + + Shift packed 16-bit integers in "a" left by "imm8" while shifting in zeros, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -dst[7:0] := a[31:24] -dst[15:8] := a[23:16] -dst[23:16] := a[15:8] -dst[31:24] := a[7:0] +FOR j := 0 to 7 + i := j*16 + IF k[j] + IF imm8[7:0] > 15 + dst[i+15:i] := 0 + ELSE + dst[i+15:i] := ZeroExtend16(a[i+15:i] << imm8[7:0]) + FI + ELSE + dst[i+15:i] := src[i+15:i] + FI +ENDFOR +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - Bit Manipulation - - Reverse the byte order of 64-bit integer "a", and store the result in "dst". This intrinsic is provided for conversion between little and big endian values. + AVX512VL + AVX512BW + Shift + + + + + Shift packed 16-bit integers in "a" left by "count" while shifting in zeros, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -dst[7:0] := a[63:56] -dst[15:8] := a[55:48] -dst[23:16] := a[47:40] -dst[31:24] := a[39:32] -dst[39:32] := a[31:24] -dst[47:40] := a[23:16] -dst[55:48] := a[15:8] -dst[63:56] := a[7:0] +FOR j := 0 to 7 + i := j*16 + IF k[j] + IF count[63:0] > 15 + dst[i+15:i] := 0 + ELSE + dst[i+15:i] := ZeroExtend16(a[i+15:i] << count[63:0]) + FI + ELSE + dst[i+15:i] := 0 + FI +ENDFOR +dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point + Integer - Cast - - Cast from type float to type unsigned __int32 without conversion. - This intrinsic is only used for compilation and does not generate any instructions, thus it has zero latency. + AVX512VL + AVX512BW + Shift + + + + + Shift packed 16-bit integers in "a" left by "imm8" while shifting in zeros, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + +FOR j := 0 to 7 + i := j*16 + IF k[j] + IF imm8[7:0] > 15 + dst[i+15:i] := 0 + ELSE + dst[i+15:i] := ZeroExtend16(a[i+15:i] << imm8[7:0]) + FI + ELSE + dst[i+15:i] := 0 + FI +ENDFOR +dst[MAX:128] := 0 + +
immintrin.h
 - - Floating Point + Integer - Cast - - Cast from type double to type unsigned __int64 without conversion. - This intrinsic is only used for compilation and does not generate any instructions, thus it has zero latency. + AVX512VL + AVX512BW + Shift + + + + + + Shift packed 16-bit integers in "a" right by the amount specified by the corresponding element in "count" while shifting in sign bits, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +FOR j := 0 to 15 + i := j*16 + IF k[j] + IF count[i+15:i] < 16 + dst[i+15:i] := SignExtend16(a[i+15:i] >> count[i+15:i]) + ELSE + dst[i+15:i] := (a[i+15] ? 0xFFFF : 0) + FI + ELSE + dst[i+15:i] := src[i+15:i] + FI +ENDFOR +dst[MAX:256] := 0 + +
immintrin.h
 - - Floating Point + Integer - Cast - - Cast from type unsigned __int32 to type float without conversion. - This intrinsic is only used for compilation and does not generate any instructions, thus it has zero latency. + AVX512VL + AVX512BW + Shift + + + + + Shift packed 16-bit integers in "a" right by the amount specified by the corresponding element in "count" while shifting in sign bits, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + +FOR j := 0 to 15 + i := j*16 + IF k[j] + IF count[i+15:i] < 16 + dst[i+15:i] := SignExtend16(a[i+15:i] >> count[i+15:i]) + ELSE + dst[i+15:i] := (a[i+15] ? 0xFFFF : 0) + FI + ELSE + dst[i+15:i] := 0 + FI +ENDFOR +dst[MAX:256] := 0 + +
immintrin.h
 - - Floating Point + Integer - Cast - - Cast from type unsigned __int64 to type double without conversion. - This intrinsic is only used for compilation and does not generate any instructions, thus it has zero latency. -
immintrin.h
- - - FXSR - OS-Targeted - - Reload the x87 FPU, MMX technology, XMM, and MXCSR registers from the 512-byte memory image at "mem_addr". This data should have been written to memory previously using the FXSAVE instruction, and in the same format as required by the operating mode. "mem_addr" must be aligned on a 16-byte boundary. + AVX512VL + AVX512BW + Shift + + + + Shift packed 16-bit integers in "a" right by the amount specified by the corresponding element in "count" while shifting in sign bits, and store the results in "dst". -(x87 FPU, MMX, XMM7-XMM0, MXCSR) := Load(MEM[mem_addr]) +FOR j := 0 to 15 + i := j*16 + IF count[i+15:i] < 16 + dst[i+15:i] := SignExtend16(a[i+15:i] >> count[i+15:i]) + ELSE + dst[i+15:i] := (a[i+15] ? 0xFFFF : 0) + FI +ENDFOR +dst[MAX:256] := 0 - +
immintrin.h
 - - FXSR - OS-Targeted - - Reload the x87 FPU, MMX technology, XMM, and MXCSR registers from the 512-byte memory image at "mem_addr". This data should have been written to memory previously using the FXSAVE64 instruction, and in the same format as required by the operating mode. "mem_addr" must be aligned on a 16-byte boundary. + + Integer + AVX512BW + Shift + + + + + + Shift packed 16-bit integers in "a" right by the amount specified by the corresponding element in "count" while shifting in sign bits, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -(x87 FPU, MMX, XMM7-XMM0, MXCSR) := Load(MEM[mem_addr]) +FOR j := 0 to 31 + i := j*16 + IF k[j] + IF count[i+15:i] < 16 + dst[i+15:i] := SignExtend16(a[i+15:i] >> count[i+15:i]) + ELSE + dst[i+15:i] := (a[i+15] ? 0xFFFF : 0) + FI + ELSE + dst[i+15:i] := src[i+15:i] + FI +ENDFOR +dst[MAX:512] := 0 - +
immintrin.h
 - - FXSR - OS-Targeted - - Save the current state of the x87 FPU, MMX technology, XMM, and MXCSR registers to a 512-byte memory location at "mem_addr". The clayout of the 512-byte region depends on the operating mode. Bytes [511:464] are available for software use and will not be overwritten by the processor. + + Integer + AVX512BW + Shift + + + + + Shift packed 16-bit integers in "a" right by the amount specified by the corresponding element in "count" while shifting in sign bits, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -MEM[mem_addr+511*8:mem_addr] := Fxsave(x87 FPU, MMX, XMM7-XMM0, MXCSR) +FOR j := 0 to 31 + i := j*16 + IF k[j] + IF count[i+15:i] < 16 + dst[i+15:i] := SignExtend16(a[i+15:i] >> count[i+15:i]) + ELSE + dst[i+15:i] := (a[i+15] ? 0xFFFF : 0) + FI + ELSE + dst[i+15:i] := 0 + FI +ENDFOR +dst[MAX:512] := 0 - +
immintrin.h
 - - FXSR - OS-Targeted - - Save the current state of the x87 FPU, MMX technology, XMM, and MXCSR registers to a 512-byte memory location at "mem_addr". The layout of the 512-byte region depends on the operating mode. Bytes [511:464] are available for software use and will not be overwritten by the processor. + + Integer + AVX512BW + Shift + + + + Shift packed 16-bit integers in "a" right by the amount specified by the corresponding element in "count" while shifting in sign bits, and store the results in "dst". -MEM[mem_addr+511*8:mem_addr] := Fxsave64(x87 FPU, MMX, XMM7-XMM0, MXCSR) +FOR j := 0 to 31 + i := j*16 + IF count[i+15:i] < 16 + dst[i+15:i] := SignExtend16(a[i+15:i] >> count[i+15:i]) + ELSE + dst[i+15:i] := (a[i+15] ? 0xFFFF : 0) + FI +ENDFOR +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer + AVX512VL + AVX512BW Shift - - - Shift the bits of unsigned 64-bit integer "a" left by the number of bits specified in "shift", rotating the most-significant bit to the least-significant bit location, and store the unsigned result in "dst". + + + + + + Shift packed 16-bit integers in "a" right by the amount specified by the corresponding element in "count" while shifting in sign bits, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -dst := a -count := shift BITWISE AND 63 -DO WHILE (count > 0) - tmp[0] := dst[63] - dst := (dst << 1) OR tmp[0] - count := count - 1 -OD +FOR j := 0 to 7 + i := j*16 + IF k[j] + IF count[i+15:i] < 16 + dst[i+15:i] := SignExtend16(a[i+15:i] >> count[i+15:i]) + ELSE + dst[i+15:i] := (a[i+15] ? 0xFFFF : 0) + FI + ELSE + dst[i+15:i] := src[i+15:i] + FI +ENDFOR +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer + AVX512VL + AVX512BW Shift - - - Shift the bits of unsigned 64-bit integer "a" right by the number of bits specified in "shift", rotating the least-significant bit to the most-significant bit location, and store the unsigned result in "dst". + + + + + Shift packed 16-bit integers in "a" right by the amount specified by the corresponding element in "count" while shifting in sign bits, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -dst := a -count := shift BITWISE AND 63 -DO WHILE (count > 0) - tmp[63] := dst[0] - dst := (dst >> 1) OR tmp[63] - count := count - 1 -OD +FOR j := 0 to 7 + i := j*16 + IF k[j] + IF count[i+15:i] < 16 + dst[i+15:i] := SignExtend16(a[i+15:i] >> count[i+15:i]) + ELSE + dst[i+15:i] := (a[i+15] ? 0xFFFF : 0) + FI + ELSE + dst[i+15:i] := 0 + FI +ENDFOR +dst[MAX:128] := 0 - +
immintrin.h
 - - - - - General Support - - Treat the processor-specific feature(s) specified in "a" as available. Multiple features may be OR'd together. See the valid feature flags below: + + Integer + AVX512VL + AVX512BW + Shift + + + + Shift packed 16-bit integers in "a" right by the amount specified by the corresponding element in "count" while shifting in sign bits, and store the results in "dst". -_FEATURE_GENERIC_IA32 -_FEATURE_FPU -_FEATURE_CMOV -_FEATURE_MMX -_FEATURE_FXSAVE -_FEATURE_SSE -_FEATURE_SSE2 -_FEATURE_SSE3 -_FEATURE_SSSE3 -_FEATURE_SSE4_1 -_FEATURE_SSE4_2 -_FEATURE_MOVBE -_FEATURE_POPCNT -_FEATURE_PCLMULQDQ -_FEATURE_AES -_FEATURE_F16C -_FEATURE_AVX -_FEATURE_RDRND -_FEATURE_FMA -_FEATURE_BMI -_FEATURE_LZCNT -_FEATURE_HLE -_FEATURE_RTM -_FEATURE_AVX2 -_FEATURE_KNCNI -_FEATURE_AVX512F -_FEATURE_ADX -_FEATURE_RDSEED -_FEATURE_AVX512ER -_FEATURE_AVX512PF -_FEATURE_AVX512CD -_FEATURE_SHA -_FEATURE_MPX +FOR j := 0 to 7 + i := j*16 + IF count[i+15:i] < 16 + dst[i+15:i] := SignExtend16(a[i+15:i] >> count[i+15:i]) + ELSE + dst[i+15:i] := (a[i+15] ? 0xFFFF : 0) + FI +ENDFOR +dst[MAX:128] := 0 +
immintrin.h
 - - General Support - - Dynamically query the processor to determine if the processor-specific feature(s) specified in "a" are available, and return true or false (1 or 0) if the set of features is available. Multiple features may be OR'd together. This intrinsic does not check the processor vendor. See the valid feature flags below: + + Integer + AVX512VL + AVX512BW + Shift + + + + + + Shift packed 16-bit integers in "a" right by "count" while shifting in sign bits, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -_FEATURE_GENERIC_IA32 -_FEATURE_FPU -_FEATURE_CMOV -_FEATURE_MMX -_FEATURE_FXSAVE -_FEATURE_SSE -_FEATURE_SSE2 -_FEATURE_SSE3 -_FEATURE_SSSE3 -_FEATURE_SSE4_1 -_FEATURE_SSE4_2 -_FEATURE_MOVBE -_FEATURE_POPCNT -_FEATURE_PCLMULQDQ -_FEATURE_AES -_FEATURE_F16C -_FEATURE_AVX -_FEATURE_RDRND -_FEATURE_FMA -_FEATURE_BMI -_FEATURE_LZCNT -_FEATURE_HLE -_FEATURE_RTM -_FEATURE_AVX2 -_FEATURE_KNCNI -_FEATURE_AVX512F -_FEATURE_ADX -_FEATURE_RDSEED -_FEATURE_AVX512ER -_FEATURE_AVX512PF -_FEATURE_AVX512CD -_FEATURE_SHA -_FEATURE_MPX +FOR j := 0 to 15 + i := j*16 + IF k[j] + IF count[63:0] > 15 + dst[i+15:i] := (a[i+15] ? 0xFFFF : 0x0) + ELSE + dst[i+15:i] := SignExtend16(a[i+15:i] >> count[63:0]) + FI + ELSE + dst[i+15:i] := src[i+15:i] + FI +ENDFOR +dst[MAX:256] := 0 +
immintrin.h
 - - Floating Point - SSE - Trigonometry - - Compute the inverse cosine of packed double-precision (64-bit) floating-point elements in "a" expressed in radians, and store the results in "dst". + + Integer + AVX512VL + AVX512BW + Shift + + + + + + Shift packed 16-bit integers in "a" right by "imm8" while shifting in sign bits, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 1 - i := j*64 - dst[i+63:i] := ACOS(a[i+63:i]) +FOR j := 0 to 15 + i := j*16 + IF k[j] + IF imm8[7:0] > 15 + dst[i+15:i] := (a[i+15] ? 0xFFFF : 0x0) + ELSE + dst[i+15:i] := SignExtend16(a[i+15:i] >> imm8[7:0]) + FI + ELSE + dst[i+15:i] := src[i+15:i] + FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:256] := 0 +
immintrin.h
 - - Floating Point - SSE - Trigonometry - - Compute the inverse cosine of packed single-precision (32-bit) floating-point elements in "a" expressed in radians, and store the results in "dst". + + Integer + AVX512VL + AVX512BW + Shift + + + + + Shift packed 16-bit integers in "a" right by "count" while shifting in sign bits, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*32 - dst[i+31:i] := ACOS(a[i+31:i]) +FOR j := 0 to 15 + i := j*16 + IF k[j] + IF count[63:0] > 15 + dst[i+15:i] := (a[i+15] ? 0xFFFF : 0x0) + ELSE + dst[i+15:i] := SignExtend16(a[i+15:i] >> count[63:0]) + FI + ELSE + dst[i+15:i] := 0 + FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:256] := 0 +
immintrin.h
 - - Floating Point - SSE - Trigonometry - - Compute the inverse hyperbolic cosine of packed double-precision (64-bit) floating-point elements in "a" expressed in radians, and store the results in "dst". + + Integer + AVX512VL + AVX512BW + Shift + + + + + Shift packed 16-bit integers in "a" right by "imm8" while shifting in sign bits, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 1 - i := j*64 - dst[i+63:i] := ACOSH(a[i+63:i]) +FOR j := 0 to 15 + i := j*16 + IF k[j] + IF imm8[7:0] > 15 + dst[i+15:i] := (a[i+15] ? 0xFFFF : 0x0) + ELSE + dst[i+15:i] := SignExtend16(a[i+15:i] >> imm8[7:0]) + FI + ELSE + dst[i+15:i] := 0 + FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:256] := 0 +
immintrin.h
 - - Floating Point - SSE - Trigonometry - - Compute the inverse hyperbolic cosine of packed single-precision (32-bit) floating-point elements in "a" expressed in radians, and store the results in "dst". + + Integer + AVX512BW + Shift + + + + + + Shift packed 16-bit integers in "a" right by "count" while shifting in sign bits, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*32 - dst[i+31:i] := ACOSH(a[i+31:i]) +FOR j := 0 to 31 + i := j*16 + IF k[j] + IF count[63:0] > 15 + dst[i+15:i] := (a[i+15] ? 0xFFFF : 0x0) + ELSE + dst[i+15:i] := SignExtend16(a[i+15:i] >> count[63:0]) + FI + ELSE + dst[i+15:i] := src[i+15:i] + FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 +
immintrin.h
 - - Floating Point - SSE - Trigonometry - - Compute the inverse sine of packed double-precision (64-bit) floating-point elements in "a" expressed in radians, and store the results in "dst". + + Integer + AVX512BW + Shift + + + + + + Shift packed 16-bit integers in "a" right by "imm8" while shifting in sign bits, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 1 - i := j*64 - dst[i+63:i] := ASIN(a[i+63:i]) +FOR j := 0 to 31 + i := j*16 + IF k[j] + IF imm8[7:0] > 15 + dst[i+15:i] := (a[i+15] ? 0xFFFF : 0x0) + ELSE + dst[i+15:i] := SignExtend16(a[i+15:i] >> imm8[7:0]) + FI + ELSE + dst[i+15:i] := src[i+15:i] + FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 +
immintrin.h
 - - Floating Point - SSE - Trigonometry - - Compute the inverse sine of packed single-precision (32-bit) floating-point elements in "a" expressed in radians, and store the results in "dst". + + Integer + AVX512BW + Shift + + + + + Shift packed 16-bit integers in "a" right by "count" while shifting in sign bits, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*32 - dst[i+31:i] := ASIN(a[i+31:i]) +FOR j := 0 to 31 + i := j*16 + IF k[j] + IF count[63:0] > 15 + dst[i+15:i] := (a[i+15] ? 0xFFFF : 0x0) + ELSE + dst[i+15:i] := SignExtend16(a[i+15:i] >> count[63:0]) + FI + ELSE + dst[i+15:i] := 0 + FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 +
immintrin.h
 - - Floating Point - SSE - Trigonometry - - Compute the inverse hyperbolic sine of packed double-precision (64-bit) floating-point elements in "a" expressed in radians, and store the results in "dst". + + Integer + AVX512BW + Shift + + + + + Shift packed 16-bit integers in "a" right by "imm8" while shifting in sign bits, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 1 - i := j*64 - dst[i+63:i] := ASINH(a[i+63:i]) +FOR j := 0 to 31 + i := j*16 + IF k[j] + IF imm8[7:0] > 15 + dst[i+15:i] := (a[i+15] ? 0xFFFF : 0x0) + ELSE + dst[i+15:i] := SignExtend16(a[i+15:i] >> imm8[7:0]) + FI + ELSE + dst[i+15:i] := 0 + FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 +
immintrin.h
 - - Floating Point - SSE - Trigonometry - - Compute the inverse hyperbolic sine of packed single-precision (32-bit) floating-point elements in "a" expressed in radians, and store the results in "dst". + + Integer + AVX512BW + Shift + + + + Shift packed 16-bit integers in "a" right by "count" while shifting in sign bits, and store the results in "dst". -FOR j := 0 to 3 - i := j*32 - dst[i+31:i] := ASINH(a[i+31:i]) +FOR j := 0 to 31 + i := j*16 + IF count[63:0] > 15 + dst[i+15:i] := (a[i+15] ? 0xFFFF : 0x0) + ELSE + dst[i+15:i] := SignExtend16(a[i+15:i] >> count[63:0]) + FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 +
immintrin.h
 - - Floating Point - SSE - Trigonometry - - Compute the inverse tangent of packed double-precision (64-bit) floating-point elements in "a" expressed in radians, and store the results in "dst". + + Integer + AVX512BW + Shift + + + + Shift packed 16-bit integers in "a" right by "imm8" while shifting in sign bits, and store the results in "dst". -FOR j := 0 to 1 - i := j*64 - dst[i+63:i] := ATAN(a[i+63:i]) +FOR j := 0 to 31 + i := j*16 + IF imm8[7:0] > 15 + dst[i+15:i] := (a[i+15] ? 0xFFFF : 0x0) + ELSE + dst[i+15:i] := SignExtend16(a[i+15:i] >> imm8[7:0]) + FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 +
immintrin.h
 - - Floating Point - SSE - Trigonometry - - Compute the inverse tangent of packed single-precision (32-bit) floating-point elements in "a" expressed in radians, and store the results in "dst". + + Integer + AVX512VL + AVX512BW + Shift + + + + + + Shift packed 16-bit integers in "a" right by "count" while shifting in sign bits, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*32 - dst[i+31:i] := ATAN(a[i+31:i]) +FOR j := 0 to 7 + i := j*16 + IF k[j] + IF count[63:0] > 15 + dst[i+15:i] := (a[i+15] ? 0xFFFF : 0x0) + ELSE + dst[i+15:i] := SignExtend16(a[i+15:i] >> count[63:0]) + FI + ELSE + dst[i+15:i] := src[i+15:i] + FI ENDFOR dst[MAX:128] := 0 +
immintrin.h
 - - Floating Point - SSE - Trigonometry - - - Compute the inverse tangent of packed double-precision (64-bit) floating-point elements in "a" divided by packed elements in "b", and store the results in "dst" expressed in radians. + + Integer + AVX512VL + AVX512BW + Shift + + + + + + Shift packed 16-bit integers in "a" right by "imm8" while shifting in sign bits, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 1 - i := j*64 - dst[i+63:i] := ATAN(a[i+63:i] / b[i+63:i]) +FOR j := 0 to 7 + i := j*16 + IF k[j] + IF imm8[7:0] > 15 + dst[i+15:i] := (a[i+15] ? 0xFFFF : 0x0) + ELSE + dst[i+15:i] := SignExtend16(a[i+15:i] >> imm8[7:0]) + FI + ELSE + dst[i+15:i] := src[i+15:i] + FI ENDFOR dst[MAX:128] := 0 +
immintrin.h
 - - Floating Point - SSE - Trigonometry - - - Compute the inverse tangent of packed single-precision (32-bit) floating-point elements in "a" divided by packed elements in "b", and store the results in "dst" expressed in radians. + + Integer + AVX512VL + AVX512BW + Shift + + + + + Shift packed 16-bit integers in "a" right by "count" while shifting in sign bits, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*32 - dst[i+31:i] := ATAN(a[i+31:i] / b[i+31:i]) +FOR j := 0 to 7 + i := j*16 + IF k[j] + IF count[63:0] > 15 + dst[i+15:i] := (a[i+15] ? 0xFFFF : 0x0) + ELSE + dst[i+15:i] := SignExtend16(a[i+15:i] >> count[63:0]) + FI + ELSE + dst[i+15:i] := 0 + FI ENDFOR dst[MAX:128] := 0 +
immintrin.h
 - - Floating Point - SSE - Trigonometry - - Compute the inverse hyperbolic tangent of packed double-precision (64-bit) floating-point elements in "a" expressed in radians, and store the results in "dst". + + Integer + AVX512VL + AVX512BW + Shift + + + + + Shift packed 16-bit integers in "a" right by "imm8" while shifting in sign bits, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 1 - i := j*64 - dst[i+63:i] := ATANH(a[i+63:i]) +FOR j := 0 to 7 + i := j*16 + IF k[j] + IF imm8[7:0] > 15 + dst[i+15:i] := (a[i+15] ? 0xFFFF : 0x0) + ELSE + dst[i+15:i] := SignExtend16(a[i+15:i] >> imm8[7:0]) + FI + ELSE + dst[i+15:i] := 0 + FI ENDFOR dst[MAX:128] := 0 +
immintrin.h
 - - Floating Point - SSE - Trigonometry - - Compute the inverse hyperbolic tangent of packed single-precision (32-bit) floating-point elements in "a" expressed in radians, and store the results in "dst". + + Integer + AVX512BW + Shift + + + + Shift 128-bit lanes in "a" right by "imm8" bytes while shifting in zeros, and store the results in "dst". -FOR j := 0 to 3 - i := j*32 - dst[i+31:i] := ATANH(a[i+31:i]) -ENDFOR -dst[MAX:128] := 0 +tmp := imm8[7:0] +IF tmp > 15 + tmp := 16 +FI +dst[127:0] := a[127:0] >> (tmp*8) +dst[255:128] := a[255:128] >> (tmp*8) +dst[383:256] := a[383:256] >> (tmp*8) +dst[511:384] := a[511:384] >> (tmp*8) +dst[MAX:512] := 0 +
immintrin.h
 - - Floating Point - SSE - Elementary Math Functions - - Compute the cube root of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst". + + Integer + AVX512VL + AVX512BW + Shift + + + + + + Shift packed 16-bit integers in "a" right by the amount specified by the corresponding element in "count" while shifting in zeros, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 1 - i := j*64 - dst[i+63:i] := CubeRoot(a[i+63:i]) +FOR j := 0 to 15 + i := j*16 + IF k[j] + IF count[i+15:i] < 16 + dst[i+15:i] := ZeroExtend16(a[i+15:i] >> count[i+15:i]) + ELSE + dst[i+15:i] := 0 + FI + ELSE + dst[i+15:i] := src[i+15:i] + FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:256] := 0 +
immintrin.h
 - - Floating Point - SSE - Elementary Math Functions - - Compute the cube root of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst". + + Integer + AVX512VL + AVX512BW + Shift + + + + + Shift packed 16-bit integers in "a" right by the amount specified by the corresponding element in "count" while shifting in zeros, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*32 - dst[i+31:i] := CubeRoot(a[i+31:i]) +FOR j := 0 to 15 + i := j*16 + IF k[j] + IF count[i+15:i] < 16 + dst[i+15:i] := ZeroExtend16(a[i+15:i] >> count[i+15:i]) + ELSE + dst[i+15:i] := 0 + FI + ELSE + dst[i+15:i] := 0 + FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:256] := 0 +
immintrin.h
 - - Floating Point - SSE - Probability/Statistics - - Compute the cumulative distribution function of packed double-precision (64-bit) floating-point elements in "a" using the normal distribution, and store the results in "dst". + + Integer + AVX512VL + AVX512BW + Shift + + + + Shift packed 16-bit integers in "a" right by the amount specified by the corresponding element in "count" while shifting in zeros, and store the results in "dst". -FOR j := 0 to 1 - i := j*64 - dst[i+63:i] := CDFNormal(a[i+63:i]) +FOR j := 0 to 15 + i := j*16 + IF count[i+15:i] < 16 + dst[i+15:i] := ZeroExtend16(a[i+15:i] >> count[i+15:i]) + ELSE + dst[i+15:i] := 0 + FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:256] := 0 +
immintrin.h
 - - Floating Point - SSE - Probability/Statistics - - Compute the cumulative distribution function of packed single-precision (32-bit) floating-point elements in "a" using the normal distribution, and store the results in "dst". + + Integer + AVX512BW + Shift + + + + + + Shift packed 16-bit integers in "a" right by the amount specified by the corresponding element in "count" while shifting in zeros, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*32 - dst[i+31:i] := CDFNormal(a[i+31:i]) +FOR j := 0 to 31 + i := j*16 + IF k[j] + IF count[i+15:i] < 16 + dst[i+15:i] := ZeroExtend16(a[i+15:i] >> count[i+15:i]) + ELSE + dst[i+15:i] := 0 + FI + ELSE + dst[i+15:i] := src[i+15:i] + FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 +
immintrin.h
 - - Floating Point - SSE - Probability/Statistics - - Compute the inverse cumulative distribution function of packed double-precision (64-bit) floating-point elements in "a" using the normal distribution, and store the results in "dst". + + Integer + AVX512BW + Shift + + + + + Shift packed 16-bit integers in "a" right by the amount specified by the corresponding element in "count" while shifting in zeros, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 1 - i := j*64 - dst[i+63:i] := InverseCDFNormal(a[i+63:i]) +FOR j := 0 to 31 + i := j*16 + IF k[j] + IF count[i+15:i] < 16 + dst[i+15:i] := ZeroExtend16(a[i+15:i] >> count[i+15:i]) + ELSE + dst[i+15:i] := 0 + FI + ELSE + dst[i+15:i] := 0 + FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 +
immintrin.h
 - - Floating Point - SSE - Probability/Statistics - - Compute the inverse cumulative distribution function of packed single-precision (32-bit) floating-point elements in "a" using the normal distribution, and store the results in "dst". + + Integer + AVX512BW + Shift + + + + Shift packed 16-bit integers in "a" right by the amount specified by the corresponding element in "count" while shifting in zeros, and store the results in "dst". -FOR j := 0 to 3 - i := j*32 - dst[i+31:i] := InverseCDFNormal(a[i+31:i]) +FOR j := 0 to 31 + i := j*16 + IF count[i+15:i] < 16 + dst[i+15:i] := ZeroExtend16(a[i+15:i] >> count[i+15:i]) + ELSE + dst[i+15:i] := 0 + FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 +
immintrin.h
 - - Floating Point - SSE - Elementary Math Functions - - Compute the exponential value of "e" raised to the power of packed complex single-precision (32-bit) floating-point elements in "a", and store the results in "dst". + + Integer + AVX512VL + AVX512BW + Shift + + + + + + Shift packed 16-bit integers in "a" right by the amount specified by the corresponding element in "count" while shifting in zeros, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*32 - dst[i+31:i] := e^(a[i+31:i]) +FOR j := 0 to 7 + i := j*16 + IF k[j] + IF count[i+15:i] < 16 + dst[i+15:i] := ZeroExtend16(a[i+15:i] >> count[i+15:i]) + ELSE + dst[i+15:i] := 0 + FI + ELSE + dst[i+15:i] := src[i+15:i] + FI ENDFOR dst[MAX:128] := 0 +
immintrin.h
 - - Floating Point - SSE - Elementary Math Functions - - Compute the natural logarithm of packed complex single-precision (32-bit) floating-point elements in "a", and store the results in "dst". + + Integer + AVX512VL + AVX512BW + Shift + + + + + Shift packed 16-bit integers in "a" right by the amount specified by the corresponding element in "count" while shifting in zeros, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*32 - dst[i+31:i] := ln(a[i+31:i]) +FOR j := 0 to 7 + i := j*16 + IF k[j] + IF count[i+15:i] < 16 + dst[i+15:i] := ZeroExtend16(a[i+15:i] >> count[i+15:i]) + ELSE + dst[i+15:i] := 0 + FI + ELSE + dst[i+15:i] := 0 + FI ENDFOR dst[MAX:128] := 0 +
immintrin.h
 - - Floating Point - SSE - Trigonometry - - Compute the cosine of packed double-precision (64-bit) floating-point elements in "a" expressed in radians, and store the results in "dst". + + Integer + AVX512VL + AVX512BW + Shift + + + + Shift packed 16-bit integers in "a" right by the amount specified by the corresponding element in "count" while shifting in zeros, and store the results in "dst". -FOR j := 0 to 1 - i := j*64 - dst[i+63:i] := COS(a[i+63:i]) +FOR j := 0 to 7 + i := j*16 + IF count[i+15:i] < 16 + dst[i+15:i] := ZeroExtend16(a[i+15:i] >> count[i+15:i]) + ELSE + dst[i+15:i] := 0 + FI ENDFOR dst[MAX:128] := 0 +
immintrin.h
 - - Floating Point - SSE - Trigonometry - - Compute the cosine of packed single-precision (32-bit) floating-point elements in "a" expressed in radians, and store the results in "dst". + + Integer + AVX512VL + AVX512BW + Shift + + + + + + Shift packed 16-bit integers in "a" right by "count" while shifting in zeros, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*32 - dst[i+31:i] := COS(a[i+31:i]) +FOR j := 0 to 15 + i := j*16 + IF k[j] + IF count[63:0] > 15 + dst[i+15:i] := 0 + ELSE + dst[i+15:i] := ZeroExtend16(a[i+15:i] >> count[63:0]) + FI + ELSE + dst[i+15:i] := src[i+15:i] + FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:256] := 0 +
immintrin.h
 - - Floating Point - SSE - Trigonometry - - Compute the cosine of packed double-precision (64-bit) floating-point elements in "a" expressed in degrees, and store the results in "dst". + + Integer + AVX512VL + AVX512BW + Shift + + + + + + Shift packed 16-bit integers in "a" right by "imm8" while shifting in zeros, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 1 - i := j*64 - dst[i+63:i] := COSD(a[i+63:i]) +FOR j := 0 to 15 + i := j*16 + IF k[j] + IF imm8[7:0] > 15 + dst[i+15:i] := 0 + ELSE + dst[i+15:i] := ZeroExtend16(a[i+15:i] >> imm8[7:0]) + FI + ELSE + dst[i+15:i] := src[i+15:i] + FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:256] := 0 +
immintrin.h
 - - Floating Point - SSE - Trigonometry - - Compute the cosine of packed single-precision (32-bit) floating-point elements in "a" expressed in degrees, and store the results in "dst". + + Integer + AVX512VL + AVX512BW + Shift + + + + + Shift packed 16-bit integers in "a" right by "count" while shifting in zeros, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*32 - dst[i+31:i] := COSD(a[i+31:i]) +FOR j := 0 to 15 + i := j*16 + IF k[j] + IF count[63:0] > 15 + dst[i+15:i] := 0 + ELSE + dst[i+15:i] := ZeroExtend16(a[i+15:i] >> count[63:0]) + FI + ELSE + dst[i+15:i] := 0 + FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:256] := 0 +
immintrin.h
 - - Floating Point - SSE - Trigonometry - - Compute the hyperbolic cosine of packed double-precision (64-bit) floating-point elements in "a" expressed in radians, and store the results in "dst". + + Integer + AVX512VL + AVX512BW + Shift + + + + + Shift packed 16-bit integers in "a" right by "imm8" while shifting in zeros, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 1 - i := j*64 - dst[i+63:i] := COSH(a[i+63:i]) +FOR j := 0 to 15 + i := j*16 + IF k[j] + IF imm8[7:0] > 15 + dst[i+15:i] := 0 + ELSE + dst[i+15:i] := ZeroExtend16(a[i+15:i] >> imm8[7:0]) + FI + ELSE + dst[i+15:i] := 0 + FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:256] := 0 +
immintrin.h
 - - Floating Point - SSE - Trigonometry - - Compute the hyperbolic cosine of packed single-precision (32-bit) floating-point elements in "a" expressed in radians, and store the results in "dst". + + Integer + AVX512BW + Shift + + + + + + Shift packed 16-bit integers in "a" right by "count" while shifting in zeros, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*32 - dst[i+31:i] := COSH(a[i+31:i]) +FOR j := 0 to 31 + i := j*16 + IF k[j] + IF count[63:0] > 15 + dst[i+15:i] := 0 + ELSE + dst[i+15:i] := ZeroExtend16(a[i+15:i] >> count[63:0]) + FI + ELSE + dst[i+15:i] := src[i+15:i] + FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 +
immintrin.h
 - - Floating Point - SSE - Elementary Math Functions - - Compute the square root of packed complex single-precision (32-bit) floating-point elements in "a", and store the results in "dst". + + Integer + AVX512BW + Shift + + + + + + Shift packed 16-bit integers in "a" right by "imm8" while shifting in zeros, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*32 - dst[i+31:i] := SQRT(a[i+31:i]) +FOR j := 0 to 31 + i := j*16 + IF k[j] + IF imm8[7:0] > 15 + dst[i+15:i] := 0 + ELSE + dst[i+15:i] := ZeroExtend16(a[i+15:i] >> imm8[7:0]) + FI + ELSE + dst[i+15:i] := src[i+15:i] + FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 +
immintrin.h
 - + Integer - SSE - Arithmetic - - - Divide packed 8-bit integers in "a" by packed elements in "b", and store the truncated results in "dst". + AVX512BW + Shift + + + + + Shift packed 16-bit integers in "a" right by "count" while shifting in zeros, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := 8*j - dst[i+7:i] := TRUNCATE(a[i+7:i] / b[i+7:i]) +FOR j := 0 to 31 + i := j*16 + IF k[j] + IF count[63:0] > 15 + dst[i+15:i] := 0 + ELSE + dst[i+15:i] := ZeroExtend16(a[i+15:i] >> count[63:0]) + FI + ELSE + dst[i+15:i] := 0 + FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 +
immintrin.h
 - + Integer - SSE - Arithmetic - - - Divide packed 16-bit integers in "a" by packed elements in "b", and store the truncated results in "dst". + AVX512BW + Shift + + + + + Shift packed 16-bit integers in "a" right by "imm8" while shifting in zeros, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := 16*j - dst[i+15:i] := TRUNCATE(a[i+15:i] / b[i+15:i]) +FOR j := 0 to 31 + i := j*16 + IF k[j] + IF imm8[7:0] > 15 + dst[i+15:i] := 0 + ELSE + dst[i+15:i] := ZeroExtend16(a[i+15:i] >> imm8[7:0]) + FI + ELSE + dst[i+15:i] := 0 + FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 +
immintrin.h
 - + Integer - SSE - Arithmetic - - - Divide packed 32-bit integers in "a" by packed elements in "b", and store the truncated results in "dst". + AVX512BW + Shift + + + + Shift packed 16-bit integers in "a" right by "count" while shifting in zeros, and store the results in "dst". -FOR j := 0 to 3 - i := 32*j - dst[i+31:i] := TRUNCATE(a[i+31:i] / b[i+31:i]) +FOR j := 0 to 31 + i := j*16 + IF count[63:0] > 15 + dst[i+15:i] := 0 + ELSE + dst[i+15:i] := ZeroExtend16(a[i+15:i] >> count[63:0]) + FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 +
immintrin.h
 - + Integer - SSE - Arithmetic - - - Divide packed 64-bit integers in "a" by packed elements in "b", and store the truncated results in "dst". + AVX512BW + Shift + + + + Shift packed 16-bit integers in "a" right by "imm8" while shifting in zeros, and store the results in "dst". -FOR j := 0 to 1 - i := 64*j - dst[i+63:i] := TRUNCATE(a[i+63:i] / b[i+63:i]) +FOR j := 0 to 31 + i := j*16 + IF imm8[7:0] > 15 + dst[i+15:i] := 0 + ELSE + dst[i+15:i] := ZeroExtend16(a[i+15:i] >> imm8[7:0]) + FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 +
immintrin.h
 - + Integer - SSE - Arithmetic - - - Divide packed unsigned 8-bit integers in "a" by packed elements in "b", and store the truncated results in "dst". + AVX512VL + AVX512BW + Shift + + + + + + Shift packed 16-bit integers in "a" right by "count" while shifting in zeros, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := 8*j - dst[i+7:i] := TRUNCATE(a[i+7:i] / b[i+7:i]) +FOR j := 0 to 7 + i := j*16 + IF k[j] + IF count[63:0] > 15 + dst[i+15:i] := 0 + ELSE + dst[i+15:i] := ZeroExtend16(a[i+15:i] >> count[63:0]) + FI + ELSE + dst[i+15:i] := src[i+15:i] + FI ENDFOR dst[MAX:128] := 0 +
immintrin.h
 - + Integer - SSE - Arithmetic - - - Divide packed unsigned 16-bit integers in "a" by packed elements in "b", and store the truncated results in "dst". + AVX512VL + AVX512BW + Shift + + + + + + Shift packed 16-bit integers in "a" right by "imm8" while shifting in zeros, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). FOR j := 0 to 7 - i := 16*j - dst[i+15:i] := TRUNCATE(a[i+15:i] / b[i+15:i]) + i := j*16 + IF k[j] + IF imm8[7:0] > 15 + dst[i+15:i] := 0 + ELSE + dst[i+15:i] := ZeroExtend16(a[i+15:i] >> imm8[7:0]) + FI + ELSE + dst[i+15:i] := src[i+15:i] + FI ENDFOR dst[MAX:128] := 0 +
immintrin.h
 - + Integer - SSE - Arithmetic - - - Divide packed unsigned 32-bit integers in "a" by packed elements in "b", and store the truncated results in "dst". + AVX512VL + AVX512BW + Shift + + + + + Shift packed 16-bit integers in "a" right by "count" while shifting in zeros, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := 32*j - dst[i+31:i] := TRUNCATE(a[i+31:i] / b[i+31:i]) +FOR j := 0 to 7 + i := j*16 + IF k[j] + IF count[63:0] > 15 + dst[i+15:i] := 0 + ELSE + dst[i+15:i] := ZeroExtend16(a[i+15:i] >> count[63:0]) + FI + ELSE + dst[i+15:i] := 0 + FI ENDFOR dst[MAX:128] := 0 +
immintrin.h
 - + Integer - SSE - Arithmetic - - - Divide packed unsigned 64-bit integers in "a" by packed elements in "b", and store the truncated results in "dst". + AVX512VL + AVX512BW + Shift + + + + + Shift packed 16-bit integers in "a" right by "imm8" while shifting in zeros, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 1 - i := 64*j - dst[i+63:i] := TRUNCATE(a[i+63:i] / b[i+63:i]) +FOR j := 0 to 7 + i := j*16 + IF k[j] + IF imm8[7:0] > 15 + dst[i+15:i] := 0 + ELSE + dst[i+15:i] := ZeroExtend16(a[i+15:i] >> imm8[7:0]) + FI + ELSE + dst[i+15:i] := 0 + FI ENDFOR dst[MAX:128] := 0 +
immintrin.h
 - - Floating Point - SSE + + Integer + AVX512VL + AVX512BW Arithmetic - - Compute the error function of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst". + + + + + + Subtract packed 8-bit integers in "b" from packed 8-bit integers in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 1 - i := j*64 - dst[i+63:i] := ERF(a[i+63:i]) +FOR j := 0 to 31 + i := j*8 + IF k[j] + dst[i+7:i] := a[i+7:i] - b[i+7:i] + ELSE + dst[i+7:i] := src[i+7:i] + FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:256] := 0 +
immintrin.h
 - - Floating Point - SSE - Probability/Statistics - - Compute the error function of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst". + + Integer + AVX512VL + AVX512BW + Arithmetic + + + + + Subtract packed 8-bit integers in "b" from packed 8-bit integers in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*32 - dst[i+31:i] := ERF(a[i+31:i]) +FOR j := 0 to 31 + i := j*8 + IF k[j] + dst[i+7:i] := a[i+7:i] - b[i+7:i] + ELSE + dst[i+7:i] := 0 + FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:256] := 0 +
immintrin.h
 - - Floating Point - SSE - Probability/Statistics - - Compute the complementary error function of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst". + + Integer + AVX512BW + Arithmetic + + + + + + Subtract packed 8-bit integers in "b" from packed 8-bit integers in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 1 - i := j*64 - dst[i+63:i] := 1.0 - ERF(a[i+63:i]) +FOR j := 0 to 63 + i := j*8 + IF k[j] + dst[i+7:i] := a[i+7:i] - b[i+7:i] + ELSE + dst[i+7:i] := src[i+7:i] + FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 +
immintrin.h
 - - Floating Point - SSE - Probability/Statistics - - Compute the complementary error function of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst". + + Integer + AVX512BW + Arithmetic + + + + + Subtract packed 8-bit integers in "b" from packed 8-bit integers in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*32 - dst[i+31:i] := 1.0 - ERF(a[i+31:i]) +FOR j := 0 to 63 + i := j*8 + IF k[j] + dst[i+7:i] := a[i+7:i] - b[i+7:i] + ELSE + dst[i+7:i] := 0 + FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 +
immintrin.h
 - - Floating Point - SSE - Probability/Statistics - - Compute the inverse complementary error function of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst". + + Integer + AVX512BW + Arithmetic + + + + Subtract packed 8-bit integers in "b" from packed 8-bit integers in "a", and store the results in "dst". -FOR j := 0 to 1 - i := j*64 - dst[i+63:i] := 1.0 / (1.0 - ERF(a[i+63:i])) +FOR j := 0 to 63 + i := j*8 + dst[i+7:i] := a[i+7:i] - b[i+7:i] ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 +
immintrin.h
 - - Floating Point - SSE - Probability/Statistics - - Compute the inverse complementary error function of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst". + + Integer + AVX512VL + AVX512BW + Arithmetic + + + + + + Subtract packed 8-bit integers in "b" from packed 8-bit integers in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*32 - dst[i+31:i] := 1.0 / (1.0 - ERF(a[i+31:i])) +FOR j := 0 to 15 + i := j*8 + IF k[j] + dst[i+7:i] := a[i+7:i] - b[i+7:i] + ELSE + dst[i+7:i] := src[i+7:i] + FI ENDFOR dst[MAX:128] := 0 +
immintrin.h
 - - Floating Point - SSE - Probability/Statistics - - Compute the inverse error function of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst". + + Integer + AVX512VL + AVX512BW + Arithmetic + + + + + Subtract packed 8-bit integers in "b" from packed 8-bit integers in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 1 - i := j*64 - dst[i+63:i] := 1.0 / ERF(a[i+63:i]) +FOR j := 0 to 15 + i := j*8 + IF k[j] + dst[i+7:i] := a[i+7:i] - b[i+7:i] + ELSE + dst[i+7:i] := 0 + FI ENDFOR dst[MAX:128] := 0 +
immintrin.h
 - - Floating Point - SSE - Probability/Statistics - - Compute the inverse error function of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst". + + Integer + AVX512VL + AVX512BW + Arithmetic + + + + + + Subtract packed signed 8-bit integers in "b" from packed 8-bit integers in "a" using saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*32 - dst[i+31:i] := 1.0 / ERF(a[i+31:i]) +FOR j := 0 to 31 + i := j*8 + IF k[j] + dst[i+7:i] := Saturate8(a[i+7:i] - b[i+7:i]) + ELSE + dst[i+7:i] := src[i+7:i] + FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:256] := 0 +
immintrin.h
 - - Floating Point - SSE - Elementary Math Functions - - Compute the exponential value of "e" raised to the power of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst". + + Integer + AVX512VL + AVX512BW + Arithmetic + + + + + Subtract packed signed 8-bit integers in "b" from packed 8-bit integers in "a" using saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 1 - i := j*64 - dst[i+63:i] := e^(a[i+63:i]) +FOR j := 0 to 31 + i := j*8 + IF k[j] + dst[i+7:i] := Saturate8(a[i+7:i] - b[i+7:i]) + ELSE + dst[i+7:i] := 0 + FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:256] := 0 +
immintrin.h
 - - Floating Point - SSE - Elementary Math Functions - - Compute the exponential value of "e" raised to the power of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst". + + Integer + AVX512BW + Arithmetic + + + + + + Subtract packed signed 8-bit integers in "b" from packed 8-bit integers in "a" using saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*32 - dst[i+31:i] := e^(a[i+31:i]) +FOR j := 0 to 63 + i := j*8 + IF k[j] + dst[i+7:i] := Saturate8(a[i+7:i] - b[i+7:i]) + ELSE + dst[i+7:i] := src[i+7:i] + FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 +
immintrin.h
 - - Floating Point - SSE - Elementary Math Functions - - Compute the exponential value of 10 raised to the power of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst". + + Integer + AVX512BW + Arithmetic + + + + + Subtract packed signed 8-bit integers in "b" from packed 8-bit integers in "a" using saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 1 - i := j*64 - dst[i+63:i] := 10^(a[i+63:i]) +FOR j := 0 to 63 + i := j*8 + IF k[j] + dst[i+7:i] := Saturate8(a[i+7:i] - b[i+7:i]) + ELSE + dst[i+7:i] := 0 + FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 +
immintrin.h
 - - Floating Point - SSE - Elementary Math Functions - - Compute the exponential value of 10 raised to the power of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst". - -FOR j := 0 to 3 - i := j*32 - dst[i+31:i] := 10^(a[i+31:i]) -ENDFOR -dst[MAX:128] := 0 - -
immintrin.h
- - - Floating Point - SSE - Elementary Math Functions - - Compute the exponential value of 2 raised to the power of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst". - -FOR j := 0 to 1 - i := j*64 - dst[i+63:i] := 2^(a[i+63:i]) -ENDFOR -dst[MAX:128] := 0 - -
immintrin.h
- - - Floating Point - SSE - Elementary Math Functions - - Compute the exponential value of 2 raised to the power of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst". + + Integer + AVX512BW + Arithmetic + + + + Subtract packed signed 8-bit integers in "b" from packed 8-bit integers in "a" using saturation, and store the results in "dst". -FOR j := 0 to 3 - i := j*32 - dst[i+31:i] := 2^(a[i+31:i]) +FOR j := 0 to 63 + i := j*8 + dst[i+7:i] := Saturate8(a[i+7:i] - b[i+7:i]) ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 +
immintrin.h
 - - Floating Point - SSE - Elementary Math Functions - - Compute the exponential value of "e" raised to the power of packed double-precision (64-bit) floating-point elements in "a", subtract one from each element, and store the results in "dst". + + Integer + AVX512VL + AVX512BW + Arithmetic + + + + + + Subtract packed signed 8-bit integers in "b" from packed 8-bit integers in "a" using saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 1 - i := j*64 - dst[i+63:i] := e^(a[i+63:i]) - 1.0 +FOR j := 0 to 15 + i := j*8 + IF k[j] + dst[i+7:i] := Saturate8(a[i+7:i] - b[i+7:i]) + ELSE + dst[i+7:i] := src[i+7:i] + FI ENDFOR dst[MAX:128] := 0 +
immintrin.h
 - - Floating Point - SSE - Elementary Math Functions - - Compute the exponential value of "e" raised to the power of packed single-precision (32-bit) floating-point elements in "a", subtract one from each element, and store the results in "dst". + + Integer + AVX512VL + AVX512BW + Arithmetic + + + + + Subtract packed signed 8-bit integers in "b" from packed 8-bit integers in "a" using saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*32 - dst[i+31:i] := e^(a[i+31:i]) - 1.0 +FOR j := 0 to 15 + i := j*8 + IF k[j] + dst[i+7:i] := Saturate8(a[i+7:i] - b[i+7:i]) + ELSE + dst[i+7:i] := 0 + FI ENDFOR dst[MAX:128] := 0 +
immintrin.h
 - - Floating Point - SSE - Trigonometry - - - Compute the length of the hypotenous of a right triangle, with the lengths of the other two sides of the triangle stored as packed double-precision (64-bit) floating-point elements in "a" and "b", and store the results in "dst". + + Integer + AVX512VL + AVX512BW + Arithmetic + + + + + + Subtract packed signed 16-bit integers in "b" from packed 16-bit integers in "a" using saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 1 - i := j*64 - dst[i+63:i] := SQRT(a[i+63:i]^2 + b[i+63:i]^2) +FOR j := 0 to 15 + i := j*16 + IF k[j] + dst[i+15:i] := Saturate16(a[i+15:i] - b[i+15:i]) + ELSE + dst[i+15:i] := src[i+15:i] + FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:256] := 0 +
immintrin.h
 - - Floating Point - SSE - Trigonometry - - - Compute the length of the hypotenous of a right triangle, with the lengths of the other two sides of the triangle stored as packed single-precision (32-bit) floating-point elements in "a" and "b", and store the results in "dst". + + Integer + AVX512VL + AVX512BW + Arithmetic + + + + + Subtract packed signed 16-bit integers in "b" from packed 16-bit integers in "a" using saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*32 - dst[i+31:i] := SQRT(a[i+31:i]^2 + b[i+31:i]^2) +FOR j := 0 to 15 + i := j*16 + IF k[j] + dst[i+15:i] := Saturate16(a[i+15:i] - b[i+15:i]) + ELSE + dst[i+15:i] := 0 + FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:256] := 0 +
immintrin.h
 - + Integer - SSE + AVX512BW Arithmetic - - - Divide packed 32-bit integers in "a" by packed elements in "b", and store the truncated results in "dst". + + + + + + Subtract packed signed 16-bit integers in "b" from packed 16-bit integers in "a" using saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := 32*j - dst[i+31:i] := TRUNCATE(a[i+31:i] / b[i+31:i]) +FOR j := 0 to 31 + i := j*16 + IF k[j] + dst[i+15:i] := Saturate16(a[i+15:i] - b[i+15:i]) + ELSE + dst[i+15:i] := src[i+15:i] + FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 +
immintrin.h
 - + Integer - SSE + AVX512BW Arithmetic - - - - Divide packed 32-bit integers in "a" by packed elements in "b", store the truncated results in "dst", and store the remainders as packed 32-bit integers into memory at "mem_addr". + + + + + Subtract packed signed 16-bit integers in "b" from packed 16-bit integers in "a" using saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := 32*j - dst[i+31:i] := TRUNCATE(a[i+31:i] / b[i+31:i]) - MEM[mem_addr+i+31:mem_addr+i] := REMAINDER(a[i+31:i] / b[i+31:i]) +FOR j := 0 to 31 + i := j*16 + IF k[j] + dst[i+15:i] := Saturate16(a[i+15:i] - b[i+15:i]) + ELSE + dst[i+15:i] := 0 + FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 +
immintrin.h
 - - Floating Point - SSE - Elementary Math Functions - - Compute the inverse cube root of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst". + + Integer + AVX512BW + Arithmetic + + + + Subtract packed signed 16-bit integers in "b" from packed 16-bit integers in "a" using saturation, and store the results in "dst". -FOR j := 0 to 1 - i := j*64 - dst[i+63:i] := InvCubeRoot(a[i+63:i]) +FOR j := 0 to 31 + i := j*16 + dst[i+15:i] := Saturate16(a[i+15:i] - b[i+15:i]) ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 +
immintrin.h
 - - Floating Point - SSE - Elementary Math Functions - - Compute the inverse cube root of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst". + + Integer + AVX512VL + AVX512BW + Arithmetic + + + + + + Subtract packed signed 16-bit integers in "b" from packed 16-bit integers in "a" using saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*32 - dst[i+31:i] := InvCubeRoot(a[i+31:i]) +FOR j := 0 to 7 + i := j*16 + IF k[j] + dst[i+15:i] := Saturate16(a[i+15:i] - b[i+15:i]) + ELSE + dst[i+15:i] := src[i+15:i] + FI ENDFOR dst[MAX:128] := 0 +
immintrin.h
 - - Floating Point - SSE - Elementary Math Functions - - Compute the inverse square root of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst". + + Integer + AVX512VL + AVX512BW + Arithmetic + + + + + Subtract packed signed 16-bit integers in "b" from packed 16-bit integers in "a" using saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 1 - i := j*64 - dst[i+63:i] := InvSQRT(a[i+63:i]) +FOR j := 0 to 7 + i := j*16 + IF k[j] + dst[i+15:i] := Saturate16(a[i+15:i] - b[i+15:i]) + ELSE + dst[i+15:i] := 0 + FI ENDFOR dst[MAX:128] := 0 +
immintrin.h
 - - Floating Point - SSE - Elementary Math Functions - - Compute the inverse square root of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst". + + Integer + AVX512VL + AVX512BW + Arithmetic + + + + + + Subtract packed unsigned 8-bit integers in "b" from packed unsigned 8-bit integers in "a" using saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*32 - dst[i+31:i] := InvSQRT(a[i+31:i]) +FOR j := 0 to 31 + i := j*8 + IF k[j] + dst[i+7:i] := SaturateU8(a[i+7:i] - b[i+7:i]) + ELSE + dst[i+7:i] := src[i+7:i] + FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:256] := 0 +
immintrin.h
 - + Integer - SSE + AVX512VL + AVX512BW Arithmetic - - - Divide packed 32-bit integers in "a" by packed elements in "b", and store the remainders as packed 32-bit integers in "dst". + + + + + Subtract packed unsigned 8-bit integers in "b" from packed unsigned 8-bit integers in "a" using saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := 32*j - dst[i+31:i] := REMAINDER(a[i+31:i] / b[i+31:i]) +FOR j := 0 to 31 + i := j*8 + IF k[j] + dst[i+7:i] := SaturateU8(a[i+7:i] - b[i+7:i]) + ELSE + dst[i+7:i] := 0 + FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:256] := 0 +
immintrin.h
 - - Floating Point - SSE - Elementary Math Functions - - Compute the natural logarithm of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst". + + Integer + AVX512BW + Arithmetic + + + + + + Subtract packed unsigned 8-bit integers in "b" from packed unsigned 8-bit integers in "a" using saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 1 - i := j*64 - dst[i+63:i] := ln(a[i+63:i]) +FOR j := 0 to 63 + i := j*8 + IF k[j] + dst[i+7:i] := SaturateU8(a[i+7:i] - b[i+7:i]) + ELSE + dst[i+7:i] := src[i+7:i] + FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 +
immintrin.h
 - - Floating Point - SSE - Elementary Math Functions - - Compute the natural logarithm of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst". + + Integer + AVX512BW + Arithmetic + + + + + Subtract packed unsigned 8-bit integers in "b" from packed unsigned 8-bit integers in "a" using saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*32 - dst[i+31:i] := ln(a[i+31:i]) +FOR j := 0 to 63 + i := j*8 + IF k[j] + dst[i+7:i] := SaturateU8(a[i+7:i] - b[i+7:i]) + ELSE + dst[i+7:i] := 0 + FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 +
immintrin.h
 - - Floating Point - SSE - Elementary Math Functions - - Compute the base-10 logarithm of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst". + + Integer + AVX512BW + Arithmetic + + + + Subtract packed unsigned 8-bit integers in "b" from packed unsigned 8-bit integers in "a" using saturation, and store the results in "dst". -FOR j := 0 to 1 - i := j*64 - dst[i+63:i] := log10(a[i+63:i]) +FOR j := 0 to 63 + i := j*8 + dst[i+7:i] := SaturateU8(a[i+7:i] - b[i+7:i]) ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 +
immintrin.h
 - - Floating Point - SSE - Elementary Math Functions - - Compute the base-10 logarithm of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst". + + Integer + AVX512VL + AVX512BW + Arithmetic + + + + + + Subtract packed unsigned 8-bit integers in "b" from packed unsigned 8-bit integers in "a" using saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*32 - dst[i+31:i] := log10(a[i+31:i]) +FOR j := 0 to 15 + i := j*8 + IF k[j] + dst[i+7:i] := SaturateU8(a[i+7:i] - b[i+7:i]) + ELSE + dst[i+7:i] := src[i+7:i] + FI ENDFOR dst[MAX:128] := 0 +
immintrin.h
 - - Floating Point - SSE - Elementary Math Functions - - Compute the natural logarithm of one plus packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst". + + Integer + AVX512VL + AVX512BW + Arithmetic + + + + + Subtract packed unsigned 8-bit integers in "b" from packed unsigned 8-bit integers in "a" using saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 1 - i := j*64 - dst[i+63:i] := ln(1.0 + a[i+63:i]) +FOR j := 0 to 15 + i := j*8 + IF k[j] + dst[i+7:i] := SaturateU8(a[i+7:i] - b[i+7:i]) + ELSE + dst[i+7:i] := 0 + FI ENDFOR dst[MAX:128] := 0 +
immintrin.h
 - - Floating Point - SSE - Elementary Math Functions - - Compute the natural logarithm of one plus packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst". + + Integer + AVX512VL + AVX512BW + Arithmetic + + + + + + Subtract packed unsigned 16-bit integers in "b" from packed unsigned 16-bit integers in "a" using saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*32 - dst[i+31:i] := ln(1.0 + a[i+31:i]) +FOR j := 0 to 15 + i := j*16 + IF k[j] + dst[i+15:i] := SaturateU16(a[i+15:i] - b[i+15:i]) + ELSE + dst[i+15:i] := src[i+15:i] + FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:256] := 0 +
immintrin.h
 - - Floating Point - SSE - Elementary Math Functions - - Compute the base-2 logarithm of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst". + + Integer + AVX512VL + AVX512BW + Arithmetic + + + + + Subtract packed unsigned 16-bit integers in "b" from packed unsigned 16-bit integers in "a" using saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 1 - i := j*64 - dst[i+63:i] := log2(a[i+63:i]) +FOR j := 0 to 15 + i := j*16 + IF k[j] + dst[i+15:i] := SaturateU16(a[i+15:i] - b[i+15:i]) + ELSE + dst[i+15:i] := 0 + FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:256] := 0 +
immintrin.h
 - - Floating Point - SSE - Elementary Math Functions - - Compute the base-2 logarithm of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst". + + Integer + AVX512BW + Arithmetic + + + + + + Subtract packed unsigned 16-bit integers in "b" from packed unsigned 16-bit integers in "a" using saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*32 - dst[i+31:i] := log2(a[i+31:i]) +FOR j := 0 to 31 + i := j*16 + IF k[j] + dst[i+15:i] := SaturateU16(a[i+15:i] - b[i+15:i]) + ELSE + dst[i+15:i] := src[i+15:i] + FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 +
immintrin.h
 - - Floating Point - SSE - Elementary Math Functions - - Convert the exponent of each packed double-precision (64-bit) floating-point element in "a" to a double-precision floating-point number representing the integer exponent, and store the results in "dst". This intrinsic essentially calculates "floor(log2(x))" for each element. + + Integer + AVX512BW + Arithmetic + + + + + Subtract packed unsigned 16-bit integers in "b" from packed unsigned 16-bit integers in "a" using saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 1 - i := j*64 - dst[i+63:i] := ConvertExpFP64(a[i+63:i]) +FOR j := 0 to 31 + i := j*16 + IF k[j] + dst[i+15:i] := SaturateU16(a[i+15:i] - b[i+15:i]) + ELSE + dst[i+15:i] := 0 + FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 +
immintrin.h
 - - Floating Point - SSE - Elementary Math Functions - - Convert the exponent of each packed single-precision (32-bit) floating-point element in "a" to a single-precision floating-point number representing the integer exponent, and store the results in "dst". This intrinsic essentially calculates "floor(log2(x))" for each element. + + Integer + AVX512BW + Arithmetic + + + + Subtract packed unsigned 16-bit integers in "b" from packed unsigned 16-bit integers in "a" using saturation, and store the results in "dst". -FOR j := 0 to 3 - i := j*32 - dst[i+31:i] := ConvertExpFP32(a[i+31:i]) +FOR j := 0 to 31 + i := j*16 + dst[i+15:i] := SaturateU16(a[i+15:i] - b[i+15:i]) ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 +
immintrin.h
 - - Floating Point - SSE - Elementary Math Functions - - - Compute the exponential value of packed double-precision (64-bit) floating-point elements in "a" raised by packed elements in "b", and store the results in "dst". + + Integer + AVX512VL + AVX512BW + Arithmetic + + + + + + Subtract packed unsigned 16-bit integers in "b" from packed unsigned 16-bit integers in "a" using saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 1 - i := j*64 - dst[i+63:i] := (a[i+63:i])^(b[i+63:i]) +FOR j := 0 to 7 + i := j*16 + IF k[j] + dst[i+15:i] := SaturateU16(a[i+15:i] - b[i+15:i]) + ELSE + dst[i+15:i] := src[i+15:i] + FI ENDFOR dst[MAX:128] := 0 +
immintrin.h
 - - Floating Point - SSE - Elementary Math Functions - - - Compute the exponential value of packed single-precision (32-bit) floating-point elements in "a" raised by packed elements in "b", and store the results in "dst". + + Integer + AVX512VL + AVX512BW + Arithmetic + + + + + Subtract packed unsigned 16-bit integers in "b" from packed unsigned 16-bit integers in "a" using saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*32 - dst[i+31:i] := (a[i+31:i])^(b[i+31:i]) +FOR j := 0 to 7 + i := j*16 + IF k[j] + dst[i+15:i] := SaturateU16(a[i+15:i] - b[i+15:i]) + ELSE + dst[i+15:i] := 0 + FI ENDFOR dst[MAX:128] := 0 +
immintrin.h
 - + Integer - SSE + AVX512VL + AVX512BW Arithmetic - - - Divide packed 8-bit integers in "a" by packed elements in "b", and store the remainders as packed 32-bit integers in "dst". + + + + + + Subtract packed 16-bit integers in "b" from packed 16-bit integers in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). FOR j := 0 to 15 - i := 8*j - dst[i+7:i] := REMAINDER(a[i+7:i] / b[i+7:i]) + i := j*16 + IF k[j] + dst[i+15:i] := a[i+15:i] - b[i+15:i] + ELSE + dst[i+15:i] := src[i+15:i] + FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:256] := 0 +
immintrin.h
 - + Integer - SSE + AVX512VL + AVX512BW Arithmetic - - - Divide packed 16-bit integers in "a" by packed elements in "b", and store the remainders as packed 32-bit integers in "dst". + + + + + Subtract packed 16-bit integers in "b" from packed 16-bit integers in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := 16*j - dst[i+15:i] := REMAINDER(a[i+15:i] / b[i+15:i]) +FOR j := 0 to 15 + i := j*16 + IF k[j] + dst[i+15:i] := a[i+15:i] - b[i+15:i] + ELSE + dst[i+15:i] := 0 + FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:256] := 0 +
immintrin.h
 - + Integer - SSE + AVX512BW Arithmetic - - - Divide packed 32-bit integers in "a" by packed elements in "b", and store the remainders as packed 32-bit integers in "dst". + + + + + + Subtract packed 16-bit integers in "b" from packed 16-bit integers in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := 32*j - dst[i+31:i] := REMAINDER(a[i+31:i] / b[i+31:i]) +FOR j := 0 to 31 + i := j*16 + IF k[j] + dst[i+15:i] := a[i+15:i] - b[i+15:i] + ELSE + dst[i+15:i] := src[i+15:i] + FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 +
immintrin.h
 - + Integer - SSE + AVX512BW Arithmetic - - - Divide packed 64-bit integers in "a" by packed elements in "b", and store the remainders as packed 32-bit integers in "dst". + + + + + Subtract packed 16-bit integers in "b" from packed 16-bit integers in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 1 - i := 64*j - dst[i+63:i] := REMAINDER(a[i+63:i] / b[i+63:i]) +FOR j := 0 to 31 + i := j*16 + IF k[j] + dst[i+15:i] := a[i+15:i] - b[i+15:i] + ELSE + dst[i+15:i] := 0 + FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 +
immintrin.h
 - + Integer - SSE + AVX512BW Arithmetic - - - Divide packed unsigned 8-bit integers in "a" by packed elements in "b", and store the remainders as packed unsigned 32-bit integers in "dst". + + + + Subtract packed 16-bit integers in "b" from packed 16-bit integers in "a", and store the results in "dst". -FOR j := 0 to 15 - i := 8*j - dst[i+7:i] := REMAINDER(a[i+7:i] / b[i+7:i]) +FOR j := 0 to 31 + i := j*16 + dst[i+15:i] := a[i+15:i] - b[i+15:i] ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 +
immintrin.h
 - + Integer - SSE + AVX512VL + AVX512BW Arithmetic - - - Divide packed unsigned 16-bit integers in "a" by packed elements in "b", and store the remainders as packed unsigned 32-bit integers in "dst". + + + + + + Subtract packed 16-bit integers in "b" from packed 16-bit integers in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). FOR j := 0 to 7 - i := 16*j - dst[i+15:i] := REMAINDER(a[i+15:i] / b[i+15:i]) + i := j*16 + IF k[j] + dst[i+15:i] := a[i+15:i] - b[i+15:i] + ELSE + dst[i+15:i] := src[i+15:i] + FI ENDFOR dst[MAX:128] := 0 +
immintrin.h
 - + Integer - SSE + AVX512VL + AVX512BW Arithmetic - - - Divide packed unsigned 32-bit integers in "a" by packed elements in "b", and store the remainders as packed unsigned 32-bit integers in "dst". + + + + + Subtract packed 16-bit integers in "b" from packed 16-bit integers in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := 32*j - dst[i+31:i] := REMAINDER(a[i+31:i] / b[i+31:i]) +FOR j := 0 to 7 + i := j*16 + IF k[j] + dst[i+15:i] := a[i+15:i] - b[i+15:i] + ELSE + dst[i+15:i] := 0 + FI ENDFOR dst[MAX:128] := 0 +
immintrin.h
 - + Integer - SSE - Arithmetic - - - Divide packed unsigned 64-bit integers in "a" by packed elements in "b", and store the remainders as packed unsigned 32-bit integers in "dst". - -FOR j := 0 to 1 - i := 64*j - dst[i+63:i] := REMAINDER(a[i+63:i] / b[i+63:i]) -ENDFOR -dst[MAX:128] := 0 - -
immintrin.h
- - - Floating Point - SSE - Trigonometry - - Compute the sine of packed double-precision (64-bit) floating-point elements in "a" expressed in radians, and store the results in "dst". - -FOR j := 0 to 1 - i := j*64 - dst[i+63:i] := SIN(a[i+63:i]) -ENDFOR -dst[MAX:128] := 0 - -
immintrin.h
- - - Floating Point - SSE - Trigonometry - - Compute the sine of packed single-precision (32-bit) floating-point elements in "a" expressed in radians, and store the results in "dst". + Mask + AVX512VL + AVX512BW + Compare + + + + + Compute the bitwise AND of packed 8-bit integers in "a" and "b", producing intermediate 8-bit values, and set the corresponding bit in result mask "k" (subject to writemask "k") if the intermediate value is non-zero. -FOR j := 0 to 3 - i := j*32 - dst[i+31:i] := SIN(a[i+31:i]) +FOR j := 0 to 31 + i := j*8 + IF k1[j] + k[j] := ((a[i+7:i] AND b[i+7:i]) != 0) ? 1 : 0 + ELSE + k[j] := 0 + FI ENDFOR -dst[MAX:128] := 0 +k[MAX:32] := 0 +
immintrin.h
 - - Floating Point - SSE - Trigonometry - - - Compute the sine and cosine of packed double-precision (64-bit) floating-point elements in "a" expressed in radians, store the sine in "dst", and store the cosine into memory at "mem_addr". + + Integer + Mask + AVX512VL + AVX512BW + Compare + + + + Compute the bitwise AND of packed 8-bit integers in "a" and "b", producing intermediate 8-bit values, and set the corresponding bit in result mask "k" if the intermediate value is non-zero. -FOR j := 0 to 1 - i := j*64 - dst[i+63:i] := SIN(a[i+63:i]) - MEM[mem_addr+i+63:mem_addr+i] := COS(a[i+63:i]) +FOR j := 0 to 31 + i := j*8 + k[j] := ((a[i+7:i] AND b[i+7:i]) != 0) ? 1 : 0 ENDFOR -dst[MAX:128] := 0 +k[MAX:32] := 0 +
immintrin.h
 - - Floating Point - SSE - Trigonometry - - - Compute the sine and cosine of packed single-precision (32-bit) floating-point elements in "a" expressed in radians, store the sine in "dst", and store the cosine into memory at "mem_addr". + + Integer + Mask + AVX512BW + Compare + + + + + Compute the bitwise AND of packed 8-bit integers in "a" and "b", producing intermediate 8-bit values, and set the corresponding bit in result mask "k" (subject to writemask "k") if the intermediate value is non-zero. -FOR j := 0 to 3 - i := j*32 - dst[i+31:i] := SIN(a[i+31:i]) - MEM[mem_addr+i+31:mem_addr+i] := COS(a[i+31:i]) +FOR j := 0 to 63 + i := j*8 + IF k1[j] + k[j] := ((a[i+7:i] AND b[i+7:i]) != 0) ? 1 : 0 + ELSE + k[j] := 0 + FI ENDFOR -dst[MAX:128] := 0 +k[MAX:64] := 0 +
immintrin.h
 - - Floating Point - SSE - Trigonometry - - Compute the sine of packed double-precision (64-bit) floating-point elements in "a" expressed in degrees, and store the results in "dst". + + Integer + Mask + AVX512BW + Compare + + + + Compute the bitwise AND of packed 8-bit integers in "a" and "b", producing intermediate 8-bit values, and set the corresponding bit in result mask "k" if the intermediate value is non-zero. -FOR j := 0 to 1 - i := j*64 - dst[i+63:i] := SIND(a[i+63:i]) +FOR j := 0 to 63 + i := j*8 + k[j] := ((a[i+7:i] AND b[i+7:i]) != 0) ? 1 : 0 ENDFOR -dst[MAX:128] := 0 +k[MAX:64] := 0 +
immintrin.h
 - - Floating Point - SSE - Trigonometry - - Compute the sine of packed single-precision (32-bit) floating-point elements in "a" expressed in degrees, and store the results in "dst". + + Integer + Mask + AVX512VL + AVX512BW + Compare + + + + + Compute the bitwise AND of packed 8-bit integers in "a" and "b", producing intermediate 8-bit values, and set the corresponding bit in result mask "k" (subject to writemask "k") if the intermediate value is non-zero. -FOR j := 0 to 3 - i := j*32 - dst[i+31:i] := SIND(a[i+31:i]) +FOR j := 0 to 15 + i := j*8 + IF k1[j] + k[j] := ((a[i+7:i] AND b[i+7:i]) != 0) ? 1 : 0 + ELSE + k[j] := 0 + FI ENDFOR -dst[MAX:128] := 0 +k[MAX:16] := 0 +
immintrin.h
 - - Floating Point - SSE - Trigonometry - - Compute the hyperbolic sine of packed double-precision (64-bit) floating-point elements in "a" expressed in radians, and store the results in "dst". + + Integer + Mask + AVX512VL + AVX512BW + Compare + + + + Compute the bitwise AND of packed 8-bit integers in "a" and "b", producing intermediate 8-bit values, and set the corresponding bit in result mask "k" if the intermediate value is non-zero. -FOR j := 0 to 1 - i := j*64 - dst[i+63:i] := SINH(a[i+63:i]) +FOR j := 0 to 15 + i := j*8 + k[j] := ((a[i+7:i] AND b[i+7:i]) != 0) ? 1 : 0 ENDFOR -dst[MAX:128] := 0 +k[MAX:16] := 0 +
immintrin.h
 - - Floating Point - SSE - Trigonometry - - Compute the hyperbolic sine of packed single-precision (32-bit) floating-point elements in "a" expressed in radians, and store the results in "dst". + + Integer + Mask + AVX512VL + AVX512BW + Compare + + + + + Compute the bitwise AND of packed 16-bit integers in "a" and "b", producing intermediate 16-bit values, and set the corresponding bit in result mask "k" (subject to writemask "k") if the intermediate value is non-zero. -FOR j := 0 to 3 - i := j*32 - dst[i+31:i] := SINH(a[i+31:i]) +FOR j := 0 to 15 + i := j*16 + IF k1[j] + k[j] := ((a[i+15:i] AND b[i+15:i]) != 0) ? 1 : 0 + ELSE + k[j] := 0 + FI ENDFOR -dst[MAX:128] := 0 +k[MAX:16] := 0 +
immintrin.h
 - - Floating Point - SSE - Special Math Functions - - Round the packed double-precision (64-bit) floating-point elements in "a" up to an integer value, and store the results as packed double-precision floating-point elements in "dst". This intrinsic may generate the "roundpd"/"vroundpd" instruction. + + Integer + Mask + AVX512VL + AVX512BW + Compare + + + + Compute the bitwise AND of packed 16-bit integers in "a" and "b", producing intermediate 16-bit values, and set the corresponding bit in result mask "k" if the intermediate value is non-zero. -FOR j := 0 to 1 - i := j*64 - dst[i+63:i] := CEIL(a[i+63:i]) +FOR j := 0 to 15 + i := j*16 + k[j] := ((a[i+15:i] AND b[i+15:i]) != 0) ? 1 : 0 ENDFOR -dst[MAX:128] := 0 +k[MAX:16] := 0 +
immintrin.h
 - - Floating Point - SSE - Special Math Functions - - Round the packed single-precision (32-bit) floating-point elements in "a" up to an integer value, and store the results as packed single-precision floating-point elements in "dst". This intrinsic may generate the "roundps"/"vroundps" instruction. + + Integer + Mask + AVX512BW + Compare + + + + + Compute the bitwise AND of packed 16-bit integers in "a" and "b", producing intermediate 16-bit values, and set the corresponding bit in result mask "k" (subject to writemask "k") if the intermediate value is non-zero. -FOR j := 0 to 3 - i := j*32 - dst[i+31:i] := CEIL(a[i+31:i]) +FOR j := 0 to 31 + i := j*16 + IF k1[j] + k[j] := ((a[i+15:i] AND b[i+15:i]) != 0) ? 1 : 0 + ELSE + k[j] := 0 + FI ENDFOR -dst[MAX:128] := 0 +k[MAX:32] := 0 +
immintrin.h
 - - Floating Point - SSE - Special Math Functions - - Round the packed double-precision (64-bit) floating-point elements in "a" down to an integer value, and store the results as packed double-precision floating-point elements in "dst". This intrinsic may generate the "roundpd"/"vroundpd" instruction. + + Integer + Mask + AVX512BW + Compare + + + + Compute the bitwise AND of packed 16-bit integers in "a" and "b", producing intermediate 16-bit values, and set the corresponding bit in result mask "k" if the intermediate value is non-zero. -FOR j := 0 to 1 - i := j*64 - dst[i+63:i] := FLOOR(a[i+63:i]) +FOR j := 0 to 31 + i := j*16 + k[j] := ((a[i+15:i] AND b[i+15:i]) != 0) ? 1 : 0 ENDFOR -dst[MAX:128] := 0 +k[MAX:32] := 0 +
immintrin.h
 - - Floating Point - SSE - Special Math Functions - - Round the packed single-precision (32-bit) floating-point elements in "a" down to an integer value, and store the results as packed single-precision floating-point elements in "dst". This intrinsic may generate the "roundps"/"vroundps" instruction. + + Integer + Mask + AVX512VL + AVX512BW + Compare + + + + + Compute the bitwise AND of packed 16-bit integers in "a" and "b", producing intermediate 16-bit values, and set the corresponding bit in result mask "k" (subject to writemask "k") if the intermediate value is non-zero. -FOR j := 0 to 3 - i := j*32 - dst[i+31:i] := FLOOR(a[i+31:i]) +FOR j := 0 to 7 + i := j*16 + IF k1[j] + k[j] := ((a[i+15:i] AND b[i+15:i]) != 0) ? 1 : 0 + ELSE + k[j] := 0 + FI ENDFOR -dst[MAX:128] := 0 +k[MAX:8] := 0 +
immintrin.h
 - - Floating Point - SSE - Special Math Functions - - Round the packed double-precision (64-bit) floating-point elements in "a" to the nearest integer value, and store the results as packed double-precision floating-point elements in "dst". This intrinsic may generate the "roundpd"/"vroundpd" instruction. + + Integer + Mask + AVX512VL + AVX512BW + Compare + + + + Compute the bitwise AND of packed 16-bit integers in "a" and "b", producing intermediate 16-bit values, and set the corresponding bit in result mask "k" if the intermediate value is non-zero. -FOR j := 0 to 1 - i := j*64 - dst[i+63:i] := ROUND(a[i+63:i]) +FOR j := 0 to 7 + i := j*16 + k[j] := ((a[i+15:i] AND b[i+15:i]) != 0) ? 1 : 0 ENDFOR -dst[MAX:128] := 0 +k[MAX:8] := 0 +
immintrin.h
 - - Floating Point - SSE - Special Math Functions - - Round the packed single-precision (32-bit) floating-point elements in "a" to the nearest integer value, and store the results as packed single-precision floating-point elements in "dst". This intrinsic may generate the "roundps"/"vroundps" instruction. + + Integer + Mask + AVX512VL + AVX512BW + Compare + + + + + Compute the bitwise NAND of packed 8-bit integers in "a" and "b", producing intermediate 8-bit values, and set the corresponding bit in result mask "k" (subject to writemask "k") if the intermediate value is zero. -FOR j := 0 to 3 - i := j*32 - dst[i+31:i] := ROUND(a[i+31:i]) +FOR j := 0 to 31 + i := j*8 + IF k1[j] + k[j] := ((a[i+7:i] AND b[i+7:i]) == 0) ? 1 : 0 + ELSE + k[j] := 0 + FI ENDFOR -dst[MAX:128] := 0 +k[MAX:32] := 0 +
immintrin.h
 - - Floating Point - SSE - Elementary Math Functions - - Compute the square root of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst". Note that this intrinsic is less efficient than "_mm_sqrt_pd". + + Integer + Mask + AVX512VL + AVX512BW + Compare + + + + Compute the bitwise NAND of packed 8-bit integers in "a" and "b", producing intermediate 8-bit values, and set the corresponding bit in result mask "k" if the intermediate value is zero. -FOR j := 0 to 1 - i := j*64 - dst[i+63:i] := SQRT(a[i+63:i]) +FOR j := 0 to 31 + i := j*8 + k[j] := ((a[i+7:i] AND b[i+7:i]) == 0) ? 1 : 0 ENDFOR -dst[MAX:128] := 0 +k[MAX:32] := 0 +
immintrin.h
 - - Floating Point - SSE - Elementary Math Functions - - Compute the square root of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst". Note that this intrinsic is less efficient than "_mm_sqrt_ps". + + Integer + Mask + AVX512BW + Compare + + + + + Compute the bitwise NAND of packed 8-bit integers in "a" and "b", producing intermediate 8-bit values, and set the corresponding bit in result mask "k" (subject to writemask "k") if the intermediate value is zero. -FOR j := 0 to 3 - i := j*32 - dst[i+31:i] := SQRT(a[i+31:i]) +FOR j := 0 to 63 + i := j*8 + IF k1[j] + k[j] := ((a[i+7:i] AND b[i+7:i]) == 0) ? 1 : 0 + ELSE + k[j] := 0 + FI ENDFOR -dst[MAX:128] := 0 +k[MAX:64] := 0 +
immintrin.h
 - - Floating Point - SSE - Trigonometry - - Compute the tangent of packed double-precision (64-bit) floating-point elements in "a" expressed in radians, and store the results in "dst". + + Integer + Mask + AVX512BW + Compare + + + + Compute the bitwise NAND of packed 8-bit integers in "a" and "b", producing intermediate 8-bit values, and set the corresponding bit in result mask "k" if the intermediate value is zero. -FOR j := 0 to 1 - i := j*64 - dst[i+63:i] := TAN(a[i+63:i]) +FOR j := 0 to 63 + i := j*8 + k[j] := ((a[i+7:i] AND b[i+7:i]) == 0) ? 1 : 0 ENDFOR -dst[MAX:128] := 0 +k[MAX:64] := 0 +
immintrin.h
 - - Floating Point - SSE - Trigonometry - - Compute the tangent of packed single-precision (32-bit) floating-point elements in "a" expressed in radians, and store the results in "dst". + + Integer + Mask + AVX512VL + AVX512BW + Compare + + + + + Compute the bitwise NAND of packed 8-bit integers in "a" and "b", producing intermediate 8-bit values, and set the corresponding bit in result mask "k" (subject to writemask "k") if the intermediate value is zero. -FOR j := 0 to 3 - i := j*32 - dst[i+31:i] := TAN(a[i+31:i]) +FOR j := 0 to 15 + i := j*8 + IF k1[j] + k[j] := ((a[i+7:i] AND b[i+7:i]) == 0) ? 1 : 0 + ELSE + k[j] := 0 + FI ENDFOR -dst[MAX:128] := 0 +k[MAX:16] := 0 +
immintrin.h
 - - Floating Point - SSE - Trigonometry - - Compute the tangent of packed double-precision (64-bit) floating-point elements in "a" expressed in degrees, and store the results in "dst". + + Integer + Mask + AVX512VL + AVX512BW + Compare + + + + Compute the bitwise NAND of packed 8-bit integers in "a" and "b", producing intermediate 8-bit values, and set the corresponding bit in result mask "k" if the intermediate value is zero. -FOR j := 0 to 1 - i := j*64 - dst[i+63:i] := TAND(a[i+63:i]) +FOR j := 0 to 15 + i := j*8 + k[j] := ((a[i+7:i] AND b[i+7:i]) == 0) ? 1 : 0 ENDFOR -dst[MAX:128] := 0 +k[MAX:16] := 0 +
immintrin.h
 - - Floating Point - SSE - Trigonometry - - Compute the tangent of packed single-precision (32-bit) floating-point elements in "a" expressed in degrees, and store the results in "dst". + + Integer + Mask + AVX512VL + AVX512BW + Compare + + + + + Compute the bitwise NAND of packed 16-bit integers in "a" and "b", producing intermediate 16-bit values, and set the corresponding bit in result mask "k" (subject to writemask "k") if the intermediate value is zero. -FOR j := 0 to 3 - i := j*32 - dst[i+31:i] := TAND(a[i+31:i]) +FOR j := 0 to 15 + i := j*16 + IF k1[j] + k[j] := ((a[i+15:i] AND b[i+15:i]) == 0) ? 1 : 0 + ELSE + k[j] := 0 + FI ENDFOR -dst[MAX:128] := 0 +k[MAX:16] := 0 +
immintrin.h
 - - Floating Point - SSE - Trigonometry - - Compute the hyperbolic tangent of packed double-precision (64-bit) floating-point elements in "a" expressed in radians, and store the results in "dst". + + Integer + Mask + AVX512VL + AVX512BW + Compare + + + + Compute the bitwise NAND of packed 16-bit integers in "a" and "b", producing intermediate 16-bit values, and set the corresponding bit in result mask "k" if the intermediate value is zero. -FOR j := 0 to 1 - i := j*64 - dst[i+63:i] := TANH(a[i+63:i]) +FOR j := 0 to 15 + i := j*16 + k[j] := ((a[i+15:i] AND b[i+15:i]) == 0) ? 1 : 0 ENDFOR -dst[MAX:128] := 0 +k[MAX:16] := 0 +
immintrin.h
 - - Floating Point - SSE - Trigonometry - - Compute the hyperbolic tangent of packed single-precision (32-bit) floating-point elements in "a" expressed in radians, and store the results in "dst". + + Integer + Mask + AVX512BW + Compare + + + + + Compute the bitwise NAND of packed 16-bit integers in "a" and "b", producing intermediate 16-bit values, and set the corresponding bit in result mask "k" (subject to writemask "k") if the intermediate value is zero. -FOR j := 0 to 3 - i := j*32 - dst[i+31:i] := TANH(a[i+31:i]) +FOR j := 0 to 31 + i := j*16 + IF k1[j] + k[j] := ((a[i+15:i] AND b[i+15:i]) == 0) ? 1 : 0 + ELSE + k[j] := 0 + FI ENDFOR -dst[MAX:128] := 0 +k[MAX:32] := 0 +
immintrin.h
 - - Floating Point - SSE - Miscellaneous - - Truncate the packed double-precision (64-bit) floating-point elements in "a", and store the results as packed double-precision floating-point elements in "dst". This intrinsic may generate the "roundpd"/"vroundpd" instruction. + + Integer + Mask + AVX512BW + Compare + + + + Compute the bitwise NAND of packed 16-bit integers in "a" and "b", producing intermediate 16-bit values, and set the corresponding bit in result mask "k" if the intermediate value is zero. -FOR j := 0 to 1 - i := j*64 - dst[i+63:i] := TRUNCATE(a[i+63:i]) +FOR j := 0 to 31 + i := j*16 + k[j] := ((a[i+15:i] AND b[i+15:i]) == 0) ? 1 : 0 ENDFOR -dst[MAX:128] := 0 +k[MAX:32] := 0 +
immintrin.h
 - - Floating Point - SSE - Miscellaneous - - Truncate the packed single-precision (32-bit) floating-point elements in "a", and store the results as packed single-precision floating-point elements in "dst". This intrinsic may generate the "roundps"/"vroundps" instruction. + + Integer + Mask + AVX512VL + AVX512BW + Compare + + + + + Compute the bitwise NAND of packed 16-bit integers in "a" and "b", producing intermediate 16-bit values, and set the corresponding bit in result mask "k" (subject to writemask "k") if the intermediate value is zero. -FOR j := 0 to 3 - i := j*32 - dst[i+31:i] := TRUNCATE(a[i+31:i]) +FOR j := 0 to 7 + i := j*16 + IF k1[j] + k[j] := ((a[i+15:i] AND b[i+15:i]) == 0) ? 1 : 0 + ELSE + k[j] := 0 + FI ENDFOR -dst[MAX:128] := 0 +k[MAX:8] := 0 +
immintrin.h
 - + Integer - SSE - Arithmetic - - - Divide packed unsigned 32-bit integers in "a" by packed elements in "b", and store the truncated results in "dst". + Mask + AVX512VL + AVX512BW + Compare + + + + Compute the bitwise NAND of packed 16-bit integers in "a" and "b", producing intermediate 16-bit values, and set the corresponding bit in result mask "k" if the intermediate value is zero. -FOR j := 0 to 3 - i := 32*j - dst[i+31:i] := TRUNCATE(a[i+31:i] / b[i+31:i]) +FOR j := 0 to 7 + i := j*16 + k[j] := ((a[i+15:i] AND b[i+15:i]) == 0) ? 1 : 0 ENDFOR -dst[MAX:128] := 0 +k[MAX:8] := 0 +
immintrin.h
 - + Integer - SSE - Arithmetic - - - - Divide packed unsigned 32-bit integers in "a" by packed elements in "b", store the truncated results in "dst", and store the remainders as packed unsigned 32-bit integers into memory at "mem_addr". - -FOR j := 0 to 3 - i := 32*j - dst[i+31:i] := TRUNCATE(a[i+31:i] / b[i+31:i]) - MEM[mem_addr+i+31:mem_addr+i] := REMAINDER(a[i+31:i] / b[i+31:i]) -ENDFOR -dst[MAX:128] := 0 - -
immintrin.h
- - - Integer - SSE - Arithmetic - - - Divide packed unsigned 32-bit integers in "a" by packed elements in "b", and store the remainders as packed unsigned 32-bit integers in "dst". - -FOR j := 0 to 3 - i := 32*j - dst[i+31:i] := REMAINDER(a[i+31:i] / b[i+31:i]) -ENDFOR -dst[MAX:128] := 0 - -
immintrin.h
- - - Floating Point - AVX - Trigonometry - - Compute the inverse cosine of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst". + AVX512VL + AVX512BW + Miscellaneous + + + + + + Unpack and interleave 8-bit integers from the high half of each 128-bit lane in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*64 - dst[i+63:i] := ACOS(a[i+63:i]) +DEFINE INTERLEAVE_HIGH_BYTES(src1[127:0], src2[127:0]) { + dst[7:0] := src1[71:64] + dst[15:8] := src2[71:64] + dst[23:16] := src1[79:72] + dst[31:24] := src2[79:72] + dst[39:32] := src1[87:80] + dst[47:40] := src2[87:80] + dst[55:48] := src1[95:88] + dst[63:56] := src2[95:88] + dst[71:64] := src1[103:96] + dst[79:72] := src2[103:96] + dst[87:80] := src1[111:104] + dst[95:88] := src2[111:104] + dst[103:96] := src1[119:112] + dst[111:104] := src2[119:112] + dst[119:112] := src1[127:120] + dst[127:120] := src2[127:120] + RETURN dst[127:0] +} +tmp_dst[127:0] := INTERLEAVE_HIGH_BYTES(a[127:0], b[127:0]) +tmp_dst[255:128] := INTERLEAVE_HIGH_BYTES(a[255:128], b[255:128]) +FOR j := 0 to 31 + i := j*8 + IF k[j] + dst[i+7:i] := tmp_dst[i+7:i] + ELSE + dst[i+7:i] := src[i+7:i] + FI ENDFOR dst[MAX:256] := 0 +
immintrin.h
 - - Floating Point - AVX - Trigonometry - - Compute the inverse cosine of packed single-precision (32-bit) floating-point elements in "a" expressed in radians, and store the results in "dst". + + Integer + AVX512VL + AVX512BW + Miscellaneous + + + + + Unpack and interleave 8-bit integers from the high half of each 128-bit lane in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := j*32 - dst[i+31:i] := ACOS(a[i+31:i]) +DEFINE INTERLEAVE_HIGH_BYTES(src1[127:0], src2[127:0]) { + dst[7:0] := src1[71:64] + dst[15:8] := src2[71:64] + dst[23:16] := src1[79:72] + dst[31:24] := src2[79:72] + dst[39:32] := src1[87:80] + dst[47:40] := src2[87:80] + dst[55:48] := src1[95:88] + dst[63:56] := src2[95:88] + dst[71:64] := src1[103:96] + dst[79:72] := src2[103:96] + dst[87:80] := src1[111:104] + dst[95:88] := src2[111:104] + dst[103:96] := src1[119:112] + dst[111:104] := src2[119:112] + dst[119:112] := src1[127:120] + dst[127:120] := src2[127:120] + RETURN dst[127:0] +} +tmp_dst[127:0] := INTERLEAVE_HIGH_BYTES(a[127:0], b[127:0]) +tmp_dst[255:128] := INTERLEAVE_HIGH_BYTES(a[255:128], b[255:128]) +FOR j := 0 to 31 + i := j*8 + IF k[j] + dst[i+7:i] := tmp_dst[i+7:i] + ELSE + dst[i+7:i] := 0 + FI ENDFOR dst[MAX:256] := 0 +
immintrin.h
 - - Floating Point - AVX - Trigonometry - - Compute the inverse hyperbolic cosine of packed double-precision (64-bit) floating-point elements in "a" expressed in radians, and store the results in "dst". + + Integer + AVX512BW + Miscellaneous + + + + + + Unpack and interleave 8-bit integers from the high half of each 128-bit lane in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*64 - dst[i+63:i] := ACOSH(a[i+63:i]) +DEFINE INTERLEAVE_HIGH_BYTES(src1[127:0], src2[127:0]) { + dst[7:0] := src1[71:64] + dst[15:8] := src2[71:64] + dst[23:16] := src1[79:72] + dst[31:24] := src2[79:72] + dst[39:32] := src1[87:80] + dst[47:40] := src2[87:80] + dst[55:48] := src1[95:88] + dst[63:56] := src2[95:88] + dst[71:64] := src1[103:96] + dst[79:72] := src2[103:96] + dst[87:80] := src1[111:104] + dst[95:88] := src2[111:104] + dst[103:96] := src1[119:112] + dst[111:104] := src2[119:112] + dst[119:112] := src1[127:120] + dst[127:120] := src2[127:120] + RETURN dst[127:0] +} +tmp_dst[127:0] := INTERLEAVE_HIGH_BYTES(a[127:0], b[127:0]) +tmp_dst[255:128] := INTERLEAVE_HIGH_BYTES(a[255:128], b[255:128]) +tmp_dst[383:256] := INTERLEAVE_HIGH_BYTES(a[383:256], b[383:256]) +tmp_dst[511:384] := INTERLEAVE_HIGH_BYTES(a[511:384], b[511:384]) +FOR j := 0 to 63 + i := j*8 + IF k[j] + dst[i+7:i] := tmp_dst[i+7:i] + ELSE + dst[i+7:i] := src[i+7:i] + FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 +
immintrin.h
 - - Floating Point - AVX - Trigonometry - - Compute the inverse hyperbolic cosine of packed single-precision (32-bit) floating-point elements in "a" expressed in radians, and store the results in "dst". + + Integer + AVX512BW + Miscellaneous + + + + + Unpack and interleave 8-bit integers from the high half of each 128-bit lane in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := j*32 - dst[i+31:i] := ACOSH(a[i+31:i]) +DEFINE INTERLEAVE_HIGH_BYTES(src1[127:0], src2[127:0]) { + dst[7:0] := src1[71:64] + dst[15:8] := src2[71:64] + dst[23:16] := src1[79:72] + dst[31:24] := src2[79:72] + dst[39:32] := src1[87:80] + dst[47:40] := src2[87:80] + dst[55:48] := src1[95:88] + dst[63:56] := src2[95:88] + dst[71:64] := src1[103:96] + dst[79:72] := src2[103:96] + dst[87:80] := src1[111:104] + dst[95:88] := src2[111:104] + dst[103:96] := src1[119:112] + dst[111:104] := src2[119:112] + dst[119:112] := src1[127:120] + dst[127:120] := src2[127:120] + RETURN dst[127:0] +} +tmp_dst[127:0] := INTERLEAVE_HIGH_BYTES(a[127:0], b[127:0]) +tmp_dst[255:128] := INTERLEAVE_HIGH_BYTES(a[255:128], b[255:128]) +tmp_dst[383:256] := INTERLEAVE_HIGH_BYTES(a[383:256], b[383:256]) +tmp_dst[511:384] := INTERLEAVE_HIGH_BYTES(a[511:384], b[511:384]) +FOR j := 0 to 63 + i := j*8 + IF k[j] + dst[i+7:i] := tmp_dst[i+7:i] + ELSE + dst[i+7:i] := 0 + FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 +
immintrin.h
 - - Floating Point - AVX - Trigonometry - - Compute the inverse sine of packed double-precision (64-bit) floating-point elements in "a" expressed in radians, and store the results in "dst". + + Integer + AVX512BW + Miscellaneous + + + + Unpack and interleave 8-bit integers from the high half of each 128-bit lane in "a" and "b", and store the results in "dst". -FOR j := 0 to 3 - i := j*64 - dst[i+63:i] := ASIN(a[i+63:i]) -ENDFOR -dst[MAX:256] := 0 +DEFINE INTERLEAVE_HIGH_BYTES(src1[127:0], src2[127:0]) { + dst[7:0] := src1[71:64] + dst[15:8] := src2[71:64] + dst[23:16] := src1[79:72] + dst[31:24] := src2[79:72] + dst[39:32] := src1[87:80] + dst[47:40] := src2[87:80] + dst[55:48] := src1[95:88] + dst[63:56] := src2[95:88] + dst[71:64] := src1[103:96] + dst[79:72] := src2[103:96] + dst[87:80] := src1[111:104] + dst[95:88] := src2[111:104] + dst[103:96] := src1[119:112] + dst[111:104] := src2[119:112] + dst[119:112] := src1[127:120] + dst[127:120] := src2[127:120] + RETURN dst[127:0] +} +dst[127:0] := INTERLEAVE_HIGH_BYTES(a[127:0], b[127:0]) +dst[255:128] := INTERLEAVE_HIGH_BYTES(a[255:128], b[255:128]) +dst[383:256] := INTERLEAVE_HIGH_BYTES(a[383:256], b[383:256]) +dst[511:384] := INTERLEAVE_HIGH_BYTES(a[511:384], b[511:384]) +dst[MAX:512] := 0 +
immintrin.h
 - - Floating Point - AVX - Trigonometry - - Compute the inverse sine of packed single-precision (32-bit) floating-point elements in "a" expressed in radians, and store the results in "dst". + + Integer + AVX512VL + AVX512BW + Miscellaneous + + + + + + Unpack and interleave 8-bit integers from the high half of "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := j*32 - dst[i+31:i] := ASIN(a[i+31:i]) +DEFINE INTERLEAVE_HIGH_BYTES(src1[127:0], src2[127:0]) { + dst[7:0] := src1[71:64] + dst[15:8] := src2[71:64] + dst[23:16] := src1[79:72] + dst[31:24] := src2[79:72] + dst[39:32] := src1[87:80] + dst[47:40] := src2[87:80] + dst[55:48] := src1[95:88] + dst[63:56] := src2[95:88] + dst[71:64] := src1[103:96] + dst[79:72] := src2[103:96] + dst[87:80] := src1[111:104] + dst[95:88] := src2[111:104] + dst[103:96] := src1[119:112] + dst[111:104] := src2[119:112] + dst[119:112] := src1[127:120] + dst[127:120] := src2[127:120] + RETURN dst[127:0] +} +tmp_dst[127:0] := INTERLEAVE_HIGH_BYTES(a[127:0], b[127:0]) +FOR j := 0 to 15 + i := j*8 + IF k[j] + dst[i+7:i] := tmp_dst[i+7:i] + ELSE + dst[i+7:i] := src[i+7:i] + FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:128] := 0 +
immintrin.h
 - - Floating Point - AVX - Trigonometry - - Compute the inverse hyperbolic sine of packed double-precision (64-bit) floating-point elements in "a" expressed in radians, and store the results in "dst". + + Integer + AVX512VL + AVX512BW + Miscellaneous + + + + + Unpack and interleave 8-bit integers from the high half of "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*64 - dst[i+63:i] := ASINH(a[i+63:i]) +DEFINE INTERLEAVE_HIGH_BYTES(src1[127:0], src2[127:0]) { + dst[7:0] := src1[71:64] + dst[15:8] := src2[71:64] + dst[23:16] := src1[79:72] + dst[31:24] := src2[79:72] + dst[39:32] := src1[87:80] + dst[47:40] := src2[87:80] + dst[55:48] := src1[95:88] + dst[63:56] := src2[95:88] + dst[71:64] := src1[103:96] + dst[79:72] := src2[103:96] + dst[87:80] := src1[111:104] + dst[95:88] := src2[111:104] + dst[103:96] := src1[119:112] + dst[111:104] := src2[119:112] + dst[119:112] := src1[127:120] + dst[127:120] := src2[127:120] + RETURN dst[127:0] +} +tmp_dst[127:0] := INTERLEAVE_HIGH_BYTES(a[127:0], b[127:0]) +FOR j := 0 to 15 + i := j*8 + IF k[j] + dst[i+7:i] := tmp_dst[i+7:i] + ELSE + dst[i+7:i] := 0 + FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:128] := 0 +
immintrin.h
 - - Floating Point - AVX - Trigonometry - - Compute the inverse hyperbolic sine of packed single-precision (32-bit) floating-point elements in "a" expressed in radians, and store the results in "dst". + + Integer + AVX512VL + AVX512BW + Miscellaneous + + + + + + Unpack and interleave 16-bit integers from the high half of each 128-bit lane in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := j*32 - dst[i+31:i] := ASINH(a[i+31:i]) +DEFINE INTERLEAVE_HIGH_WORDS(src1[127:0], src2[127:0]) { + dst[15:0] := src1[79:64] + dst[31:16] := src2[79:64] + dst[47:32] := src1[95:80] + dst[63:48] := src2[95:80] + dst[79:64] := src1[111:96] + dst[95:80] := src2[111:96] + dst[111:96] := src1[127:112] + dst[127:112] := src2[127:112] + RETURN dst[127:0] +} +tmp_dst[127:0] := INTERLEAVE_HIGH_WORDS(a[127:0], b[127:0]) +tmp_dst[255:128] := INTERLEAVE_HIGH_WORDS(a[255:128], b[255:128]) +FOR j := 0 to 15 + i := j*16 + IF k[j] + dst[i+15:i] := tmp_dst[i+15:i] + ELSE + dst[i+15:i] := src[i+15:i] + FI ENDFOR dst[MAX:256] := 0 +
immintrin.h
 - - Floating Point - AVX - Trigonometry - - Compute the inverse tangent of packed double-precision (64-bit) floating-point elements in "a" expressed in radians, and store the results in "dst". + + Integer + AVX512VL + AVX512BW + Miscellaneous + + + + + Unpack and interleave 16-bit integers from the high half of each 128-bit lane in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*64 - dst[i+63:i] := ATAN(a[i+63:i]) +DEFINE INTERLEAVE_HIGH_WORDS(src1[127:0], src2[127:0]) { + dst[15:0] := src1[79:64] + dst[31:16] := src2[79:64] + dst[47:32] := src1[95:80] + dst[63:48] := src2[95:80] + dst[79:64] := src1[111:96] + dst[95:80] := src2[111:96] + dst[111:96] := src1[127:112] + dst[127:112] := src2[127:112] + RETURN dst[127:0] +} +tmp_dst[127:0] := INTERLEAVE_HIGH_WORDS(a[127:0], b[127:0]) +tmp_dst[255:128] := INTERLEAVE_HIGH_WORDS(a[255:128], b[255:128]) +FOR j := 0 to 15 + i := j*16 + IF k[j] + dst[i+15:i] := tmp_dst[i+15:i] + ELSE + dst[i+15:i] := 0 + FI ENDFOR dst[MAX:256] := 0 +
immintrin.h
 - - Floating Point - AVX - Trigonometry - - Compute the inverse tangent of packed single-precision (32-bit) floating-point elements in "a" expressed in radians, and store the results in "dst". + + Integer + AVX512BW + Miscellaneous + + + + + + Unpack and interleave 16-bit integers from the high half of each 128-bit lane in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := j*32 - dst[i+31:i] := ATAN(a[i+31:i]) +DEFINE INTERLEAVE_HIGH_WORDS(src1[127:0], src2[127:0]) { + dst[15:0] := src1[79:64] + dst[31:16] := src2[79:64] + dst[47:32] := src1[95:80] + dst[63:48] := src2[95:80] + dst[79:64] := src1[111:96] + dst[95:80] := src2[111:96] + dst[111:96] := src1[127:112] + dst[127:112] := src2[127:112] + RETURN dst[127:0] +} +tmp_dst[127:0] := INTERLEAVE_HIGH_WORDS(a[127:0], b[127:0]) +tmp_dst[255:128] := INTERLEAVE_HIGH_WORDS(a[255:128], b[255:128]) +tmp_dst[383:256] := INTERLEAVE_HIGH_WORDS(a[383:256], b[383:256]) +tmp_dst[511:384] := INTERLEAVE_HIGH_WORDS(a[511:384], b[511:384]) +FOR j := 0 to 31 + i := j*16 + IF k[j] + dst[i+15:i] := tmp_dst[i+15:i] + ELSE + dst[i+15:i] := src[i+15:i] + FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 +
immintrin.h
 - - Floating Point - AVX - Trigonometry - - - Compute the inverse tangent of packed double-precision (64-bit) floating-point elements in "a" divided by packed elements in "b", and store the results in "dst" expressed in radians. + + Integer + AVX512BW + Miscellaneous + + + + + Unpack and interleave 16-bit integers from the high half of each 128-bit lane in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*64 - dst[i+63:i] := ATAN(a[i+63:i] / b[i+63:i]) +DEFINE INTERLEAVE_HIGH_WORDS(src1[127:0], src2[127:0]) { + dst[15:0] := src1[79:64] + dst[31:16] := src2[79:64] + dst[47:32] := src1[95:80] + dst[63:48] := src2[95:80] + dst[79:64] := src1[111:96] + dst[95:80] := src2[111:96] + dst[111:96] := src1[127:112] + dst[127:112] := src2[127:112] + RETURN dst[127:0] +} +tmp_dst[127:0] := INTERLEAVE_HIGH_WORDS(a[127:0], b[127:0]) +tmp_dst[255:128] := INTERLEAVE_HIGH_WORDS(a[255:128], b[255:128]) +tmp_dst[383:256] := INTERLEAVE_HIGH_WORDS(a[383:256], b[383:256]) +tmp_dst[511:384] := INTERLEAVE_HIGH_WORDS(a[511:384], b[511:384]) +FOR j := 0 to 31 + i := j*16 + IF k[j] + dst[i+15:i] := tmp_dst[i+15:i] + ELSE + dst[i+15:i] := 0 + FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 +
immintrin.h
 - - Floating Point - AVX - Trigonometry - - - Compute the inverse tangent of packed single-precision (32-bit) floating-point elements in "a" divided by packed elements in "b", and store the results in "dst" expressed in radians. + + Integer + AVX512BW + Miscellaneous + + + + Unpack and interleave 16-bit integers from the high half of each 128-bit lane in "a" and "b", and store the results in "dst". -FOR j := 0 to 7 - i := j*32 - dst[i+31:i] := ATAN(a[i+31:i] / b[i+31:i]) -ENDFOR -dst[MAX:256] := 0 +DEFINE INTERLEAVE_HIGH_WORDS(src1[127:0], src2[127:0]) { + dst[15:0] := src1[79:64] + dst[31:16] := src2[79:64] + dst[47:32] := src1[95:80] + dst[63:48] := src2[95:80] + dst[79:64] := src1[111:96] + dst[95:80] := src2[111:96] + dst[111:96] := src1[127:112] + dst[127:112] := src2[127:112] + RETURN dst[127:0] +} +dst[127:0] := INTERLEAVE_HIGH_WORDS(a[127:0], b[127:0]) +dst[255:128] := INTERLEAVE_HIGH_WORDS(a[255:128], b[255:128]) +dst[383:256] := INTERLEAVE_HIGH_WORDS(a[383:256], b[383:256]) +dst[511:384] := INTERLEAVE_HIGH_WORDS(a[511:384], b[511:384]) +dst[MAX:512] := 0 +
immintrin.h
 - - Floating Point - AVX - Trigonometry - - Compute the inverse hyperbolic tangent of packed double-precision (64-bit) floating-point elements in "a" expressed in radians, and store the results in "dst". + + Integer + AVX512VL + AVX512BW + Miscellaneous + + + + + + Unpack and interleave 16-bit integers from the high half of "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*64 - dst[i+63:i] := ATANH(a[i+63:i]) +DEFINE INTERLEAVE_HIGH_WORDS(src1[127:0], src2[127:0]) { + dst[15:0] := src1[79:64] + dst[31:16] := src2[79:64] + dst[47:32] := src1[95:80] + dst[63:48] := src2[95:80] + dst[79:64] := src1[111:96] + dst[95:80] := src2[111:96] + dst[111:96] := src1[127:112] + dst[127:112] := src2[127:112] + RETURN dst[127:0] +} +tmp_dst[127:0] := INTERLEAVE_HIGH_WORDS(a[127:0], b[127:0]) +FOR j := 0 to 7 + i := j*16 + IF k[j] + dst[i+15:i] := tmp_dst[i+15:i] + ELSE + dst[i+15:i] := src[i+15:i] + FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:128] := 0 +
immintrin.h
 - - Floating Point - AVX - Trigonometry - - Compute the inverse hyperbolic tangent of packed single-precision (32-bit) floating-point elements in "a" expressed in radians, and store the results in "dst". + + Integer + AVX512VL + AVX512BW + Miscellaneous + + + + + Unpack and interleave 16-bit integers from the high half of "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). +DEFINE INTERLEAVE_HIGH_WORDS(src1[127:0], src2[127:0]) { + dst[15:0] := src1[79:64] + dst[31:16] := src2[79:64] + dst[47:32] := src1[95:80] + dst[63:48] := src2[95:80] + dst[79:64] := src1[111:96] + dst[95:80] := src2[111:96] + dst[111:96] := src1[127:112] + dst[127:112] := src2[127:112] + RETURN dst[127:0] +} +tmp_dst[127:0] := INTERLEAVE_HIGH_WORDS(a[127:0], b[127:0]) FOR j := 0 to 7 - i := j*32 - dst[i+31:i] := ATANH(a[i+31:i]) + i := j*16 + IF k[j] + dst[i+15:i] := tmp_dst[i+15:i] + ELSE + dst[i+15:i] := 0 + FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:128] := 0 +
immintrin.h
 - - Floating Point - AVX - Elementary Math Functions - - Compute the cube root of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst". + + Integer + AVX512VL + AVX512BW + Miscellaneous + + + + + + Unpack and interleave 8-bit integers from the low half of each 128-bit lane in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*64 - dst[i+63:i] := CubeRoot(a[i+63:i]) -ENDFOR -dst[MAX:256] := 0 - -
immintrin.h
- - - Floating Point - AVX - Elementary Math Functions - - Compute the cube root of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst". - -FOR j := 0 to 7 - i := j*32 - dst[i+31:i] := CubeRoot(a[i+31:i]) +DEFINE INTERLEAVE_BYTES(src1[127:0], src2[127:0]) { + dst[7:0] := src1[7:0] + dst[15:8] := src2[7:0] + dst[23:16] := src1[15:8] + dst[31:24] := src2[15:8] + dst[39:32] := src1[23:16] + dst[47:40] := src2[23:16] + dst[55:48] := src1[31:24] + dst[63:56] := src2[31:24] + dst[71:64] := src1[39:32] + dst[79:72] := src2[39:32] + dst[87:80] := src1[47:40] + dst[95:88] := src2[47:40] + dst[103:96] := src1[55:48] + dst[111:104] := src2[55:48] + dst[119:112] := src1[63:56] + dst[127:120] := src2[63:56] + RETURN dst[127:0] +} +tmp_dst[127:0] := INTERLEAVE_BYTES(a[127:0], b[127:0]) +tmp_dst[255:128] := INTERLEAVE_BYTES(a[255:128], b[255:128]) +FOR j := 0 to 31 + i := j*8 + IF k[j] + dst[i+7:i] := tmp_dst[i+7:i] + ELSE + dst[i+7:i] := src[i+7:i] + FI ENDFOR dst[MAX:256] := 0 +
immintrin.h
 - - Floating Point - AVX - Probability/Statistics - - Compute the cumulative distribution function of packed double-precision (64-bit) floating-point elements in "a" using the normal distribution, and store the results in "dst". + + Integer + AVX512VL + AVX512BW + Miscellaneous + + + + + Unpack and interleave 8-bit integers from the low half of each 128-bit lane in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*64 - dst[i+63:i] := CDFNormal(a[i+63:i]) +DEFINE INTERLEAVE_BYTES(src1[127:0], src2[127:0]) { + dst[7:0] := src1[7:0] + dst[15:8] := src2[7:0] + dst[23:16] := src1[15:8] + dst[31:24] := src2[15:8] + dst[39:32] := src1[23:16] + dst[47:40] := src2[23:16] + dst[55:48] := src1[31:24] + dst[63:56] := src2[31:24] + dst[71:64] := src1[39:32] + dst[79:72] := src2[39:32] + dst[87:80] := src1[47:40] + dst[95:88] := src2[47:40] + dst[103:96] := src1[55:48] + dst[111:104] := src2[55:48] + dst[119:112] := src1[63:56] + dst[127:120] := src2[63:56] + RETURN dst[127:0] +} +tmp_dst[127:0] := INTERLEAVE_BYTES(a[127:0], b[127:0]) +tmp_dst[255:128] := INTERLEAVE_BYTES(a[255:128], b[255:128]) +FOR j := 0 to 31 + i := j*8 + IF k[j] + dst[i+7:i] := tmp_dst[i+7:i] + ELSE + dst[i+7:i] := 0 + FI ENDFOR dst[MAX:256] := 0 +
immintrin.h
 - - Floating Point - AVX - Probability/Statistics - - Compute the cumulative distribution function of packed single-precision (32-bit) floating-point elements in "a" using the normal distribution, and store the results in "dst". + + Integer + AVX512BW + Miscellaneous + + + + + + Unpack and interleave 8-bit integers from the low half of each 128-bit lane in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := j*32 - dst[i+31:i] := CDFNormal(a[i+31:i]) +DEFINE INTERLEAVE_BYTES(src1[127:0], src2[127:0]) { + dst[7:0] := src1[7:0] + dst[15:8] := src2[7:0] + dst[23:16] := src1[15:8] + dst[31:24] := src2[15:8] + dst[39:32] := src1[23:16] + dst[47:40] := src2[23:16] + dst[55:48] := src1[31:24] + dst[63:56] := src2[31:24] + dst[71:64] := src1[39:32] + dst[79:72] := src2[39:32] + dst[87:80] := src1[47:40] + dst[95:88] := src2[47:40] + dst[103:96] := src1[55:48] + dst[111:104] := src2[55:48] + dst[119:112] := src1[63:56] + dst[127:120] := src2[63:56] + RETURN dst[127:0] +} +tmp_dst[127:0] := INTERLEAVE_BYTES(a[127:0], b[127:0]) +tmp_dst[255:128] := INTERLEAVE_BYTES(a[255:128], b[255:128]) +tmp_dst[383:256] := INTERLEAVE_BYTES(a[383:256], b[383:256]) +tmp_dst[511:384] := INTERLEAVE_BYTES(a[511:384], b[511:384]) +FOR j := 0 to 63 + i := j*8 + IF k[j] + dst[i+7:i] := tmp_dst[i+7:i] + ELSE + dst[i+7:i] := src[i+7:i] + FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 +
immintrin.h
 - - Floating Point - AVX - Probability/Statistics - - Compute the inverse cumulative distribution function of packed double-precision (64-bit) floating-point elements in "a" using the normal distribution, and store the results in "dst". + + Integer + AVX512BW + Miscellaneous + + + + + Unpack and interleave 8-bit integers from the low half of each 128-bit lane in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*64 - dst[i+63:i] := InverseCDFNormal(a[i+63:i]) +DEFINE INTERLEAVE_BYTES(src1[127:0], src2[127:0]) { + dst[7:0] := src1[7:0] + dst[15:8] := src2[7:0] + dst[23:16] := src1[15:8] + dst[31:24] := src2[15:8] + dst[39:32] := src1[23:16] + dst[47:40] := src2[23:16] + dst[55:48] := src1[31:24] + dst[63:56] := src2[31:24] + dst[71:64] := src1[39:32] + dst[79:72] := src2[39:32] + dst[87:80] := src1[47:40] + dst[95:88] := src2[47:40] + dst[103:96] := src1[55:48] + dst[111:104] := src2[55:48] + dst[119:112] := src1[63:56] + dst[127:120] := src2[63:56] + RETURN dst[127:0] +} +tmp_dst[127:0] := INTERLEAVE_BYTES(a[127:0], b[127:0]) +tmp_dst[255:128] := INTERLEAVE_BYTES(a[255:128], b[255:128]) +tmp_dst[383:256] := INTERLEAVE_BYTES(a[383:256], b[383:256]) +tmp_dst[511:384] := INTERLEAVE_BYTES(a[511:384], b[511:384]) +FOR j := 0 to 63 + i := j*8 + IF k[j] + dst[i+7:i] := tmp_dst[i+7:i] + ELSE + dst[i+7:i] := 0 + FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 +
immintrin.h
 - - Floating Point - AVX - Probability/Statistics - - Compute the inverse cumulative distribution function of packed single-precision (32-bit) floating-point elements in "a" using the normal distribution, and store the results in "dst". + + Integer + AVX512BW + Miscellaneous + + + + Unpack and interleave 8-bit integers from the low half of each 128-bit lane in "a" and "b", and store the results in "dst". -FOR j := 0 to 7 - i := j*32 - dst[i+31:i] := InverseCDFNormal(a[i+31:i]) -ENDFOR -dst[MAX:256] := 0 +DEFINE INTERLEAVE_BYTES(src1[127:0], src2[127:0]) { + dst[7:0] := src1[7:0] + dst[15:8] := src2[7:0] + dst[23:16] := src1[15:8] + dst[31:24] := src2[15:8] + dst[39:32] := src1[23:16] + dst[47:40] := src2[23:16] + dst[55:48] := src1[31:24] + dst[63:56] := src2[31:24] + dst[71:64] := src1[39:32] + dst[79:72] := src2[39:32] + dst[87:80] := src1[47:40] + dst[95:88] := src2[47:40] + dst[103:96] := src1[55:48] + dst[111:104] := src2[55:48] + dst[119:112] := src1[63:56] + dst[127:120] := src2[63:56] + RETURN dst[127:0] +} +dst[127:0] := INTERLEAVE_BYTES(a[127:0], b[127:0]) +dst[255:128] := INTERLEAVE_BYTES(a[255:128], b[255:128]) +dst[383:256] := INTERLEAVE_BYTES(a[383:256], b[383:256]) +dst[511:384] := INTERLEAVE_BYTES(a[511:384], b[511:384]) +dst[MAX:512] := 0 +
immintrin.h
 - - Floating Point - AVX - Elementary Math Functions - - Compute the exponential value of "e" raised to the power of packed complex single-precision (32-bit) floating-point elements in "a", and store the results in "dst". + + Integer + AVX512VL + AVX512BW + Miscellaneous + + + + + + Unpack and interleave 8-bit integers from the low half of "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := j*32 - dst[i+31:i] := e^(a[i+31:i]) +DEFINE INTERLEAVE_BYTES(src1[127:0], src2[127:0]) { + dst[7:0] := src1[7:0] + dst[15:8] := src2[7:0] + dst[23:16] := src1[15:8] + dst[31:24] := src2[15:8] + dst[39:32] := src1[23:16] + dst[47:40] := src2[23:16] + dst[55:48] := src1[31:24] + dst[63:56] := src2[31:24] + dst[71:64] := src1[39:32] + dst[79:72] := src2[39:32] + dst[87:80] := src1[47:40] + dst[95:88] := src2[47:40] + dst[103:96] := src1[55:48] + dst[111:104] := src2[55:48] + dst[119:112] := src1[63:56] + dst[127:120] := src2[63:56] + RETURN dst[127:0] +} +tmp_dst[127:0] := INTERLEAVE_BYTES(a[127:0], b[127:0]) +FOR j := 0 to 15 + i := j*8 + IF k[j] + dst[i+7:i] := tmp_dst[i+7:i] + ELSE + dst[i+7:i] := src[i+7:i] + FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:128] := 0 +
immintrin.h
 - - Floating Point - AVX - Elementary Math Functions - - Compute the natural logarithm of packed complex single-precision (32-bit) floating-point elements in "a", and store the results in "dst". + + Integer + AVX512VL + AVX512BW + Miscellaneous + + + + + Unpack and interleave 8-bit integers from the low half of "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := j*32 - dst[i+31:i] := ln(a[i+31:i]) +DEFINE INTERLEAVE_BYTES(src1[127:0], src2[127:0]) { + dst[7:0] := src1[7:0] + dst[15:8] := src2[7:0] + dst[23:16] := src1[15:8] + dst[31:24] := src2[15:8] + dst[39:32] := src1[23:16] + dst[47:40] := src2[23:16] + dst[55:48] := src1[31:24] + dst[63:56] := src2[31:24] + dst[71:64] := src1[39:32] + dst[79:72] := src2[39:32] + dst[87:80] := src1[47:40] + dst[95:88] := src2[47:40] + dst[103:96] := src1[55:48] + dst[111:104] := src2[55:48] + dst[119:112] := src1[63:56] + dst[127:120] := src2[63:56] + RETURN dst[127:0] +} +tmp_dst[127:0] := INTERLEAVE_BYTES(a[127:0], b[127:0]) +FOR j := 0 to 15 + i := j*8 + IF k[j] + dst[i+7:i] := tmp_dst[i+7:i] + ELSE + dst[i+7:i] := 0 + FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:128] := 0 +
immintrin.h
 - - Floating Point - AVX - Trigonometry - - Compute the cosine of packed double-precision (64-bit) floating-point elements in "a" expressed in radians, and store the results in "dst". + + Integer + AVX512VL + AVX512BW + Miscellaneous + + + + + + Unpack and interleave 16-bit integers from the low half of each 128-bit lane in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*64 - dst[i+63:i] := COS(a[i+63:i]) +DEFINE INTERLEAVE_WORDS(src1[127:0], src2[127:0]) { + dst[15:0] := src1[15:0] + dst[31:16] := src2[15:0] + dst[47:32] := src1[31:16] + dst[63:48] := src2[31:16] + dst[79:64] := src1[47:32] + dst[95:80] := src2[47:32] + dst[111:96] := src1[63:48] + dst[127:112] := src2[63:48] + RETURN dst[127:0] +} +tmp_dst[127:0] := INTERLEAVE_WORDS(a[127:0], b[127:0]) +tmp_dst[255:128] := INTERLEAVE_WORDS(a[255:128], b[255:128]) +FOR j := 0 to 15 + i := j*16 + IF k[j] + dst[i+15:i] := tmp_dst[i+15:i] + ELSE + dst[i+15:i] := src[i+15:i] + FI ENDFOR dst[MAX:256] := 0 +
immintrin.h
 - - Floating Point - AVX - Trigonometry - - Compute the cosine of packed single-precision (32-bit) floating-point elements in "a" expressed in radians, and store the results in "dst". + + Integer + AVX512VL + AVX512BW + Miscellaneous + + + + + Unpack and interleave 16-bit integers from the low half of each 128-bit lane in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := j*32 - dst[i+31:i] := COS(a[i+31:i]) +DEFINE INTERLEAVE_WORDS(src1[127:0], src2[127:0]) { + dst[15:0] := src1[15:0] + dst[31:16] := src2[15:0] + dst[47:32] := src1[31:16] + dst[63:48] := src2[31:16] + dst[79:64] := src1[47:32] + dst[95:80] := src2[47:32] + dst[111:96] := src1[63:48] + dst[127:112] := src2[63:48] + RETURN dst[127:0] +} +tmp_dst[127:0] := INTERLEAVE_WORDS(a[127:0], b[127:0]) +tmp_dst[255:128] := INTERLEAVE_WORDS(a[255:128], b[255:128]) +FOR j := 0 to 15 + i := j*16 + IF k[j] + dst[i+15:i] := tmp_dst[i+15:i] + ELSE + dst[i+15:i] := 0 + FI ENDFOR dst[MAX:256] := 0 +
immintrin.h
 - - Floating Point - AVX - Trigonometry - - Compute the cosine of packed double-precision (64-bit) floating-point elements in "a" expressed in degrees, and store the results in "dst". + + Integer + AVX512BW + Miscellaneous + + + + + + Unpack and interleave 16-bit integers from the low half of each 128-bit lane in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*64 - dst[i+63:i] := COSD(a[i+63:i]) +DEFINE INTERLEAVE_WORDS(src1[127:0], src2[127:0]) { + dst[15:0] := src1[15:0] + dst[31:16] := src2[15:0] + dst[47:32] := src1[31:16] + dst[63:48] := src2[31:16] + dst[79:64] := src1[47:32] + dst[95:80] := src2[47:32] + dst[111:96] := src1[63:48] + dst[127:112] := src2[63:48] + RETURN dst[127:0] +} +tmp_dst[127:0] := INTERLEAVE_WORDS(a[127:0], b[127:0]) +tmp_dst[255:128] := INTERLEAVE_WORDS(a[255:128], b[255:128]) +tmp_dst[383:256] := INTERLEAVE_WORDS(a[383:256], b[383:256]) +tmp_dst[511:384] := INTERLEAVE_WORDS(a[511:384], b[511:384]) +FOR j := 0 to 31 + i := j*16 + IF k[j] + dst[i+15:i] := tmp_dst[i+15:i] + ELSE + dst[i+15:i] := src[i+15:i] + FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 +
immintrin.h
 - - Floating Point - AVX - Trigonometry - - Compute the cosine of packed single-precision (32-bit) floating-point elements in "a" expressed in degrees, and store the results in "dst". + + Integer + AVX512BW + Miscellaneous + + + + + Unpack and interleave 16-bit integers from the low half of each 128-bit lane in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := j*32 - dst[i+31:i] := COSD(a[i+31:i]) +DEFINE INTERLEAVE_WORDS(src1[127:0], src2[127:0]) { + dst[15:0] := src1[15:0] + dst[31:16] := src2[15:0] + dst[47:32] := src1[31:16] + dst[63:48] := src2[31:16] + dst[79:64] := src1[47:32] + dst[95:80] := src2[47:32] + dst[111:96] := src1[63:48] + dst[127:112] := src2[63:48] + RETURN dst[127:0] +} +tmp_dst[127:0] := INTERLEAVE_WORDS(a[127:0], b[127:0]) +tmp_dst[255:128] := INTERLEAVE_WORDS(a[255:128], b[255:128]) +tmp_dst[383:256] := INTERLEAVE_WORDS(a[383:256], b[383:256]) +tmp_dst[511:384] := INTERLEAVE_WORDS(a[511:384], b[511:384]) +FOR j := 0 to 31 + i := j*16 + IF k[j] + dst[i+15:i] := tmp_dst[i+15:i] + ELSE + dst[i+15:i] := 0 + FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 +
immintrin.h
 - - Floating Point - AVX - Trigonometry - - Compute the hyperbolic cosine of packed double-precision (64-bit) floating-point elements in "a" expressed in radians, and store the results in "dst". + + Integer + AVX512BW + Miscellaneous + + + + Unpack and interleave 16-bit integers from the low half of each 128-bit lane in "a" and "b", and store the results in "dst". -FOR j := 0 to 3 - i := j*64 - dst[i+63:i] := COSH(a[i+63:i]) -ENDFOR -dst[MAX:256] := 0 +DEFINE INTERLEAVE_WORDS(src1[127:0], src2[127:0]) { + dst[15:0] := src1[15:0] + dst[31:16] := src2[15:0] + dst[47:32] := src1[31:16] + dst[63:48] := src2[31:16] + dst[79:64] := src1[47:32] + dst[95:80] := src2[47:32] + dst[111:96] := src1[63:48] + dst[127:112] := src2[63:48] + RETURN dst[127:0] +} +dst[127:0] := INTERLEAVE_WORDS(a[127:0], b[127:0]) +dst[255:128] := INTERLEAVE_WORDS(a[255:128], b[255:128]) +dst[383:256] := INTERLEAVE_WORDS(a[383:256], b[383:256]) +dst[511:384] := INTERLEAVE_WORDS(a[511:384], b[511:384]) +dst[MAX:512] := 0 +
immintrin.h
 - - Floating Point - AVX - Trigonometry - - Compute the hyperbolic cosine of packed single-precision (32-bit) floating-point elements in "a" expressed in radians, and store the results in "dst". + + Integer + AVX512VL + AVX512BW + Miscellaneous + + + + + + Unpack and interleave 16-bit integers from the low half of "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). +DEFINE INTERLEAVE_WORDS(src1[127:0], src2[127:0]) { + dst[15:0] := src1[15:0] + dst[31:16] := src2[15:0] + dst[47:32] := src1[31:16] + dst[63:48] := src2[31:16] + dst[79:64] := src1[47:32] + dst[95:80] := src2[47:32] + dst[111:96] := src1[63:48] + dst[127:112] := src2[63:48] + RETURN dst[127:0] +} +tmp_dst[127:0] := INTERLEAVE_WORDS(a[127:0], b[127:0]) FOR j := 0 to 7 - i := j*32 - dst[i+31:i] := COSH(a[i+31:i]) + i := j*16 + IF k[j] + dst[i+15:i] := tmp_dst[i+15:i] + ELSE + dst[i+15:i] := src[i+15:i] + FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:128] := 0 +
immintrin.h
 - - Floating Point - AVX - Elementary Math Functions - - Compute the square root of packed complex single-precision (32-bit) floating-point elements in "a", and store the results in "dst". + + Integer + AVX512VL + AVX512BW + Miscellaneous + + + + + Unpack and interleave 16-bit integers from the low half of "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). +DEFINE INTERLEAVE_WORDS(src1[127:0], src2[127:0]) { + dst[15:0] := src1[15:0] + dst[31:16] := src2[15:0] + dst[47:32] := src1[31:16] + dst[63:48] := src2[31:16] + dst[79:64] := src1[47:32] + dst[95:80] := src2[47:32] + dst[111:96] := src1[63:48] + dst[127:112] := src2[63:48] + RETURN dst[127:0] +} +tmp_dst[127:0] := INTERLEAVE_WORDS(a[127:0], b[127:0]) FOR j := 0 to 7 - i := j*32 - dst[i+31:i] := SQRT(a[i+31:i]) + i := j*16 + IF k[j] + dst[i+15:i] := tmp_dst[i+15:i] + ELSE + dst[i+15:i] := 0 + FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:128] := 0 +
immintrin.h
 - + Integer - AVX - Arithmetic - - - Divide packed 8-bit integers in "a" by packed elements in "b", and store the truncated results in "dst". + AVX512BW + Store + + + + Store 512-bits (composed of 32 packed 16-bit integers) from "a" into memory. + "mem_addr" does not need to be aligned on any particular boundary. -FOR j := 0 to 31 - i := 8*j - dst[i+7:i] := TRUNCATE(a[i+7:i] / b[i+7:i]) -ENDFOR -dst[MAX:256] := 0 +MEM[mem_addr+511:mem_addr] := a[511:0] +
immintrin.h
 - + Integer - AVX - Arithmetic - - - Divide packed 16-bit integers in "a" by packed elements in "b", and store the truncated results in "dst". + AVX512BW + Store + + + + Store 512-bits (composed of 64 packed 8-bit integers) from "a" into memory. + "mem_addr" does not need to be aligned on any particular boundary. -FOR j := 0 to 15 - i := 16*j - dst[i+15:i] := TRUNCATE(a[i+15:i] / b[i+15:i]) -ENDFOR -dst[MAX:256] := 0 +MEM[mem_addr+511:mem_addr] := a[511:0] +
immintrin.h
 - + Integer - AVX - Arithmetic - - - Divide packed 32-bit integers in "a" by packed elements in "b", and store the truncated results in "dst". + AVX512VL + AVX512BW + Store + + + + Store 256-bits (composed of 16 packed 16-bit integers) from "a" into memory. + "mem_addr" does not need to be aligned on any particular boundary. -FOR j := 0 to 7 - i := 32*j - dst[i+31:i] := TRUNCATE(a[i+31:i] / b[i+31:i]) -ENDFOR -dst[MAX:256] := 0 +MEM[mem_addr+255:mem_addr] := a[255:0] +
immintrin.h
 - + Integer - AVX - Arithmetic - - - Divide packed 64-bit integers in "a" by packed elements in "b", and store the truncated results in "dst". + AVX512VL + AVX512BW + Store + + + + Store 256-bits (composed of 32 packed 8-bit integers) from "a" into memory. + "mem_addr" does not need to be aligned on any particular boundary. -FOR j := 0 to 3 - i := 64*j - dst[i+63:i] := TRUNCATE(a[i+63:i] / b[i+63:i]) -ENDFOR -dst[MAX:256] := 0 +MEM[mem_addr+255:mem_addr] := a[255:0] +
immintrin.h
 - + Integer - AVX - Arithmetic - - - Divide packed unsigned 8-bit integers in "a" by packed elements in "b", and store the truncated results in "dst". + AVX512VL + AVX512BW + Store + + + + Store 128-bits (composed of 8 packed 16-bit integers) from "a" into memory. + "mem_addr" does not need to be aligned on any particular boundary. -FOR j := 0 to 31 - i := 8*j - dst[i+7:i] := TRUNCATE(a[i+7:i] / b[i+7:i]) -ENDFOR -dst[MAX:256] := 0 +MEM[mem_addr+127:mem_addr] := a[127:0] +
immintrin.h
 - + Integer - AVX - Arithmetic - - - Divide packed unsigned 16-bit integers in "a" by packed elements in "b", and store the truncated results in "dst". + AVX512VL + AVX512BW + Store + + + + Store 128-bits (composed of 16 packed 8-bit integers) from "a" into memory. + "mem_addr" does not need to be aligned on any particular boundary. -FOR j := 0 to 15 - i := 16*j - dst[i+15:i] := TRUNCATE(a[i+15:i] / b[i+15:i]) -ENDFOR -dst[MAX:256] := 0 +MEM[mem_addr+127:mem_addr] := a[127:0] +
immintrin.h
 - + Integer - AVX - Arithmetic - - - Divide packed unsigned 32-bit integers in "a" by packed elements in "b", and store the truncated results in "dst". + AVX512BW + Load + + + Load 512-bits (composed of 32 packed 16-bit integers) from memory into "dst". + "mem_addr" does not need to be aligned on any particular boundary. -FOR j := 0 to 7 - i := 32*j - dst[i+31:i] := TRUNCATE(a[i+31:i] / b[i+31:i]) -ENDFOR -dst[MAX:256] := 0 +dst[511:0] := MEM[mem_addr+511:mem_addr] +dst[MAX:512] := 0 +
immintrin.h
 - + Integer - AVX - Arithmetic - - - Divide packed unsigned 64-bit integers in "a" by packed elements in "b", and store the truncated results in "dst". + AVX512BW + Load + + + Load 512-bits (composed of 64 packed 8-bit integers) from memory into "dst". + "mem_addr" does not need to be aligned on any particular boundary. -FOR j := 0 to 3 - i := 64*j - dst[i+63:i] := TRUNCATE(a[i+63:i] / b[i+63:i]) -ENDFOR -dst[MAX:256] := 0 +dst[511:0] := MEM[mem_addr+511:mem_addr] +dst[MAX:512] := 0 +
immintrin.h
 - - Floating Point - AVX - Probability/Statistics - - Compute the error function of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst". + + Integer + AVX512VL + AVX512BW + Load + + + Load 256-bits (composed of 16 packed 16-bit integers) from memory into "dst". + "mem_addr" does not need to be aligned on any particular boundary. -FOR j := 0 to 3 - i := j*64 - dst[i+63:i] := ERF(a[i+63:i]) -ENDFOR +dst[255:0] := MEM[mem_addr+255:mem_addr] dst[MAX:256] := 0 +
immintrin.h
 - - Floating Point - AVX - Probability/Statistics - - Compute the error function of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst". + + Integer + AVX512VL + AVX512BW + Load + + + Load 256-bits (composed of 32 packed 8-bit integers) from memory into "dst". + "mem_addr" does not need to be aligned on any particular boundary. -FOR j := 0 to 7 - i := j*32 - dst[i+31:i] := ERF(a[i+31:i]) -ENDFOR +dst[255:0] := MEM[mem_addr+255:mem_addr] dst[MAX:256] := 0 +
immintrin.h
 - - Floating Point - AVX - Probability/Statistics - - Compute the complementary error function of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst". + + Integer + AVX512VL + AVX512BW + Load + + + Load 128-bits (composed of 8 packed 16-bit integers) from memory into "dst". + "mem_addr" does not need to be aligned on any particular boundary. -FOR j := 0 to 3 - i := j*64 - dst[i+63:i] := 1.0 - ERF(a[i+63:i]) -ENDFOR -dst[MAX:256] := 0 +dst[127:0] := MEM[mem_addr+127:mem_addr] +dst[MAX:128] := 0 +
immintrin.h
 - - Floating Point - AVX - Probability/Statistics - - Compute the complementary error function of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst". + + Integer + AVX512VL + AVX512BW + Load + + + Load 128-bits (composed of 16 packed 8-bit integers) from memory into "dst". + "mem_addr" does not need to be aligned on any particular boundary. -FOR j := 0 to 7 - i := j*32 - dst[i+31:i] := 1.0 - ERF(a[i+31:i]) -ENDFOR -dst[MAX:256] := 0 +dst[127:0] := MEM[mem_addr+127:mem_addr] +dst[MAX:128] := 0 +
immintrin.h
 - - Floating Point - AVX - Probability/Statistics - - Compute the inverse complementary error function of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst". + + Mask + AVX512BW + Mask + + + + Add 32-bit masks in "a" and "b", and store the result in "k". -FOR j := 0 to 3 - i := j*64 - dst[i+63:i] := 1.0 / (1.0 - ERF(a[i+63:i])) -ENDFOR -dst[MAX:256] := 0 +k[31:0] := a[31:0] + b[31:0] +k[MAX:32] := 0 +
immintrin.h
 - - Floating Point - AVX - Probability/Statistics - - Compute the inverse complementary error function of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst". + + Mask + AVX512BW + Mask + + + + Add 64-bit masks in "a" and "b", and store the result in "k". -FOR j := 0 to 7 - i := j*32 - dst[i+31:i] := 1.0 / (1.0 - ERF(a[i+31:i])) -ENDFOR -dst[MAX:256] := 0 +k[63:0] := a[63:0] + b[63:0] +k[MAX:64] := 0 +
immintrin.h
 - - Floating Point - AVX - Probability/Statistics - - Compute the inverse error function of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst". + + Mask + AVX512BW + Mask + + + + Compute the bitwise AND of 32-bit masks "a" and "b", and store the result in "k". -FOR j := 0 to 3 - i := j*64 - dst[i+63:i] := 1.0 / ERF(a[i+63:i]) -ENDFOR -dst[MAX:256] := 0 +k[31:0] := a[31:0] AND b[31:0] +k[MAX:32] := 0 +
immintrin.h
 - - Floating Point - AVX - Probability/Statistics - - Compute the inverse error function of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst". + + Mask + AVX512BW + Mask + + + + Compute the bitwise AND of 64-bit masks "a" and "b", and store the result in "k". -FOR j := 0 to 7 - i := j*32 - dst[i+31:i] := 1.0 / ERF(a[i+31:i]) -ENDFOR -dst[MAX:256] := 0 +k[63:0] := a[63:0] AND b[63:0] +k[MAX:64] := 0 +
immintrin.h
 - - Floating Point - AVX - Elementary Math Functions - - Compute the exponential value of "e" raised to the power of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst". + + Mask + AVX512BW + Mask + + + + Compute the bitwise NOT of 32-bit masks "a" and then AND with "b", and store the result in "k". -FOR j := 0 to 3 - i := j*64 - dst[i+63:i] := e^(a[i+63:i]) -ENDFOR -dst[MAX:256] := 0 +k[31:0] := (NOT a[31:0]) AND b[31:0] +k[MAX:32] := 0 +
immintrin.h
 - - Floating Point - AVX - Elementary Math Functions - - Compute the exponential value of "e" raised to the power of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst". + + Mask + AVX512BW + Mask + + + + Compute the bitwise NOT of 64-bit masks "a" and then AND with "b", and store the result in "k". -FOR j := 0 to 7 - i := j*32 - dst[i+31:i] := e^(a[i+31:i]) -ENDFOR -dst[MAX:256] := 0 +k[63:0] := (NOT a[63:0]) AND b[63:0] +k[MAX:64] := 0 +
immintrin.h
 - - Floating Point - AVX - Elementary Math Functions - - Compute the exponential value of 10 raised to the power of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst". + + Mask + AVX512BW + Mask + + + Compute the bitwise NOT of 32-bit mask "a", and store the result in "k". -FOR j := 0 to 3 - i := j*64 - dst[i+63:i] := 10^(a[i+63:i]) -ENDFOR -dst[MAX:256] := 0 +k[31:0] := NOT a[31:0] +k[MAX:32] := 0 +
immintrin.h
 - - Floating Point - AVX - Elementary Math Functions - - Compute the exponential value of 10 raised to the power of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst". + + Mask + AVX512BW + Mask + + + Compute the bitwise NOT of 64-bit mask "a", and store the result in "k". -FOR j := 0 to 7 - i := j*32 - dst[i+31:i] := 10^(a[i+31:i]) -ENDFOR -dst[MAX:256] := 0 +k[63:0] := NOT a[63:0] +k[MAX:64] := 0 +
immintrin.h
 - - Floating Point - AVX - Elementary Math Functions - - Compute the exponential value of 2 raised to the power of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst". + + Mask + AVX512BW + Mask + + + + Compute the bitwise OR of 32-bit masks "a" and "b", and store the result in "k". -FOR j := 0 to 3 - i := j*64 - dst[i+63:i] := 2^(a[i+63:i]) -ENDFOR -dst[MAX:256] := 0 +k[31:0] := a[31:0] OR b[31:0] +k[MAX:32] := 0 +
immintrin.h
 - - Floating Point - AVX - Elementary Math Functions - - Compute the exponential value of 2 raised to the power of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst". + + Mask + AVX512BW + Mask + + + + Compute the bitwise OR of 64-bit masks "a" and "b", and store the result in "k". -FOR j := 0 to 7 - i := j*32 - dst[i+31:i] := 2^(a[i+31:i]) -ENDFOR -dst[MAX:256] := 0 +k[63:0] := a[63:0] OR b[63:0] +k[MAX:64] := 0 +
immintrin.h
 - - Floating Point - AVX - Elementary Math Functions - - Compute the exponential value of "e" raised to the power of packed double-precision (64-bit) floating-point elements in "a", subtract one from each element, and store the results in "dst". + + Mask + AVX512BW + Mask + + + + Compute the bitwise XNOR of 32-bit masks "a" and "b", and store the result in "k". -FOR j := 0 to 3 - i := j*64 - dst[i+63:i] := e^(a[i+63:i]) - 1.0 -ENDFOR -dst[MAX:256] := 0 +k[31:0] := NOT (a[31:0] XOR b[31:0]) +k[MAX:32] := 0 +
immintrin.h
 - - Floating Point - AVX - Elementary Math Functions - - Compute the exponential value of "e" raised to the power of packed single-precision (32-bit) floating-point elements in "a", subtract one from each element, and store the results in "dst". + + Mask + AVX512BW + Mask + + + + Compute the bitwise XNOR of 64-bit masks "a" and "b", and store the result in "k". -FOR j := 0 to 7 - i := j*32 - dst[i+31:i] := e^(a[i+31:i]) - 1.0 -ENDFOR -dst[MAX:256] := 0 +k[63:0] := NOT (a[63:0] XOR b[63:0]) +k[MAX:64] := 0 +
immintrin.h
 - - Floating Point - AVX - Trigonometry - - - Compute the length of the hypotenous of a right triangle, with the lengths of the other two sides of the triangle stored as packed double-precision (64-bit) floating-point elements in "a" and "b", and store the results in "dst". + + Mask + AVX512BW + Mask + + + + Compute the bitwise XOR of 32-bit masks "a" and "b", and store the result in "k". -FOR j := 0 to 3 - i := j*64 - dst[i+63:i] := SQRT(a[i+63:i]^2 + b[i+63:i]^2) -ENDFOR -dst[MAX:256] := 0 +k[31:0] := a[31:0] XOR b[31:0] +k[MAX:32] := 0 +
immintrin.h
 - - Floating Point - AVX - Trigonometry - - - Compute the length of the hypotenous of a right triangle, with the lengths of the other two sides of the triangle stored as packed single-precision (32-bit) floating-point elements in "a" and "b", and store the results in "dst". + + Mask + AVX512BW + Mask + + + + Compute the bitwise XOR of 64-bit masks "a" and "b", and store the result in "k". -FOR j := 0 to 7 - i := j*32 - dst[i+31:i] := SQRT(a[i+31:i]^2 + b[i+31:i]^2) -ENDFOR -dst[MAX:256] := 0 +k[63:0] := a[63:0] XOR b[63:0] +k[MAX:64] := 0 +
immintrin.h
 - - Integer - AVX - Arithmetic - - - Divide packed 32-bit integers in "a" by packed elements in "b", and store the truncated results in "dst". - -FOR j := 0 to 7 - i := 32*j - dst[i+31:i] := TRUNCATE(a[i+31:i] / b[i+31:i]) -ENDFOR -dst[MAX:256] := 0 + + Mask + AVX512BW + Mask + + + + Shift the bits of 32-bit mask "a" left by "count" while shifting in zeros, and store the least significant 32 bits of the result in "k". + +k[MAX:0] := 0 +IF count[7:0] <= 31 + k[31:0] := a[31:0] << count[7:0] +FI +
immintrin.h
 - - Integer - AVX - Arithmetic - - - - Divide packed 32-bit integers in "a" by packed elements in "b", store the truncated results in "dst", and store the remainders as packed 32-bit integers into memory at "mem_addr". - -FOR j := 0 to 7 - i := 32*j - dst[i+31:i] := TRUNCATE(a[i+31:i] / b[i+31:i]) - MEM[mem_addr+i+31:mem_addr+i] := REMAINDER(a[i+31:i] / b[i+31:i]) -ENDFOR -dst[MAX:256] := 0 + + Mask + AVX512BW + Mask + + + + Shift the bits of 64-bit mask "a" left by "count" while shifting in zeros, and store the least significant 64 bits of the result in "k". + +k[MAX:0] := 0 +IF count[7:0] <= 63 + k[63:0] := a[63:0] << count[7:0] +FI +
immintrin.h
 - - Floating Point - AVX - Elementary Math Functions - - Compute the inverse cube root of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst". - -FOR j := 0 to 3 - i := j*64 - dst[i+63:i] := InvCubeRoot(a[i+63:i]) -ENDFOR -dst[MAX:256] := 0 + + Mask + AVX512BW + Mask + + + + Shift the bits of 32-bit mask "a" right by "count" while shifting in zeros, and store the least significant 32 bits of the result in "k". + +k[MAX:0] := 0 +IF count[7:0] <= 31 + k[31:0] := a[31:0] >> count[7:0] +FI +
immintrin.h
 - - Floating Point - AVX - Elementary Math Functions - - Compute the inverse cube root of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst". - -FOR j := 0 to 7 - i := j*32 - dst[i+31:i] := InvCubeRoot(a[i+31:i]) -ENDFOR -dst[MAX:256] := 0 + + Mask + AVX512BW + Mask + + + + Shift the bits of 64-bit mask "a" right by "count" while shifting in zeros, and store the least significant 64 bits of the result in "k". + +k[MAX:0] := 0 +IF count[7:0] <= 63 + k[63:0] := a[63:0] >> count[7:0] +FI +
immintrin.h
 - - Floating Point - AVX - Elementary Math Functions - - Compute the inverse square root of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst". + + Mask + AVX512BW + Load + + + Load 32-bit mask from memory into "k". -FOR j := 0 to 3 - i := j*64 - dst[i+63:i] := InvSQRT(a[i+63:i]) -ENDFOR -dst[MAX:256] := 0 +k[31:0] := MEM[mem_addr+31:mem_addr] +
immintrin.h
 - - Floating Point - AVX - Elementary Math Functions - - Compute the inverse square root of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst". + + Mask + AVX512BW + Load + + + Load 64-bit mask from memory into "k". -FOR j := 0 to 7 - i := j*32 - dst[i+31:i] := InvSQRT(a[i+31:i]) -ENDFOR -dst[MAX:256] := 0 +k[63:0] := MEM[mem_addr+63:mem_addr] +
immintrin.h
 - - Integer - AVX - Arithmetic - - - Divide packed 32-bit integers in "a" by packed elements in "b", and store the remainders as packed 32-bit integers in "dst". + + Mask + AVX512BW + Store + + + + Store 32-bit mask from "a" into memory. -FOR j := 0 to 7 - i := 32*j - dst[i+31:i] := REMAINDER(a[i+31:i] / b[i+31:i]) -ENDFOR -dst[MAX:256] := 0 +MEM[mem_addr+31:mem_addr] := a[31:0] +
immintrin.h
 - - Floating Point - AVX - Elementary Math Functions - - Compute the natural logarithm of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst". + + Mask + AVX512BW + Store + + + + Store 64-bit mask from "a" into memory. -FOR j := 0 to 3 - i := j*64 - dst[i+63:i] := ln(a[i+63:i]) -ENDFOR -dst[MAX:256] := 0 +MEM[mem_addr+63:mem_addr] := a[63:0] +
immintrin.h
 - - Floating Point - AVX - Elementary Math Functions - - Compute the natural logarithm of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst". - -FOR j := 0 to 7 - i := j*32 - dst[i+31:i] := ln(a[i+31:i]) -ENDFOR -dst[MAX:256] := 0 + + Mask + AVX512BW + Mask + + + + + Compute the bitwise OR of 32-bit masks "a" and "b". If the result is all zeros, store 1 in "dst", otherwise store 0 in "dst". If the result is all ones, store 1 in "all_ones", otherwise store 0 in "all_ones". + +tmp[31:0] := a[31:0] OR b[31:0] +IF tmp[31:0] == 0x0 + dst := 1 +ELSE + dst := 0 +FI +IF tmp[31:0] == 0xFFFFFFFF + MEM[all_ones+7:all_ones] := 1 +ELSE + MEM[all_ones+7:all_ones] := 0 +FI +
immintrin.h
 - - Floating Point - AVX - Elementary Math Functions - - Compute the base-10 logarithm of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst". + + Mask + AVX512BW + Mask + + + + Compute the bitwise OR of 32-bit masks "a" and "b". If the result is all zeroes, store 1 in "dst", otherwise store 0 in "dst". -FOR j := 0 to 3 - i := j*64 - dst[i+63:i] := log10(a[i+63:i]) -ENDFOR -dst[MAX:256] := 0 +tmp[31:0] := a[31:0] OR b[31:0] +IF tmp[31:0] == 0x0 + dst := 1 +ELSE + dst := 0 +FI +
immintrin.h
 - - Floating Point - AVX - Elementary Math Functions - - Compute the base-10 logarithm of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst". + + Mask + AVX512BW + Mask + + + + Compute the bitwise OR of 32-bit masks "a" and "b". If the result is all ones, store 1 in "dst", otherwise store 0 in "dst". -FOR j := 0 to 7 - i := j*32 - dst[i+31:i] := log10(a[i+31:i]) -ENDFOR -dst[MAX:256] := 0 +tmp[31:0] := a[31:0] OR b[31:0] +IF tmp[31:0] == 0xFFFFFFFF + dst := 1 +ELSE + dst := 0 +FI +
immintrin.h
 - - Floating Point - AVX - Elementary Math Functions - - Compute the natural logarithm of one plus packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst". - -FOR j := 0 to 3 - i := j*64 - dst[i+63:i] := ln(1.0 + a[i+63:i]) -ENDFOR -dst[MAX:256] := 0 + + Mask + AVX512BW + Mask + + + + + Compute the bitwise OR of 64-bit masks "a" and "b". If the result is all zeros, store 1 in "dst", otherwise store 0 in "dst". If the result is all ones, store 1 in "all_ones", otherwise store 0 in "all_ones". + +tmp[63:0] := a[63:0] OR b[63:0] +IF tmp[63:0] == 0x0 + dst := 1 +ELSE + dst := 0 +FI +IF tmp[7:0] == 0xFFFFFFFFFFFFFFFF + MEM[all_ones+7:all_ones] := 1 +ELSE + MEM[all_ones+7:all_ones] := 0 +FI +
immintrin.h
 - - Floating Point - AVX - Elementary Math Functions - - Compute the natural logarithm of one plus packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst". + + Mask + AVX512BW + Mask + + + + Compute the bitwise OR of 64-bit masks "a" and "b". If the result is all zeroes, store 1 in "dst", otherwise store 0 in "dst". -FOR j := 0 to 7 - i := j*32 - dst[i+31:i] := ln(1.0 + a[i+31:i]) -ENDFOR -dst[MAX:256] := 0 +tmp[63:0] := a[63:0] OR b[63:0] +IF tmp[63:0] == 0x0 + dst := 1 +ELSE + dst := 0 +FI +
immintrin.h
 - - Floating Point - AVX - Elementary Math Functions - - Compute the base-2 logarithm of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst". + + Mask + AVX512BW + Mask + + + + Compute the bitwise OR of 64-bit masks "a" and "b". If the result is all ones, store 1 in "dst", otherwise store 0 in "dst". -FOR j := 0 to 3 - i := j*64 - dst[i+63:i] := log2(a[i+63:i]) -ENDFOR -dst[MAX:256] := 0 +tmp[63:0] := a[63:0] OR b[63:0] +IF tmp[63:0] == 0xFFFFFFFFFFFFFFFF + dst := 1 +ELSE + dst := 0 +FI +
immintrin.h
 - - Floating Point - AVX - Elementary Math Functions - - Compute the base-2 logarithm of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst". - -FOR j := 0 to 7 - i := j*32 - dst[i+31:i] := log2(a[i+31:i]) -ENDFOR -dst[MAX:256] := 0 + + Mask + AVX512BW + Mask + + + + + Compute the bitwise AND of 32-bit masks "a" and "b", and if the result is all zeros, store 1 in "dst", otherwise store 0 in "dst". Compute the bitwise NOT of "a" and then AND with "b", if the result is all zeros, store 1 in "and_not", otherwise store 0 in "and_not". + +tmp1[31:0] := a[31:0] AND b[31:0] +IF tmp1[31:0] == 0x0 + dst := 1 +ELSE + dst := 0 +FI +tmp2[31:0] := (NOT a[31:0]) AND b[31:0] +IF tmp2[31:0] == 0x0 + MEM[and_not+7:and_not] := 1 +ELSE + MEM[and_not+7:and_not] := 0 +FI +
immintrin.h
 - - Floating Point - AVX - Elementary Math Functions - - Convert the exponent of each packed double-precision (64-bit) floating-point element in "a" to a double-precision floating-point number representing the integer exponent, and store the results in "dst". This intrinsic essentially calculates "floor(log2(x))" for each element. + + Mask + AVX512BW + Mask + + + + Compute the bitwise AND of 32-bit masks "a" and "b", and if the result is all zeros, store 1 in "dst", otherwise store 0 in "dst". -FOR j := 0 to 3 - i := j*64 - dst[i+63:i] := ConvertExpFP64(a[i+63:i]) -ENDFOR -dst[MAX:256] := 0 - +tmp[31:0] := a[31:0] AND b[31:0] +IF tmp[31:0] == 0x0 + dst := 1 +ELSE + dst := 0 +FI + +
immintrin.h
 - - Floating Point - AVX - Elementary Math Functions - - Convert the exponent of each packed single-precision (32-bit) floating-point element in "a" to a single-precision floating-point number representing the integer exponent, and store the results in "dst". This intrinsic essentially calculates "floor(log2(x))" for each element. + + Mask + AVX512BW + Mask + + + + Compute the bitwise NOT of 32-bit mask "a" and then AND with "b", if the result is all zeroes, store 1 in "dst", otherwise store 0 in "dst". -FOR j := 0 to 7 - i := j*32 - dst[i+31:i] := ConvertExpFP32(a[i+31:i]) -ENDFOR -dst[MAX:256] := 0 +tmp[31:0] := (NOT a[31:0]) AND b[31:0] +IF tmp[31:0] == 0x0 + dst := 1 +ELSE + dst := 0 +FI +
immintrin.h
 - - Floating Point - AVX - Elementary Math Functions - - - Compute the exponential value of packed double-precision (64-bit) floating-point elements in "a" raised by packed elements in "b", and store the results in "dst". - -FOR j := 0 to 3 - i := j*64 - dst[i+63:i] := (a[i+63:i])^(b[i+63:i]) -ENDFOR -dst[MAX:256] := 0 + + Mask + AVX512BW + Mask + + + + + Compute the bitwise AND of 64-bit masks "a" and "b", and if the result is all zeros, store 1 in "dst", otherwise store 0 in "dst". Compute the bitwise NOT of "a" and then AND with "b", if the result is all zeros, store 1 in "and_not", otherwise store 0 in "and_not". + +tmp1[63:0] := a[63:0] AND b[63:0] +IF tmp1[63:0] == 0x0 + dst := 1 +ELSE + dst := 0 +FI +tmp2[63:0] := (NOT a[63:0]) AND b[63:0] +IF tmp2[63:0] == 0x0 + MEM[and_not+7:and_not] := 1 +ELSE + MEM[and_not+7:and_not] := 0 +FI +
immintrin.h
 - - Floating Point - AVX - Elementary Math Functions - - - Compute the exponential value of packed single-precision (32-bit) floating-point elements in "a" raised by packed elements in "b", and store the results in "dst". + + Mask + AVX512BW + Mask + + + + Compute the bitwise AND of 64-bit masks "a" and "b", and if the result is all zeros, store 1 in "dst", otherwise store 0 in "dst". -FOR j := 0 to 7 - i := j*32 - dst[i+31:i] := (a[i+31:i])^(b[i+31:i]) -ENDFOR -dst[MAX:256] := 0 +tmp[63:0] := a[63:0] AND b[63:0] +IF tmp[63:0] == 0x0 + dst := 1 +ELSE + dst := 0 +FI +
immintrin.h
 - - Integer - AVX - Arithmetic - - - Divide packed 8-bit integers in "a" by packed elements in "b", and store the remainders as packed 32-bit integers in "dst". + + Mask + AVX512BW + Mask + + + + Compute the bitwise NOT of 64-bit mask "a" and then AND with "b", if the result is all zeroes, store 1 in "dst", otherwise store 0 in "dst". -FOR j := 0 to 31 - i := 8*j - dst[i+7:i] := REMAINDER(a[i+7:i] / b[i+7:i]) -ENDFOR -dst[MAX:256] := 0 +tmp[63:0] := (NOT a[63:0]) AND b[63:0] +IF tmp[63:0] == 0x0 + dst := 1 +ELSE + dst := 0 +FI +
immintrin.h
 - - Integer - AVX - Arithmetic - - - Divide packed 16-bit integers in "a" by packed elements in "b", and store the remainders as packed 32-bit integers in "dst". + + AVX512BW + Mask + + + Convert 32-bit mask "a" into an integer value, and store the result in "dst". -FOR j := 0 to 15 - i := 16*j - dst[i+15:i] := REMAINDER(a[i+15:i] / b[i+15:i]) -ENDFOR -dst[MAX:256] := 0 +dst := ZeroExtend32(a[31:0]) +
immintrin.h
 - - Integer - AVX - Arithmetic - - - Divide packed 32-bit integers in "a" by packed elements in "b", and store the remainders as packed 32-bit integers in "dst". + + AVX512BW + Mask + + + Convert 64-bit mask "a" into an integer value, and store the result in "dst". -FOR j := 0 to 7 - i := 32*j - dst[i+31:i] := REMAINDER(a[i+31:i] / b[i+31:i]) -ENDFOR -dst[MAX:256] := 0 +dst := ZeroExtend64(a[63:0]) +
immintrin.h
 - - Integer - AVX - Arithmetic - - - Divide packed 64-bit integers in "a" by packed elements in "b", and store the remainders as packed 32-bit integers in "dst". + + AVX512BW + Mask + + + Convert integer value "a" into an 32-bit mask, and store the result in "k". -FOR j := 0 to 3 - i := 64*j - dst[i+63:i] := REMAINDER(a[i+63:i] / b[i+63:i]) -ENDFOR -dst[MAX:256] := 0 +k := ZeroExtend32(a[31:0]) +
immintrin.h
 - - Integer - AVX - Arithmetic - - - Divide packed unsigned 8-bit integers in "a" by packed elements in "b", and store the remainders as packed unsigned 32-bit integers in "dst". + + AVX512BW + Mask + + + Convert integer value "a" into an 64-bit mask, and store the result in "k". -FOR j := 0 to 31 - i := 8*j - dst[i+7:i] := REMAINDER(a[i+7:i] / b[i+7:i]) -ENDFOR -dst[MAX:256] := 0 +k := ZeroExtend64(a[63:0]) +
immintrin.h
 - + Integer - AVX - Arithmetic - - - Divide packed unsigned 16-bit integers in "a" by packed elements in "b", and store the remainders as packed unsigned 32-bit integers in "dst". + AVX512VL + AVX512CD + Miscellaneous + + + Broadcast the low 8-bits from input mask "k" to all 64-bit elements of "dst". -FOR j := 0 to 15 - i := 16*j - dst[i+15:i] := REMAINDER(a[i+15:i] / b[i+15:i]) +FOR j := 0 to 3 + i := j*64 + dst[i+63:i] := ZeroExtend64(k[7:0]) ENDFOR dst[MAX:256] := 0 +
immintrin.h
 - + Integer - AVX - Arithmetic - - - Divide packed unsigned 32-bit integers in "a" by packed elements in "b", and store the remainders as packed unsigned 32-bit integers in "dst". + AVX512VL + AVX512CD + Miscellaneous + + + Broadcast the low 8-bits from input mask "k" to all 64-bit elements of "dst". -FOR j := 0 to 7 - i := 32*j - dst[i+31:i] := REMAINDER(a[i+31:i] / b[i+31:i]) +FOR j := 0 to 1 + i := j*64 + dst[i+63:i] := ZeroExtend64(k[7:0]) ENDFOR -dst[MAX:256] := 0 +dst[MAX:128] := 0 +
immintrin.h
 - + Integer - AVX - Arithmetic - - - Divide packed unsigned 64-bit integers in "a" by packed elements in "b", and store the remainders as packed unsigned 32-bit integers in "dst". + AVX512VL + AVX512CD + Miscellaneous + + + Broadcast the low 16-bits from input mask "k" to all 32-bit elements of "dst". -FOR j := 0 to 3 - i := 64*j - dst[i+63:i] := REMAINDER(a[i+63:i] / b[i+63:i]) +FOR j := 0 to 7 + i := j*32 + dst[i+31:i] := ZeroExtend32(k[15:0]) ENDFOR dst[MAX:256] := 0 +
immintrin.h
 - - Floating Point - AVX - Trigonometry - - Compute the sine of packed double-precision (64-bit) floating-point elements in "a" expressed in radians, and store the results in "dst". + + Integer + AVX512VL + AVX512CD + Miscellaneous + + + Broadcast the low 16-bits from input mask "k" to all 32-bit elements of "dst". FOR j := 0 to 3 - i := j*64 - dst[i+63:i] := SIN(a[i+63:i]) + i := j*32 + dst[i+31:i] := ZeroExtend32(k[15:0]) ENDFOR -dst[MAX:256] := 0 +dst[MAX:128] := 0 +
immintrin.h
 - - Floating Point - AVX - Trigonometry - - Compute the sine of packed single-precision (32-bit) floating-point elements in "a" expressed in radians, and store the results in "dst". + + Integer + AVX512VL + AVX512CD + Compare + + + Test each 32-bit element of "a" for equality with all other elements in "a" closer to the least significant bit. Each element's comparison forms a zero extended bit vector in "dst". FOR j := 0 to 7 i := j*32 - dst[i+31:i] := SIN(a[i+31:i]) + FOR k := 0 to j-1 + m := k*32 + dst[i+k] := (a[i+31:i] == a[m+31:m]) ? 1 : 0 + ENDFOR + dst[i+31:i+j] := 0 ENDFOR dst[MAX:256] := 0 +
immintrin.h
 - - Floating Point - AVX - Trigonometry - - - Compute the sine and cosine of packed double-precision (64-bit) floating-point elements in "a" expressed in radians, store the sine in "dst", and store the cosine into memory at "mem_addr". + + Integer + AVX512VL + AVX512CD + Compare + + + + + Test each 32-bit element of "a" for equality with all other elements in "a" closer to the least significant bit using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). Each element's comparison forms a zero extended bit vector in "dst". -FOR j := 0 to 3 - i := j*64 - dst[i+63:i] := SIN(a[i+63:i]) - MEM[mem_addr+i+63:mem_addr+i] := COS(a[i+63:i]) +FOR j := 0 to 7 + i := j*32 + IF k[j] + FOR l := 0 to j-1 + m := l*32 + dst[i+l] := (a[i+31:i] == a[m+31:m]) ? 1 : 0 + ENDFOR + dst[i+31:i+j] := 0 + ELSE + dst[i+31:i] := src[i+31:i] + FI ENDFOR dst[MAX:256] := 0 +
immintrin.h
 - - Floating Point - AVX - Trigonometry - - - Compute the sine and cosine of packed single-precision (32-bit) floating-point elements in "a" expressed in radians, store the sine in "dst", and store the cosine into memory at "mem_addr". + + Integer + AVX512VL + AVX512CD + Compare + + + + Test each 32-bit element of "a" for equality with all other elements in "a" closer to the least significant bit using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). Each element's comparison forms a zero extended bit vector in "dst". FOR j := 0 to 7 i := j*32 - dst[i+31:i] := SIN(a[i+31:i]) - MEM[mem_addr+i+31:mem_addr+i] := COS(a[i+31:i]) + IF k[j] + FOR l := 0 to j-1 + m := l*32 + dst[i+l] := (a[i+31:i] == a[m+31:m]) ? 1 : 0 + ENDFOR + dst[i+31:i+j] := 0 + ELSE + dst[i+31:i] := 0 + FI ENDFOR dst[MAX:256] := 0 +
immintrin.h
 - - Floating Point - AVX - Trigonometry - - Compute the sine of packed double-precision (64-bit) floating-point elements in "a" expressed in degrees, and store the results in "dst". + + Integer + AVX512VL + AVX512CD + Compare + + + Test each 32-bit element of "a" for equality with all other elements in "a" closer to the least significant bit. Each element's comparison forms a zero extended bit vector in "dst". FOR j := 0 to 3 - i := j*64 - dst[i+63:i] := SIND(a[i+63:i]) + i := j*32 + FOR k := 0 to j-1 + m := k*32 + dst[i+k] := (a[i+31:i] == a[m+31:m]) ? 1 : 0 + ENDFOR + dst[i+31:i+j] := 0 ENDFOR -dst[MAX:256] := 0 +dst[MAX:128] := 0 +
immintrin.h
 - - Floating Point - AVX - Trigonometry - - Compute the sine of packed single-precision (32-bit) floating-point elements in "a" expressed in degrees, and store the results in "dst". + + Integer + AVX512VL + AVX512CD + Compare + + + + + Test each 32-bit element of "a" for equality with all other elements in "a" closer to the least significant bit using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). Each element's comparison forms a zero extended bit vector in "dst". -FOR j := 0 to 7 +FOR j := 0 to 3 i := j*32 - dst[i+31:i] := SIND(a[i+31:i]) + IF k[j] + FOR l := 0 to j-1 + m := l*32 + dst[i+l] := (a[i+31:i] == a[m+31:m]) ? 1 : 0 + ENDFOR + dst[i+31:i+j] := 0 + ELSE + dst[i+31:i] := src[i+31:i] + FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:128] := 0 +
immintrin.h
 - - Floating Point - AVX - Trigonometry - - Compute the hyperbolic sine of packed double-precision (64-bit) floating-point elements in "a" expressed in radians, and store the results in "dst". + + Integer + AVX512VL + AVX512CD + Compare + + + + Test each 32-bit element of "a" for equality with all other elements in "a" closer to the least significant bit using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). Each element's comparison forms a zero extended bit vector in "dst". FOR j := 0 to 3 - i := j*64 - dst[i+63:i] := SINH(a[i+63:i]) + i := j*32 + IF k[j] + FOR l := 0 to j-1 + m := l*32 + dst[i+l] := (a[i+31:i] == a[m+31:m]) ? 1 : 0 + ENDFOR + dst[i+31:i+j] := 0 + ELSE + dst[i+31:i] := 0 + FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:128] := 0 +
immintrin.h
 - - Floating Point - AVX - Trigonometry - - Compute the hyperbolic sine of packed single-precision (32-bit) floating-point elements in "a" expressed in radians, and store the results in "dst". + + Integer + AVX512VL + AVX512CD + Compare + + + Test each 64-bit element of "a" for equality with all other elements in "a" closer to the least significant bit. Each element's comparison forms a zero extended bit vector in "dst". -FOR j := 0 to 7 - i := j*32 - dst[i+31:i] := SINH(a[i+31:i]) +FOR j := 0 to 3 + i := j*64 + FOR k := 0 to j-1 + m := k*64 + dst[i+k] := (a[i+63:i] == a[m+63:m]) ? 1 : 0 + ENDFOR + dst[i+63:i+j] := 0 ENDFOR dst[MAX:256] := 0 +
immintrin.h
 - - Floating Point - AVX - Special Math Functions - - Round the packed double-precision (64-bit) floating-point elements in "a" up to an integer value, and store the results as packed double-precision floating-point elements in "dst". This intrinsic may generate the "roundpd"/"vroundpd" instruction. + + Integer + AVX512VL + AVX512CD + Compare + + + + + Test each 64-bit element of "a" for equality with all other elements in "a" closer to the least significant bit using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). Each element's comparison forms a zero extended bit vector in "dst". FOR j := 0 to 3 i := j*64 - dst[i+63:i] := CEIL(a[i+63:i]) + IF k[j] + FOR l := 0 to j-1 + m := l*64 + dst[i+l] := (a[i+63:i] == a[m+63:m]) ? 1 : 0 + ENDFOR + dst[i+63:i+j] := 0 + ELSE + dst[i+63:i] := src[i+63:i] + FI ENDFOR dst[MAX:256] := 0 +
immintrin.h
 - - Floating Point - AVX - Special Math Functions - - Round the packed single-precision (32-bit) floating-point elements in "a" up to an integer value, and store the results as packed single-precision floating-point elements in "dst". This intrinsic may generate the "roundps"/"vroundps" instruction. + + Integer + AVX512VL + AVX512CD + Compare + + + + Test each 64-bit element of "a" for equality with all other elements in "a" closer to the least significant bit using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). Each element's comparison forms a zero extended bit vector in "dst". -FOR j := 0 to 7 - i := j*32 - dst[i+31:i] := CEIL(a[i+31:i]) +FOR j := 0 to 3 + i := j*64 + IF k[j] + FOR l := 0 to j-1 + m := l*64 + dst[i+l] := (a[i+63:i] == a[m+63:m]) ? 1 : 0 + ENDFOR + dst[i+63:i+j] := 0 + ELSE + dst[i+63:i] := 0 + FI ENDFOR dst[MAX:256] := 0 +
immintrin.h
 - - Floating Point - AVX - Special Math Functions - - Round the packed double-precision (64-bit) floating-point elements in "a" down to an integer value, and store the results as packed double-precision floating-point elements in "dst". This intrinsic may generate the "roundpd"/"vroundpd" instruction. + + Integer + AVX512VL + AVX512CD + Compare + + + Test each 64-bit element of "a" for equality with all other elements in "a" closer to the least significant bit. Each element's comparison forms a zero extended bit vector in "dst". -FOR j := 0 to 3 +FOR j := 0 to 1 i := j*64 - dst[i+63:i] := FLOOR(a[i+63:i]) + FOR k := 0 to j-1 + m := k*64 + dst[i+k] := (a[i+63:i] == a[m+63:m]) ? 1 : 0 + ENDFOR + dst[i+63:i+j] := 0 ENDFOR -dst[MAX:256] := 0 +dst[MAX:128] := 0 +
immintrin.h
 - - Floating Point - AVX - Special Math Functions - - Round the packed single-precision (32-bit) floating-point elements in "a" down to an integer value, and store the results as packed single-precision floating-point elements in "dst". This intrinsic may generate the "roundps"/"vroundps" instruction. + + Integer + AVX512VL + AVX512CD + Compare + + + + + Test each 64-bit element of "a" for equality with all other elements in "a" closer to the least significant bit using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). Each element's comparison forms a zero extended bit vector in "dst". -FOR j := 0 to 7 - i := j*32 - dst[i+31:i] := FLOOR(a[i+31:i]) +FOR j := 0 to 1 + i := j*64 + IF k[j] + FOR l := 0 to j-1 + m := l*64 + dst[i+l] := (a[i+63:i] == a[m+63:m]) ? 1 : 0 + ENDFOR + dst[i+63:i+j] := 0 + ELSE + dst[i+63:i] := src[i+63:i] + FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:128] := 0 +
immintrin.h
 - - Floating Point - AVX - Special Math Functions - - Round the packed double-precision (64-bit) floating-point elements in "a" to the nearest integer value, and store the results as packed double-precision floating-point elements in "dst". This intrinsic may generate the "roundpd"/"vroundpd" instruction. + + Integer + AVX512VL + AVX512CD + Compare + + + + Test each 64-bit element of "a" for equality with all other elements in "a" closer to the least significant bit using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). Each element's comparison forms a zero extended bit vector in "dst". -FOR j := 0 to 3 +FOR j := 0 to 1 i := j*64 - dst[i+63:i] := ROUND(a[i+63:i]) + IF k[j] + FOR l := 0 to j-1 + m := l*64 + dst[i+l] := (a[i+63:i] == a[m+63:m]) ? 1 : 0 + ENDFOR + dst[i+63:i+j] := 0 + ELSE + dst[i+63:i] := 0 + FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:128] := 0 +
immintrin.h
 - - Floating Point - AVX - Special Math Functions - - Round the packed single-precision (32-bit) floating-point elements in "a" to the nearest integer value, and store the results as packed single-precision floating-point elements in "dst". This intrinsic may generate the "roundps"/"vroundps" instruction. + + Integer + AVX512VL + AVX512CD + Bit Manipulation + + + Counts the number of leading zero bits in each packed 32-bit integer in "a", and store the results in "dst". FOR j := 0 to 7 i := j*32 - dst[i+31:i] := ROUND(a[i+31:i]) + tmp := 31 + dst[i+31:i] := 0 + DO WHILE (tmp >= 0 AND a[i+tmp] == 0) + tmp := tmp - 1 + dst[i+31:i] := dst[i+31:i] + 1 + OD ENDFOR dst[MAX:256] := 0 +
immintrin.h
 - - Floating Point - AVX - Elementary Math Functions - - Compute the square root of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst". Note that this intrinsic is less efficient than "_mm_sqrt_pd". + + Integer + AVX512VL + AVX512CD + Bit Manipulation + + + + + Counts the number of leading zero bits in each packed 32-bit integer in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*64 - dst[i+63:i] := SQRT(a[i+63:i]) +FOR j := 0 to 7 + i := j*32 + IF k[j] + tmp := 31 + dst[i+31:i] := 0 + DO WHILE (tmp >= 0 AND a[i+tmp] == 0) + tmp := tmp - 1 + dst[i+31:i] := dst[i+31:i] + 1 + OD + ELSE + dst[i+31:i] := src[i+31:i] + FI ENDFOR dst[MAX:256] := 0 +
immintrin.h
 - - Floating Point - AVX - Elementary Math Functions - - Compute the square root of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst". Note that this intrinsic is less efficient than "_mm_sqrt_ps". + + Integer + AVX512VL + AVX512CD + Bit Manipulation + + + + Counts the number of leading zero bits in each packed 32-bit integer in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). FOR j := 0 to 7 i := j*32 - dst[i+31:i] := SQRT(a[i+31:i]) + IF k[j] + tmp := 31 + dst[i+31:i] := 0 + DO WHILE (tmp >= 0 AND a[i+tmp] == 0) + tmp := tmp - 1 + dst[i+31:i] := dst[i+31:i] + 1 + OD + ELSE + dst[i+31:i] := 0 + FI ENDFOR dst[MAX:256] := 0 +
immintrin.h
 - - Floating Point - AVX - Trigonometry - - Compute the tangent of packed double-precision (64-bit) floating-point elements in "a" expressed in radians, and store the results in "dst". + + Integer + AVX512VL + AVX512CD + Bit Manipulation + + + Counts the number of leading zero bits in each packed 32-bit integer in "a", and store the results in "dst". FOR j := 0 to 3 - i := j*64 - dst[i+63:i] := TAN(a[i+63:i]) + i := j*32 + tmp := 31 + dst[i+31:i] := 0 + DO WHILE (tmp >= 0 AND a[i+tmp] == 0) + tmp := tmp - 1 + dst[i+31:i] := dst[i+31:i] + 1 + OD ENDFOR -dst[MAX:256] := 0 +dst[MAX:128] := 0 +
immintrin.h
 - - Floating Point - AVX - Trigonometry - - Compute the tangent of packed single-precision (32-bit) floating-point elements in "a" expressed in radians, and store the results in "dst". + + Integer + AVX512VL + AVX512CD + Bit Manipulation + + + + + Counts the number of leading zero bits in each packed 32-bit integer in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 7 +FOR j := 0 to 3 i := j*32 - dst[i+31:i] := TAN(a[i+31:i]) -ENDFOR -dst[MAX:256] := 0 - -
immintrin.h
- - - Floating Point - AVX - Trigonometry - - Compute the tangent of packed double-precision (64-bit) floating-point elements in "a" expressed in degrees, and store the results in "dst". - -FOR j := 0 to 3 - i := j*64 - dst[i+63:i] := TAND(a[i+63:i]) + IF k[j] + tmp := 31 + dst[i+31:i] := 0 + DO WHILE (tmp >= 0 AND a[i+tmp] == 0) + tmp := tmp - 1 + dst[i+31:i] := dst[i+31:i] + 1 + OD + ELSE + dst[i+31:i] := src[i+31:i] + FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:128] := 0 +
immintrin.h
 - - Floating Point - AVX - Trigonometry - - Compute the tangent of packed single-precision (32-bit) floating-point elements in "a" expressed in degrees, and store the results in "dst". + + Integer + AVX512VL + AVX512CD + Bit Manipulation + + + + Counts the number of leading zero bits in each packed 32-bit integer in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 +FOR j := 0 to 3 i := j*32 - dst[i+31:i] := TAND(a[i+31:i]) + IF k[j] + tmp := 31 + dst[i+31:i] := 0 + DO WHILE (tmp >= 0 AND a[i+tmp] == 0) + tmp := tmp - 1 + dst[i+31:i] := dst[i+31:i] + 1 + OD + ELSE + dst[i+31:i] := 0 + FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:128] := 0 +
immintrin.h
 - - Floating Point - AVX - Trigonometry - - Compute the hyperbolic tangent of packed double-precision (64-bit) floating-point elements in "a" expressed in radians, and store the results in "dst". + + Integer + AVX512VL + AVX512CD + Bit Manipulation + + + Counts the number of leading zero bits in each packed 64-bit integer in "a", and store the results in "dst". FOR j := 0 to 3 i := j*64 - dst[i+63:i] := TANH(a[i+63:i]) + tmp := 63 + dst[i+63:i] := 0 + DO WHILE (tmp >= 0 AND a[i+tmp] == 0) + tmp := tmp - 1 + dst[i+63:i] := dst[i+63:i] + 1 + OD ENDFOR dst[MAX:256] := 0 +
immintrin.h
 - - Floating Point - AVX - Trigonometry - - Compute the hyperbolic tangent of packed single-precision (32-bit) floating-point elements in "a" expressed in radians, and store the results in "dst". + + Integer + AVX512VL + AVX512CD + Bit Manipulation + + + + + Counts the number of leading zero bits in each packed 64-bit integer in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := j*32 - dst[i+31:i] := TANH(a[i+31:i]) +FOR j := 0 to 3 + i := j*64 + IF k[j] + tmp := 63 + dst[i+63:i] := 0 + DO WHILE (tmp >= 0 AND a[i+tmp] == 0) + tmp := tmp - 1 + dst[i+63:i] := dst[i+63:i] + 1 + OD + ELSE + dst[i+63:i] := src[i+63:i] + FI ENDFOR dst[MAX:256] := 0 +
immintrin.h
 - - Floating Point - AVX - Miscellaneous - - Truncate the packed double-precision (64-bit) floating-point elements in "a", and store the results as packed double-precision floating-point elements in "dst". This intrinsic may generate the "roundpd"/"vroundpd" instruction. + + Integer + AVX512VL + AVX512CD + Bit Manipulation + + + + Counts the number of leading zero bits in each packed 64-bit integer in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). FOR j := 0 to 3 i := j*64 - dst[i+63:i] := TRUNCATE(a[i+63:i]) + IF k[j] + tmp := 63 + dst[i+63:i] := 0 + DO WHILE (tmp >= 0 AND a[i+tmp] == 0) + tmp := tmp - 1 + dst[i+63:i] := dst[i+63:i] + 1 + OD + ELSE + dst[i+63:i] := 0 + FI ENDFOR dst[MAX:256] := 0 +
immintrin.h
 - - Floating Point - AVX - Miscellaneous - - Truncate the packed single-precision (32-bit) floating-point elements in "a", and store the results as packed single-precision floating-point elements in "dst". This intrinsic may generate the "roundps"/"vroundps" instruction. + + Integer + AVX512VL + AVX512CD + Bit Manipulation + + + Counts the number of leading zero bits in each packed 64-bit integer in "a", and store the results in "dst". -FOR j := 0 to 7 - i := j*32 - dst[i+31:i] := TRUNCATE(a[i+31:i]) +FOR j := 0 to 1 + i := j*64 + tmp := 63 + dst[i+63:i] := 0 + DO WHILE (tmp >= 0 AND a[i+tmp] == 0) + tmp := tmp - 1 + dst[i+63:i] := dst[i+63:i] + 1 + OD ENDFOR -dst[MAX:256] := 0 +dst[MAX:128] := 0 +
immintrin.h
 - + Integer - AVX - Arithmetic - - - Divide packed unsigned 32-bit integers in "a" by packed elements in "b", and store the truncated results in "dst". + AVX512VL + AVX512CD + Bit Manipulation + + + + + Counts the number of leading zero bits in each packed 64-bit integer in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := 32*j - dst[i+31:i] := TRUNCATE(a[i+31:i] / b[i+31:i]) +FOR j := 0 to 1 + i := j*64 + IF k[j] + tmp := 63 + dst[i+63:i] := 0 + DO WHILE (tmp >= 0 AND a[i+tmp] == 0) + tmp := tmp - 1 + dst[i+63:i] := dst[i+63:i] + 1 + OD + ELSE + dst[i+63:i] := src[i+63:i] + FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:128] := 0 +
immintrin.h
 - + Integer - AVX - Arithmetic - - - - Divide packed unsigned 32-bit integers in "a" by packed elements in "b", store the truncated results in "dst", and store the remainders as packed unsigned 32-bit integers into memory at "mem_addr". + AVX512VL + AVX512CD + Bit Manipulation + + + + Counts the number of leading zero bits in each packed 64-bit integer in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := 32*j - dst[i+31:i] := TRUNCATE(a[i+31:i] / b[i+31:i]) - MEM[mem_addr+i+31:mem_addr+i] := REMAINDER(a[i+31:i] / b[i+31:i]) +FOR j := 0 to 1 + i := j*64 + IF k[j] + tmp := 63 + dst[i+63:i] := 0 + DO WHILE (tmp >= 0 AND a[i+tmp] == 0) + tmp := tmp - 1 + dst[i+63:i] := dst[i+63:i] + 1 + OD + ELSE + dst[i+63:i] := 0 + FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:128] := 0 +
immintrin.h
 - + Integer - AVX - Arithmetic - - - Divide packed unsigned 32-bit integers in "a" by packed elements in "b", and store the remainders as packed unsigned 32-bit integers in "dst". + AVX512CD + Swizzle + + + Broadcast the low 8-bits from input mask "k" to all 64-bit elements of "dst". FOR j := 0 to 7 - i := 32*j - dst[i+31:i] := REMAINDER(a[i+31:i] / b[i+31:i]) + i := j*64 + dst[i+63:i] := ZeroExtend64(k[7:0]) ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 +
immintrin.h
 - + Integer - POPCNT - Bit Manipulation - - - Count the number of bits set to 1 in 32-bit integer "a", and return that count in "dst". - + AVX512CD + Swizzle + + + Broadcast the low 16-bits from input mask "k" to all 32-bit elements of "dst". -dst := 0 -FOR i := 0 to 31 - IF a[i] - dst := dst + 1 - FI +FOR j := 0 to 15 + i := j*32 + dst[i+31:i] := ZeroExtend32(k[15:0]) ENDFOR +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - POPCNT - Bit Manipulation - - - Count the number of bits set to 1 in 64-bit integer "a", and return that count in "dst". - + AVX512CD + Compare + + + Test each 32-bit element of "a" for equality with all other elements in "a" closer to the least significant bit. Each element's comparison forms a zero extended bit vector in "dst". -dst := 0 -FOR i := 0 to 63 - IF a[i] - dst := dst + 1 - FI +FOR j := 0 to 15 + i := j*32 + FOR k := 0 to j-1 + m := k*32 + dst[i+k] := (a[i+31:i] == a[m+31:m]) ? 1 : 0 + ENDFOR + dst[i+31:i+j] := 0 ENDFOR +dst[MAX:512] := 0 - +
immintrin.h
 - - General Support - - Read the Performance Monitor Counter (PMC) specified by "a", and store up to 64-bits in "dst". The width of performance counters is implementation specific. + + Integer + AVX512CD + Compare + + + + + Test each 32-bit element of "a" for equality with all other elements in "a" closer to the least significant bit using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). Each element's comparison forms a zero extended bit vector in "dst". -dst[63:0] := ReadPMC(a) +FOR j := 0 to 15 + i := j*32 + IF k[j] + FOR l := 0 to j-1 + m := l*32 + dst[i+l] := (a[i+31:i] == a[m+31:m]) ? 1 : 0 + ENDFOR + dst[i+31:i+j] := 0 + ELSE + dst[i+31:i] := src[i+31:i] + FI +ENDFOR +dst[MAX:512] := 0 - +
immintrin.h
 - - TSC - General Support - - Copy the current 64-bit value of the processor's time-stamp counter into "dst". + + Integer + AVX512CD + Compare + + + + Test each 32-bit element of "a" for equality with all other elements in "a" closer to the least significant bit using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). Each element's comparison forms a zero extended bit vector in "dst". -dst[63:0] := TimeStampCounter +FOR j := 0 to 15 + i := j*32 + IF k[j] + FOR l := 0 to j-1 + m := l*32 + dst[i+l] := (a[i+31:i] == a[m+31:m]) ? 1 : 0 + ENDFOR + dst[i+31:i+j] := 0 + ELSE + dst[i+31:i] := 0 + FI +ENDFOR +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - Shift - - - Shift the bits of unsigned 32-bit integer "a" left by the number of bits specified in "shift", rotating the most-significant bit to the least-significant bit location, and store the unsigned result in "dst". + AVX512CD + Compare + + + Test each 64-bit element of "a" for equality with all other elements in "a" closer to the least significant bit. Each element's comparison forms a zero extended bit vector in "dst". -dst := a -count := shift BITWISE AND 31 -DO WHILE (count > 0) - tmp[0] := dst[31] - dst := (dst << 1) OR tmp[0] - count := count - 1 -OD +FOR j := 0 to 7 + i := j*64 + FOR k := 0 to j-1 + m := k*64 + dst[i+k] := (a[i+63:i] == a[m+63:m]) ? 1 : 0 + ENDFOR + dst[i+63:i+j] := 0 +ENDFOR +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - Shift - - - Shift the bits of unsigned 32-bit integer "a" right by the number of bits specified in "shift", rotating the least-significant bit to the most-significant bit location, and store the unsigned result in "dst". + AVX512CD + Compare + + + + + Test each 64-bit element of "a" for equality with all other elements in "a" closer to the least significant bit using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). Each element's comparison forms a zero extended bit vector in "dst". -dst := a -count := shift BITWISE AND 31 -DO WHILE (count > 0) - tmp[31] := dst[0] - dst := (dst >> 1) OR tmp[31] - count := count - 1 -OD +FOR j := 0 to 7 + i := j*64 + IF k[j] + FOR l := 0 to j-1 + m := l*64 + dst[i+l] := (a[i+63:i] == a[m+63:m]) ? 1 : 0 + ENDFOR + dst[i+63:i+j] := 0 + ELSE + dst[i+63:i] := src[i+63:i] + FI +ENDFOR +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - Shift - - - Shift the bits of unsigned 16-bit integer "a" left by the number of bits specified in "shift", rotating the most-significant bit to the least-significant bit location, and store the unsigned result in "dst". + AVX512CD + Compare + + + + Test each 64-bit element of "a" for equality with all other elements in "a" closer to the least significant bit using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). Each element's comparison forms a zero extended bit vector in "dst". -dst := a -count := shift BITWISE AND 15 -DO WHILE (count > 0) - tmp[0] := dst[15] - dst := (dst << 1) OR tmp[0] - count := count - 1 -OD +FOR j := 0 to 7 + i := j*64 + IF k[j] + FOR l := 0 to j-1 + m := l*64 + dst[i+l] := (a[i+63:i] == a[m+63:m]) ? 1 : 0 + ENDFOR + dst[i+63:i+j] := 0 + ELSE + dst[i+63:i] := 0 + FI +ENDFOR +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - Shift - - - Shift the bits of unsigned 16-bit integer "a" right by the number of bits specified in "shift", rotating the least-significant bit to the most-significant bit location, and store the unsigned result in "dst". + AVX512CD + Bit Manipulation + + + Counts the number of leading zero bits in each packed 32-bit integer in "a", and store the results in "dst". -dst := a -count := shift BITWISE AND 15 -DO WHILE (count > 0) - tmp[15] := dst[0] - dst := (dst >> 1) OR tmp[15] - count := count - 1 -OD +FOR j := 0 to 15 + i := j*32 + tmp := 31 + dst[i+31:i] := 0 + DO WHILE (tmp >= 0 AND a[i+tmp] == 0) + tmp := tmp - 1 + dst[i+31:i] := dst[i+31:i] + 1 + OD +ENDFOR +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - Shift - - - Shift the bits of unsigned 64-bit integer "a" left by the number of bits specified in "shift", rotating the most-significant bit to the least-significant bit location, and store the unsigned result in "dst". + AVX512CD + Bit Manipulation + + + + + Counts the number of leading zero bits in each packed 32-bit integer in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -dst := a -count := shift BITWISE AND 63 -DO WHILE (count > 0) - tmp[0] := dst[63] - dst := (dst << 1) OR tmp[0] - count := count - 1 -OD +FOR j := 0 to 15 + i := j*32 + IF k[j] + tmp := 31 + dst[i+31:i] := 0 + DO WHILE (tmp >= 0 AND a[i+tmp] == 0) + tmp := tmp - 1 + dst[i+31:i] := dst[i+31:i] + 1 + OD + ELSE + dst[i+31:i] := src[i+31:i] + FI +ENDFOR +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - Shift - - - Shift the bits of unsigned 64-bit integer "a" right by the number of bits specified in "shift", rotating the least-significant bit to the most-significant bit location, and store the unsigned result in "dst". + AVX512CD + Bit Manipulation + + + + Counts the number of leading zero bits in each packed 32-bit integer in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -dst := a -count := shift BITWISE AND 63 -DO WHILE (count > 0) - tmp[63] := dst[0] - dst := (dst >> 1) OR tmp[63] - count := count - 1 -OD +FOR j := 0 to 15 + i := j*32 + IF k[j] + tmp := 31 + dst[i+31:i] := 0 + DO WHILE (tmp >= 0 AND a[i+tmp] == 0) + tmp := tmp - 1 + dst[i+31:i] := dst[i+31:i] + 1 + OD + ELSE + dst[i+31:i] := 0 + FI +ENDFOR +dst[MAX:512] := 0 - +
immintrin.h
 - - XSAVE - OS-Targeted - - Copy up to 64-bits from the value of the extended control register (XCR) specified by "a" into "dst". Currently only XFEATURE_ENABLED_MASK XCR is supported. + + Integer + AVX512CD + Bit Manipulation + + + Counts the number of leading zero bits in each packed 64-bit integer in "a", and store the results in "dst". -dst[63:0] := XCR[a] +FOR j := 0 to 7 + i := j*64 + tmp := 63 + dst[i+63:i] := 0 + DO WHILE (tmp >= 0 AND a[i+tmp] == 0) + tmp := tmp - 1 + dst[i+63:i] := dst[i+63:i] + 1 + OD +ENDFOR +dst[MAX:512] := 0 - +
immintrin.h
 - - XSAVE - OS-Targeted - - - Perform a full or partial restore of the enabled processor states using the state information stored in memory at "mem_addr". State is restored based on bits [62:0] in "rs_mask", "XCR0", and "mem_addr.HEADER.XSTATE_BV". "mem_addr" must be aligned on a 64-byte boundary. + + Integer + AVX512CD + Bit Manipulation + + + + + Counts the number of leading zero bits in each packed 64-bit integer in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -st_mask = mem_addr.HEADER.XSTATE_BV[62:0] -FOR i := 0 to 62 - IF (rs_mask[i] AND XCR0[i]) - IF st_mask[i] - CASE (i) OF - 0: ProcessorState[x87 FPU] := mem_addr.FPUSSESave_Area[FPU] - 1: ProcessorState[SSE] := mem_addr.FPUSSESaveArea[SSE] - DEFAULT: ProcessorState[i] := mem_addr.Ext_Save_Area[i] - ESAC - ELSE - // ProcessorExtendedState := Processor Supplied Values - CASE (i) OF - 1: MXCSR := mem_addr.FPUSSESave_Area[SSE] - ESAC - FI +FOR j := 0 to 7 + i := j*64 + IF k[j] + tmp := 63 + dst[i+63:i] := 0 + DO WHILE (tmp >= 0 AND a[i+tmp] == 0) + tmp := tmp - 1 + dst[i+63:i] := dst[i+63:i] + 1 + OD + ELSE + dst[i+63:i] := src[i+63:i] FI - i := i + 1 ENDFOR +dst[MAX:512] := 0 - +
immintrin.h
 - - XSAVE - OS-Targeted - - - Perform a full or partial restore of the enabled processor states using the state information stored in memory at "mem_addr". State is restored based on bits [62:0] in "rs_mask", "XCR0", and "mem_addr.HEADER.XSTATE_BV". "mem_addr" must be aligned on a 64-byte boundary. + + Integer + AVX512CD + Bit Manipulation + + + + Counts the number of leading zero bits in each packed 64-bit integer in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -st_mask = mem_addr.HEADER.XSTATE_BV[62:0] -FOR i := 0 to 62 - IF (rs_mask[i] AND XCR0[i]) - IF st_mask[i] - CASE (i) OF - 0: ProcessorState[x87 FPU] := mem_addr.FPUSSESave_Area[FPU] - 1: ProcessorState[SSE] := mem_addr.FPUSSESaveArea[SSE] - DEFAULT: ProcessorState[i] := mem_addr.Ext_Save_Area[i] - ESAC - ELSE - // ProcessorExtendedState := Processor Supplied Values - CASE (i) OF - 1: MXCSR := mem_addr.FPUSSESave_Area[SSE] - ESAC - FI +FOR j := 0 to 7 + i := j*64 + IF k[j] + tmp := 63 + dst[i+63:i] := 0 + DO WHILE (tmp >= 0 AND a[i+tmp] == 0) + tmp := tmp - 1 + dst[i+63:i] := dst[i+63:i] + 1 + OD + ELSE + dst[i+63:i] := 0 FI - i := i + 1 ENDFOR +dst[MAX:512] := 0 - +
immintrin.h
 - - XSAVE - OS-Targeted - - - Perform a full or partial save of the enabled processor states to memory at "mem_addr". State is saved based on bits [62:0] in "save_mask" and "XCR0". "mem_addr" must be aligned on a 64-byte boundary. + + Floating Point + AVX512VL + AVX512DQ + Logical + + + + + + Compute the bitwise NOT of packed double-precision (64-bit) floating-point elements in "a" and then AND with "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -mask[62:0] := save_mask[62:0] BITWISE AND XCR0[62:0] -FOR i := 0 to 62 - IF mask[i] - CASE (i) OF - 0: mem_addr.FPUSSESave_Area[FPU] := ProcessorState[x87 FPU] - 1: mem_addr.FPUSSESaveArea[SSE] := ProcessorState[SSE] - DEFAULT: mem_addr.Ext_Save_Area[i] := ProcessorState[i] - ESAC - mem_addr.HEADER.XSTATE_BV[i] := INIT_FUNCTION[i] +FOR j := 0 to 3 + i := j*64 + IF k[j] + dst[i+63:i] := ((NOT a[i+63:i]) AND b[i+63:i]) + ELSE + dst[i+63:i] := src[i+63:i] FI - i := i + 1 ENDFOR +dst[MAX:256] := 0 - +
immintrin.h
 - - XSAVE - OS-Targeted - - - Perform a full or partial save of the enabled processor states to memory at "mem_addr". State is saved based on bits [62:0] in "save_mask" and "XCR0". "mem_addr" must be aligned on a 64-byte boundary. + + Floating Point + AVX512VL + AVX512DQ + Logical + + + + + Compute the bitwise NOT of packed double-precision (64-bit) floating-point elements in "a" and then AND with "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -mask[62:0] := save_mask[62:0] BITWISE AND XCR0[62:0] -FOR i := 0 to 62 - IF mask[i] - CASE (i) OF - 0: mem_addr.FPUSSESave_Area[FPU] := ProcessorState[x87 FPU] - 1: mem_addr.FPUSSESaveArea[SSE] := ProcessorState[SSE] - DEFAULT: mem_addr.Ext_Save_Area[i] := ProcessorState[i] - ESAC - mem_addr.HEADER.XSTATE_BV[i] := INIT_FUNCTION[i] +FOR j := 0 to 3 + i := j*64 + IF k[j] + dst[i+63:i] := ((NOT a[i+63:i]) AND b[i+63:i]) + ELSE + dst[i+63:i] := 0 FI - i := i + 1 ENDFOR +dst[MAX:256] := 0 - +
immintrin.h
 - - XSAVE - XSAVEOPT - OS-Targeted - - - Perform a full or partial save of the enabled processor states to memory at "mem_addr". State is saved based on bits [62:0] in "save_mask" and "XCR0". "mem_addr" must be aligned on a 64-byte boundary. The hardware may optimize the manner in which data is saved. The performance of this instruction will be equal to or better than using the XSAVE instruction. - -mask[62:0] := save_mask[62:0] BITWISE AND XCR0[62:0] -FOR i := 0 to 62 - IF mask[i] - CASE (i) OF - 0: mem_addr.FPUSSESave_Area[FPU] := ProcessorState[x87 FPU] - 1: mem_addr.FPUSSESaveArea[SSE] := ProcessorState[SSE] - 2: mem_addr.EXT_SAVE_Area2[YMM] := ProcessorState[YMM] - DEFAULT: mem_addr.Ext_Save_Area[i] := ProcessorState[i] - ESAC - mem_addr.HEADER.XSTATE_BV[i] := INIT_FUNCTION[i] - FI - i := i + 1 + + Floating Point + AVX512DQ + Logical + + + + Compute the bitwise NOT of packed double-precision (64-bit) floating-point elements in "a" and then AND with "b", and store the results in "dst". + +FOR j := 0 to 7 + i := j*64 + dst[i+63:i] := ((NOT a[i+63:i]) AND b[i+63:i]) ENDFOR +dst[MAX:512] := 0 - +
immintrin.h
 - - XSAVE - XSAVEOPT - OS-Targeted - - - Perform a full or partial save of the enabled processor states to memory at "mem_addr". State is saved based on bits [62:0] in "save_mask" and "XCR0". "mem_addr" must be aligned on a 64-byte boundary. The hardware may optimize the manner in which data is saved. The performance of this instruction will be equal to or better than using the XSAVE64 instruction. + + Floating Point + AVX512DQ + Logical + + + + + + Compute the bitwise NOT of packed double-precision (64-bit) floating-point elements in "a" and then AND with "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -mask[62:0] := save_mask[62:0] BITWISE AND XCR0[62:0] -FOR i := 0 to 62 - IF mask[i] - CASE (i) OF - 0: mem_addr.FPUSSESave_Area[FPU] := ProcessorState[x87 FPU] - 1: mem_addr.FPUSSESaveArea[SSE] := ProcessorState[SSE] - 2: mem_addr.EXT_SAVE_Area2[YMM] := ProcessorState[YMM] - DEFAULT: mem_addr.Ext_Save_Area[i] := ProcessorState[i] - ESAC - mem_addr.HEADER.XSTATE_BV[i] := INIT_FUNCTION[i] +FOR j := 0 to 7 + i := j*64 + IF k[j] + dst[i+63:i] := ((NOT a[i+63:i]) AND b[i+63:i]) + ELSE + dst[i+63:i] := src[i+63:i] FI - i := i + 1 ENDFOR +dst[MAX:512] := 0 - -
immintrin.h
- - - XSAVE - OS-Targeted - - - Copy 64-bits from "val" to the extended control register (XCR) specified by "a". Currently only XFEATURE_ENABLED_MASK XCR is supported. - -XCR[a] := val[63:0] - - +
immintrin.h
 - - - - - Integer - SSE2 - Load - - Load unaligned 32-bit integer from memory into the first element of "dst". + + Floating Point + AVX512DQ + Logical + + + + + Compute the bitwise NOT of packed double-precision (64-bit) floating-point elements in "a" and then AND with "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -dst[31:0] := MEM[mem_addr+31:mem_addr] -dst[MAX:32] := 0 +FOR j := 0 to 7 + i := j*64 + IF k[j] + dst[i+63:i] := ((NOT a[i+63:i]) AND b[i+63:i]) + ELSE + dst[i+63:i] := 0 + FI +ENDFOR +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - SSE2 - Store - - - Store 32-bit integer from the first element of "a" into memory. "mem_addr" does not need to be aligned on any particular boundary. + + Floating Point + AVX512VL + AVX512DQ + Logical + + + + + + Compute the bitwise NOT of packed double-precision (64-bit) floating-point elements in "a" and then AND with "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -MEM[mem_addr+31:mem_addr] := a[31:0] +FOR j := 0 to 1 + i := j*64 + IF k[j] + dst[i+63:i] := ((NOT a[i+63:i]) AND b[i+63:i]) + ELSE + dst[i+63:i] := src[i+63:i] + FI +ENDFOR +dst[MAX:128] := 0 - +
immintrin.h
 - - Integer - SSE - Store - - - Store 16-bit integer from the first element of "a" into memory. "mem_addr" does not need to be aligned on any particular boundary. + + Floating Point + AVX512VL + AVX512DQ + Logical + + + + + Compute the bitwise NOT of packed double-precision (64-bit) floating-point elements in "a" and then AND with "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -MEM[mem_addr+15:mem_addr] := a[15:0] +FOR j := 0 to 1 + i := j*64 + IF k[j] + dst[i+63:i] := ((NOT a[i+63:i]) AND b[i+63:i]) + ELSE + dst[i+63:i] := 0 + FI +ENDFOR +dst[MAX:128] := 0 +
immintrin.h
 - - Integer - SSE - Load - - Load unaligned 64-bit integer from memory into the first element of "dst". + + Floating Point + AVX512VL + AVX512DQ + Logical + + + + + + Compute the bitwise NOT of packed single-precision (32-bit) floating-point elements in "a" and then AND with "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -dst[63:0] := MEM[mem_addr+63:mem_addr] -dst[MAX:64] := 0 +FOR j := 0 to 7 + i := j*32 + IF k[j] + dst[i+31:i] := ((NOT a[i+31:i]) AND b[i+31:i]) + ELSE + dst[i+31:i] := src[i+31:i] + FI +ENDFOR +dst[MAX:256] := 0 - +
immintrin.h
 - - Integer - SSE - Store - - - Store 64-bit integer from the first element of "a" into memory. "mem_addr" does not need to be aligned on any particular boundary. + + Floating Point + AVX512VL + AVX512DQ + Logical + + + + + Compute the bitwise NOT of packed single-precision (32-bit) floating-point elements in "a" and then AND with "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -MEM[mem_addr+63:mem_addr] := a[63:0] +FOR j := 0 to 7 + i := j*32 + IF k[j] + dst[i+31:i] := ((NOT a[i+31:i]) AND b[i+31:i]) + ELSE + dst[i+31:i] := 0 + FI +ENDFOR +dst[MAX:256] := 0 - +
immintrin.h
 - - Integer - SSE - Load - - Load unaligned 16-bit integer from memory into the first element of "dst". + + Floating Point + AVX512DQ + Logical + + + + Compute the bitwise NOT of packed single-precision (32-bit) floating-point elements in "a" and then AND with "b", and store the results in "dst". -dst[15:0] := MEM[mem_addr+15:mem_addr] -dst[MAX:16] := 0 +FOR j := 0 to 15 + i := j*32 + dst[i+31:i] := ((NOT a[i+31:i]) AND b[i+31:i]) +ENDFOR +dst[MAX:512] := 0 +
immintrin.h
 - - Integer - FSGSBASE - General Support - Read the FS segment base register and store the 32-bit result in "dst". + + Floating Point + AVX512DQ + Logical + + + + + + Compute the bitwise NOT of packed single-precision (32-bit) floating-point elements in "a" and then AND with "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -dst[31:0] := FS_Segment_Base_Register; -dst[63:32] := 0 +FOR j := 0 to 15 + i := j*32 + IF k[j] + dst[i+31:i] := ((NOT a[i+31:i]) AND b[i+31:i]) + ELSE + dst[i+31:i] := src[i+31:i] + FI +ENDFOR +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - FSGSBASE - General Support - Read the FS segment base register and store the 64-bit result in "dst". + + Floating Point + AVX512DQ + Logical + + + + + Compute the bitwise NOT of packed single-precision (32-bit) floating-point elements in "a" and then AND with "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -dst[63:0] := FS_Segment_Base_Register; +FOR j := 0 to 15 + i := j*32 + IF k[j] + dst[i+31:i] := ((NOT a[i+31:i]) AND b[i+31:i]) + ELSE + dst[i+31:i] := 0 + FI +ENDFOR +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - FSGSBASE - General Support - Read the GS segment base register and store the 32-bit result in "dst". + + Floating Point + AVX512VL + AVX512DQ + Logical + + + + + + Compute the bitwise NOT of packed single-precision (32-bit) floating-point elements in "a" and then AND with "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -dst[31:0] := GS_Segment_Base_Register; -dst[63:32] := 0 +FOR j := 0 to 3 + i := j*32 + IF k[j] + dst[i+31:i] := ((NOT a[i+31:i]) AND b[i+31:i]) + ELSE + dst[i+31:i] := src[i+31:i] + FI +ENDFOR +dst[MAX:128] := 0 - +
immintrin.h
 - - Integer - FSGSBASE - General Support - Read the GS segment base register and store the 64-bit result in "dst". + + Floating Point + AVX512VL + AVX512DQ + Logical + + + + + Compute the bitwise NOT of packed single-precision (32-bit) floating-point elements in "a" and then AND with "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -dst[63:0] := GS_Segment_Base_Register; +FOR j := 0 to 3 + i := j*32 + IF k[j] + dst[i+31:i] := ((NOT a[i+31:i]) AND b[i+31:i]) + ELSE + dst[i+31:i] := 0 + FI +ENDFOR +dst[MAX:128] := 0 - +
immintrin.h
 - - Integer - RDRAND - Random - - Read a hardware generated 16-bit random value and store the result in "val". Return 1 if a random value was generated, and 0 otherwise. - + + Floating Point + AVX512VL + AVX512DQ + Logical + + + + + + Compute the bitwise AND of packed double-precision (64-bit) floating-point elements in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -IF HW_RND_GEN.ready = 1 - val[15:0] := HW_RND_GEN.data; - RETURN 1; -ELSE - val[15:0] := 0; - RETURN 0; -FI +FOR j := 0 to 3 + i := j*64 + IF k[j] + dst[i+63:i] := (a[i+63:i] AND b[i+63:i]) + ELSE + dst[i+63:i] := src[i+63:i] + FI +ENDFOR +dst[MAX:256] := 0 - +
immintrin.h
 - - Integer - RDRAND - Random - - Read a hardware generated 32-bit random value and store the result in "val". Return 1 if a random value was generated, and 0 otherwise. - + + Floating Point + AVX512VL + AVX512DQ + Logical + + + + + Compute the bitwise AND of packed double-precision (64-bit) floating-point elements in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -IF HW_RND_GEN.ready = 1 - val[31:0] := HW_RND_GEN.data; - RETURN 1; -ELSE - val[31:0] := 0; - RETURN 0; -FI +FOR j := 0 to 3 + i := j*64 + IF k[j] + dst[i+63:i] := (a[i+63:i] AND b[i+63:i]) + ELSE + dst[i+63:i] := 0 + FI +ENDFOR +dst[MAX:256] := 0 - +
immintrin.h
 - - Integer - RDRAND - Random - - Read a hardware generated 64-bit random value and store the result in "val". Return 1 if a random value was generated, and 0 otherwise. - + + Floating Point + AVX512DQ + Logical + + + + Compute the bitwise AND of packed double-precision (64-bit) floating-point elements in "a" and "b", and store the results in "dst". -IF HW_RND_GEN.ready = 1 - val[63:0] := HW_RND_GEN.data; - RETURN 1; -ELSE - val[63:0] := 0; - RETURN 0; -FI +FOR j := 0 to 7 + i := j*64 + dst[i+63:i] := (a[i+63:i] AND b[i+63:i]) +ENDFOR +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - FSGSBASE - General Support - - Write the unsigned 32-bit integer "a" to the FS segment base register. + + Floating Point + AVX512DQ + Logical + + + + + + Compute the bitwise AND of packed double-precision (64-bit) floating-point elements in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FS_Segment_Base_Register[31:0] := a[31:0]; -FS_Segment_Base_Register[63:32] := 0 +FOR j := 0 to 7 + i := j*64 + IF k[j] + dst[i+63:i] := (a[i+63:i] AND b[i+63:i]) + ELSE + dst[i+63:i] := src[i+63:i] + FI +ENDFOR +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - FSGSBASE - General Support - - Write the unsigned 64-bit integer "a" to the FS segment base register. + + Floating Point + AVX512DQ + Logical + + + + + Compute the bitwise AND of packed double-precision (64-bit) floating-point elements in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FS_Segment_Base_Register[63:0] := a[63:0]; +FOR j := 0 to 7 + i := j*64 + IF k[j] + dst[i+63:i] := (a[i+63:i] AND b[i+63:i]) + ELSE + dst[i+63:i] := 0 + FI +ENDFOR +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - FSGSBASE - General Support - - Write the unsigned 32-bit integer "a" to the GS segment base register. + + Floating Point + AVX512VL + AVX512DQ + Logical + + + + + + Compute the bitwise AND of packed double-precision (64-bit) floating-point elements in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -GS_Segment_Base_Register[31:0] := a[31:0]; -GS_Segment_Base_Register[63:32] := 0 +FOR j := 0 to 1 + i := j*64 + IF k[j] + dst[i+63:i] := (a[i+63:i] AND b[i+63:i]) + ELSE + dst[i+63:i] := src[i+63:i] + FI +ENDFOR +dst[MAX:128] := 0 - +
immintrin.h
 - - Integer - FSGSBASE - General Support - - Write the unsigned 64-bit integer "a" to the GS segment base register. + + Floating Point + AVX512VL + AVX512DQ + Logical + + + + + Compute the bitwise AND of packed double-precision (64-bit) floating-point elements in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -GS_Segment_Base_Register[63:0] := a[63:0]; +FOR j := 0 to 1 + i := j*64 + IF k[j] + dst[i+63:i] := (a[i+63:i] AND b[i+63:i]) + ELSE + dst[i+63:i] := 0 + FI +ENDFOR +dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point - FP16C - Convert - - Convert packed half-precision (16-bit) floating-point elements in "a" to packed single-precision (32-bit) floating-point elements, and store the results in "dst". + AVX512VL + AVX512DQ + Logical + + + + + + Compute the bitwise AND of packed single-precision (32-bit) floating-point elements in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). FOR j := 0 to 7 i := j*32 - m := j*16 - dst[i+31:i] := Convert_FP16_To_FP32(a[m+15:m]) + IF k[j] + dst[i+31:i] := (a[i+31:i] AND b[i+31:i]) + ELSE + dst[i+31:i] := src[i+31:i] + FI ENDFOR dst[MAX:256] := 0 - +
immintrin.h
 - + Floating Point - FP16C - Convert - - - Convert packed single-precision (32-bit) floating-point elements in "a" to packed half-precision (16-bit) floating-point elements, and store the results in "dst". - [round_note] - + AVX512VL + AVX512DQ + Logical + + + + + Compute the bitwise AND of packed single-precision (32-bit) floating-point elements in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). FOR j := 0 to 7 - i := 16*j - l := 32*j - dst[i+15:i] := Convert_FP32_To_FP16FP(a[l+31:l]) + i := j*32 + IF k[j] + dst[i+31:i] := (a[i+31:i] AND b[i+31:i]) + ELSE + dst[i+31:i] := 0 + FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - - RDSEED - Random - - Read a 16-bit NIST SP800-90B and SP800-90C compliant random value and store in "val". Return 1 if a random value was generated, and 0 otherwise. + + Floating Point + AVX512DQ + Logical + + + + Compute the bitwise AND of packed single-precision (32-bit) floating-point elements in "a" and "b", and store the results in "dst". -IF HW_NRND_GEN.ready = 1 THEN - val[15:0] := HW_NRND_GEN.data - RETURN 1 -ELSE - val[15:0] := 0 - RETURN 0 -FI +FOR j := 0 to 15 + i := j*32 + dst[i+31:i] := (a[i+31:i] AND b[i+31:i]) +ENDFOR +dst[MAX:512] := 0 - +
immintrin.h
 - - RDSEED - Random - - Read a 32-bit NIST SP800-90B and SP800-90C compliant random value and store in "val". Return 1 if a random value was generated, and 0 otherwise. + + Floating Point + AVX512DQ + Logical + + + + + + Compute the bitwise AND of packed single-precision (32-bit) floating-point elements in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -IF HW_NRND_GEN.ready = 1 THEN - val[31:0] := HW_NRND_GEN.data - RETURN 1 -ELSE - val[31:0] := 0 - RETURN 0 -FI +FOR j := 0 to 15 + i := j*32 + IF k[j] + dst[i+31:i] := (a[i+31:i] AND b[i+31:i]) + ELSE + dst[i+31:i] := src[i+31:i] + FI +ENDFOR +dst[MAX:512] := 0 - +
immintrin.h
 - - RDSEED - Random - - Read a 64-bit NIST SP800-90B and SP800-90C compliant random value and store in "val". Return 1 if a random value was generated, and 0 otherwise. + + Floating Point + AVX512DQ + Logical + + + + + Compute the bitwise AND of packed single-precision (32-bit) floating-point elements in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -IF HW_NRND_GEN.ready = 1 THEN - val[63:0] := HW_NRND_GEN.data - RETURN 1 -ELSE - val[63:0] := 0 - RETURN 0 -FI +FOR j := 0 to 15 + i := j*32 + IF k[j] + dst[i+31:i] := (a[i+31:i] AND b[i+31:i]) + ELSE + dst[i+31:i] := 0 + FI +ENDFOR +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - Arithmetic - - - - - Add unsigned 32-bit integers "a" and "b" with unsigned 8-bit carry-in "c_in" (carry flag), and store the unsigned 32-bit result in "out", and the carry-out in "dst" (carry or overflow flag). + + Floating Point + AVX512VL + AVX512DQ + Logical + + + + + + Compute the bitwise AND of packed single-precision (32-bit) floating-point elements in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -dst:out[31:0] := a[31:0] + b[31:0] + c_in; +FOR j := 0 to 3 + i := j*32 + IF k[j] + dst[i+31:i] := (a[i+31:i] AND b[i+31:i]) + ELSE + dst[i+31:i] := src[i+31:i] + FI +ENDFOR +dst[MAX:128] := 0 - +
immintrin.h
 - - Integer - Arithmetic - - - - - Add unsigned 64-bit integers "a" and "b" with unsigned 8-bit carry-in "c_in" (carry flag), and store the unsigned 64-bit result in "out", and the carry-out in "dst" (carry or overflow flag). + + Floating Point + AVX512VL + AVX512DQ + Logical + + + + + Compute the bitwise AND of packed single-precision (32-bit) floating-point elements in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -dst:out[63:0] := a[63:0] + b[63:0] + c_in; +FOR j := 0 to 3 + i := j*32 + IF k[j] + dst[i+31:i] := (a[i+31:i] AND b[i+31:i]) + ELSE + dst[i+31:i] := 0 + FI +ENDFOR +dst[MAX:128] := 0 - +
immintrin.h
 - - Integer - Arithmetic - - - - - Add unsigned 8-bit borrow "b_in" (carry flag) to unsigned 32-bit integer "b", and subtract the result from unsigned 32-bit integer "a". Store the unsigned 32-bit result in "out", and the carry-out in "dst" (carry or overflow flag). + + Floating Point + AVX512VL + AVX512DQ + Miscellaneous + + + Broadcast the lower 2 packed single-precision (32-bit) floating-point elements from "a" to all elements of "dst". -dst:out[31:0] := (a[31:0] - (b[31:0] + b_in)); +FOR j := 0 to 7 + i := j*32 + n := (j % 2)*32 + dst[i+31:i] := a[n+31:n] +ENDFOR +dst[MAX:256] := 0 - +
immintrin.h
 - - Integer - Arithmetic - - - - - Add unsigned 8-bit borrow "b_in" (carry flag) to unsigned 64-bit integer "b", and subtract the result from unsigned 64-bit integer "a". Store the unsigned 64-bit result in "out", and the carry-out in "dst" (carry or overflow flag). + + Floating Point + AVX512VL + AVX512DQ + Miscellaneous + + + + + Broadcast the lower 2 packed single-precision (32-bit) floating-point elements from "a" to all elements of "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -dst:out[63:0] := (a[63:0] - (b[63:0] + b_in)); +FOR j := 0 to 7 + i := j*32 + n := (j % 2)*32 + IF k[j] + dst[i+31:i] := a[n+31:n] + ELSE + dst[i+31:i] := src[i+31:i] + FI +ENDFOR +dst[MAX:256] := 0 - +
immintrin.h
 - - Integer - ADX - Arithmetic - - - - - Add unsigned 32-bit integers "a" and "b" with unsigned 8-bit carry-in "c_in" (carry or overflow flag), and store the unsigned 32-bit result in "out", and the carry-out in "dst" (carry or overflow flag). + + Floating Point + AVX512VL + AVX512DQ + Miscellaneous + + + + Broadcast the lower 2 packed single-precision (32-bit) floating-point elements from "a" to all elements of "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -dst:out[31:0] := a[31:0] + b[31:0] + c_in; +FOR j := 0 to 7 + i := j*32 + n := (j % 2)*32 + IF k[j] + dst[i+31:i] := a[n+31:n] + ELSE + dst[i+31:i] := 0 + FI +ENDFOR +dst[MAX:256] := 0 - - +
immintrin.h
 - - Integer - ADX - Arithmetic - - - - - Add unsigned 64-bit integers "a" and "b" with unsigned 8-bit carry-in "c_in" (carry or overflow flag), and store the unsigned 64-bit result in "out", and the carry-out in "dst" (carry or overflow flag). + + Floating Point + AVX512DQ + Miscellaneous + + + Broadcast the lower 2 packed single-precision (32-bit) floating-point elements from "a" to all elements of "dst". -dst:out[63:0] := a[63:0] + b[63:0] + c_in; +FOR j := 0 to 15 + i := j*32 + n := (j % 2)*32 + dst[i+31:i] := a[n+31:n] +ENDFOR +dst[MAX:512] := 0 - - +
immintrin.h
 - - Integer - SHA - Cryptography - - - Perform an intermediate calculation for the next four SHA1 message values (unsigned 32-bit integers) using previous message values from "a" and "b", and store the result in "dst". + + Floating Point + AVX512DQ + Miscellaneous + + + + + Broadcast the lower 2 packed single-precision (32-bit) floating-point elements from "a" to all elements of "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -W0 := a[127:96]; -W1 := a[95:64]; -W2 := a[63:32]; -W3 := a[31:0]; -W4 := b[127:96]; -W5 := b[95:64]; - -dst[127:96] := W2 XOR W0; -dst[95:64] := W3 XOR W1; -dst[63:32] := W4 XOR W2; -dst[31:0] := W5 XOR W3; +FOR j := 0 to 15 + i := j*32 + n := (j % 2)*32 + IF k[j] + dst[i+31:i] := a[n+31:n] + ELSE + dst[i+31:i] := src[i+31:i] + FI +ENDFOR +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - SHA - Cryptography - - - Perform the final calculation for the next four SHA1 message values (unsigned 32-bit integers) using the intermediate result in "a" and the previous message values in "b", and store the result in "dst". + + Floating Point + AVX512DQ + Miscellaneous + + + + Broadcast the lower 2 packed single-precision (32-bit) floating-point elements from "a" to all elements of "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -W13 := b[95:64]; -W14 := b[63:32]; -W15 := b[31:0]; -W16 := (a[127:96] XOR W13) <<< 1; -W17 := (a[95:64] XOR W14) <<< 1; -W18 := (a[63:32] XOR W15) <<< 1; -W19 := (a[31:0] XOR W16) <<< 1; - -dst[127:96] := W16; -dst[95:64] := W17; -dst[63:32] := W18; -dst[31:0] := W19; +FOR j := 0 to 15 + i := j*32 + n := (j % 2)*32 + IF k[j] + dst[i+31:i] := a[n+31:n] + ELSE + dst[i+31:i] := 0 + FI +ENDFOR +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - SHA - Cryptography - - - Calculate SHA1 state variable E after four rounds of operation from the current SHA1 state variable "a", add that value to the scheduled values (unsigned 32-bit integers) in "b", and store the result in "dst". + + Floating Point + AVX512DQ + Miscellaneous + + + Broadcast the 8 packed single-precision (32-bit) floating-point elements from "a" to all elements of "dst". -tmp := (a[127:96] <<< 30); -dst[127:96] := b[127:96] + tmp; -dst[95:64] := b[95:64]; -dst[63:32] := b[63:32]; -dst[31:0] := b[31:0]; +FOR j := 0 to 15 + i := j*32 + n := (j % 8)*32 + dst[i+31:i] := a[n+31:n] +ENDFOR +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - SHA - Cryptography - - - - Perform four rounds of SHA1 operation using an initial SHA1 state (A,B,C,D) from "a" and some pre-computed sum of the next 4 round message values (unsigned 32-bit integers), and state variable E from "b", and store the updated SHA1 state (A,B,C,D) in "dst". "func" contains the logic functions and round constants. - -IF (func[1:0] = 0) THEN - f() := f0(), K := K0; -ELSE IF (func[1:0] = 1) THEN - f() := f1(), K := K1; -ELSE IF (func[1:0] = 2) THEN - f() := f2(), K := K2; -ELSE IF (func[1:0] = 3) THEN - f() := f3(), K := K3; -FI; - -A := a[127:96]; -B := a[95:64]; -C := a[63:32]; -D := a[31:0]; - -W[0] := b[127:96]; -W[1] := b[95:64]; -W[2] := b[63:32]; -W[3] := b[31:0]; - -A[1] := f(B, C, D) + (A <<< 5) + W[0] + K; -B[1] := A; -C[1] := B <<< 30; -D[1] := C; -E[1] := D; - -FOR i = 1 to 3 - A[i+1] := f(B[i], C[i], D[i]) + (A[i] <<< 5) + W[i] + E[i] + K; - B[i+1] := A[i]; - C[i+1] := B[i] <<< 30; - D[i+1] := C[i]; - E[i+1] := D[i]; -ENDFOR; - -dst[127:96] := A[4]; -dst[95:64] := B[4]; -dst[63:32] := C[4]; -dst[31:0] := D[4]; - - -
immintrin.h
- - - Integer - SHA - Cryptography - - - Perform an intermediate calculation for the next four SHA256 message values (unsigned 32-bit integers) using previous message values from "a" and "b", and store the result in "dst". + + Floating Point + AVX512DQ + Miscellaneous + + + + + Broadcast the 8 packed single-precision (32-bit) floating-point elements from "a" to all elements of "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -W4 := b[31:0]; -W3 := a[127:96]; -W2 := a[95:64]; -W1 := a[63:32]; -W0 := a[31:0]; - -dst[127:96] := W3 + sigma0(W4); -dst[95:64] := W2 + sigma0(W3); -dst[63:32] := W1 + sigma0(W2); -dst[31:0] := W0 + sigma0(W1); +FOR j := 0 to 15 + i := j*32 + n := (j % 8)*32 + IF k[j] + dst[i+31:i] := a[n+31:n] + ELSE + dst[i+31:i] := src[i+31:i] + FI +ENDFOR +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - SHA - Cryptography - - - Perform the final calculation for the next four SHA256 message values (unsigned 32-bit integers) using previous message values from "a" and "b", and store the result in "dst"." + + Floating Point + AVX512DQ + Miscellaneous + + + + Broadcast the 8 packed single-precision (32-bit) floating-point elements from "a" to all elements of "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -W14 := b[95:64]; -W15 := b[127:96]; -W16 := a[31:0] + sigma1(W14); -W17 := a[63:32] + sigma1(W15); -W18 := a[95:64] + sigma1(W16); -W19 := a[127:96] + sigma1(W17); - -dst[127:96] := W19; -dst[95:64] := W18; -dst[63:32] := W17; -dst[31:0] := W16; +FOR j := 0 to 15 + i := j*32 + n := (j % 8)*32 + IF k[j] + dst[i+31:i] := a[n+31:n] + ELSE + dst[i+31:i] := 0 + FI +ENDFOR +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - SHA - Cryptography - - - - Perform 2 rounds of SHA256 operation using an initial SHA256 state (C,D,G,H) from "a", an initial SHA256 state (A,B,E,F) from "b", and a pre-computed sum of the next 2 round message values (unsigned 32-bit integers) and the corresponding round constants from "k", and store the updated SHA256 state (A,B,E,F) in "dst". - -A[0] := b[127:96]; -B[0] := b[95:64]; -C[0] := a[127:96]; -D[0] := a[95:64]; -E[0] := b[63:32]; -F[0] := b[31:0]; -G[0] := a[63:32]; -H[0] := a[31:0]; - -W_K0 := k[31:0]; -W_K1 := k[63:32]; - -FOR i = 0 to 1 - A_(i+1) := Ch(E[i], F[i], G[i]) + sum1(E[i]) + WKi + H[i] + Maj(A[i], B[i], C[i]) + sum0(A[i]); - B_(i+1) := A[i]; - C_(i+1) := B[i]; - D_(i+1) := C[i]; - E_(i+1) := Ch(E[i], F[i], G[i]) + sum1(E[i]) + WKi + H[i] + D[i]; - F_(i+1) := E[i]; - G_(i+1) := F[i]; - H_(i+1) := G[i]; -ENDFOR; - -dst[127:96] := A[2]; -dst[95:64] := B[2]; -dst[63:32] := E[2]; -dst[31:0] := F[2]; - - -
immintrin.h
- - - MPX + + Floating Point + AVX512VL + AVX512DQ Miscellaneous - - - Make a pointer with the value of "srcmem" and bounds set to ["srcmem", "srcmem" + "size" - 1], and store the result in "dst". + + + Broadcast the 2 packed double-precision (64-bit) floating-point elements from "a" to all elements of "dst". -dst := srcmem; -dst.LB := srcmem.LB; -dst.UB := srcmem + size - 1; +FOR j := 0 to 3 + i := j*64 + n := (j % 2)*64 + dst[i+63:i] := a[n+63:n] +ENDFOR +dst[MAX:256] := 0 - +
immintrin.h
 - - MPX + + Floating Point + AVX512VL + AVX512DQ Miscellaneous - - - - Narrow the bounds for pointer "q" to the intersection of the bounds of "r" and the bounds ["q", "q" + "size" - 1], and store the result in "dst". + + + + + Broadcast the 2 packed double-precision (64-bit) floating-point elements from "a" to all elements of "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -dst := q; -IF r.LB > (q + size - 1) OR r.UB < q THEN - dst.LB := 1; - dst.UB := 0; -ELSE - dst.LB := MAX(r.LB, q); - dst.UB := MIN(r.UB, (q + size - 1)); -FI; +FOR j := 0 to 3 + i := j*64 + n := (j % 2)*64 + IF k[j] + dst[i+63:i] := a[n+63:n] + ELSE + dst[i+63:i] := src[i+63:i] + FI +ENDFOR +dst[MAX:256] := 0 +
immintrin.h
 - - MPX + + Floating Point + AVX512VL + AVX512DQ Miscellaneous - - - Make a pointer with the value of "q" and bounds set to the bounds of "r" (e.g. copy the bounds of "r" to pointer "q"), and store the result in "dst". + + + + Broadcast the 2 packed double-precision (64-bit) floating-point elements from "a" to all elements of "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -dst := q; -dst.LB := r.LB; -dst.UB := r.UB; +FOR j := 0 to 3 + i := j*64 + n := (j % 2)*64 + IF k[j] + dst[i+63:i] := a[n+63:n] + ELSE + dst[i+63:i] := 0 + FI +ENDFOR +dst[MAX:256] := 0 +
immintrin.h
 - - MPX + + Floating Point + AVX512DQ Miscellaneous - - Make a pointer with the value of "q" and open bounds, which allow the pointer to access the entire virtual address space, and store the result in "dst". + + + Broadcast the 2 packed double-precision (64-bit) floating-point elements from "a" to all elements of "dst". -dst := q; -dst.LB := 0; -dst.UB := 0; +FOR j := 0 to 7 + i := j*64 + n := (j % 2)*64 + dst[i+63:i] := a[n+63:n] +ENDFOR +dst[MAX:512] := 0 +
immintrin.h
 - - MPX + + Floating Point + AVX512DQ Miscellaneous - - - Stores the bounds of "ptr_val" pointer in memory at address "ptr_addr". + + + + + Broadcast the 2 packed double-precision (64-bit) floating-point elements from "a" to all elements of "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -MEM[ptr_addr].LB := ptr_val.LB; -MEM[ptr_addr].UB := ptr_val.UB; +FOR j := 0 to 7 + i := j*64 + n := (j % 2)*64 + IF k[j] + dst[i+63:i] := a[n+63:n] + ELSE + dst[i+63:i] := src[i+63:i] + FI +ENDFOR +dst[MAX:512] := 0 - +
immintrin.h
 - - MPX + + Floating Point + AVX512DQ Miscellaneous - - Checks if "q" is within its lower bound, and throws a #BR if not. + + + + Broadcast the 2 packed double-precision (64-bit) floating-point elements from "a" to all elements of "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -IF q < q.LB THEN - #BR; -FI; +FOR j := 0 to 7 + i := j*64 + n := (j % 2)*64 + IF k[j] + dst[i+63:i] := a[n+63:n] + ELSE + dst[i+63:i] := 0 + FI +ENDFOR +dst[MAX:512] := 0 - +
immintrin.h
 - - MPX + + AVX512VL + AVX512DQ Miscellaneous - - Checks if "q" is within its upper bound, and throws a #BR if not. + + + Broadcast the lower 2 packed 32-bit integers from "a" to all elements of "dst. -IF q > q.UB THEN - #BR; -FI; +FOR j := 0 to 7 + i := j*32 + n := (j % 2)*32 + dst[i+31:i] := a[n+31:n] +ENDFOR +dst[MAX:256] := 0 - - +
immintrin.h
 - - MPX + + AVX512VL + AVX512DQ Miscellaneous - - - Checks if ["q", "q" + "size" - 1] is within the lower and upper bounds of "q" and throws a #BR if not. + + + + + Broadcast the lower 2 packed 32-bit integers from "a" to all elements of "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -IF (q + size - 1) < q.LB OR (q + size - 1) > q.UB THEN - #BR; -FI; +FOR j := 0 to 7 + i := j*32 + n := (j % 2)*32 + IF k[j] + dst[i+31:i] := a[n+31:n] + ELSE + dst[i+31:i] := src[i+31:i] + FI +ENDFOR +dst[MAX:256] := 0 - - +
immintrin.h
 - - MPX + + AVX512VL + AVX512DQ Miscellaneous - - Return the lower bound of "q". + + + + Broadcast the lower 2 packed 32-bit integers from "a" to all elements of "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -dst := q.LB +FOR j := 0 to 7 + i := j*32 + n := (j % 2)*32 + IF k[j] + dst[i+31:i] := a[n+31:n] + ELSE + dst[i+31:i] := 0 + FI +ENDFOR +dst[MAX:256] := 0 +
immintrin.h
 - - MPX + + AVX512DQ Miscellaneous - - Return the upper bound of "q". + + + Broadcast the lower 2 packed 32-bit integers from "a" to all elements of "dst. -dst := q.UB +FOR j := 0 to 15 + i := j*32 + n := (j % 2)*32 + dst[i+31:i] := a[n+31:n] +ENDFOR +dst[MAX:512] := 0 +
immintrin.h
 - - Load - - Loads a big-endian word (16-bit) value from address "ptr" and stores the result in "dst". - + + AVX512DQ + Miscellaneous + + + + + Broadcast the lower 2 packed 32-bit integers from "a" to all elements of "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -addr := MEM[ptr] -FOR j := 0 to 1 - i := j*8 - dst[i+7:i] := addr[15-i:15-i-7] +FOR j := 0 to 15 + i := j*32 + n := (j % 2)*32 + IF k[j] + dst[i+31:i] := a[n+31:n] + ELSE + dst[i+31:i] := src[i+31:i] + FI ENDFOR +dst[MAX:512] := 0 +
immintrin.h
 - - Load - - Loads a big-endian double word (32-bit) value from address "ptr" and stores the result in "dst". - + + AVX512DQ + Miscellaneous + + + + Broadcast the lower 2 packed 32-bit integers from "a" to all elements of "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -addr := MEM[ptr] -FOR j := 0 to 4 - i := j*8 - dst[i+7:i] := addr[31-i:31-i-7] +FOR j := 0 to 15 + i := j*32 + n := (j % 2)*32 + IF k[j] + dst[i+31:i] := a[n+31:n] + ELSE + dst[i+31:i] := 0 + FI ENDFOR +dst[MAX:512] := 0 +
immintrin.h
 - - Load - - Loads a big-endian quad word (64-bit) value from address "ptr" and stores the result in "dst". - + + AVX512VL + AVX512DQ + Miscellaneous + + + Broadcast the lower 2 packed 32-bit integers from "a" to all elements of "dst. -addr := MEM[ptr] -FOR j := 0 to 8 - i := j*8 - dst[i+7:i] := addr[63-i:63-i-7] +FOR j := 0 to 3 + i := j*32 + n := (j % 2)*32 + dst[i+31:i] := a[n+31:n] ENDFOR +dst[MAX:128] := 0 +
immintrin.h
 - - Store - - - Stores word-sized (16-bit) "data" to address "ptr" in big-endian format. - + + AVX512VL + AVX512DQ + Miscellaneous + + + + + Broadcast the lower 2 packed 32-bit integers from "a" to all elements of "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -addr := MEM[ptr] -FOR j := 0 to 1 - i := j*8 - addr[i+7:i] := data[15-i:15-i-7] +FOR j := 0 to 3 + i := j*32 + n := (j % 2)*32 + IF k[j] + dst[i+31:i] := a[n+31:n] + ELSE + dst[i+31:i] := src[i+31:i] + FI ENDFOR +dst[MAX:128] := 0 +
immintrin.h
 - - Store - - - Stores double word-sized (32-bit) "data" to address "ptr" in big-endian format. - + + AVX512VL + AVX512DQ + Miscellaneous + + + + Broadcast the lower 2 packed 32-bit integers from "a" to all elements of "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -addr := MEM[ptr] -FOR j := 0 to 4 - i := j*8 - addr[i+7:i] := data[31-i:31-i-7] +FOR j := 0 to 3 + i := j*32 + n := (j % 2)*32 + IF k[j] + dst[i+31:i] := a[n+31:n] + ELSE + dst[i+31:i] := 0 + FI ENDFOR +dst[MAX:128] := 0 +
immintrin.h
 - - Store - - - Stores quad word-sized (64-bit) "data" to address "ptr" in big-endian format. - + + AVX512DQ + Miscellaneous + + + Broadcast the 8 packed 32-bit integers from "a" to all elements of "dst". -addr := MEM[ptr] -FOR j := 0 to 7 - i := j*8 - addr[i+7:i] := data[63-i:63-i-7] +FOR j := 0 to 15 + i := j*32 + n := (j % 8)*32 + dst[i+31:i] := a[n+31:n] ENDFOR +dst[MAX:512] := 0 +
immintrin.h
 - - XSAVE - XSAVEC - OS-Targeted - - - Perform a full or partial save of the enabled processor states to memory at "mem_addr"; xsavec differs from xsave in that it uses compaction and that it may use init optimization. State is saved based on bits [62:0] in "save_mask" and "XCR0". "mem_addr" must be aligned on a 64-byte boundary. + + AVX512DQ + Miscellaneous + + + + + Broadcast the 8 packed 32-bit integers from "a" to all elements of "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -mask[62:0] := save_mask[62:0] BITWISE AND XCR0[62:0] -FOR i := 0 to 62 - IF mask[i] - CASE (i) OF - 0: mem_addr.FPUSSESave_Area[FPU] := ProcessorState[x87 FPU] - 1: mem_addr.FPUSSESaveArea[SSE] := ProcessorState[SSE] - DEFAULT: mem_addr.Ext_Save_Area[i] := ProcessorState[i] - ESAC - mem_addr.HEADER.XSTATE_BV[i] := INIT_FUNCTION[i] +FOR j := 0 to 15 + i := j*32 + n := (j % 8)*32 + IF k[j] + dst[i+31:i] := a[n+31:n] + ELSE + dst[i+31:i] := src[i+31:i] FI - i := i + 1 ENDFOR +dst[MAX:512] := 0 - +
immintrin.h
 - - XSAVE - XSS - OS-Targeted - - - Perform a full or partial save of the enabled processor states to memory at "mem_addr"; xsaves differs from xsave in that it can save state components corresponding to bits set in IA32_XSS MSR and that it may use the modified optimization. State is saved based on bits [62:0] in "save_mask" and "XCR0". "mem_addr" must be aligned on a 64-byte boundary. + + AVX512DQ + Miscellaneous + + + + Broadcast the 8 packed 32-bit integers from "a" to all elements of "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -mask[62:0] := save_mask[62:0] BITWISE AND XCR0[62:0] -FOR i := 0 to 62 - IF mask[i] - CASE (i) OF - 0: mem_addr.FPUSSESave_Area[FPU] := ProcessorState[x87 FPU] - 1: mem_addr.FPUSSESaveArea[SSE] := ProcessorState[SSE] - DEFAULT: mem_addr.Ext_Save_Area[i] := ProcessorState[i] - ESAC - mem_addr.HEADER.XSTATE_BV[i] := INIT_FUNCTION[i] +FOR j := 0 to 15 + i := j*32 + n := (j % 8)*32 + IF k[j] + dst[i+31:i] := a[n+31:n] + ELSE + dst[i+31:i] := 0 FI - i := i + 1 ENDFOR +dst[MAX:512] := 0 - +
immintrin.h
 - - XSAVE - XSAVEC - OS-Targeted - - - Perform a full or partial save of the enabled processor states to memory at "mem_addr"; xsavec differs from xsave in that it uses compaction and that it may use init optimization. State is saved based on bits [62:0] in "save_mask" and "XCR0". "mem_addr" must be aligned on a 64-byte boundary. + + AVX512VL + AVX512DQ + Miscellaneous + + + Broadcast the 2 packed 64-bit integers from "a" to all elements of "dst". -mask[62:0] := save_mask[62:0] BITWISE AND XCR0[62:0] -FOR i := 0 to 62 - IF mask[i] - CASE (i) OF - 0: mem_addr.FPUSSESave_Area[FPU] := ProcessorState[x87 FPU] - 1: mem_addr.FPUSSESaveArea[SSE] := ProcessorState[SSE] - DEFAULT: mem_addr.Ext_Save_Area[i] := ProcessorState[i] - ESAC - mem_addr.HEADER.XSTATE_BV[i] := INIT_FUNCTION[i] - FI - i := i + 1 +FOR j := 0 to 3 + i := j*64 + n := (j % 2)*64 + dst[i+63:i] := a[n+63:n] ENDFOR +dst[MAX:256] := 0 - +
immintrin.h
 - - XSAVE - XSS - OS-Targeted - - - Perform a full or partial save of the enabled processor states to memory at "mem_addr"; xsaves differs from xsave in that it can save state components corresponding to bits set in IA32_XSS MSR and that it may use the modified optimization. State is saved based on bits [62:0] in "save_mask" and "XCR0". "mem_addr" must be aligned on a 64-byte boundary. + + AVX512VL + AVX512DQ + Miscellaneous + + + + + Broadcast the 2 packed 64-bit integers from "a" to all elements of "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -mask[62:0] := save_mask[62:0] BITWISE AND XCR0[62:0] -FOR i := 0 to 62 - IF mask[i] - CASE (i) OF - 0: mem_addr.FPUSSESave_Area[FPU] := ProcessorState[x87 FPU] - 1: mem_addr.FPUSSESaveArea[SSE] := ProcessorState[SSE] - DEFAULT: mem_addr.Ext_Save_Area[i] := ProcessorState[i] - ESAC - mem_addr.HEADER.XSTATE_BV[i] := INIT_FUNCTION[i] +FOR j := 0 to 3 + i := j*64 + n := (j % 2)*64 + IF k[j] + dst[i+63:i] := a[n+63:n] + ELSE + dst[i+63:i] := src[i+63:i] FI - i := i + 1 ENDFOR +dst[MAX:256] := 0 - +
immintrin.h
 - - XSAVE - XSS - OS-Targeted - - - Perform a full or partial restore of the enabled processor states using the state information stored in memory at "mem_addr". xrstors differs from xrstor in that it can restore state components corresponding to bits set in the IA32_XSS MSR; xrstors cannot restore from an xsave area in which the extended region is in the standard form. State is restored based on bits [62:0] in "rs_mask", "XCR0", and "mem_addr.HEADER.XSTATE_BV". "mem_addr" must be aligned on a 64-byte boundary. + + AVX512VL + AVX512DQ + Miscellaneous + + + + Broadcast the 2 packed 64-bit integers from "a" to all elements of "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -st_mask = mem_addr.HEADER.XSTATE_BV[62:0] -FOR i := 0 to 62 - IF (rs_mask[i] AND XCR0[i]) - IF st_mask[i] - CASE (i) OF - 0: ProcessorState[x87 FPU] := mem_addr.FPUSSESave_Area[FPU] - 1: ProcessorState[SSE] := mem_addr.FPUSSESaveArea[SSE] - DEFAULT: ProcessorState[i] := mem_addr.Ext_Save_Area[i] - ESAC - ELSE - // ProcessorExtendedState := Processor Supplied Values - CASE (i) OF - 1: MXCSR := mem_addr.FPUSSESave_Area[SSE] - ESAC - FI +FOR j := 0 to 3 + i := j*64 + n := (j % 2)*64 + IF k[j] + dst[i+63:i] := a[n+63:n] + ELSE + dst[i+63:i] := 0 FI - i := i + 1 ENDFOR +dst[MAX:256] := 0 - +
immintrin.h
 - - XSAVE - XSS - OS-Targeted - - - Perform a full or partial restore of the enabled processor states using the state information stored in memory at "mem_addr". xrstors differs from xrstor in that it can restore state components corresponding to bits set in the IA32_XSS MSR; xrstors cannot restore from an xsave area in which the extended region is in the standard form. State is restored based on bits [62:0] in "rs_mask", "XCR0", and "mem_addr.HEADER.XSTATE_BV". "mem_addr" must be aligned on a 64-byte boundary. + + AVX512DQ + Miscellaneous + + + Broadcast the 2 packed 64-bit integers from "a" to all elements of "dst". -st_mask = mem_addr.HEADER.XSTATE_BV[62:0] -FOR i := 0 to 62 - IF (rs_mask[i] AND XCR0[i]) - IF st_mask[i] - CASE (i) OF - 0: ProcessorState[x87 FPU] := mem_addr.FPUSSESave_Area[FPU] - 1: ProcessorState[SSE] := mem_addr.FPUSSESaveArea[SSE] - DEFAULT: ProcessorState[i] := mem_addr.Ext_Save_Area[i] - ESAC - ELSE - // ProcessorExtendedState := Processor Supplied Values - CASE (i) OF - 1: MXCSR := mem_addr.FPUSSESave_Area[SSE] - ESAC - FI +FOR j := 0 to 7 + i := j*64 + n := (j % 2)*64 + dst[i+63:i] := a[n+63:n] +ENDFOR +dst[MAX:512] := 0 + + +
immintrin.h
+ + + AVX512DQ + Miscellaneous + + + + + Broadcast the 2 packed 64-bit integers from "a" to all elements of "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +FOR j := 0 to 7 + i := j*64 + n := (j % 2)*64 + IF k[j] + dst[i+63:i] := a[n+63:n] + ELSE + dst[i+63:i] := src[i+63:i] FI - i := i + 1 ENDFOR +dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point - AVX - Convert - - Copy the lower single-precision (32-bit) floating-point element of "a" to "dst". - dst[31:0] := a[31:0] - + + AVX512DQ + Miscellaneous + + + + Broadcast the 2 packed 64-bit integers from "a" to all elements of "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + +FOR j := 0 to 7 + i := j*64 + n := (j % 2)*64 + IF k[j] + dst[i+63:i] := a[n+63:n] + ELSE + dst[i+63:i] := 0 + FI +ENDFOR +dst[MAX:512] := 0 + +
immintrin.h
 - + Floating Point - AVX2 + Integer + AVX512VL + AVX512DQ Convert - - Copy the lower double-precision (64-bit) floating-point element of "a" to "dst". - dst[63:0] := a[63:0] - + + + Convert packed double-precision (64-bit) floating-point elements in "a" to packed 64-bit integers, and store the results in "dst". + +FOR j := 0 to 3 + i := j*64 + dst[i+63:i] := Convert_FP64_To_Int64(a[i+63:i]) +ENDFOR +dst[MAX:256] := 0 + +
immintrin.h
 - + + Floating Point Integer - AVX2 + AVX512VL + AVX512DQ Convert - - Copy the lower 32-bit integer in "a" to "dst". + + + + + Convert packed double-precision (64-bit) floating-point elements in "a" to packed 64-bit integers, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -dst[31:0] := a[31:0] +FOR j := 0 to 3 + i := j*64 + IF k[j] + dst[i+63:i] := Convert_FP64_To_Int64(a[i+63:i]) + ELSE + dst[i+63:i] := src[i+63:i] + FI +ENDFOR +dst[MAX:256] := 0 - +
immintrin.h
 - - - - Mask - AVX512F - Mask - - - Compute the bitwise NOT of 16-bit masks "a" and then AND with "b", and store the result in "k". - -k[15:0] := (NOT a[15:0]) AND b[15:0] -k[MAX:16] := 0 - - -
immintrin.h
- - - Mask - KNCNI - Mask - - - Compute the bitwise NOT of 16-bit masks "a" and then AND with "b", and store the result in "k". - -k[15:0] := (NOT a[15:0]) AND b[15:0] -k[MAX:16] := 0 - - -
immintrin.h
- - - Mask - AVX512F - Mask - - - Compute the bitwise AND of 16-bit masks "a" and "b", and store the result in "k". + + Floating Point + Integer + AVX512VL + AVX512DQ + Convert + + + + Convert packed double-precision (64-bit) floating-point elements in "a" to packed 64-bit integers, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -k[15:0] := a[15:0] AND b[15:0] -k[MAX:16] := 0 +FOR j := 0 to 3 + i := j*64 + IF k[j] + dst[i+63:i] := Convert_FP64_To_Int64(a[i+63:i]) + ELSE + dst[i+63:i] := 0 + FI +ENDFOR +dst[MAX:256] := 0 - +
immintrin.h
 - - Mask - KNCNI - Mask - - - Compute the bitwise AND of 16-bit masks "a" and "b", and store the result in "k". + + Floating Point + Integer + AVX512DQ + Convert + + + + Convert packed double-precision (64-bit) floating-point elements in "a" to packed 64-bit integers, and store the results in "dst". + [round_note] -k[15:0] := a[15:0] AND b[15:0] -k[MAX:16] := 0 +FOR j := 0 to 7 + i := j*64 + dst[i+63:i] := Convert_FP64_To_Int64(a[i+63:i]) +ENDFOR +dst[MAX:512] := 0 - +
immintrin.h
 - - Mask - AVX512F - Mask - - Copy 16-bit mask "a" to "k". + + Floating Point + Integer + AVX512DQ + Convert + + + Convert packed double-precision (64-bit) floating-point elements in "a" to packed 64-bit integers, and store the results in "dst". -k[15:0] := a[15:0] -k[MAX:16] := 0 +FOR j := 0 to 7 + i := j*64 + dst[i+63:i] := Convert_FP64_To_Int64(a[i+63:i]) +ENDFOR +dst[MAX:512] := 0 - +
immintrin.h
 - - Mask - KNCNI - Mask - - Copy 16-bit mask "a" to "k". + + Floating Point + Integer + AVX512DQ + Convert + + + + + + Convert packed double-precision (64-bit) floating-point elements in "a" to packed 64-bit integers, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + [round_note] -k[15:0] := a[15:0] -k[MAX:16] := 0 +FOR j := 0 to 7 + i := j*64 + IF k[j] + dst[i+63:i] := Convert_FP64_To_Int64(a[i+63:i]) + ELSE + dst[i+63:i] := src[i+63:i] + FI +ENDFOR +dst[MAX:512] := 0 - +
immintrin.h
 - - Mask - AVX512F - Mask - - Compute the bitwise NOT of 16-bit mask "a", and store the result in "k". + + Floating Point + Integer + AVX512DQ + Convert + + + + + Convert packed double-precision (64-bit) floating-point elements in "a" to packed 64-bit integers, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -k[15:0] := NOT a[15:0] -k[MAX:16] := 0 +FOR j := 0 to 7 + i := j*64 + IF k[j] + dst[i+63:i] := Convert_FP64_To_Int64(a[i+63:i]) + ELSE + dst[i+63:i] := src[i+63:i] + FI +ENDFOR +dst[MAX:512] := 0 - +
immintrin.h
 - - Mask - KNCNI - Mask - - Compute the bitwise NOT of 16-bit mask "a", and store the result in "k". + + Floating Point + Integer + AVX512DQ + Convert + + + + + Convert packed double-precision (64-bit) floating-point elements in "a" to packed 64-bit integers, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + [round_note] -k[15:0] := NOT a[15:0] -k[MAX:16] := 0 +FOR j := 0 to 7 + i := j*64 + IF k[j] + dst[i+63:i] := Convert_FP64_To_Int64(a[i+63:i]) + ELSE + dst[i+63:i] := 0 + FI +ENDFOR +dst[MAX:512] := 0 - +
immintrin.h
 - - Mask - AVX512F - Mask - - - Compute the bitwise OR of 16-bit masks "a" and "b", and store the result in "k". + + Floating Point + Integer + AVX512DQ + Convert + + + + Convert packed double-precision (64-bit) floating-point elements in "a" to packed 64-bit integers, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -k[15:0] := a[15:0] OR b[15:0] -k[MAX:16] := 0 +FOR j := 0 to 7 + i := j*64 + IF k[j] + dst[i+63:i] := Convert_FP64_To_Int64(a[i+63:i]) + ELSE + dst[i+63:i] := 0 + FI +ENDFOR +dst[MAX:512] := 0 - +
immintrin.h
 - - Mask - KNCNI - Mask - - - Compute the bitwise OR of 16-bit masks "a" and "b", and store the result in "k". + + Floating Point + Integer + AVX512VL + AVX512DQ + Convert + + + Convert packed double-precision (64-bit) floating-point elements in "a" to packed 64-bit integers, and store the results in "dst". -k[15:0] := a[15:0] OR b[15:0] -k[MAX:16] := 0 +FOR j := 0 to 1 + i := j*64 + dst[i+63:i] := Convert_FP64_To_Int64(a[i+63:i]) +ENDFOR +dst[MAX:128] := 0 - +
immintrin.h
 - - Mask - AVX512F - Mask - - - Unpack and interleave 8 bits from masks "a" and "b", and store the 16-bit result in "k". + + Floating Point + Integer + AVX512VL + AVX512DQ + Convert + + + + + Convert packed double-precision (64-bit) floating-point elements in "a" to packed 64-bit integers, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -k[7:0] := b[7:0] -k[15:8] := a[7:0] -k[MAX:16] := 0 +FOR j := 0 to 1 + i := j*64 + IF k[j] + dst[i+63:i] := Convert_FP64_To_Int64(a[i+63:i]) + ELSE + dst[i+63:i] := src[i+63:i] + FI +ENDFOR +dst[MAX:128] := 0 - +
immintrin.h
 - - Mask - AVX512F - Mask - - - Compute the bitwise XNOR of 16-bit masks "a" and "b", and store the result in "k". + + Floating Point + Integer + AVX512VL + AVX512DQ + Convert + + + + Convert packed double-precision (64-bit) floating-point elements in "a" to packed 64-bit integers, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -k[15:0] := NOT (a[15:0] XOR b[15:0]) -k[MAX:16] := 0 +FOR j := 0 to 1 + i := j*64 + IF k[j] + dst[i+63:i] := Convert_FP64_To_Int64(a[i+63:i]) + ELSE + dst[i+63:i] := 0 + FI +ENDFOR +dst[MAX:128] := 0 - +
immintrin.h
 - - Mask - KNCNI - Mask - - - Compute the bitwise XNOR of 16-bit masks "a" and "b", and store the result in "k". + + Floating Point + Integer + AVX512VL + AVX512DQ + Convert + + + Convert packed double-precision (64-bit) floating-point elements in "a" to packed unsigned 64-bit integers, and store the results in "dst". -k[15:0] := NOT (a[15:0] XOR b[15:0]) -k[MAX:16] := 0 +FOR j := 0 to 3 + i := j*64 + dst[i+63:i] := Convert_FP64_To_UInt64(a[i+63:i]) +ENDFOR +dst[MAX:256] := 0 - +
immintrin.h
 - - Mask - AVX512F - Mask - - - Compute the bitwise XOR of 16-bit masks "a" and "b", and store the result in "k". + + Floating Point + Integer + AVX512VL + AVX512DQ + Convert + + + + + Convert packed double-precision (64-bit) floating-point elements in "a" to packed unsigned 64-bit integers, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -k[15:0] := a[15:0] XOR b[15:0] -k[MAX:16] := 0 +FOR j := 0 to 3 + i := j*64 + IF k[j] + dst[i+63:i] := Convert_FP64_To_UInt64(a[i+63:i]) + ELSE + dst[i+63:i] := src[i+63:i] + FI +ENDFOR +dst[MAX:256] := 0 - +
immintrin.h
 - - Mask - KNCNI - Mask - - - Compute the bitwise XOR of 16-bit masks "a" and "b", and store the result in "k". + + Floating Point + Integer + AVX512VL + AVX512DQ + Convert + + + + Convert packed double-precision (64-bit) floating-point elements in "a" to packed unsigned 64-bit integers, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -k[15:0] := a[15:0] XOR b[15:0] -k[MAX:16] := 0 +FOR j := 0 to 3 + i := j*64 + IF k[j] + dst[i+63:i] := Convert_FP64_To_UInt64(a[i+63:i]) + ELSE + dst[i+63:i] := 0 + FI +ENDFOR +dst[MAX:256] := 0 - +
immintrin.h
 - + Floating Point - AVX512F/KNCNI - Arithmetic - - - Add packed double-precision (64-bit) floating-point elements in "a" and "b", and store the results in "dst". + Integer + AVX512DQ + Convert + + + + Convert packed double-precision (64-bit) floating-point elements in "a" to packed unsigned 64-bit integers, and store the results in "dst". + [round_note] FOR j := 0 to 7 i := j*64 - dst[i+63:i] := a[i+63:i] + b[i+63:i] + dst[i+63:i] := Convert_FP64_To_UInt64(a[i+63:i]) ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - AVX512F/KNCNI - Arithmetic - - - - Add packed double-precision (64-bit) floating-point elements in "a" and "b", and store the results in "dst". - [round_note] - + Integer + AVX512DQ + Convert + + + Convert packed double-precision (64-bit) floating-point elements in "a" to packed unsigned 64-bit integers, and store the results in "dst". FOR j := 0 to 7 i := j*64 - dst[i+63:i] := a[i+63:i] + b[i+63:i] + dst[i+63:i] := Convert_FP64_To_UInt64(a[i+63:i]) ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - AVX512F/KNCNI - Arithmetic - - - - - Add packed double-precision (64-bit) floating-point elements in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Integer + AVX512DQ + Convert + + + + + + Convert packed double-precision (64-bit) floating-point elements in "a" to packed unsigned 64-bit integers, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + [round_note] FOR j := 0 to 7 i := j*64 IF k[j] - dst[i+63:i] := a[i+63:i] + b[i+63:i] + dst[i+63:i] := Convert_FP64_To_UInt64(a[i+63:i]) ELSE dst[i+63:i] := src[i+63:i] FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - AVX512F/KNCNI - Arithmetic - - - - - - Add packed double-precision (64-bit) floating-point elements in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - [round_note] - + Integer + AVX512DQ + Convert + + + + + Convert packed double-precision (64-bit) floating-point elements in "a" to packed unsigned 64-bit integers, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). FOR j := 0 to 7 i := j*64 IF k[j] - dst[i+63:i] := a[i+63:i] + b[i+63:i] + dst[i+63:i] := Convert_FP64_To_UInt64(a[i+63:i]) ELSE dst[i+63:i] := src[i+63:i] FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - AVX512F - Arithmetic - - - - Add packed double-precision (64-bit) floating-point elements in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Integer + AVX512DQ + Convert + + + + + Convert packed double-precision (64-bit) floating-point elements in "a" to packed unsigned 64-bit integers, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + [round_note] FOR j := 0 to 7 i := j*64 IF k[j] - dst[i+63:i] := a[i+63:i] + b[i+63:i] + dst[i+63:i] := Convert_FP64_To_UInt64(a[i+63:i]) ELSE dst[i+63:i] := 0 FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - AVX512F - Arithmetic - - - - - Add packed double-precision (64-bit) floating-point elements in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - [round_note] - + Integer + AVX512DQ + Convert + + + + Convert packed double-precision (64-bit) floating-point elements in "a" to packed unsigned 64-bit integers, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). FOR j := 0 to 7 i := j*64 IF k[j] - dst[i+63:i] := a[i+63:i] + b[i+63:i] + dst[i+63:i] := Convert_FP64_To_UInt64(a[i+63:i]) ELSE dst[i+63:i] := 0 FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - AVX512F/KNCNI - Arithmetic - - - Add packed single-precision (32-bit) floating-point elements in "a" and "b", and store the results in "dst". + Integer + AVX512VL + AVX512DQ + Convert + + + Convert packed double-precision (64-bit) floating-point elements in "a" to packed unsigned 64-bit integers, and store the results in "dst". -FOR j := 0 to 15 - i := j*32 - dst[i+31:i] := a[i+31:i] + b[i+31:i] +FOR j := 0 to 1 + i := j*64 + dst[i+63:i] := Convert_FP64_To_UInt64(a[i+63:i]) ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point - AVX512F/KNCNI - Arithmetic - - - - Add packed single-precision (32-bit) floating-point elements in "a" and "b", and store the results in "dst". - [round_note] - + Integer + AVX512VL + AVX512DQ + Convert + + + + + Convert packed double-precision (64-bit) floating-point elements in "a" to packed unsigned 64-bit integers, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*32 - dst[i+31:i] := a[i+31:i] + b[i+31:i] +FOR j := 0 to 1 + i := j*64 + IF k[j] + dst[i+63:i] := Convert_FP64_To_UInt64(a[i+63:i]) + ELSE + dst[i+63:i] := src[i+63:i] + FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point - AVX512F/KNCNI - Arithmetic - - - - - Add packed single-precision (32-bit) floating-point elements in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Integer + AVX512VL + AVX512DQ + Convert + + + + Convert packed double-precision (64-bit) floating-point elements in "a" to packed unsigned 64-bit integers, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*32 +FOR j := 0 to 1 + i := j*64 IF k[j] - dst[i+31:i] := a[i+31:i] + b[i+31:i] + dst[i+63:i] := Convert_FP64_To_UInt64(a[i+63:i]) ELSE - dst[i+31:i] := src[i+31:i] + dst[i+63:i] := 0 FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point - AVX512F/KNCNI - Arithmetic - - - - - - Add packed single-precision (32-bit) floating-point elements in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - [round_note] - + Integer + AVX512VL + AVX512DQ + Convert + + + Convert packed single-precision (32-bit) floating-point elements in "a" to packed 64-bit integers, and store the results in "dst". -FOR j := 0 to 15 - i := j*32 - IF k[j] - dst[i+31:i] := a[i+31:i] + b[i+31:i] - ELSE - dst[i+31:i] := src[i+31:i] - FI +FOR j := 0 to 3 + i := j*64 + l := j*32 + dst[i+63:i] := Convert_FP32_To_Int64(a[l+31:l]) ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Floating Point - AVX512F - Arithmetic - - - - Add packed single-precision (32-bit) floating-point elements in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Integer + AVX512VL + AVX512DQ + Convert + + + + + Convert packed single-precision (32-bit) floating-point elements in "a" to packed 64-bit integers, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*32 +FOR j := 0 to 3 + i := j*64 + l := j*32 IF k[j] - dst[i+31:i] := a[i+31:i] + b[i+31:i] + dst[i+63:i] := Convert_FP32_To_Int64(a[l+31:l]) ELSE - dst[i+31:i] := 0 + dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Floating Point - AVX512F - Arithmetic - - - - - Add packed single-precision (32-bit) floating-point elements in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - [round_note] - + Integer + AVX512VL + AVX512DQ + Convert + + + + Convert packed single-precision (32-bit) floating-point elements in "a" to packed 64-bit integers, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*32 +FOR j := 0 to 3 + i := j*64 + l := j*32 IF k[j] - dst[i+31:i] := a[i+31:i] + b[i+31:i] + dst[i+63:i] := Convert_FP32_To_Int64(a[l+31:l]) ELSE - dst[i+31:i] := 0 + dst[i+63:i] := 0 FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Floating Point - AVX512F - Arithmetic - - - - Add the lower double-precision (64-bit) floating-point element in "a" and "b", store the result in the lower element of "dst", and copy the upper element from "a" to the upper element of "dst". - [round_note] - + Integer + AVX512DQ + Convert + + + + Convert packed single-precision (32-bit) floating-point elements in "a" to packed 64-bit integers, and store the results in "dst". + [round_note] -dst[63:0] := a[63:0] + b[63:0] -dst[127:64] := a[127:64] -dst[MAX:128] := 0 +FOR j := 0 to 7 + i := j*64 + l := j*32 + dst[i+63:i] := Convert_FP32_To_Int64(a[l+31:l]) +ENDFOR +dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - AVX512F - Arithmetic - - - - - - Add the lower double-precision (64-bit) floating-point element in "a" and "b", store the result in the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". - [round_note] - + Integer + AVX512DQ + Convert + + + Convert packed single-precision (32-bit) floating-point elements in "a" to packed 64-bit integers, and store the results in "dst". -IF k[0] - dst[63:0] := a[63:0] + b[63:0] -ELSE - dst[63:0] := src[63:0] -FI -dst[127:64] := a[127:64] -dst[MAX:128] := 0 +FOR j := 0 to 7 + i := j*64 + l := j*32 + dst[i+63:i] := Convert_FP32_To_Int64(a[l+31:l]) +ENDFOR +dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - AVX512F - Arithmetic - - - - - Add the lower double-precision (64-bit) floating-point element in "a" and "b", store the result in the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". + Integer + AVX512DQ + Convert + + + + + + Convert packed single-precision (32-bit) floating-point elements in "a" to packed 64-bit integers, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + [round_note] -IF k[0] - dst[63:0] := a[63:0] + b[63:0] -ELSE - dst[63:0] := src[63:0] -FI -dst[127:64] := a[127:64] -dst[MAX:128] := 0 +FOR j := 0 to 7 + i := j*64 + l := j*32 + IF k[j] + dst[i+63:i] := Convert_FP32_To_Int64(a[l+31:l]) + ELSE + dst[i+63:i] := src[i+63:i] + FI +ENDFOR +dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - AVX512F - Arithmetic - - - - - Add the lower double-precision (64-bit) floating-point element in "a" and "b", store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". - [round_note] - + Integer + AVX512DQ + Convert + + + + + Convert packed single-precision (32-bit) floating-point elements in "a" to packed 64-bit integers, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -IF k[0] - dst[63:0] := a[63:0] + b[63:0] -ELSE - dst[63:0] := 0 -FI -dst[127:64] := a[127:64] -dst[MAX:128] := 0 +FOR j := 0 to 7 + i := j*64 + l := j*32 + IF k[j] + dst[i+63:i] := Convert_FP32_To_Int64(a[l+31:l]) + ELSE + dst[i+63:i] := src[i+63:i] + FI +ENDFOR +dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - AVX512F - Arithmetic - - - - Add the lower double-precision (64-bit) floating-point element in "a" and "b", store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". + Integer + AVX512DQ + Convert + + + + + Convert packed single-precision (32-bit) floating-point elements in "a" to packed 64-bit integers, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + [round_note] -IF k[0] - dst[63:0] := a[63:0] + b[63:0] -ELSE - dst[63:0] := 0 -FI -dst[127:64] := a[127:64] -dst[MAX:128] := 0 +FOR j := 0 to 7 + i := j*64 + l := j*32 + IF k[j] + dst[i+63:i] := Convert_FP32_To_Int64(a[l+31:l]) + ELSE + dst[i+63:i] := 0 + FI +ENDFOR +dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - AVX512F - Arithmetic - - - - Add the lower single-precision (32-bit) floating-point element in "a" and "b", store the result in the lower element of "dst", and copy the upper 3 packed elements from "a" to the upper elements of "dst". - [round_note] - + Integer + AVX512DQ + Convert + + + + Convert packed single-precision (32-bit) floating-point elements in "a" to packed 64-bit integers, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -dst[31:0] := a[31:0] + b[31:0] -dst[127:32] := a[127:32] -dst[MAX:128] := 0 +FOR j := 0 to 7 + i := j*64 + l := j*32 + IF k[j] + dst[i+63:i] := Convert_FP32_To_Int64(a[l+31:l]) + ELSE + dst[i+63:i] := 0 + FI +ENDFOR +dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - AVX512F - Arithmetic - - - - - - Add the lower single-precision (32-bit) floating-point element in "a" and "b", store the result in the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". - [round_note] - + Integer + AVX512VL + AVX512DQ + Convert + + + Convert packed single-precision (32-bit) floating-point elements in "a" to packed 64-bit integers, and store the results in "dst". -IF k[0] - dst[31:0] := a[31:0] + b[31:0] -ELSE - dst[31:0] := src[31:0] -FI -dst[127:32] := a[127:32] +FOR j := 0 to 1 + i := j*64 + l := j*32 + dst[i+63:i] := Convert_FP32_To_Int64(a[l+31:l]) +ENDFOR dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point - AVX512F - Arithmetic - - - - - Add the lower single-precision (32-bit) floating-point element in "a" and "b", store the result in the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". + Integer + AVX512VL + AVX512DQ + Convert + + + + + Convert packed single-precision (32-bit) floating-point elements in "a" to packed 64-bit integers, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -IF k[0] - dst[31:0] := a[31:0] + b[31:0] -ELSE - dst[31:0] := src[31:0] -FI -dst[127:32] := a[127:32] +FOR j := 0 to 1 + i := j*64 + l := j*32 + IF k[j] + dst[i+63:i] := Convert_FP32_To_Int64(a[l+31:l]) + ELSE + dst[i+63:i] := src[i+63:i] + FI +ENDFOR dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point - AVX512F - Arithmetic - - - - - Add the lower single-precision (32-bit) floating-point element in "a" and "b", store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". - [round_note] - + Integer + AVX512VL + AVX512DQ + Convert + + + + Convert packed single-precision (32-bit) floating-point elements in "a" to packed 64-bit integers, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -IF k[0] - dst[31:0] := a[31:0] + b[31:0] -ELSE - dst[31:0] := 0 -FI -dst[127:32] := a[127:32] +FOR j := 0 to 1 + i := j*64 + l := j*32 + IF k[j] + dst[i+63:i] := Convert_FP32_To_Int64(a[l+31:l]) + ELSE + dst[i+63:i] := 0 + FI +ENDFOR dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point - AVX512F - Arithmetic - - - - Add the lower single-precision (32-bit) floating-point element in "a" and "b", store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". + Integer + AVX512VL + AVX512DQ + Convert + + + Convert packed single-precision (32-bit) floating-point elements in "a" to packed unsigned 64-bit integers, and store the results in "dst". -IF k[0] - dst[31:0] := a[31:0] + b[31:0] -ELSE - dst[31:0] := 0 -FI -dst[127:32] := a[127:32] -dst[MAX:128] := 0 +FOR j := 0 to 3 + i := j*64 + l := j*32 + dst[i+63:i] := Convert_FP32_To_UInt64(a[l+31:l]) +ENDFOR +dst[MAX:256] := 0 - +
immintrin.h
 - + + Floating Point Integer - AVX512F/KNCNI - Miscellaneous - - - - Concatenate "a" and "b" into a 128-byte immediate result, shift the result right by "count" 32-bit elements, and store the low 64 bytes (16 elements) in "dst". + AVX512VL + AVX512DQ + Convert + + + + + Convert packed single-precision (32-bit) floating-point elements in "a" to packed unsigned 64-bit integers, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -temp[1023:512] := a[511:0] -temp[511:0] := b[511:0] -temp[1023:0] := temp[1023:0] >> (32*count) -dst[511:0] := temp[511:0] -dst[MAX:512] := 0 +FOR j := 0 to 3 + i := j*64 + l := j*32 + IF k[j] + dst[i+63:i] := Convert_FP32_To_UInt64(a[l+31:l]) + ELSE + dst[i+63:i] := src[i+63:i] + FI +ENDFOR +dst[MAX:256] := 0 - +
immintrin.h
 - + + Floating Point Integer - AVX512F/KNCNI - Miscellaneous - - - - - - Concatenate "a" and "b" into a 128-byte immediate result, shift the result right by "count" 32-bit elements, and store the low 64 bytes (16 elements) in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + AVX512VL + AVX512DQ + Convert + + + + Convert packed single-precision (32-bit) floating-point elements in "a" to packed unsigned 64-bit integers, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -temp[1023:512] := a[511:0] -temp[511:0] := b[511:0] -temp[1023:0] := temp[1023:0] >> (32*count) -FOR j := 0 to 15 - i := j*32 +FOR j := 0 to 3 + i := j*64 + l := j*32 IF k[j] - dst[i+31:i] := temp[i+31:i] + dst[i+63:i] := Convert_FP32_To_UInt64(a[l+31:l]) ELSE - dst[i+31:i] := src[i+31:i] + dst[i+63:i] := 0 FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + + Floating Point Integer - AVX512F - Miscellaneous - - - - - Concatenate "a" and "b" into a 128-byte immediate result, shift the result right by "count" 32-bit elements, and stores the low 64 bytes (16 elements) in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + AVX512DQ + Convert + + + + Convert packed single-precision (32-bit) floating-point elements in "a" to packed unsigned 64-bit integers, and store the results in "dst". + [round_note] -temp[1023:512] := a[511:0] -temp[511:0] := b[511:0] -temp[1023:0] := temp[1023:0] >> (32*count) -FOR j := 0 to 15 - i := j*32 - IF k[j] - dst[i+31:i] := temp[i+31:i] - ELSE - dst[i+31:i] := 0 - FI +FOR j := 0 to 7 + i := j*64 + l := j*32 + dst[i+63:i] := Convert_FP32_To_UInt64(a[l+31:l]) ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + + Floating Point Integer - AVX512F - Miscellaneous - - - - Concatenate "a" and "b" into a 128-byte immediate result, shift the result right by "count" 64-bit elements, and store the low 64 bytes (8 elements) in "dst". + AVX512DQ + Convert + + + Convert packed single-precision (32-bit) floating-point elements in "a" to packed unsigned 64-bit integers, and store the results in "dst". -temp[1023:512] := a[511:0] -temp[511:0] := b[511:0] -temp[1023:0] := temp[1023:0] >> (64*count) -dst[511:0] := temp[511:0] +FOR j := 0 to 7 + i := j*64 + l := j*32 + dst[i+63:i] := Convert_FP32_To_UInt64(a[l+31:l]) +ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + + Floating Point Integer - AVX512F - Miscellaneous - - - - - - Concatenate "a" and "b" into a 128-byte immediate result, shift the result right by "count" 64-bit elements, and store the low 64 bytes (8 elements) in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + AVX512DQ + Convert + + + + + + Convert packed single-precision (32-bit) floating-point elements in "a" to packed unsigned 64-bit integers, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + [round_note] -temp[1023:512] := a[511:0] -temp[511:0] := b[511:0] -temp[1023:0] := temp[1023:0] >> (64*count) FOR j := 0 to 7 i := j*64 + l := j*32 IF k[j] - dst[i+63:i] := temp[i+63:i] + dst[i+63:i] := Convert_FP32_To_UInt64(a[l+31:l]) ELSE dst[i+63:i] := src[i+63:i] FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + + Floating Point Integer - AVX512F - Miscellaneous - - - - - Concatenate "a" and "b" into a 128-byte immediate result, shift the result right by "count" 64-bit elements, and stores the low 64 bytes (8 elements) in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + AVX512DQ + Convert + + + + + Convert packed single-precision (32-bit) floating-point elements in "a" to packed unsigned 64-bit integers, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -temp[1023:512] := a[511:0] -temp[511:0] := b[511:0] -temp[1023:0] := temp[1023:0] >> (64*count) FOR j := 0 to 7 i := j*64 + l := j*32 IF k[j] - dst[i+63:i] := temp[i+63:i] + dst[i+63:i] := Convert_FP32_To_UInt64(a[l+31:l]) ELSE - dst[i+63:i] := 0 + dst[i+63:i] := src[i+63:i] FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - AVX512F/KNCNI - Swizzle - - - - Blend packed double-precision (64-bit) floating-point elements from "a" and "b" using control mask "k", and store the results in "dst". + Integer + AVX512DQ + Convert + + + + + Convert packed single-precision (32-bit) floating-point elements in "a" to packed unsigned 64-bit integers, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + [round_note] FOR j := 0 to 7 i := j*64 + l := j*32 IF k[j] - dst[i+63:i] := b[i+63:i] + dst[i+63:i] := Convert_FP32_To_UInt64(a[l+31:l]) ELSE - dst[i+63:i] := a[i+63:i] + dst[i+63:i] := 0 FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - AVX512F/KNCNI - Swizzle - - - - Blend packed single-precision (32-bit) floating-point elements from "a" and "b" using control mask "k", and store the results in "dst". + Integer + AVX512DQ + Convert + + + + Convert packed single-precision (32-bit) floating-point elements in "a" to packed unsigned 64-bit integers, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*32 +FOR j := 0 to 7 + i := j*64 + l := j*32 IF k[j] - dst[i+31:i] := b[i+31:i] + dst[i+63:i] := Convert_FP32_To_UInt64(a[l+31:l]) ELSE - dst[i+31:i] := a[i+31:i] + dst[i+63:i] := 0 FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - AVX512F - Swizzle - - Broadcast the 4 packed single-precision (32-bit) floating-point elements from "a" to all elements of "dst". + Integer + AVX512VL + AVX512DQ + Convert + + + Convert packed single-precision (32-bit) floating-point elements in "a" to packed unsigned 64-bit integers, and store the results in "dst". -FOR j := 0 to 15 - i := j*32 - n := (j mod 4)*32 - dst[i+31:i] := a[n+31:n] +FOR j := 0 to 1 + i := j*64 + l := j*32 + dst[i+63:i] := Convert_FP32_To_UInt64(a[l+31:l]) ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point - AVX512F - Swizzle - - - - Broadcast the 4 packed single-precision (32-bit) floating-point elements from "a" to all elements of "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Integer + AVX512VL + AVX512DQ + Convert + + + + + Convert packed single-precision (32-bit) floating-point elements in "a" to packed unsigned 64-bit integers, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*32 - n := (j mod 4)*32 +FOR j := 0 to 1 + i := j*64 + l := j*32 IF k[j] - dst[i+31:i] := a[n+31:n] + dst[i+63:i] := Convert_FP32_To_UInt64(a[l+31:l]) ELSE - dst[i+31:i] := src[i+31:i] + dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point - AVX512F - Swizzle - - - Broadcast the 4 packed single-precision (32-bit) floating-point elements from "a" to all elements of "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Integer + AVX512VL + AVX512DQ + Convert + + + + Convert packed single-precision (32-bit) floating-point elements in "a" to packed unsigned 64-bit integers, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*32 - n := (j mod 4)*32 +FOR j := 0 to 1 + i := j*64 + l := j*32 IF k[j] - dst[i+31:i] := a[n+31:n] + dst[i+63:i] := Convert_FP32_To_UInt64(a[l+31:l]) ELSE - dst[i+31:i] := 0 + dst[i+63:i] := 0 FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point - AVX512F - Swizzle - - Broadcast the 4 packed double-precision (64-bit) floating-point elements from "a" to all elements of "dst". + Integer + AVX512VL + AVX512DQ + Convert + + + Convert packed signed 64-bit integers in "a" to packed double-precision (64-bit) floating-point elements, and store the results in "dst". -FOR j := 0 to 7 +FOR j := 0 to 3 i := j*64 - n := (j mod 4)*64 - dst[i+63:i] := a[n+63:n] + dst[i+63:i] := Convert_Int64_To_FP64(a[i+63:i]) ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Floating Point - AVX512F - Swizzle - - - - Broadcast the 4 packed double-precision (64-bit) floating-point elements from "a" to all elements of "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Integer + AVX512VL + AVX512DQ + Convert + + + + + Convert packed signed 64-bit integers in "a" to packed double-precision (64-bit) floating-point elements, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 7 +FOR j := 0 to 3 i := j*64 - n := (j mod 4)*64 IF k[j] - dst[i+63:i] := a[n+63:n] + dst[i+63:i] := Convert_Int64_To_FP64(a[i+63:i]) ELSE - dst[i+63:i] := src[n+63:n] + dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Floating Point - AVX512F - Swizzle - - - Broadcast the 4 packed double-precision (64-bit) floating-point elements from "a" to all elements of "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Integer + AVX512VL + AVX512DQ + Convert + + + + Convert packed signed 64-bit integers in "a" to packed double-precision (64-bit) floating-point elements, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 +FOR j := 0 to 3 i := j*64 - n := (j mod 4)*64 IF k[j] - dst[i+63:i] := a[n+63:n] + dst[i+63:i] := Convert_Int64_To_FP64(a[i+63:i]) ELSE dst[i+63:i] := 0 FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + + Floating Point Integer - AVX512F - Swizzle - - Broadcast the 4 packed 32-bit integers from "a" to all elements of "dst". + AVX512DQ + Convert + + + + Convert packed signed 64-bit integers in "a" to packed double-precision (64-bit) floating-point elements, and store the results in "dst". + [round_note] -FOR j := 0 to 15 - i := j*32 - n := (j mod 4)*32 - dst[i+31:i] := a[n+31:n] +FOR j := 0 to 7 + i := j*64 + dst[i+63:i] := Convert_Int64_To_FP64(a[i+63:i]) ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + + Floating Point Integer - AVX512F - Swizzle - - - - Broadcast the 4 packed 32-bit integers from "a" to all elements of "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + AVX512DQ + Convert + + + Convert packed signed 64-bit integers in "a" to packed double-precision (64-bit) floating-point elements, and store the results in "dst". -FOR j := 0 to 15 - i := j*32 - n := (j mod 4)*32 - IF k[j] - dst[i+31:i] := a[n+31:n] - ELSE - dst[i+31:i] := src[n+31:n] - FI +FOR j := 0 to 7 + i := j*64 + dst[i+63:i] := Convert_Int64_To_FP64(a[i+63:i]) ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + + Floating Point Integer - AVX512F - Swizzle - - - Broadcast the 4 packed 32-bit integers from "a" to all elements of "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + AVX512DQ + Convert + + + + + + Convert packed signed 64-bit integers in "a" to packed double-precision (64-bit) floating-point elements, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + [round_note] -FOR j := 0 to 15 - i := j*32 - n := (j mod 4)*32 +FOR j := 0 to 7 + i := j*64 IF k[j] - dst[i+31:i] := a[n+31:n] + dst[i+63:i] := Convert_Int64_To_FP64(a[i+63:i]) ELSE - dst[i+31:i] := 0 + dst[i+63:i] := src[i+63:i] FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + + Floating Point Integer - AVX512F - Swizzle - - Broadcast the 4 packed 64-bit integers from "a" to all elements of "dst". + AVX512DQ + Convert + + + + + Convert packed signed 64-bit integers in "a" to packed double-precision (64-bit) floating-point elements, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). FOR j := 0 to 7 i := j*64 - n := (j mod 4)*64 - dst[i+63:i] := a[n+63:n] + IF k[j] + dst[i+63:i] := Convert_Int64_To_FP64(a[i+63:i]) + ELSE + dst[i+63:i] := src[i+63:i] + FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + + Floating Point Integer - AVX512F - Swizzle - - - - Broadcast the 4 packed 64-bit integers from "a" to all elements of "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + AVX512DQ + Convert + + + + + Convert packed signed 64-bit integers in "a" to packed double-precision (64-bit) floating-point elements, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + [round_note] FOR j := 0 to 7 i := j*64 - n := (j mod 4)*64 IF k[j] - dst[i+63:i] := a[n+63:n] + dst[i+63:i] := Convert_Int64_To_FP64(a[i+63:i]) ELSE - dst[i+63:i] := src[n+63:n] + dst[i+63:i] := 0 FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + + Floating Point Integer - AVX512F - Swizzle - - - Broadcast the 4 packed 64-bit integers from "a" to all elements of "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + AVX512DQ + Convert + + + + Convert packed signed 64-bit integers in "a" to packed double-precision (64-bit) floating-point elements, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). FOR j := 0 to 7 i := j*64 - n := (j mod 4)*64 IF k[j] - dst[i+63:i] := a[n+63:n] + dst[i+63:i] := Convert_Int64_To_FP64(a[i+63:i]) ELSE dst[i+63:i] := 0 FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - AVX512F - Swizzle - - Broadcast the low double-precision (64-bit) floating-point element from "a" to all elements of "dst". + Integer + AVX512VL + AVX512DQ + Convert + + + Convert packed signed 64-bit integers in "a" to packed double-precision (64-bit) floating-point elements, and store the results in "dst". -FOR j := 0 to 7 +FOR j := 0 to 1 i := j*64 - dst[i+63:i] := a[63:0] + dst[i+63:i] := Convert_Int64_To_FP64(a[i+63:i]) ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point - AVX512F - Swizzle - - - - Broadcast the low double-precision (64-bit) floating-point element from "a" to all elements of "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Integer + AVX512VL + AVX512DQ + Convert + + + + + Convert packed signed 64-bit integers in "a" to packed double-precision (64-bit) floating-point elements, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 7 +FOR j := 0 to 1 i := j*64 IF k[j] - dst[i+63:i] := a[63:0] + dst[i+63:i] := Convert_Int64_To_FP64(a[i+63:i]) ELSE dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point - AVX512F - Swizzle - - - Broadcast the low double-precision (64-bit) floating-point element from "a" to all elements of "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Integer + AVX512VL + AVX512DQ + Convert + + + + Convert packed signed 64-bit integers in "a" to packed double-precision (64-bit) floating-point elements, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 +FOR j := 0 to 1 i := j*64 IF k[j] - dst[i+63:i] := a[63:0] + dst[i+63:i] := Convert_Int64_To_FP64(a[i+63:i]) ELSE dst[i+63:i] := 0 FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point - AVX512F - Swizzle - - Broadcast the low single-precision (32-bit) floating-point element from "a" to all elements of "dst". + Integer + AVX512VL + AVX512DQ + Convert + + + Convert packed signed 64-bit integers in "a" to packed single-precision (32-bit) floating-point elements, and store the results in "dst". -FOR j := 0 to 15 - i := j*32 - dst[i+31:i] := a[31:0] +FOR j := 0 to 3 + i := j*64 + l := j*32 + dst[l+31:l] := Convert_Int64_To_FP32(a[i+63:i]) ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point - AVX512F - Swizzle - - - - Broadcast the low single-precision (32-bit) floating-point element from "a" to all elements of "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Integer + AVX512VL + AVX512DQ + Convert + + + + + Convert packed signed 64-bit integers in "a" to packed single-precision (32-bit) floating-point elements, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*32 +FOR j := 0 to 3 + i := j*64 + l := j*32 IF k[j] - dst[i+31:i] := a[31:0] + dst[l+31:l] := Convert_Int64_To_FP32(a[i+63:i]) ELSE - dst[i+31:i] := src[i+31:i] + dst[l+31:l] := src[l+31:l] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point - AVX512F - Swizzle - - - Broadcast the low single-precision (32-bit) floating-point element from "a" to all elements of "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Integer + AVX512VL + AVX512DQ + Convert + + + + Convert packed signed 64-bit integers in "a" to packed single-precision (32-bit) floating-point elements, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*32 +FOR j := 0 to 3 + i := j*64 + l := j*32 IF k[j] - dst[i+31:i] := a[31:0] + dst[l+31:l] := Convert_Int64_To_FP32(a[i+63:i]) ELSE - dst[i+31:i] := 0 + dst[l+31:l] := 0 FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point - Mask - AVX512F/KNCNI - Compare - - - - Compare packed double-precision (64-bit) floating-point elements in "a" and "b" based on the comparison operand specified by "imm8", and store the results in mask vector "k". + Integer + AVX512DQ + Convert + + + + Convert packed signed 64-bit integers in "a" to packed single-precision (32-bit) floating-point elements, and store the results in "dst". + [round_note] -CASE (imm8[7:0]) OF -0: OP := _CMP_EQ_OQ -1: OP := _CMP_LT_OS -2: OP := _CMP_LE_OS -3: OP := _CMP_UNORD_Q -4: OP := _CMP_NEQ_UQ -5: OP := _CMP_NLT_US -6: OP := _CMP_NLE_US -7: OP := _CMP_ORD_Q -8: OP := _CMP_EQ_UQ -9: OP := _CMP_NGE_US -10: OP := _CMP_NGT_US -11: OP := _CMP_FALSE_OQ -12: OP := _CMP_NEQ_OQ -13: OP := _CMP_GE_OS -14: OP := _CMP_GT_OS -15: OP := _CMP_TRUE_UQ -16: OP := _CMP_EQ_OS -17: OP := _CMP_LT_OQ -18: OP := _CMP_LE_OQ -19: OP := _CMP_UNORD_S -20: OP := _CMP_NEQ_US -21: OP := _CMP_NLT_UQ -22: OP := _CMP_NLE_UQ -23: OP := _CMP_ORD_S -24: OP := _CMP_EQ_US -25: OP := _CMP_NGE_UQ -26: OP := _CMP_NGT_UQ -27: OP := _CMP_FALSE_OS -28: OP := _CMP_NEQ_OS -29: OP := _CMP_GE_OQ -30: OP := _CMP_GT_OQ -31: OP := _CMP_TRUE_US -ESAC FOR j := 0 to 7 i := j*64 - k[j] := (a[i+63:i] OP b[i+63:i]) ? 1 : 0 + l := j*32 + dst[l+31:l] := Convert_Int64_To_FP32(a[i+63:i]) ENDFOR -k[MAX:8] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Floating Point - Mask - AVX512F/KNCNI - Compare - - - - - Compare packed double-precision (64-bit) floating-point elements in "a" and "b" based on the comparison operand specified by "imm8", and store the results in mask vector "k". - Pass __MM_FROUND_NO_EXC to "sae" to suppress all exceptions. + Integer + AVX512DQ + Convert + + + Convert packed signed 64-bit integers in "a" to packed single-precision (32-bit) floating-point elements, and store the results in "dst". -CASE (imm8[7:0]) OF -0: OP := _CMP_EQ_OQ -1: OP := _CMP_LT_OS -2: OP := _CMP_LE_OS -3: OP := _CMP_UNORD_Q -4: OP := _CMP_NEQ_UQ -5: OP := _CMP_NLT_US -6: OP := _CMP_NLE_US -7: OP := _CMP_ORD_Q -8: OP := _CMP_EQ_UQ -9: OP := _CMP_NGE_US -10: OP := _CMP_NGT_US -11: OP := _CMP_FALSE_OQ -12: OP := _CMP_NEQ_OQ -13: OP := _CMP_GE_OS -14: OP := _CMP_GT_OS -15: OP := _CMP_TRUE_UQ -16: OP := _CMP_EQ_OS -17: OP := _CMP_LT_OQ -18: OP := _CMP_LE_OQ -19: OP := _CMP_UNORD_S -20: OP := _CMP_NEQ_US -21: OP := _CMP_NLT_UQ -22: OP := _CMP_NLE_UQ -23: OP := _CMP_ORD_S -24: OP := _CMP_EQ_US -25: OP := _CMP_NGE_UQ -26: OP := _CMP_NGT_UQ -27: OP := _CMP_FALSE_OS -28: OP := _CMP_NEQ_OS -29: OP := _CMP_GE_OQ -30: OP := _CMP_GT_OQ -31: OP := _CMP_TRUE_US -ESAC FOR j := 0 to 7 i := j*64 - k[j] := (a[i+63:i] OP b[i+63:i]) ? 1 : 0 + l := j*32 + dst[l+31:l] := Convert_Int64_To_FP32(a[i+63:i]) ENDFOR -k[MAX:8] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Floating Point - Mask - AVX512F/KNCNI - Compare - - - Compare packed double-precision (64-bit) floating-point elements in "a" and "b" for equality, and store the results in mask vector "k". + Integer + AVX512DQ + Convert + + + + + + Convert packed signed 64-bit integers in "a" to packed single-precision (32-bit) floating-point elements, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + [round_note] FOR j := 0 to 7 i := j*64 - k[j] := (a[i+63:i] == b[i+63:i]) ? 1 : 0 -ENDFOR -k[MAX:8] := 0 + l := j*32 + IF k[j] + dst[l+31:l] := Convert_Int64_To_FP32(a[i+63:i]) + ELSE + dst[l+31:l] := src[l+31:l] + FI +ENDFOR +dst[MAX:256] := 0 - +
immintrin.h
 - + Floating Point - Mask - AVX512F/KNCNI - Compare - - - Compare packed double-precision (64-bit) floating-point elements in "a" and "b" for less-than-or-equal, and store the results in mask vector "k". + Integer + AVX512DQ + Convert + + + + + Convert packed signed 64-bit integers in "a" to packed single-precision (32-bit) floating-point elements, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). FOR j := 0 to 7 i := j*64 - k[j] := (a[i+63:i] <= b[i+63:i]) ? 1 : 0 + l := j*32 + IF k[j] + dst[l+31:l] := Convert_Int64_To_FP32(a[i+63:i]) + ELSE + dst[l+31:l] := src[l+31:l] + FI ENDFOR -k[MAX:8] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Floating Point - Mask - AVX512F/KNCNI - Compare - - - Compare packed double-precision (64-bit) floating-point elements in "a" and "b" for less-than, and store the results in mask vector "k". + Integer + AVX512DQ + Convert + + + + + Convert packed signed 64-bit integers in "a" to packed single-precision (32-bit) floating-point elements, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + [round_note] FOR j := 0 to 7 i := j*64 - k[j] := (a[i+63:i] < b[i+63:i]) ? 1 : 0 + l := j*32 + IF k[j] + dst[l+31:l] := Convert_Int64_To_FP32(a[i+63:i]) + ELSE + dst[l+31:l] := 0 + FI ENDFOR -k[MAX:8] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Floating Point - Mask - AVX512F/KNCNI - Compare - - - Compare packed double-precision (64-bit) floating-point elements in "a" and "b" for not-equal, and store the results in mask vector "k". + Integer + AVX512DQ + Convert + + + + Convert packed signed 64-bit integers in "a" to packed single-precision (32-bit) floating-point elements, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). FOR j := 0 to 7 i := j*64 - k[j] := (a[i+63:i] != b[i+63:i]) ? 1 : 0 + l := j*32 + IF k[j] + dst[l+31:l] := Convert_Int64_To_FP32(a[i+63:i]) + ELSE + dst[l+31:l] := 0 + FI ENDFOR -k[MAX:8] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Floating Point - Mask - AVX512F/KNCNI - Compare - - - Compare packed double-precision (64-bit) floating-point elements in "a" and "b" for not-less-than-or-equal, and store the results in mask vector "k". + Integer + AVX512VL + AVX512DQ + Convert + + + Convert packed signed 64-bit integers in "a" to packed single-precision (32-bit) floating-point elements, and store the results in "dst". -FOR j := 0 to 7 +FOR j := 0 to 1 i := j*64 - k[j] := !(a[i+63:i] <= b[i+63:i]) ? 1 : 0 + l := j*32 + dst[l+31:l] := Convert_Int64_To_FP32(a[i+63:i]) ENDFOR -k[MAX:8] := 0 +dst[MAX:64] := 0 - +
immintrin.h
 - + Floating Point - Mask - AVX512F/KNCNI - Compare - - - Compare packed double-precision (64-bit) floating-point elements in "a" and "b" for not-less-than, and store the results in mask vector "k". + Integer + AVX512VL + AVX512DQ + Convert + + + + + Convert packed signed 64-bit integers in "a" to packed single-precision (32-bit) floating-point elements, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 7 +FOR j := 0 to 1 i := j*64 - k[j] := !(a[i+63:i] < b[i+63:i]) ? 1 : 0 + l := j*32 + IF k[j] + dst[l+31:l] := Convert_Int64_To_FP32(a[i+63:i]) + ELSE + dst[l+31:l] := src[l+31:l] + FI ENDFOR -k[MAX:8] := 0 +dst[MAX:64] := 0 - +
immintrin.h
 - + Floating Point - Mask - AVX512F/KNCNI - Compare - - - Compare packed double-precision (64-bit) floating-point elements in "a" and "b" to see if neither is NaN, and store the results in mask vector "k". + Integer + AVX512VL + AVX512DQ + Convert + + + + Convert packed signed 64-bit integers in "a" to packed single-precision (32-bit) floating-point elements, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 +FOR j := 0 to 1 i := j*64 - k[j] := (a[i+63:i] != NaN AND b[i+63:i] != NaN) ? 1 : 0 + l := j*32 + IF k[j] + dst[l+31:l] := Convert_Int64_To_FP32(a[i+63:i]) + ELSE + dst[l+31:l] := 0 + FI ENDFOR -k[MAX:8] := 0 +dst[MAX:64] := 0 - +
immintrin.h
 - + Floating Point - Mask - AVX512F/KNCNI - Compare - - - Compare packed double-precision (64-bit) floating-point elements in "a" and "b" to see if either is NaN, and store the results in mask vector "k". + Integer + AVX512VL + AVX512DQ + Convert + + + Convert packed double-precision (64-bit) floating-point elements in "a" to packed 64-bit integers with truncation, and store the results in "dst". -FOR j := 0 to 7 +FOR j := 0 to 3 i := j*64 - k[j] := (a[i+63:i] == NaN OR b[i+63:i] == NaN) ? 1 : 0 + dst[i+63:i] := Convert_FP64_To_Int64_Truncate(a[i+63:i]) ENDFOR -k[MAX:8] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Floating Point - Mask - AVX512F/KNCNI - Compare - - - - - Compare packed double-precision (64-bit) floating-point elements in "a" and "b" based on the comparison operand specified by "imm8", and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). + Integer + AVX512VL + AVX512DQ + Convert + + + + + Convert packed double-precision (64-bit) floating-point elements in "a" to packed 64-bit integers with truncation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -CASE (imm8[7:0]) OF -0: OP := _CMP_EQ_OQ -1: OP := _CMP_LT_OS -2: OP := _CMP_LE_OS -3: OP := _CMP_UNORD_Q -4: OP := _CMP_NEQ_UQ -5: OP := _CMP_NLT_US -6: OP := _CMP_NLE_US -7: OP := _CMP_ORD_Q -8: OP := _CMP_EQ_UQ -9: OP := _CMP_NGE_US -10: OP := _CMP_NGT_US -11: OP := _CMP_FALSE_OQ -12: OP := _CMP_NEQ_OQ -13: OP := _CMP_GE_OS -14: OP := _CMP_GT_OS -15: OP := _CMP_TRUE_UQ -16: OP := _CMP_EQ_OS -17: OP := _CMP_LT_OQ -18: OP := _CMP_LE_OQ -19: OP := _CMP_UNORD_S -20: OP := _CMP_NEQ_US -21: OP := _CMP_NLT_UQ -22: OP := _CMP_NLE_UQ -23: OP := _CMP_ORD_S -24: OP := _CMP_EQ_US -25: OP := _CMP_NGE_UQ -26: OP := _CMP_NGT_UQ -27: OP := _CMP_FALSE_OS -28: OP := _CMP_NEQ_OS -29: OP := _CMP_GE_OQ -30: OP := _CMP_GT_OQ -31: OP := _CMP_TRUE_US -ESAC -FOR j := 0 to 7 +FOR j := 0 to 3 i := j*64 - IF k1[j] - k[j] := ( a[i+63:i] OP b[i+63:i] ) ? 1 : 0 + IF k[j] + dst[i+63:i] := Convert_FP64_To_Int64_Truncate(a[i+63:i]) ELSE - k[j] := 0 + dst[i+63:i] := src[i+63:i] FI ENDFOR -k[MAX:8] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Floating Point - Mask - AVX512F/KNCNI - Compare - - - - - - Compare packed double-precision (64-bit) floating-point elements in "a" and "b" based on the comparison operand specified by "imm8", and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). - Pass __MM_FROUND_NO_EXC to "sae" to suppress all exceptions. + Integer + AVX512VL + AVX512DQ + Convert + + + + Convert packed double-precision (64-bit) floating-point elements in "a" to packed 64-bit integers with truncation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -CASE (imm8[7:0]) OF -0: OP := _CMP_EQ_OQ -1: OP := _CMP_LT_OS -2: OP := _CMP_LE_OS -3: OP := _CMP_UNORD_Q -4: OP := _CMP_NEQ_UQ -5: OP := _CMP_NLT_US -6: OP := _CMP_NLE_US -7: OP := _CMP_ORD_Q -8: OP := _CMP_EQ_UQ -9: OP := _CMP_NGE_US -10: OP := _CMP_NGT_US -11: OP := _CMP_FALSE_OQ -12: OP := _CMP_NEQ_OQ -13: OP := _CMP_GE_OS -14: OP := _CMP_GT_OS -15: OP := _CMP_TRUE_UQ -16: OP := _CMP_EQ_OS -17: OP := _CMP_LT_OQ -18: OP := _CMP_LE_OQ -19: OP := _CMP_UNORD_S -20: OP := _CMP_NEQ_US -21: OP := _CMP_NLT_UQ -22: OP := _CMP_NLE_UQ -23: OP := _CMP_ORD_S -24: OP := _CMP_EQ_US -25: OP := _CMP_NGE_UQ -26: OP := _CMP_NGT_UQ -27: OP := _CMP_FALSE_OS -28: OP := _CMP_NEQ_OS -29: OP := _CMP_GE_OQ -30: OP := _CMP_GT_OQ -31: OP := _CMP_TRUE_US -ESAC -FOR j := 0 to 7 +FOR j := 0 to 3 i := j*64 - IF k1[j] - k[j] := ( a[i+63:i] OP b[i+63:i] ) ? 1 : 0 - ELSE - k[j] := 0 + IF k[j] + dst[i+63:i] := Convert_FP64_To_Int64_Truncate(a[i+63:i]) + ELSE + dst[i+63:i] := 0 FI ENDFOR -k[MAX:8] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Floating Point - Mask - AVX512F/KNCNI - Compare - - - - Compare packed double-precision (64-bit) floating-point elements in "a" and "b" for equality, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). + Integer + AVX512DQ + Convert + + + + Convert packed double-precision (64-bit) floating-point elements in "a" to packed 64-bit integers with truncation, and store the results in "dst". [sae_note] FOR j := 0 to 7 i := j*64 - IF k1[j] - k[j] := (a[i+63:i] == b[i+63:i]) ? 1 : 0 - ELSE - k[j] := 0 - FI -ENDFOR -k[MAX:8] := 0 + dst[i+63:i] := Convert_FP64_To_Int64_Truncate(a[i+63:i]) +ENDFOR +dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - Mask - AVX512F/KNCNI - Compare - - - - Compare packed double-precision (64-bit) floating-point elements in "a" and "b" for less-than-or-equal, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). + Integer + AVX512DQ + Convert + + + Convert packed double-precision (64-bit) floating-point elements in "a" to packed 64-bit integers with truncation, and store the results in "dst". FOR j := 0 to 7 i := j*64 - IF k1[j] - k[j] := (a[i+63:i] <= b[i+63:i]) ? 1 : 0 - ELSE - k[j] := 0 - FI + dst[i+63:i] := Convert_FP64_To_Int64_Truncate(a[i+63:i]) ENDFOR -k[MAX:8] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - Mask - AVX512F/KNCNI - Compare - - - - Compare packed double-precision (64-bit) floating-point elements in "a" and "b" for less-than, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). + Integer + AVX512DQ + Convert + + + + + + Convert packed double-precision (64-bit) floating-point elements in "a" to packed 64-bit integers with truncation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). [sae_note] FOR j := 0 to 7 i := j*64 - IF k1[j] - k[j] := (a[i+63:i] < b[i+63:i]) ? 1 : 0 - ELSE - k[j] := 0 + IF k[j] + dst[i+63:i] := Convert_FP64_To_Int64_Truncate(a[i+63:i]) + ELSE + dst[i+63:i] := src[i+63:i] FI ENDFOR -k[MAX:8] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - Mask - AVX512F/KNCNI - Compare - - - - Compare packed double-precision (64-bit) floating-point elements in "a" and "b" for not-equal, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). + Integer + AVX512DQ + Convert + + + + + Convert packed double-precision (64-bit) floating-point elements in "a" to packed 64-bit integers with truncation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). FOR j := 0 to 7 i := j*64 - IF k1[j] - k[j] := (a[i+63:i] != b[i+63:i]) ? 1 : 0 - ELSE - k[j] := 0 + IF k[j] + dst[i+63:i] := Convert_FP64_To_Int64_Truncate(a[i+63:i]) + ELSE + dst[i+63:i] := src[i+63:i] FI ENDFOR -k[MAX:8] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - Mask - AVX512F/KNCNI - Compare - - - - Compare packed double-precision (64-bit) floating-point elements in "a" and "b" for not-less-than-or-equal, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). + Integer + AVX512DQ + Convert + + + + + Convert packed double-precision (64-bit) floating-point elements in "a" to packed 64-bit integers with truncation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). [sae_note] FOR j := 0 to 7 i := j*64 - IF k1[j] - k[j] := !(a[i+63:i] <= b[i+63:i]) ? 1 : 0 - ELSE - k[j] := 0 + IF k[j] + dst[i+63:i] := Convert_FP64_To_Int64_Truncate(a[i+63:i]) + ELSE + dst[i+63:i] := 0 FI ENDFOR -k[MAX:8] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - Mask - AVX512F/KNCNI - Compare - - - - Compare packed double-precision (64-bit) floating-point elements in "a" and "b" for not-less-than, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). + Integer + AVX512DQ + Convert + + + + Convert packed double-precision (64-bit) floating-point elements in "a" to packed 64-bit integers with truncation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). FOR j := 0 to 7 i := j*64 - IF k1[j] - k[j] := !(a[i+63:i] < b[i+63:i]) ? 1 : 0 - ELSE - k[j] := 0 + IF k[j] + dst[i+63:i] := Convert_FP64_To_Int64_Truncate(a[i+63:i]) + ELSE + dst[i+63:i] := 0 FI ENDFOR -k[MAX:8] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - Mask - AVX512F/KNCNI - Compare - - - - Compare packed double-precision (64-bit) floating-point elements in "a" and "b" to see if neither is NaN, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). + Integer + AVX512VL + AVX512DQ + Convert + + + Convert packed double-precision (64-bit) floating-point elements in "a" to packed 64-bit integers with truncation, and store the results in "dst". -FOR j := 0 to 7 +FOR j := 0 to 1 i := j*64 - IF k1[j] - k[j] := (a[i+63:i] != NaN AND b[i+63:i] != NaN) ? 1 : 0 - ELSE - k[j] := 0 + dst[i+63:i] := Convert_FP64_To_Int64_Truncate(a[i+63:i]) +ENDFOR +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Floating Point + Integer + AVX512VL + AVX512DQ + Convert + + + + + Convert packed double-precision (64-bit) floating-point elements in "a" to packed 64-bit integers with truncation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +FOR j := 0 to 1 + i := j*64 + IF k[j] + dst[i+63:i] := Convert_FP64_To_Int64_Truncate(a[i+63:i]) + ELSE + dst[i+63:i] := src[i+63:i] FI ENDFOR -k[MAX:8] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point - Mask - AVX512F/KNCNI - Compare - - - - Compare packed double-precision (64-bit) floating-point elements in "a" and "b" to see if either is NaN, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). + Integer + AVX512VL + AVX512DQ + Convert + + + + Convert packed double-precision (64-bit) floating-point elements in "a" to packed 64-bit integers with truncation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 +FOR j := 0 to 1 i := j*64 - IF k1[j] - k[j] := (a[i+63:i] == NaN OR b[i+63:i] == NaN) ? 1 : 0 - ELSE - k[j] := 0 + IF k[j] + dst[i+63:i] := Convert_FP64_To_Int64_Truncate(a[i+63:i]) + ELSE + dst[i+63:i] := 0 FI ENDFOR -k[MAX:8] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point - Mask - AVX512F/KNCNI - Compare - - - - Compare packed single-precision (32-bit) floating-point elements in "a" and "b" based on the comparison operand specified by "imm8", and store the results in mask vector "k". + Integer + AVX512VL + AVX512DQ + Convert + + + Convert packed double-precision (64-bit) floating-point elements in "a" to packed unsigned 64-bit integers with truncation, and store the results in "dst". -CASE (imm8[7:0]) OF -0: OP := _CMP_EQ_OQ -1: OP := _CMP_LT_OS -2: OP := _CMP_LE_OS -3: OP := _CMP_UNORD_Q -4: OP := _CMP_NEQ_UQ -5: OP := _CMP_NLT_US -6: OP := _CMP_NLE_US -7: OP := _CMP_ORD_Q -8: OP := _CMP_EQ_UQ -9: OP := _CMP_NGE_US -10: OP := _CMP_NGT_US -11: OP := _CMP_FALSE_OQ -12: OP := _CMP_NEQ_OQ -13: OP := _CMP_GE_OS -14: OP := _CMP_GT_OS -15: OP := _CMP_TRUE_UQ -16: OP := _CMP_EQ_OS -17: OP := _CMP_LT_OQ -18: OP := _CMP_LE_OQ -19: OP := _CMP_UNORD_S -20: OP := _CMP_NEQ_US -21: OP := _CMP_NLT_UQ -22: OP := _CMP_NLE_UQ -23: OP := _CMP_ORD_S -24: OP := _CMP_EQ_US -25: OP := _CMP_NGE_UQ -26: OP := _CMP_NGT_UQ -27: OP := _CMP_FALSE_OS -28: OP := _CMP_NEQ_OS -29: OP := _CMP_GE_OQ -30: OP := _CMP_GT_OQ -31: OP := _CMP_TRUE_US -ESAC -FOR j := 0 to 15 - i := j*32 - k[j] := (a[i+31:i] OP b[i+31:i]) ? 1 : 0 +FOR j := 0 to 3 + i := j*64 + dst[i+63:i] := Convert_FP64_To_UInt64_Truncate(a[i+63:i]) ENDFOR -k[MAX:16] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Floating Point - Mask - AVX512F/KNCNI - Compare - - - - - Compare packed single-precision (32-bit) floating-point elements in "a" and "b" based on the comparison operand specified by "imm8", and store the results in mask vector "k". - Pass __MM_FROUND_NO_EXC to "sae" to suppress all exceptions. + Integer + AVX512VL + AVX512DQ + Convert + + + + + Convert packed double-precision (64-bit) floating-point elements in "a" to packed unsigned 64-bit integers with truncation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -CASE (imm8[7:0]) OF -0: OP := _CMP_EQ_OQ -1: OP := _CMP_LT_OS -2: OP := _CMP_LE_OS -3: OP := _CMP_UNORD_Q -4: OP := _CMP_NEQ_UQ -5: OP := _CMP_NLT_US -6: OP := _CMP_NLE_US -7: OP := _CMP_ORD_Q -8: OP := _CMP_EQ_UQ -9: OP := _CMP_NGE_US -10: OP := _CMP_NGT_US -11: OP := _CMP_FALSE_OQ -12: OP := _CMP_NEQ_OQ -13: OP := _CMP_GE_OS -14: OP := _CMP_GT_OS -15: OP := _CMP_TRUE_UQ -16: OP := _CMP_EQ_OS -17: OP := _CMP_LT_OQ -18: OP := _CMP_LE_OQ -19: OP := _CMP_UNORD_S -20: OP := _CMP_NEQ_US -21: OP := _CMP_NLT_UQ -22: OP := _CMP_NLE_UQ -23: OP := _CMP_ORD_S -24: OP := _CMP_EQ_US -25: OP := _CMP_NGE_UQ -26: OP := _CMP_NGT_UQ -27: OP := _CMP_FALSE_OS -28: OP := _CMP_NEQ_OS -29: OP := _CMP_GE_OQ -30: OP := _CMP_GT_OQ -31: OP := _CMP_TRUE_US -ESAC -FOR j := 0 to 15 - i := j*32 - k[j] := (a[i+31:i] OP b[i+31:i]) ? 1 : 0 +FOR j := 0 to 3 + i := j*64 + IF k[j] + dst[i+63:i] := Convert_FP64_To_UInt64_Truncate(a[i+63:i]) + ELSE + dst[i+63:i] := src[i+63:i] + FI ENDFOR -k[MAX:16] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Floating Point - Mask - AVX512F/KNCNI - Compare - - - Compare packed single-precision (32-bit) floating-point elements in "a" and "b" for equality, and store the results in mask vector "k". + Integer + AVX512VL + AVX512DQ + Convert + + + + Convert packed double-precision (64-bit) floating-point elements in "a" to packed unsigned 64-bit integers with truncation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*32 - k[j] := (a[i+31:i] == b[i+31:i]) ? 1 : 0 -ENDFOR -k[MAX:16] := 0 +FOR j := 0 to 3 + i := j*64 + IF k[j] + dst[i+63:i] := Convert_FP64_To_UInt64_Truncate(a[i+63:i]) + ELSE + dst[i+63:i] := 0 + FI +ENDFOR +dst[MAX:256] := 0 - +
immintrin.h
 - + Floating Point - Mask - AVX512F/KNCNI - Compare - - - Compare packed single-precision (32-bit) floating-point elements in "a" and "b" for less-than-or-equal, and store the results in mask vector "k". + Integer + AVX512DQ + Convert + + + + Convert packed double-precision (64-bit) floating-point elements in "a" to packed unsigned 64-bit integers with truncation, and store the results in "dst". [sae_note] -FOR j := 0 to 15 - i := j*32 - k[j] := (a[i+31:i] <= b[i+31:i]) ? 1 : 0 +FOR j := 0 to 7 + i := j*64 + dst[i+63:i] := Convert_FP64_To_UInt64_Truncate(a[i+63:i]) ENDFOR -k[MAX:16] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - Mask - AVX512F/KNCNI - Compare - - - Compare packed single-precision (32-bit) floating-point elements in "a" and "b" for less-than, and store the results in mask vector "k". + Integer + AVX512DQ + Convert + + + Convert packed double-precision (64-bit) floating-point elements in "a" to packed unsigned 64-bit integers with truncation, and store the results in "dst". -FOR j := 0 to 15 - i := j*32 - k[j] := (a[i+31:i] < b[i+31:i]) ? 1 : 0 +FOR j := 0 to 7 + i := j*64 + dst[i+63:i] := Convert_FP64_To_UInt64_Truncate(a[i+63:i]) ENDFOR -k[MAX:16] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - Mask - AVX512F/KNCNI - Compare - - - Compare packed single-precision (32-bit) floating-point elements in "a" and "b" for not-equal, and store the results in mask vector "k". + Integer + AVX512DQ + Convert + + + + + + Convert packed double-precision (64-bit) floating-point elements in "a" to packed unsigned 64-bit integers with truncation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). [sae_note] -FOR j := 0 to 15 - i := j*32 - k[j] := (a[i+31:i] != b[i+31:i]) ? 1 : 0 +FOR j := 0 to 7 + i := j*64 + IF k[j] + dst[i+63:i] := Convert_FP64_To_UInt64_Truncate(a[i+63:i]) + ELSE + dst[i+63:i] := src[i+63:i] + FI ENDFOR -k[MAX:16] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - Mask - AVX512F/KNCNI - Compare - - - Compare packed single-precision (32-bit) floating-point elements in "a" and "b" for not-less-than-or-equal, and store the results in mask vector "k". + Integer + AVX512DQ + Convert + + + + + Convert packed double-precision (64-bit) floating-point elements in "a" to packed unsigned 64-bit integers with truncation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*32 - k[j] := !(a[i+31:i] <= b[i+31:i]) ? 1 : 0 +FOR j := 0 to 7 + i := j*64 + IF k[j] + dst[i+63:i] := Convert_FP64_To_UInt64_Truncate(a[i+63:i]) + ELSE + dst[i+63:i] := src[i+63:i] + FI ENDFOR -k[MAX:16] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - Mask - AVX512F/KNCNI - Compare - - - Compare packed single-precision (32-bit) floating-point elements in "a" and "b" for not-less-than, and store the results in mask vector "k". + Integer + AVX512DQ + Convert + + + + + Convert packed double-precision (64-bit) floating-point elements in "a" to packed unsigned 64-bit integers with truncation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). [sae_note] -FOR j := 0 to 15 - i := j*32 - k[j] := !(a[i+31:i] < b[i+31:i]) ? 1 : 0 +FOR j := 0 to 7 + i := j*64 + IF k[j] + dst[i+63:i] := Convert_FP64_To_UInt64_Truncate(a[i+63:i]) + ELSE + dst[i+63:i] := 0 + FI ENDFOR -k[MAX:16] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - Mask - AVX512F/KNCNI - Compare - - - Compare packed single-precision (32-bit) floating-point elements in "a" and "b" to see if neither is NaN, and store the results in mask vector "k". + Integer + AVX512DQ + Convert + + + + Convert packed double-precision (64-bit) floating-point elements in "a" to packed unsigned 64-bit integers with truncation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*32 - k[j] := (a[i+31:i] != NaN AND b[i+31:i] != NaN) ? 1 : 0 +FOR j := 0 to 7 + i := j*64 + IF k[j] + dst[i+63:i] := Convert_FP64_To_UInt64_Truncate(a[i+63:i]) + ELSE + dst[i+63:i] := 0 + FI ENDFOR -k[MAX:16] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - Mask - AVX512F/KNCNI - Compare - - - Compare packed single-precision (32-bit) floating-point elements in "a" and "b" to see if either is NaN, and store the results in mask vector "k". + Integer + AVX512VL + AVX512DQ + Convert + + + Convert packed double-precision (64-bit) floating-point elements in "a" to packed unsigned 64-bit integers with truncation, and store the results in "dst". -FOR j := 0 to 15 - i := j*32 - k[j] := (a[i+31:i] == NaN OR b[i+31:i] == NaN) ? 1 : 0 +FOR j := 0 to 1 + i := j*64 + dst[i+63:i] := Convert_FP64_To_UInt64_Truncate(a[i+63:i]) ENDFOR -k[MAX:16] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point - Mask - AVX512F/KNCNI - Compare - - - - - Compare packed single-precision (32-bit) floating-point elements in "a" and "b" based on the comparison operand specified by "imm8", and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). + Integer + AVX512VL + AVX512DQ + Convert + + + + + Convert packed double-precision (64-bit) floating-point elements in "a" to packed unsigned 64-bit integers with truncation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -CASE (imm8[7:0]) OF -0: OP := _CMP_EQ_OQ -1: OP := _CMP_LT_OS -2: OP := _CMP_LE_OS -3: OP := _CMP_UNORD_Q -4: OP := _CMP_NEQ_UQ -5: OP := _CMP_NLT_US -6: OP := _CMP_NLE_US -7: OP := _CMP_ORD_Q -8: OP := _CMP_EQ_UQ -9: OP := _CMP_NGE_US -10: OP := _CMP_NGT_US -11: OP := _CMP_FALSE_OQ -12: OP := _CMP_NEQ_OQ -13: OP := _CMP_GE_OS -14: OP := _CMP_GT_OS -15: OP := _CMP_TRUE_UQ -16: OP := _CMP_EQ_OS -17: OP := _CMP_LT_OQ -18: OP := _CMP_LE_OQ -19: OP := _CMP_UNORD_S -20: OP := _CMP_NEQ_US -21: OP := _CMP_NLT_UQ -22: OP := _CMP_NLE_UQ -23: OP := _CMP_ORD_S -24: OP := _CMP_EQ_US -25: OP := _CMP_NGE_UQ -26: OP := _CMP_NGT_UQ -27: OP := _CMP_FALSE_OS -28: OP := _CMP_NEQ_OS -29: OP := _CMP_GE_OQ -30: OP := _CMP_GT_OQ -31: OP := _CMP_TRUE_US -ESAC -FOR j := 0 to 15 - i := j*32 - IF k1[j] - k[j] := ( a[i+31:i] OP b[i+31:i] ) ? 1 : 0 - ELSE - k[j] := 0 +FOR j := 0 to 1 + i := j*64 + IF k[j] + dst[i+63:i] := Convert_FP64_To_UInt64_Truncate(a[i+63:i]) + ELSE + dst[i+63:i] := src[i+63:i] FI ENDFOR -k[MAX:16] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point - Mask - AVX512F/KNCNI - Compare - - - - - - Compare packed single-precision (32-bit) floating-point elements in "a" and "b" based on the comparison operand specified by "imm8", and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). - Pass __MM_FROUND_NO_EXC to "sae" to suppress all exceptions. + Integer + AVX512VL + AVX512DQ + Convert + + + + Convert packed double-precision (64-bit) floating-point elements in "a" to packed unsigned 64-bit integers with truncation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -CASE (imm8[7:0]) OF -0: OP := _CMP_EQ_OQ -1: OP := _CMP_LT_OS -2: OP := _CMP_LE_OS -3: OP := _CMP_UNORD_Q -4: OP := _CMP_NEQ_UQ -5: OP := _CMP_NLT_US -6: OP := _CMP_NLE_US -7: OP := _CMP_ORD_Q -8: OP := _CMP_EQ_UQ -9: OP := _CMP_NGE_US -10: OP := _CMP_NGT_US -11: OP := _CMP_FALSE_OQ -12: OP := _CMP_NEQ_OQ -13: OP := _CMP_GE_OS -14: OP := _CMP_GT_OS -15: OP := _CMP_TRUE_UQ -16: OP := _CMP_EQ_OS -17: OP := _CMP_LT_OQ -18: OP := _CMP_LE_OQ -19: OP := _CMP_UNORD_S -20: OP := _CMP_NEQ_US -21: OP := _CMP_NLT_UQ -22: OP := _CMP_NLE_UQ -23: OP := _CMP_ORD_S -24: OP := _CMP_EQ_US -25: OP := _CMP_NGE_UQ -26: OP := _CMP_NGT_UQ -27: OP := _CMP_FALSE_OS -28: OP := _CMP_NEQ_OS -29: OP := _CMP_GE_OQ -30: OP := _CMP_GT_OQ -31: OP := _CMP_TRUE_US -ESAC -FOR j := 0 to 15 - i := j*32 - IF k1[j] - k[j] := ( a[i+31:i] OP b[i+31:i] ) ? 1 : 0 - ELSE - k[j] := 0 +FOR j := 0 to 1 + i := j*64 + IF k[j] + dst[i+63:i] := Convert_FP64_To_UInt64_Truncate(a[i+63:i]) + ELSE + dst[i+63:i] := 0 FI ENDFOR -k[MAX:16] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point - Mask - AVX512F/KNCNI - Compare - - - - Compare packed single-precision (32-bit) floating-point elements in "a" and "b" for equality, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). - y -FOR j := 0 to 15 - i := j*32 - IF k1[j] - k[j] := (a[i+31:i] == b[i+31:i]) ? 1 : 0 - ELSE - k[j] := 0 - FI -ENDFOR -k[MAX:16] := 0 + Integer + AVX512VL + AVX512DQ + Convert + + + Convert packed single-precision (32-bit) floating-point elements in "a" to packed 64-bit integers with truncation, and store the results in "dst". + +FOR j := 0 to 3 + i := j*64 + l := j*32 + dst[i+63:i] := Convert_FP32_To_Int64_Truncate(a[l+31:l]) +ENDFOR +dst[MAX:256] := 0 - +
immintrin.h
 - + Floating Point - Mask - AVX512F/KNCNI - Compare - - - - Compare packed single-precision (32-bit) floating-point elements in "a" and "b" for less-than-or-equal, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). + Integer + AVX512VL + AVX512DQ + Convert + + + + + Convert packed single-precision (32-bit) floating-point elements in "a" to packed 64-bit integers with truncation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*32 - IF k1[j] - k[j] := (a[i+31:i] <= b[i+31:i]) ? 1 : 0 - ELSE - k[j] := 0 +FOR j := 0 to 3 + i := j*64 + l := j*32 + IF k[j] + dst[i+63:i] := Convert_FP32_To_Int64_Truncate(a[l+31:l]) + ELSE + dst[i+63:i] := src[i+63:i] FI ENDFOR -k[MAX:16] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Floating Point - Mask - AVX512F/KNCNI - Compare - - - - Compare packed single-precision (32-bit) floating-point elements in "a" and "b" for less-than, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). + Integer + AVX512VL + AVX512DQ + Convert + + + + Convert packed single-precision (32-bit) floating-point elements in "a" to packed 64-bit integers with truncation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*32 - IF k1[j] - k[j] := (a[i+31:i] < b[i+31:i]) ? 1 : 0 - ELSE - k[j] := 0 +FOR j := 0 to 3 + i := j*64 + l := j*32 + IF k[j] + dst[i+63:i] := Convert_FP32_To_Int64_Truncate(a[l+31:l]) + ELSE + dst[i+63:i] := 0 FI ENDFOR -k[MAX:16] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Floating Point - Mask - AVX512F/KNCNI - Compare - - - - Compare packed single-precision (32-bit) floating-point elements in "a" and "b" for not-equal, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). + Integer + AVX512DQ + Convert + + + + Convert packed single-precision (32-bit) floating-point elements in "a" to packed 64-bit integers with truncation, and store the results in "dst". [sae_note] -FOR j := 0 to 15 - i := j*32 - IF k1[j] - k[j] := (a[i+31:i] != b[i+31:i]) ? 1 : 0 - ELSE - k[j] := 0 - FI +FOR j := 0 to 7 + i := j*64 + l := j*32 + dst[i+63:i] := Convert_FP32_To_Int64_Truncate(a[l+31:l]) ENDFOR -k[MAX:16] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - Mask - AVX512F/KNCNI - Compare - - - - Compare packed single-precision (32-bit) floating-point elements in "a" and "b" for not-less-than-or-equal, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). + Integer + AVX512DQ + Convert + + + Convert packed single-precision (32-bit) floating-point elements in "a" to packed 64-bit integers with truncation, and store the results in "dst". -FOR j := 0 to 15 - i := j*32 - IF k1[j] - k[j] := !(a[i+31:i] <= b[i+31:i]) ? 1 : 0 - ELSE - k[j] := 0 +FOR j := 0 to 7 + i := j*64 + l := j*32 + dst[i+63:i] := Convert_FP32_To_Int64_Truncate(a[l+31:l]) +ENDFOR +dst[MAX:512] := 0 + + +
immintrin.h
+ + + Floating Point + Integer + AVX512DQ + Convert + + + + + + Convert packed single-precision (32-bit) floating-point elements in "a" to packed 64-bit integers with truncation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). [sae_note] + +FOR j := 0 to 7 + i := j*64 + l := j*32 + IF k[j] + dst[i+63:i] := Convert_FP32_To_Int64_Truncate(a[l+31:l]) + ELSE + dst[i+63:i] := src[i+63:i] FI ENDFOR -k[MAX:16] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - Mask - AVX512F/KNCNI - Compare - - - - Compare packed single-precision (32-bit) floating-point elements in "a" and "b" for not-less-than, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). + Integer + AVX512DQ + Convert + + + + + Convert packed single-precision (32-bit) floating-point elements in "a" to packed 64-bit integers with truncation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*32 - IF k1[j] - k[j] := !(a[i+31:i] < b[i+31:i]) ? 1 : 0 - ELSE - k[j] := 0 +FOR j := 0 to 7 + i := j*64 + l := j*32 + IF k[j] + dst[i+63:i] := Convert_FP32_To_Int64_Truncate(a[l+31:l]) + ELSE + dst[i+63:i] := src[i+63:i] FI ENDFOR -k[MAX:16] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - Mask - AVX512F/KNCNI - Compare - - - - Compare packed single-precision (32-bit) floating-point elements in "a" and "b" to see if neither is NaN, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). + Integer + AVX512DQ + Convert + + + + + Convert packed single-precision (32-bit) floating-point elements in "a" to packed 64-bit integers with truncation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). [sae_note] -FOR j := 0 to 15 - i := j*32 - IF k1[j] - k[j] := (a[i+31:i] != NaN AND b[i+31:i] != NaN) ? 1 : 0 - ELSE - k[j] := 0 +FOR j := 0 to 7 + i := j*64 + l := j*32 + IF k[j] + dst[i+63:i] := Convert_FP32_To_Int64_Truncate(a[l+31:l]) + ELSE + dst[i+63:i] := 0 FI ENDFOR -k[MAX:16] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - Mask - AVX512F/KNCNI - Compare - - - - Compare packed single-precision (32-bit) floating-point elements in "a" and "b" to see if either is NaN, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). + Integer + AVX512DQ + Convert + + + + Convert packed single-precision (32-bit) floating-point elements in "a" to packed 64-bit integers with truncation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*32 - IF k1[j] - k[j] := (a[i+31:i] == NaN OR b[i+31:i] == NaN) ? 1 : 0 - ELSE - k[j] := 0 +FOR j := 0 to 7 + i := j*64 + l := j*32 + IF k[j] + dst[i+63:i] := Convert_FP32_To_Int64_Truncate(a[l+31:l]) + ELSE + dst[i+63:i] := 0 FI ENDFOR -k[MAX:16] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - Mask - AVX512F - Compare - - - - - Compare the lower double-precision (64-bit) floating-point element in "a" and "b" based on the comparison operand specified by "imm8", and store the result in mask vector "k". - Pass __MM_FROUND_NO_EXC to "sae" to suppress all exceptions. + Integer + AVX512VL + AVX512DQ + Convert + + + Convert packed single-precision (32-bit) floating-point elements in "a" to packed 64-bit integers with truncation, and store the results in "dst". -CASE (imm8[7:0]) OF -0: OP := _CMP_EQ_OQ -1: OP := _CMP_LT_OS -2: OP := _CMP_LE_OS -3: OP := _CMP_UNORD_Q -4: OP := _CMP_NEQ_UQ -5: OP := _CMP_NLT_US -6: OP := _CMP_NLE_US -7: OP := _CMP_ORD_Q -8: OP := _CMP_EQ_UQ -9: OP := _CMP_NGE_US -10: OP := _CMP_NGT_US -11: OP := _CMP_FALSE_OQ -12: OP := _CMP_NEQ_OQ -13: OP := _CMP_GE_OS -14: OP := _CMP_GT_OS -15: OP := _CMP_TRUE_UQ -16: OP := _CMP_EQ_OS -17: OP := _CMP_LT_OQ -18: OP := _CMP_LE_OQ -19: OP := _CMP_UNORD_S -20: OP := _CMP_NEQ_US -21: OP := _CMP_NLT_UQ -22: OP := _CMP_NLE_UQ -23: OP := _CMP_ORD_S -24: OP := _CMP_EQ_US -25: OP := _CMP_NGE_UQ -26: OP := _CMP_NGT_UQ -27: OP := _CMP_FALSE_OS -28: OP := _CMP_NEQ_OS -29: OP := _CMP_GE_OQ -30: OP := _CMP_GT_OQ -31: OP := _CMP_TRUE_US -ESAC - -k[0] := ( a[63:0] OP b[63:0] ) ? 1 : 0 - -k[MAX:1] := 0 +FOR j := 0 to 1 + i := j*64 + l := j*32 + dst[i+63:i] := Convert_FP32_To_Int64_Truncate(a[l+31:l]) +ENDFOR +dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point - Mask - AVX512F - Compare - - - - Compare the lower double-precision (64-bit) floating-point element in "a" and "b" based on the comparison operand specified by "imm8", and store the result in mask vector "k". + Integer + AVX512VL + AVX512DQ + Convert + + + + + Convert packed single-precision (32-bit) floating-point elements in "a" to packed 64-bit integers with truncation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -CASE (imm8[7:0]) OF -0: OP := _CMP_EQ_OQ -1: OP := _CMP_LT_OS -2: OP := _CMP_LE_OS -3: OP := _CMP_UNORD_Q -4: OP := _CMP_NEQ_UQ -5: OP := _CMP_NLT_US -6: OP := _CMP_NLE_US -7: OP := _CMP_ORD_Q -8: OP := _CMP_EQ_UQ -9: OP := _CMP_NGE_US -10: OP := _CMP_NGT_US -11: OP := _CMP_FALSE_OQ -12: OP := _CMP_NEQ_OQ -13: OP := _CMP_GE_OS -14: OP := _CMP_GT_OS -15: OP := _CMP_TRUE_UQ -16: OP := _CMP_EQ_OS -17: OP := _CMP_LT_OQ -18: OP := _CMP_LE_OQ -19: OP := _CMP_UNORD_S -20: OP := _CMP_NEQ_US -21: OP := _CMP_NLT_UQ -22: OP := _CMP_NLE_UQ -23: OP := _CMP_ORD_S -24: OP := _CMP_EQ_US -25: OP := _CMP_NGE_UQ -26: OP := _CMP_NGT_UQ -27: OP := _CMP_FALSE_OS -28: OP := _CMP_NEQ_OS -29: OP := _CMP_GE_OQ -30: OP := _CMP_GT_OQ -31: OP := _CMP_TRUE_US -ESAC - -k[0] := ( a[63:0] OP b[63:0] ) ? 1 : 0 - -k[MAX:1] := 0 - - -
immintrin.h
- - - Floating Point - Mask - AVX512F - Compare - - - - - - Compare the lower double-precision (64-bit) floating-point element in "a" and "b" based on the comparison operand specified by "imm8", and store the result in mask vector "k" using zeromask "k1" (the element is zeroed out when mask bit 0 is not set). - Pass __MM_FROUND_NO_EXC to "sae" to suppress all exceptions. - -CASE (imm8[7:0]) OF -0: OP := _CMP_EQ_OQ -1: OP := _CMP_LT_OS -2: OP := _CMP_LE_OS -3: OP := _CMP_UNORD_Q -4: OP := _CMP_NEQ_UQ -5: OP := _CMP_NLT_US -6: OP := _CMP_NLE_US -7: OP := _CMP_ORD_Q -8: OP := _CMP_EQ_UQ -9: OP := _CMP_NGE_US -10: OP := _CMP_NGT_US -11: OP := _CMP_FALSE_OQ -12: OP := _CMP_NEQ_OQ -13: OP := _CMP_GE_OS -14: OP := _CMP_GT_OS -15: OP := _CMP_TRUE_UQ -16: OP := _CMP_EQ_OS -17: OP := _CMP_LT_OQ -18: OP := _CMP_LE_OQ -19: OP := _CMP_UNORD_S -20: OP := _CMP_NEQ_US -21: OP := _CMP_NLT_UQ -22: OP := _CMP_NLE_UQ -23: OP := _CMP_ORD_S -24: OP := _CMP_EQ_US -25: OP := _CMP_NGE_UQ -26: OP := _CMP_NGT_UQ -27: OP := _CMP_FALSE_OS -28: OP := _CMP_NEQ_OS -29: OP := _CMP_GE_OQ -30: OP := _CMP_GT_OQ -31: OP := _CMP_TRUE_US -ESAC - -IF k1[0] - k[0] := ( a[63:0] OP b[63:0] ) ? 1 : 0 -ELSE - k[0] := 0 -FI -k[MAX:1] := 0 +FOR j := 0 to 1 + i := j*64 + l := j*32 + IF k[j] + dst[i+63:i] := Convert_FP32_To_Int64_Truncate(a[l+31:l]) + ELSE + dst[i+63:i] := src[i+63:i] + FI +ENDFOR +dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point - Mask - AVX512F - Compare - - - - - Compare the lower double-precision (64-bit) floating-point element in "a" and "b" based on the comparison operand specified by "imm8", and store the result in mask vector "k" using zeromask "k1" (the element is zeroed out when mask bit 0 is not set). + Integer + AVX512VL + AVX512DQ + Convert + + + + Convert packed single-precision (32-bit) floating-point elements in "a" to packed 64-bit integers with truncation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -CASE (imm8[7:0]) OF -0: OP := _CMP_EQ_OQ -1: OP := _CMP_LT_OS -2: OP := _CMP_LE_OS -3: OP := _CMP_UNORD_Q -4: OP := _CMP_NEQ_UQ -5: OP := _CMP_NLT_US -6: OP := _CMP_NLE_US -7: OP := _CMP_ORD_Q -8: OP := _CMP_EQ_UQ -9: OP := _CMP_NGE_US -10: OP := _CMP_NGT_US -11: OP := _CMP_FALSE_OQ -12: OP := _CMP_NEQ_OQ -13: OP := _CMP_GE_OS -14: OP := _CMP_GT_OS -15: OP := _CMP_TRUE_UQ -16: OP := _CMP_EQ_OS -17: OP := _CMP_LT_OQ -18: OP := _CMP_LE_OQ -19: OP := _CMP_UNORD_S -20: OP := _CMP_NEQ_US -21: OP := _CMP_NLT_UQ -22: OP := _CMP_NLE_UQ -23: OP := _CMP_ORD_S -24: OP := _CMP_EQ_US -25: OP := _CMP_NGE_UQ -26: OP := _CMP_NGT_UQ -27: OP := _CMP_FALSE_OS -28: OP := _CMP_NEQ_OS -29: OP := _CMP_GE_OQ -30: OP := _CMP_GT_OQ -31: OP := _CMP_TRUE_US -ESAC - -IF k1[0] - k[0] := ( a[63:0] OP b[63:0] ) ? 1 : 0 -ELSE - k[0] := 0 -FI -k[MAX:1] := 0 +FOR j := 0 to 1 + i := j*64 + l := j*32 + IF k[j] + dst[i+63:i] := Convert_FP32_To_Int64_Truncate(a[l+31:l]) + ELSE + dst[i+63:i] := 0 + FI +ENDFOR +dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point - Mask - AVX512F - Compare - - - - - Compare the lower single-precision (32-bit) floating-point element in "a" and "b" based on the comparison operand specified by "imm8", and store the result in mask vector "k". - Pass __MM_FROUND_NO_EXC to "sae" to suppress all exceptions. + Integer + AVX512VL + AVX512DQ + Convert + + + Convert packed single-precision (32-bit) floating-point elements in "a" to packed unsigned 64-bit integers with truncation, and store the results in "dst". -CASE (imm8[7:0]) OF -0: OP := _CMP_EQ_OQ -1: OP := _CMP_LT_OS -2: OP := _CMP_LE_OS -3: OP := _CMP_UNORD_Q -4: OP := _CMP_NEQ_UQ -5: OP := _CMP_NLT_US -6: OP := _CMP_NLE_US -7: OP := _CMP_ORD_Q -8: OP := _CMP_EQ_UQ -9: OP := _CMP_NGE_US -10: OP := _CMP_NGT_US -11: OP := _CMP_FALSE_OQ -12: OP := _CMP_NEQ_OQ -13: OP := _CMP_GE_OS -14: OP := _CMP_GT_OS -15: OP := _CMP_TRUE_UQ -16: OP := _CMP_EQ_OS -17: OP := _CMP_LT_OQ -18: OP := _CMP_LE_OQ -19: OP := _CMP_UNORD_S -20: OP := _CMP_NEQ_US -21: OP := _CMP_NLT_UQ -22: OP := _CMP_NLE_UQ -23: OP := _CMP_ORD_S -24: OP := _CMP_EQ_US -25: OP := _CMP_NGE_UQ -26: OP := _CMP_NGT_UQ -27: OP := _CMP_FALSE_OS -28: OP := _CMP_NEQ_OS -29: OP := _CMP_GE_OQ -30: OP := _CMP_GT_OQ -31: OP := _CMP_TRUE_US -ESAC - -k[0] := ( a[31:0] OP b[31:0] ) ? 1 : 0 - -k[MAX:1] := 0 +FOR j := 0 to 3 + i := j*64 + l := j*32 + dst[i+63:i] := Convert_FP32_To_UInt64_Truncate(a[l+31:l]) +ENDFOR +dst[MAX:256] := 0 - +
immintrin.h
 - + Floating Point - Mask - AVX512F - Compare - - - - Compare the lower single-precision (32-bit) floating-point element in "a" and "b" based on the comparison operand specified by "imm8", and store the result in mask vector "k". + Integer + AVX512VL + AVX512DQ + Convert + + + + + Convert packed single-precision (32-bit) floating-point elements in "a" to packed unsigned 64-bit integers with truncation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -CASE (imm8[7:0]) OF -0: OP := _CMP_EQ_OQ -1: OP := _CMP_LT_OS -2: OP := _CMP_LE_OS -3: OP := _CMP_UNORD_Q -4: OP := _CMP_NEQ_UQ -5: OP := _CMP_NLT_US -6: OP := _CMP_NLE_US -7: OP := _CMP_ORD_Q -8: OP := _CMP_EQ_UQ -9: OP := _CMP_NGE_US -10: OP := _CMP_NGT_US -11: OP := _CMP_FALSE_OQ -12: OP := _CMP_NEQ_OQ -13: OP := _CMP_GE_OS -14: OP := _CMP_GT_OS -15: OP := _CMP_TRUE_UQ -16: OP := _CMP_EQ_OS -17: OP := _CMP_LT_OQ -18: OP := _CMP_LE_OQ -19: OP := _CMP_UNORD_S -20: OP := _CMP_NEQ_US -21: OP := _CMP_NLT_UQ -22: OP := _CMP_NLE_UQ -23: OP := _CMP_ORD_S -24: OP := _CMP_EQ_US -25: OP := _CMP_NGE_UQ -26: OP := _CMP_NGT_UQ -27: OP := _CMP_FALSE_OS -28: OP := _CMP_NEQ_OS -29: OP := _CMP_GE_OQ -30: OP := _CMP_GT_OQ -31: OP := _CMP_TRUE_US -ESAC - -k[0] := ( a[31:0] OP b[31:0] ) ? 1 : 0 - -k[MAX:1] := 0 +FOR j := 0 to 3 + i := j*64 + l := j*32 + IF k[j] + dst[i+63:i] := Convert_FP32_To_UInt64_Truncate(a[l+31:l]) + ELSE + dst[i+63:i] := src[i+63:i] + FI +ENDFOR +dst[MAX:256] := 0 - +
immintrin.h
 - + Floating Point - Mask - AVX512F - Compare - - - - - - Compare the lower single-precision (32-bit) floating-point element in "a" and "b" based on the comparison operand specified by "imm8", and store the result in mask vector "k" using zeromask "k1" (the element is zeroed out when mask bit 0 is not set). - Pass __MM_FROUND_NO_EXC to "sae" to suppress all exceptions. + Integer + AVX512VL + AVX512DQ + Convert + + + + Convert packed single-precision (32-bit) floating-point elements in "a" to packed unsigned 64-bit integers with truncation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -CASE (imm8[7:0]) OF -0: OP := _CMP_EQ_OQ -1: OP := _CMP_LT_OS -2: OP := _CMP_LE_OS -3: OP := _CMP_UNORD_Q -4: OP := _CMP_NEQ_UQ -5: OP := _CMP_NLT_US -6: OP := _CMP_NLE_US -7: OP := _CMP_ORD_Q -8: OP := _CMP_EQ_UQ -9: OP := _CMP_NGE_US -10: OP := _CMP_NGT_US -11: OP := _CMP_FALSE_OQ -12: OP := _CMP_NEQ_OQ -13: OP := _CMP_GE_OS -14: OP := _CMP_GT_OS -15: OP := _CMP_TRUE_UQ -16: OP := _CMP_EQ_OS -17: OP := _CMP_LT_OQ -18: OP := _CMP_LE_OQ -19: OP := _CMP_UNORD_S -20: OP := _CMP_NEQ_US -21: OP := _CMP_NLT_UQ -22: OP := _CMP_NLE_UQ -23: OP := _CMP_ORD_S -24: OP := _CMP_EQ_US -25: OP := _CMP_NGE_UQ -26: OP := _CMP_NGT_UQ -27: OP := _CMP_FALSE_OS -28: OP := _CMP_NEQ_OS -29: OP := _CMP_GE_OQ -30: OP := _CMP_GT_OQ -31: OP := _CMP_TRUE_US -ESAC - -IF k1[0] - k[0] := ( a[31:0] OP b[31:0] ) ? 1 : 0 -ELSE - k[0] := 0 -FI -k[MAX:1] := 0 +FOR j := 0 to 3 + i := j*64 + l := j*32 + IF k[j] + dst[i+63:i] := Convert_FP32_To_UInt64_Truncate(a[l+31:l]) + ELSE + dst[i+63:i] := 0 + FI +ENDFOR +dst[MAX:256] := 0 - +
immintrin.h
 - + Floating Point - Mask - AVX512F - Compare - - - - - Compare the lower single-precision (32-bit) floating-point element in "a" and "b" based on the comparison operand specified by "imm8", and store the result in mask vector "k" using zeromask "k1" (the element is zeroed out when mask bit 0 is not set). + Integer + AVX512DQ + Convert + + + + Convert packed single-precision (32-bit) floating-point elements in "a" to packed unsigned 64-bit integers with truncation, and store the results in "dst". [sae_note] -CASE (imm8[7:0]) OF -0: OP := _CMP_EQ_OQ -1: OP := _CMP_LT_OS -2: OP := _CMP_LE_OS -3: OP := _CMP_UNORD_Q -4: OP := _CMP_NEQ_UQ -5: OP := _CMP_NLT_US -6: OP := _CMP_NLE_US -7: OP := _CMP_ORD_Q -8: OP := _CMP_EQ_UQ -9: OP := _CMP_NGE_US -10: OP := _CMP_NGT_US -11: OP := _CMP_FALSE_OQ -12: OP := _CMP_NEQ_OQ -13: OP := _CMP_GE_OS -14: OP := _CMP_GT_OS -15: OP := _CMP_TRUE_UQ -16: OP := _CMP_EQ_OS -17: OP := _CMP_LT_OQ -18: OP := _CMP_LE_OQ -19: OP := _CMP_UNORD_S -20: OP := _CMP_NEQ_US -21: OP := _CMP_NLT_UQ -22: OP := _CMP_NLE_UQ -23: OP := _CMP_ORD_S -24: OP := _CMP_EQ_US -25: OP := _CMP_NGE_UQ -26: OP := _CMP_NGT_UQ -27: OP := _CMP_FALSE_OS -28: OP := _CMP_NEQ_OS -29: OP := _CMP_GE_OQ -30: OP := _CMP_GT_OQ -31: OP := _CMP_TRUE_US -ESAC - -IF k1[0] - k[0] := ( a[31:0] OP b[31:0] ) ? 1 : 0 -ELSE - k[0] := 0 -FI -k[MAX:1] := 0 +FOR j := 0 to 7 + i := j*64 + l := j*32 + dst[i+63:i] := Convert_FP32_To_UInt64_Truncate(a[l+31:l]) +ENDFOR +dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - AVX512F - Compare - - - - - Compare the lower double-precision (64-bit) floating-point element in "a" and "b" based on the comparison operand specified by "imm8", and return the boolean result (0 or 1). - Pass __MM_FROUND_NO_EXC to "sae" to suppress all exceptions. + Integer + AVX512DQ + Convert + + + Convert packed single-precision (32-bit) floating-point elements in "a" to packed unsigned 64-bit integers with truncation, and store the results in "dst". -CASE (imm8[7:0]) OF -0: OP := _CMP_EQ_OQ -1: OP := _CMP_LT_OS -2: OP := _CMP_LE_OS -3: OP := _CMP_UNORD_Q -4: OP := _CMP_NEQ_UQ -5: OP := _CMP_NLT_US -6: OP := _CMP_NLE_US -7: OP := _CMP_ORD_Q -8: OP := _CMP_EQ_UQ -9: OP := _CMP_NGE_US -10: OP := _CMP_NGT_US -11: OP := _CMP_FALSE_OQ -12: OP := _CMP_NEQ_OQ -13: OP := _CMP_GE_OS -14: OP := _CMP_GT_OS -15: OP := _CMP_TRUE_UQ -16: OP := _CMP_EQ_OS -17: OP := _CMP_LT_OQ -18: OP := _CMP_LE_OQ -19: OP := _CMP_UNORD_S -20: OP := _CMP_NEQ_US -21: OP := _CMP_NLT_UQ -22: OP := _CMP_NLE_UQ -23: OP := _CMP_ORD_S -24: OP := _CMP_EQ_US -25: OP := _CMP_NGE_UQ -26: OP := _CMP_NGT_UQ -27: OP := _CMP_FALSE_OS -28: OP := _CMP_NEQ_OS -29: OP := _CMP_GE_OQ -30: OP := _CMP_GT_OQ -31: OP := _CMP_TRUE_US -ESAC - -RETURN ( a[63:0] OP b[63:0] ) ? 1 : 0 +FOR j := 0 to 7 + i := j*64 + l := j*32 + dst[i+63:i] := Convert_FP32_To_UInt64_Truncate(a[l+31:l]) +ENDFOR +dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - AVX512F - Compare - - - - - Compare the lower single-precision (32-bit) floating-point element in "a" and "b" based on the comparison operand specified by "imm8", and return the boolean result (0 or 1). - Pass __MM_FROUND_NO_EXC to "sae" to suppress all exceptions. + Integer + AVX512DQ + Convert + + + + + + Convert packed single-precision (32-bit) floating-point elements in "a" to packed unsigned 64-bit integers with truncation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). [sae_note] -CASE (imm8[7:0]) OF -0: OP := _CMP_EQ_OQ -1: OP := _CMP_LT_OS -2: OP := _CMP_LE_OS -3: OP := _CMP_UNORD_Q -4: OP := _CMP_NEQ_UQ -5: OP := _CMP_NLT_US -6: OP := _CMP_NLE_US -7: OP := _CMP_ORD_Q -8: OP := _CMP_EQ_UQ -9: OP := _CMP_NGE_US -10: OP := _CMP_NGT_US -11: OP := _CMP_FALSE_OQ -12: OP := _CMP_NEQ_OQ -13: OP := _CMP_GE_OS -14: OP := _CMP_GT_OS -15: OP := _CMP_TRUE_UQ -16: OP := _CMP_EQ_OS -17: OP := _CMP_LT_OQ -18: OP := _CMP_LE_OQ -19: OP := _CMP_UNORD_S -20: OP := _CMP_NEQ_US -21: OP := _CMP_NLT_UQ -22: OP := _CMP_NLE_UQ -23: OP := _CMP_ORD_S -24: OP := _CMP_EQ_US -25: OP := _CMP_NGE_UQ -26: OP := _CMP_NGT_UQ -27: OP := _CMP_FALSE_OS -28: OP := _CMP_NEQ_OS -29: OP := _CMP_GE_OQ -30: OP := _CMP_GT_OQ -31: OP := _CMP_TRUE_US -ESAC - -RETURN ( a[31:0] OP b[31:0] ) ? 1 : 0 +FOR j := 0 to 7 + i := j*64 + l := j*32 + IF k[j] + dst[i+63:i] := Convert_FP32_To_UInt64_Truncate(a[l+31:l]) + ELSE + dst[i+63:i] := src[i+63:i] + FI +ENDFOR +dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - AVX512F - Swizzle - - - - Contiguously store the active double-precision (64-bit) floating-point elements in "a" (those with their respective bit set in writemask "k") to "dst", and pass through the remaining elements from "src". + Integer + AVX512DQ + Convert + + + + + Convert packed single-precision (32-bit) floating-point elements in "a" to packed unsigned 64-bit integers with truncation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -size := 64 -m := 0 FOR j := 0 to 7 i := j*64 + l := j*32 IF k[j] - dst[m+size-1:m] := a[i+63:i] - m := m + size + dst[i+63:i] := Convert_FP32_To_UInt64_Truncate(a[l+31:l]) + ELSE + dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[511:m] := src[511:m] dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - AVX512F - Store - Swizzle - - - - Contiguously store the active double-precision (64-bit) floating-point elements in "a" (those with their respective bit set in writemask "k") to unaligned memory at "base_addr". + Integer + AVX512DQ + Convert + + + + + Convert packed single-precision (32-bit) floating-point elements in "a" to packed unsigned 64-bit integers with truncation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). [sae_note] -size := 64 -m := base_addr FOR j := 0 to 7 i := j*64 + l := j*32 IF k[j] - MEM[m+size-1:m] := a[i+63:i] - m := m + size + dst[i+63:i] := Convert_FP32_To_UInt64_Truncate(a[l+31:l]) + ELSE + dst[i+63:i] := 0 FI ENDFOR +dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - AVX512F - Swizzle - - - Contiguously store the active double-precision (64-bit) floating-point elements in "a" (those with their respective bit set in zeromask "k") to "dst", and set the remaining elements to zero. + Integer + AVX512DQ + Convert + + + + Convert packed single-precision (32-bit) floating-point elements in "a" to packed unsigned 64-bit integers with truncation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -size := 64 -m := 0 FOR j := 0 to 7 i := j*64 + l := j*32 IF k[j] - dst[m+size-1:m] := a[i+63:i] - m := m + size + dst[i+63:i] := Convert_FP32_To_UInt64_Truncate(a[l+31:l]) + ELSE + dst[i+63:i] := 0 FI ENDFOR -dst[511:m] := 0 dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - AVX512F - Swizzle - - - - Contiguously store the active single-precision (32-bit) floating-point elements in "a" (those with their respective bit set in writemask "k") to "dst", and pass through the remaining elements from "src". + Integer + AVX512VL + AVX512DQ + Convert + + + Convert packed single-precision (32-bit) floating-point elements in "a" to packed unsigned 64-bit integers with truncation, and store the results in "dst". -size := 32 -m := 0 -FOR j := 0 to 15 - i := j*32 - IF k[j] - dst[m+size-1:m] := a[i+31:i] - m := m + size - FI +FOR j := 0 to 1 + i := j*64 + l := j*32 + dst[i+63:i] := Convert_FP32_To_UInt64_Truncate(a[l+31:l]) ENDFOR -dst[511:m] := src[511:m] -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point - AVX512F - Store - Swizzle - - - - Contiguously store the active single-precision (32-bit) floating-point elements in "a" (those with their respective bit set in writemask "k") to unaligned memory at "base_addr". + Integer + AVX512VL + AVX512DQ + Convert + + + + + Convert packed single-precision (32-bit) floating-point elements in "a" to packed unsigned 64-bit integers with truncation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -size := 32 -m := base_addr -FOR j := 0 to 15 - i := j*32 +FOR j := 0 to 1 + i := j*64 + l := j*32 IF k[j] - MEM[m+size-1:m] := a[i+31:i] - m := m + size + dst[i+63:i] := Convert_FP32_To_UInt64_Truncate(a[l+31:l]) + ELSE + dst[i+63:i] := src[i+63:i] FI ENDFOR +dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point - AVX512F - Swizzle - - - Contiguously store the active single-precision (32-bit) floating-point elements in "a" (those with their respective bit set in zeromask "k") to "dst", and set the remaining elements to zero. + Integer + AVX512VL + AVX512DQ + Convert + + + + Convert packed single-precision (32-bit) floating-point elements in "a" to packed unsigned 64-bit integers with truncation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -size := 32 -m := 0 -FOR j := 0 to 15 - i := j*32 +FOR j := 0 to 1 + i := j*64 + l := j*32 IF k[j] - dst[m+size-1:m] := a[i+31:i] - m := m + size + dst[i+63:i] := Convert_FP32_To_UInt64_Truncate(a[l+31:l]) + ELSE + dst[i+63:i] := 0 FI ENDFOR -dst[511:m] := 0 -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point Integer - AVX512F + AVX512VL + AVX512DQ Convert - - Convert packed 32-bit integers in "a" to packed double-precision (64-bit) floating-point elements, and store the results in "dst". + + + Convert packed unsigned 64-bit integers in "a" to packed double-precision (64-bit) floating-point elements, and store the results in "dst". -FOR j := 0 to 7 - i := j*32 - m := j*64 - dst[m+63:m] := Convert_Int32_To_FP64(a[i+31:i]) +FOR j := 0 to 3 + i := j*64 + dst[i+63:i] := Convert_Int64_To_FP64(a[i+63:i]) ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Floating Point Integer - AVX512F + AVX512VL + AVX512DQ Convert - - - - Convert packed 32-bit integers in "a" to packed double-precision (64-bit) floating-point elements, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + + + + Convert packed unsigned 64-bit integers in "a" to packed double-precision (64-bit) floating-point elements, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := j*32 - m := j*64 +FOR j := 0 to 3 + i := j*64 IF k[j] - dst[m+63:m] := Convert_Int32_To_FP64(a[i+31:i]) + dst[i+63:i] := Convert_Int64_To_FP64(a[i+63:i]) ELSE - dst[m+63:m] := src[m+63:m] + dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Floating Point Integer - AVX512F + AVX512VL + AVX512DQ Convert - - - Convert packed 32-bit integers in "a" to packed double-precision (64-bit) floating-point elements, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + + + Convert packed unsigned 64-bit integers in "a" to packed double-precision (64-bit) floating-point elements, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := j*32 - m := j*64 +FOR j := 0 to 3 + i := j*64 IF k[j] - dst[m+63:m] := Convert_Int32_To_FP64(a[i+31:i]) + dst[i+63:i] := Convert_Int64_To_FP64(a[i+63:i]) ELSE - dst[m+63:m] := 0 + dst[i+63:i] := 0 FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Floating Point Integer - AVX512F + AVX512DQ Convert - - - Convert packed 32-bit integers in "a" to packed single-precision (32-bit) floating-point elements, and store the results in "dst". - [round_note] - + + + + Convert packed unsigned 64-bit integers in "a" to packed double-precision (64-bit) floating-point elements, and store the results in "dst". + [round_note] -FOR j := 0 to 15 - i := 32*j - dst[i+31:i] := Convert_Int32_To_FP32(a[i+31:i]) +FOR j := 0 to 7 + i := j*64 + dst[i+63:i] := Convert_Int64_To_FP64(a[i+63:i]) ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point Integer - AVX512F + AVX512DQ Convert - - Convert packed 32-bit integers in "a" to packed single-precision (32-bit) floating-point elements, and store the results in "dst". + + + Convert packed unsigned 64-bit integers in "a" to packed double-precision (64-bit) floating-point elements, and store the results in "dst". -FOR j := 0 to 15 - i := 32*j - dst[i+31:i] := Convert_Int32_To_FP32(a[i+31:i]) +FOR j := 0 to 7 + i := j*64 + dst[i+63:i] := Convert_Int64_To_FP64(a[i+63:i]) ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point Integer - AVX512F + AVX512DQ Convert - - - - - Convert packed 32-bit integers in "a" to packed single-precision (32-bit) floating-point elements, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - [round_note] - + + + + + + Convert packed unsigned 64-bit integers in "a" to packed double-precision (64-bit) floating-point elements, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + [round_note] -FOR j := 0 to 15 - i := j*32 +FOR j := 0 to 7 + i := j*64 IF k[j] - dst[i+31:i] := Convert_Int32_To_FP32(a[i+31:i]) + dst[i+63:i] := Convert_Int64_To_FP64(a[i+63:i]) ELSE - dst[i+31:i] := src[i+31:i] + dst[i+63:i] := src[i+63:i] FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point Integer - AVX512F + AVX512DQ Convert - - - - Convert packed 32-bit integers in "a" to packed single-precision (32-bit) floating-point elements, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + + + + Convert packed unsigned 64-bit integers in "a" to packed double-precision (64-bit) floating-point elements, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*32 +FOR j := 0 to 7 + i := j*64 IF k[j] - dst[i+31:i] := Convert_Int32_To_FP32(a[i+31:i]) + dst[i+63:i] := Convert_Int64_To_FP64(a[i+63:i]) ELSE - dst[i+31:i] := src[i+31:i] + dst[i+63:i] := src[i+63:i] FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point Integer - AVX512F + AVX512DQ Convert - - - - Convert packed 32-bit integers in "a" to packed single-precision (32-bit) floating-point elements, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - [round_note] - + + + + + Convert packed unsigned 64-bit integers in "a" to packed double-precision (64-bit) floating-point elements, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + [round_note] -FOR j := 0 to 15 - i := 32*j +FOR j := 0 to 7 + i := j*64 IF k[j] - dst[i+31:i] := Convert_Int32_To_FP32(a[i+31:i]) + dst[i+63:i] := Convert_Int64_To_FP64(a[i+63:i]) ELSE - dst[i+31:i] := 0 + dst[i+63:i] := 0 FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point Integer - AVX512F + AVX512DQ Convert - - - Convert packed 32-bit integers in "a" to packed single-precision (32-bit) floating-point elements, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + + + Convert packed unsigned 64-bit integers in "a" to packed double-precision (64-bit) floating-point elements, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := 32*j +FOR j := 0 to 7 + i := j*64 IF k[j] - dst[i+31:i] := Convert_Int32_To_FP32(a[i+31:i]) + dst[i+63:i] := Convert_Int64_To_FP64(a[i+63:i]) ELSE - dst[i+31:i] := 0 + dst[i+63:i] := 0 FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point Integer - AVX512F + AVX512VL + AVX512DQ Convert - - - Convert packed double-precision (64-bit) floating-point elements in "a" to packed 32-bit integers, and store the results in "dst". - [round_note] - + + + Convert packed unsigned 64-bit integers in "a" to packed double-precision (64-bit) floating-point elements, and store the results in "dst". -FOR j := 0 to 7 - i := 32*j - k := 64*j - dst[i+31:i] := Convert_FP64_To_Int32(a[k+63:k]) +FOR j := 0 to 1 + i := j*64 + dst[i+63:i] := Convert_Int64_To_FP64(a[i+63:i]) ENDFOR -dst[MAX:256] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point Integer - AVX512F + AVX512VL + AVX512DQ Convert - - Convert packed double-precision (64-bit) floating-point elements in "a" to packed 32-bit integers, and store the results in "dst". + + + + + Convert packed unsigned 64-bit integers in "a" to packed double-precision (64-bit) floating-point elements, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := 32*j - k := 64*j - dst[i+31:i] := Convert_FP64_To_Int32(a[k+63:k]) +FOR j := 0 to 1 + i := j*64 + IF k[j] + dst[i+63:i] := Convert_Int64_To_FP64(a[i+63:i]) + ELSE + dst[i+63:i] := src[i+63:i] + FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point Integer - AVX512F + AVX512VL + AVX512DQ Convert - - - - - Convert packed double-precision (64-bit) floating-point elements in "a" to packed 32-bit integers, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - [round_note] - + + + + Convert packed unsigned 64-bit integers in "a" to packed double-precision (64-bit) floating-point elements, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := j*32 - l := j*64 +FOR j := 0 to 1 + i := j*64 IF k[j] - dst[i+31:i] := Convert_FP64_To_Int32(a[l+63:l]) + dst[i+63:i] := Convert_Int64_To_FP64(a[i+63:i]) ELSE - dst[i+31:i] := src[i+31:i] + dst[i+63:i] := 0 FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point Integer - AVX512F + AVX512VL + AVX512DQ Convert - - - - Convert packed double-precision (64-bit) floating-point elements in "a" to packed 32-bit integers, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + + Convert packed unsigned 64-bit integers in "a" to packed single-precision (32-bit) floating-point elements, and store the results in "dst". -FOR j := 0 to 7 - i := j*32 - l := j*64 - IF k[j] - dst[i+31:i] := Convert_FP64_To_Int32(a[l+63:l]) - ELSE - dst[i+31:i] := src[i+31:i] - FI +FOR j := 0 to 3 + i := j*64 + l := j*32 + dst[l+31:l] := Convert_Int64_To_FP32(a[i+63:i]) ENDFOR -dst[MAX:256] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point Integer - AVX512F + AVX512VL + AVX512DQ Convert - - - - Convert packed double-precision (64-bit) floating-point elements in "a" to packed 32-bit integers, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - [round_note] - + + + + + Convert packed unsigned 64-bit integers in "a" to packed single-precision (32-bit) floating-point elements, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := 32*j - l := 64*j +FOR j := 0 to 3 + i := j*64 + l := j*32 IF k[j] - dst[i+31:i] := Convert_FP64_To_Int32(a[l+63:l]) + dst[l+31:l] := Convert_Int64_To_FP32(a[i+63:i]) ELSE - dst[i+31:i] := 0 + dst[l+31:l] := src[l+31:l] FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point Integer - AVX512F + AVX512VL + AVX512DQ Convert - - - Convert packed double-precision (64-bit) floating-point elements in "a" to packed 32-bit integers, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + + + Convert packed unsigned 64-bit integers in "a" to packed single-precision (32-bit) floating-point elements, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := 32*j - l := 64*j +FOR j := 0 to 3 + i := j*64 + l := j*32 IF k[j] - dst[i+31:i] := Convert_FP64_To_Int32(a[l+63:l]) + dst[l+31:l] := Convert_Int64_To_FP32(a[i+63:i]) ELSE - dst[i+31:i] := 0 + dst[l+31:l] := 0 FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point - AVX512F + Integer + AVX512DQ Convert - - - Convert packed double-precision (64-bit) floating-point elements in "a" to packed single-precision (32-bit) floating-point elements, and store the results in "dst". - [round_note] - + + + + Convert packed unsigned 64-bit integers in "a" to packed single-precision (32-bit) floating-point elements, and store the results in "dst". + [round_note] FOR j := 0 to 7 - i := 32*j - k := 64*j - dst[i+31:i] := Convert_FP64_To_FP32(a[k+63:k]) + i := j*64 + l := j*32 + dst[l+31:l] := Convert_Int64_To_FP32(a[i+63:i]) ENDFOR dst[MAX:256] := 0 - +
immintrin.h
 - + Floating Point - AVX512F + Integer + AVX512DQ Convert - - Convert packed double-precision (64-bit) floating-point elements in "a" to packed single-precision (32-bit) floating-point elements, and store the results in "dst". + + + Convert packed unsigned 64-bit integers in "a" to packed single-precision (32-bit) floating-point elements, and store the results in "dst". FOR j := 0 to 7 - i := 32*j - k := 64*j - dst[i+31:i] := Convert_FP64_To_FP32(a[k+63:k]) + i := j*64 + l := j*32 + dst[l+31:l] := Convert_Int64_To_FP32(a[i+63:i]) ENDFOR dst[MAX:256] := 0 - +
immintrin.h
 - + Floating Point - AVX512F + Integer + AVX512DQ Convert - - - - - Convert packed double-precision (64-bit) floating-point elements in "a" to packed single-precision (32-bit) floating-point elements, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - [round_note] - + + + + + + Convert packed unsigned 64-bit integers in "a" to packed single-precision (32-bit) floating-point elements, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + [round_note] FOR j := 0 to 7 - i := j*32 - l := j*64 + i := j*64 + l := j*32 IF k[j] - dst[i+31:i] := Convert_FP64_To_FP32(a[l+63:l]) + dst[l+31:l] := Convert_Int64_To_FP32(a[i+63:i]) ELSE - dst[i+31:i] := src[i+31:i] + dst[l+31:l] := src[l+31:l] FI ENDFOR dst[MAX:256] := 0 - +
immintrin.h
 - + Floating Point - AVX512F + Integer + AVX512DQ Convert - - - - Convert packed double-precision (64-bit) floating-point elements in "a" to packed single-precision (32-bit) floating-point elements, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + + + + Convert packed unsigned 64-bit integers in "a" to packed single-precision (32-bit) floating-point elements, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). FOR j := 0 to 7 - i := 32*j - l := 64*j + i := j*64 + l := j*32 IF k[j] - dst[i+31:i] := Convert_FP64_To_FP32(a[l+63:l]) + dst[l+31:l] := Convert_Int64_To_FP32(a[i+63:i]) ELSE - dst[i+31:i] := src[i+31:i] + dst[l+31:l] := src[l+31:l] FI ENDFOR dst[MAX:256] := 0 - +
immintrin.h
 - + Floating Point - AVX512F + Integer + AVX512DQ Convert - - - - Convert packed double-precision (64-bit) floating-point elements in "a" to packed single-precision (32-bit) floating-point elements, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - [round_note] - + + + + + Convert packed unsigned 64-bit integers in "a" to packed single-precision (32-bit) floating-point elements, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + [round_note] FOR j := 0 to 7 - i := j*32 - l := j*64 + i := j*64 + l := j*32 IF k[j] - dst[i+31:i] := Convert_FP64_To_FP32(a[l+63:l]) + dst[l+31:l] := Convert_Int64_To_FP32(a[i+63:i]) ELSE - dst[i+31:i] := 0 + dst[l+31:l] := 0 FI ENDFOR dst[MAX:256] := 0 - +
immintrin.h
 - + Floating Point - AVX512F + Integer + AVX512DQ Convert - - - Convert packed double-precision (64-bit) floating-point elements in "a" to packed single-precision (32-bit) floating-point elements, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + + + Convert packed unsigned 64-bit integers in "a" to packed single-precision (32-bit) floating-point elements, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). FOR j := 0 to 7 - i := j*32 - l := j*64 + i := j*64 + l := j*32 IF k[j] - dst[i+31:i] := Convert_FP64_To_FP32(a[l+63:l]) + dst[l+31:l] := Convert_Int64_To_FP32(a[i+63:i]) ELSE - dst[i+31:i] := 0 + dst[l+31:l] := 0 FI ENDFOR dst[MAX:256] := 0 - +
immintrin.h
 - + Floating Point Integer - AVX512F + AVX512VL + AVX512DQ Convert - - - Convert packed double-precision (64-bit) floating-point elements in "a" to packed unsigned 32-bit integers, and store the results in "dst". - [round_note] - + + + Convert packed unsigned 64-bit integers in "a" to packed single-precision (32-bit) floating-point elements, and store the results in "dst". -FOR j := 0 to 7 - i := 32*j - k := 64*j - dst[i+31:i] := Convert_FP64_To_UnsignedInt32(a[k+63:k]) +FOR j := 0 to 1 + i := j*64 + l := j*32 + dst[l+31:l] := Convert_Int64_To_FP32(a[i+63:i]) ENDFOR -dst[MAX:256] := 0 +dst[MAX:64] := 0 - +
immintrin.h
 - + Floating Point Integer - AVX512F + AVX512VL + AVX512DQ Convert - - Convert packed double-precision (64-bit) floating-point elements in "a" to packed unsigned 32-bit integers, and store the results in "dst". + + + + + Convert packed unsigned 64-bit integers in "a" to packed single-precision (32-bit) floating-point elements, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := 32*j - k := 64*j - dst[i+31:i] := Convert_FP64_To_UnsignedInt32(a[k+63:k]) +FOR j := 0 to 1 + i := j*64 + l := j*32 + IF k[j] + dst[l+31:l] := Convert_Int64_To_FP32(a[i+63:i]) + ELSE + dst[l+31:l] := src[l+31:l] + FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:64] := 0 - +
immintrin.h
 - + Floating Point Integer - AVX512F + AVX512VL + AVX512DQ Convert - - - - - Convert packed double-precision (64-bit) floating-point elements in "a" to packed unsigned 32-bit integers, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - [round_note] - + + + + Convert packed unsigned 64-bit integers in "a" to packed single-precision (32-bit) floating-point elements, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := j*32 - l := j*64 +FOR j := 0 to 1 + i := j*64 + l := j*32 IF k[j] - dst[i+31:i] := Convert_FP64_To_UnsignedInt32(a[l+63:l]) + dst[l+31:l] := Convert_Int64_To_FP32(a[i+63:i]) ELSE - dst[i+31:i] := src[i+31:i] + dst[l+31:l] := 0 FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:64] := 0 - +
immintrin.h
 - + Floating Point - Integer - AVX512F - Convert - - - - Convert packed double-precision (64-bit) floating-point elements in "a" to packed unsigned 32-bit integers, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + AVX512DQ + Miscellaneous + + + + Extract 256 bits (composed of 8 packed single-precision (32-bit) floating-point elements) from "a", selected with "imm8", and store the result in "dst". -FOR j := 0 to 7 - i := j*32 - l := j*64 - IF k[j] - dst[i+31:i] := Convert_FP64_To_UnsignedInt32(a[l+63:l]) - ELSE - dst[i+31:i] := src[i+31:i] - FI -ENDFOR +CASE imm8[0] OF +0: dst[255:0] := a[255:0] +1: dst[255:0] := a[511:256] +ESAC dst[MAX:256] := 0 - +
immintrin.h
 - + Floating Point - Integer - AVX512F - Convert - - - - Convert packed double-precision (64-bit) floating-point elements in "a" to packed unsigned 32-bit integers, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - [round_note] - + AVX512DQ + Miscellaneous + + + + + + Extract 256 bits (composed of 8 packed single-precision (32-bit) floating-point elements) from "a", selected with "imm8", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). +CASE imm8[0] OF +0: tmp[255:0] := a[255:0] +1: tmp[255:0] := a[511:256] +ESAC FOR j := 0 to 7 - i := 32*j - l := 64*j + i := j*32 IF k[j] - dst[i+31:i] := Convert_FP64_To_UnsignedInt32(a[l+63:l]) + dst[i+31:i] := tmp[i+31:i] ELSE - dst[i+31:i] := 0 + dst[i+31:i] := src[i+31:i] FI ENDFOR dst[MAX:256] := 0 - +
immintrin.h
 - + Floating Point - Integer - AVX512F - Convert - - - Convert packed double-precision (64-bit) floating-point elements in "a" to packed unsigned 32-bit integers, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + AVX512DQ + Miscellaneous + + + + + Extract 256 bits (composed of 8 packed single-precision (32-bit) floating-point elements) from "a", selected with "imm8", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). +CASE imm8[0] OF +0: tmp[255:0] := a[255:0] +1: tmp[255:0] := a[511:256] +ESAC FOR j := 0 to 7 - i := 32*j - l := 64*j + i := j*32 IF k[j] - dst[i+31:i] := Convert_FP64_To_UnsignedInt32(a[l+63:l]) + dst[i+31:i] := tmp[i+31:i] ELSE dst[i+31:i] := 0 FI ENDFOR dst[MAX:256] := 0 - +
immintrin.h
 - + Floating Point - AVX512F - Convert - - - Convert packed half-precision (16-bit) floating-point elements in "a" to packed single-precision (32-bit) floating-point elements, and store the results in "dst". - Pass __MM_FROUND_NO_EXC to "sae" to suppress all exceptions. - + AVX512VL + AVX512DQ + Miscellaneous + + + + Extract 128 bits (composed of 2 packed double-precision (64-bit) floating-point elements) from "a", selected with "imm8", and store the result in "dst". -FOR j := 0 to 15 - i := j*32 - m := j*16 - dst[i+31:i] := Convert_FP16_To_FP32(a[m+15:m]) -ENDFOR -dst[MAX:512] := 0 +CASE imm8[0] OF +0: dst[127:0] := a[127:0] +1: dst[127:0] := a[255:128] +ESAC +dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point - AVX512F - Convert - - Convert packed half-precision (16-bit) floating-point elements in "a" to packed single-precision (32-bit) floating-point elements, and store the results in "dst". + AVX512VL + AVX512DQ + Miscellaneous + + + + + + Extract 128 bits (composed of 2 packed double-precision (64-bit) floating-point elements) from "a", selected with "imm8", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*32 - m := j*16 - dst[i+31:i] := Convert_FP16_To_FP32(a[m+15:m]) +CASE imm8[0] OF +0: tmp[127:0] := a[127:0] +1: tmp[127:0] := a[255:128] +ESAC +FOR j := 0 to 1 + i := j*64 + IF k[j] + dst[i+63:i] := tmp[i+63:i] + ELSE + dst[i+63:i] := src[i+63:i] + FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point - AVX512F - Convert - - - - - Convert packed half-precision (16-bit) floating-point elements in "a" to packed single-precision (32-bit) floating-point elements, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - Pass __MM_FROUND_NO_EXC to "sae" to suppress all exceptions. + AVX512VL + AVX512DQ + Miscellaneous + + + + + Extract 128 bits (composed of 2 packed double-precision (64-bit) floating-point elements) from "a", selected with "imm8", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*32 - m := j*16 +CASE imm8[0] OF +0: tmp[127:0] := a[127:0] +1: tmp[127:0] := a[255:128] +ESAC +FOR j := 0 to 1 + i := j*64 IF k[j] - dst[i+31:i] := Convert_FP16_To_FP32(a[m+15:m]) + dst[i+63:i] := tmp[i+63:i] ELSE - dst[i+31:i] := src[i+31:i] + dst[i+63:i] := 0 FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point - AVX512F - Convert - - - - Convert packed half-precision (16-bit) floating-point elements in "a" to packed single-precision (32-bit) floating-point elements, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + AVX512DQ + Miscellaneous + + + + Extract 128 bits (composed of 2 packed double-precision (64-bit) floating-point elements) from "a", selected with "imm8", and store the result in "dst". -FOR j := 0 to 15 - i := j*32 - m := j*16 - IF k[j] - dst[i+31:i] := Convert_FP16_To_FP32(a[m+15:m]) - ELSE - dst[i+31:i] := src[i+31:i] - FI -ENDFOR -dst[MAX:512] := 0 +CASE imm8[1:0] OF +0: dst[127:0] := a[127:0] +1: dst[127:0] := a[255:128] +2: dst[127:0] := a[383:256] +3: dst[127:0] := a[511:384] +ESAC +dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point - AVX512F - Convert - - - - Convert packed half-precision (16-bit) floating-point elements in "a" to packed single-precision (32-bit) floating-point elements, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - Pass __MM_FROUND_NO_EXC to "sae" to suppress all exceptions. - + AVX512DQ + Miscellaneous + + + + + + Extract 128 bits (composed of 2 packed double-precision (64-bit) floating-point elements) from "a", selected with "imm8", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*32 - m := j*16 +CASE imm8[1:0] OF +0: tmp[127:0] := a[127:0] +1: tmp[127:0] := a[255:128] +2: tmp[127:0] := a[383:256] +3: tmp[127:0] := a[511:384] +ESAC +FOR j := 0 to 1 + i := j*64 IF k[j] - dst[i+31:i] := Convert_FP16_To_FP32(a[m+15:m]) + dst[i+63:i] := tmp[i+63:i] ELSE - dst[i+31:i] := 0 + dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point - AVX512F - Convert - - - Convert packed half-precision (16-bit) floating-point elements in "a" to packed single-precision (32-bit) floating-point elements, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - -FOR j := 0 to 15 - i := j*32 - m := j*16 + AVX512DQ + Miscellaneous + + + + + Extract 128 bits (composed of 2 packed double-precision (64-bit) floating-point elements) from "a", selected with "imm8", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + +CASE imm8[1:0] OF +0: tmp[127:0] := a[127:0] +1: tmp[127:0] := a[255:128] +2: tmp[127:0] := a[383:256] +3: tmp[127:0] := a[511:384] +ESAC +FOR j := 0 to 1 + i := j*64 IF k[j] - dst[i+31:i] := Convert_FP16_To_FP32(a[m+15:m]) + dst[i+63:i] := tmp[i+63:i] ELSE - dst[i+31:i] := 0 + dst[i+63:i] := 0 FI ENDFOR -dst[MAX:512] := 0 - - -
immintrin.h
- - - Floating Point - Integer - AVX512F - Convert - - - Convert packed single-precision (32-bit) floating-point elements in "a" to packed 32-bit integers, and store the results in "dst". - [round_note] - - -FOR j := 0 to 15 - i := 32*j - dst[i+31:i] := Convert_FP32_To_Int32(a[i+31:i]) -ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point + Integer - AVX512F - Convert - - Convert packed single-precision (32-bit) floating-point elements in "a" to packed 32-bit integers, and store the results in "dst". + AVX512DQ + Miscellaneous + + + + Extract 256 bits (composed of 8 packed 32-bit integers) from "a", selected with "imm8", and store the result in "dst". -FOR j := 0 to 15 - i := 32*j - dst[i+31:i] := Convert_FP32_To_Int32(a[i+31:i]) -ENDFOR -dst[MAX:512] := 0 +CASE imm8[0] OF +0: dst[255:0] := a[255:0] +1: dst[255:0] := a[511:256] +ESAC +dst[MAX:256] := 0 - +
immintrin.h
 - - Floating Point + Integer - AVX512F - Convert - - - - - Convert packed single-precision (32-bit) floating-point elements in "a" to packed 32-bit integers, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - [round_note] - + AVX512DQ + Miscellaneous + + + + + + Extract 256 bits (composed of 8 packed 32-bit integers) from "a", selected with "imm8", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 15 +CASE imm8[0] OF +0: tmp[255:0] := a[255:0] +1: tmp[255:0] := a[511:256] +ESAC +FOR j := 0 to 7 i := j*32 IF k[j] - dst[i+31:i] := Convert_FP32_To_Int32(a[i+31:i]) + dst[i+31:i] := tmp[i+31:i] ELSE dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - - Floating Point + Integer - AVX512F - Convert - - - - Convert packed single-precision (32-bit) floating-point elements in "a" to packed 32-bit integers, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + AVX512DQ + Miscellaneous + + + + + Extract 256 bits (composed of 8 packed 32-bit integers) from "a", selected with "imm8", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 15 +CASE imm8[0] OF +0: tmp[255:0] := a[255:0] +1: tmp[255:0] := a[511:256] +ESAC +FOR j := 0 to 7 i := j*32 IF k[j] - dst[i+31:i] := Convert_FP32_To_Int32(a[i+31:i]) + dst[i+31:i] := tmp[i+31:i] ELSE - dst[i+31:i] := src[i+31:i] + dst[i+31:i] := 0 FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - - Floating Point + Integer - AVX512F - Convert - - - - Convert packed single-precision (32-bit) floating-point elements in "a" to packed 32-bit integers, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - [round_note] - + AVX512VL + AVX512DQ + Miscellaneous + + + + Extract 128 bits (composed of 2 packed 64-bit integers) from "a", selected with "imm8", and store the result in "dst". -FOR j := 0 to 15 - i := 32*j - IF k[j] - dst[i+31:i] := Convert_FP32_To_Int32(a[i+31:i]) - ELSE - dst[i+31:i] := 0 - FI -ENDFOR -dst[MAX:512] := 0 +CASE imm8[0] OF +0: dst[127:0] := a[127:0] +1: dst[127:0] := a[255:128] +ESAC +dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point + Integer - AVX512F - Convert - - - Convert packed single-precision (32-bit) floating-point elements in "a" to packed 32-bit integers, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + AVX512VL + AVX512DQ + Miscellaneous + + + + + + Extract 128 bits (composed of 2 packed 64-bit integers) from "a", selected with "imm8", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := 32*j +CASE imm8[0] OF +0: tmp[127:0] := a[127:0] +1: tmp[127:0] := a[255:128] +ESAC +FOR j := 0 to 1 + i := j*64 IF k[j] - dst[i+31:i] := Convert_FP32_To_Int32(a[i+31:i]) + dst[i+63:i] := tmp[i+63:i] ELSE - dst[i+31:i] := 0 + dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point - AVX512F - Convert - - - Convert packed single-precision (32-bit) floating-point elements in "a" to packed double-precision (64-bit) floating-point elements, and store the results in "dst". - Pass __MM_FROUND_NO_EXC to "sae" to suppress all exceptions. - + + Integer + AVX512VL + AVX512DQ + Miscellaneous + + + + + Extract 128 bits (composed of 2 packed 64-bit integers) from "a", selected with "imm8", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := 64*j - k := 32*j - dst[i+63:i] := Convert_FP32_To_FP64(a[k+31:k]) +CASE imm8[0] OF +0: tmp[127:0] := a[127:0] +1: tmp[127:0] := a[255:128] +ESAC +FOR j := 0 to 1 + i := j*64 + IF k[j] + dst[i+63:i] := tmp[i+63:i] + ELSE + dst[i+63:i] := 0 + FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point - AVX512F - Convert - - Convert packed single-precision (32-bit) floating-point elements in "a" to packed double-precision (64-bit) floating-point elements, and store the results in "dst". + + Integer + AVX512DQ + Miscellaneous + + + + Extract 128 bits (composed of 2 packed 64-bit integers) from "a", selected with "imm8", and store the result in "dst". -FOR j := 0 to 7 - i := 64*j - k := 32*j - dst[i+63:i] := Convert_FP32_To_FP64(a[k+31:k]) -ENDFOR -dst[MAX:512] := 0 +CASE imm8[1:0] OF +0: dst[127:0] := a[127:0] +1: dst[127:0] := a[255:128] +2: dst[127:0] := a[383:256] +3: dst[127:0] := a[511:384] +ESAC +dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point - AVX512F - Convert - - - - - Convert packed single-precision (32-bit) floating-point elements in "a" to packed double-precision (64-bit) floating-point elements, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - Pass __MM_FROUND_NO_EXC to "sae" to suppress all exceptions. - + + Integer + AVX512DQ + Miscellaneous + + + + + + Extract 128 bits (composed of 2 packed 64-bit integers) from "a", selected with "imm8", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := 64*j - l := 32*j +CASE imm8[1:0] OF +0: tmp[127:0] := a[127:0] +1: tmp[127:0] := a[255:128] +2: tmp[127:0] := a[383:256] +3: tmp[127:0] := a[511:384] +ESAC +FOR j := 0 to 1 + i := j*64 IF k[j] - dst[i+63:i] := Convert_FP32_To_FP64(a[l+31:l]) + dst[i+63:i] := tmp[i+63:i] ELSE dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point - AVX512F - Convert - - - - Convert packed single-precision (32-bit) floating-point elements in "a" to packed double-precision (64-bit) floating-point elements, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Integer + AVX512DQ + Miscellaneous + + + + + Extract 128 bits (composed of 2 packed 64-bit integers) from "a", selected with "imm8", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := 64*j - l := 32*j +CASE imm8[1:0] OF +0: tmp[127:0] := a[127:0] +1: tmp[127:0] := a[255:128] +2: tmp[127:0] := a[383:256] +3: tmp[127:0] := a[511:384] +ESAC +FOR j := 0 to 1 + i := j*64 IF k[j] - dst[i+63:i] := Convert_FP32_To_FP64(a[l+31:l]) + dst[i+63:i] := tmp[i+63:i] ELSE - dst[i+63:i] := src[i+63:i] + dst[i+63:i] := 0 FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point - AVX512F - Convert - - - - Convert packed single-precision (32-bit) floating-point elements in "a" to packed double-precision (64-bit) floating-point elements, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - Pass __MM_FROUND_NO_EXC to "sae" to suppress all exceptions. - - -FOR j := 0 to 7 - i := 64*j - l := 32*j - IF k[j] - dst[i+63:i] := Convert_FP32_To_FP64(a[l+31:l]) - ELSE - dst[i+63:i] := 0 - FI + Mask + AVX512VL + AVX512DQ + Miscellaneous + + + + Test packed double-precision (64-bit) floating-point elements in "a" for special categories specified by "imm8", and store the results in mask vector "k". + [fpclass_note] + FOR j := 0 to 3 + i := j*64 + k[j] := CheckFPClass_FP64(a[i+63:i], imm8[7:0]) ENDFOR -dst[MAX:512] := 0 +k[MAX:4] := 0 - +
immintrin.h
 - + Floating Point - AVX512F - Convert - - - Convert packed single-precision (32-bit) floating-point elements in "a" to packed double-precision (64-bit) floating-point elements, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - -FOR j := 0 to 7 - i := 64*j - l := 32*j - IF k[j] - dst[i+63:i] := Convert_FP32_To_FP64(a[l+31:l]) + Mask + AVX512VL + AVX512DQ + Miscellaneous + + + + + Test packed double-precision (64-bit) floating-point elements in "a" for special categories specified by "imm8", and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). + [fpclass_note] + FOR j := 0 to 3 + i := j*64 + IF k1[j] + k[j] := CheckFPClass_FP64(a[i+63:i], imm8[7:0]) ELSE - dst[i+63:i] := 0 + k[j] := 0 FI ENDFOR -dst[MAX:512] := 0 +k[MAX:4] := 0 - +
immintrin.h
 - + Floating Point - AVX512F - Convert - - - Convert packed single-precision (32-bit) floating-point elements in "a" to packed half-precision (16-bit) floating-point elements, and store the results in "dst". - [round_note] - - -FOR j := 0 to 15 - i := 16*j - l := 32*j - dst[i+15:i] := Convert_FP32_To_FP16FP(a[l+31:l]) + Mask + AVX512DQ + Miscellaneous + + + + Test packed double-precision (64-bit) floating-point elements in "a" for special categories specified by "imm8", and store the results in mask vector "k". + [fpclass_note] + FOR j := 0 to 7 + i := j*64 + k[j] := CheckFPClass_FP64(a[i+63:i], imm8[7:0]) ENDFOR -dst[MAX:256] := 0 +k[MAX:8] := 0 - +
immintrin.h
 - + Floating Point - AVX512F - Convert - - - Convert packed single-precision (32-bit) floating-point elements in "a" to packed half-precision (16-bit) floating-point elements, and store the results in "dst". - [round_note] - - -FOR j := 0 to 15 - i := 16*j - l := 32*j - dst[i+15:i] := Convert_FP32_To_FP16FP(a[l+31:l]) + Mask + AVX512DQ + Miscellaneous + + + + + Test packed double-precision (64-bit) floating-point elements in "a" for special categories specified by "imm8", and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). + [fpclass_note] + FOR j := 0 to 7 + i := j*64 + IF k1[j] + k[j] := CheckFPClass_FP64(a[i+63:i], imm8[7:0]) + ELSE + k[j] := 0 + FI ENDFOR -dst[MAX:256] := 0 +k[MAX:8] := 0 - +
immintrin.h
 - + Floating Point - AVX512F - Convert - - - - - Convert packed single-precision (32-bit) floating-point elements in "a" to packed half-precision (16-bit) floating-point elements, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - [round_note] - - -FOR j := 0 to 15 - i := 16*j - l := 32*j - IF k[j] - dst[i+15:i] := Convert_FP32_To_FP16FP(a[l+31:l]) - ELSE - dst[i+15:i] := src[i+15:i] - FI + Mask + AVX512VL + AVX512DQ + Miscellaneous + + + + Test packed double-precision (64-bit) floating-point elements in "a" for special categories specified by "imm8", and store the results in mask vector "k". + [fpclass_note] + FOR j := 0 to 1 + i := j*64 + k[j] := CheckFPClass_FP64(a[i+63:i], imm8[7:0]) ENDFOR -dst[MAX:256] := 0 +k[MAX:2] := 0 - +
immintrin.h
 - + Floating Point - AVX512F - Convert - - - - - Convert packed single-precision (32-bit) floating-point elements in "a" to packed half-precision (16-bit) floating-point elements, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - [round_note] - - -FOR j := 0 to 15 - i := 16*j - l := 32*j - IF k[j] - dst[i+15:i] := Convert_FP32_To_FP16FP(a[l+31:l]) + Mask + AVX512VL + AVX512DQ + Miscellaneous + + + + + Test packed double-precision (64-bit) floating-point elements in "a" for special categories specified by "imm8", and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). + [fpclass_note] + FOR j := 0 to 1 + i := j*64 + IF k1[j] + k[j] := CheckFPClass_FP64(a[i+63:i], imm8[7:0]) ELSE - dst[i+15:i] := src[i+15:i] + k[j] := 0 FI ENDFOR -dst[MAX:256] := 0 +k[MAX:2] := 0 - +
immintrin.h
 - + Floating Point - AVX512F - Convert - - - - Convert packed single-precision (32-bit) floating-point elements in "a" to packed half-precision (16-bit) floating-point elements, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - [round_note] - - -FOR j := 0 to 15 - i := 16*j - l := 32*j - IF k[j] - dst[i+15:i] := Convert_FP32_To_FP16FP(a[l+31:l]) - ELSE - dst[i+15:i] := 0 - FI + Mask + AVX512VL + AVX512DQ + Miscellaneous + + + + Test packed single-precision (32-bit) floating-point elements in "a" for special categories specified by "imm8", and store the results in mask vector "k". + [fpclass_note] + FOR j := 0 to 7 + i := j*32 + k[j] := CheckFPClass_FP32(a[i+31:i], imm8[7:0]) ENDFOR -dst[MAX:256] := 0 +k[MAX:8] := 0 - +
immintrin.h
 - + Floating Point - AVX512F - Convert - - - - Convert packed single-precision (32-bit) floating-point elements in "a" to packed half-precision (16-bit) floating-point elements, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - [round_note] - - -FOR j := 0 to 15 - i := 16*j - l := 32*j - IF k[j] - dst[i+15:i] := Convert_FP32_To_FP16FP(a[l+31:l]) + Mask + AVX512VL + AVX512DQ + Miscellaneous + + + + + Test packed single-precision (32-bit) floating-point elements in "a" for special categories specified by "imm8", and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). + [fpclass_note] + FOR j := 0 to 7 + i := j*32 + IF k1[j] + k[j] := CheckFPClass_FP32(a[i+31:i], imm8[7:0]) ELSE - dst[i+15:i] := 0 + k[j] := 0 FI ENDFOR -dst[MAX:256] := 0 +k[MAX:8] := 0 - +
immintrin.h
 - + Floating Point - Integer - AVX512F - Convert - - - Convert packed single-precision (32-bit) floating-point elements in "a" to packed unsigned 32-bit integers, and store the results in "dst". - [round_note] - - -FOR j := 0 to 15 - i := 32*j - dst[i+31:i] := Convert_FP32_To_UnsignedInt32(a[i+31:i]) + Mask + AVX512DQ + Miscellaneous + + + + Test packed single-precision (32-bit) floating-point elements in "a" for special categories specified by "imm8", and store the results in mask vector "k". + [fpclass_note] + FOR j := 0 to 15 + i := j*32 + k[j] := CheckFPClass_FP32(a[i+31:i], imm8[7:0]) ENDFOR -dst[MAX:512] := 0 +k[MAX:16] := 0 - +
immintrin.h
 - + Floating Point - Integer - AVX512F - Convert - - Convert packed single-precision (32-bit) floating-point elements in "a" to packed unsigned 32-bit integers, and store the results in "dst". - -FOR j := 0 to 15 - i := 32*j - dst[i+31:i] := Convert_FP32_To_UnsignedInt32(a[i+31:i]) + Mask + AVX512DQ + Miscellaneous + + + + + Test packed single-precision (32-bit) floating-point elements in "a" for special categories specified by "imm8", and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). + [fpclass_note] + FOR j := 0 to 15 + i := j*32 + IF k1[j] + k[j] := CheckFPClass_FP32(a[i+31:i], imm8[7:0]) + ELSE + k[j] := 0 + FI ENDFOR -dst[MAX:512] := 0 +k[MAX:16] := 0 - +
immintrin.h
 - + Floating Point - Integer - AVX512F - Convert - - - - - Convert packed single-precision (32-bit) floating-point elements in "a" to packed unsigned 32-bit integers, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - [round_note] - - -FOR j := 0 to 15 + Mask + AVX512VL + AVX512DQ + Miscellaneous + + + + Test packed single-precision (32-bit) floating-point elements in "a" for special categories specified by "imm8", and store the results in mask vector "k". + [fpclass_note] + FOR j := 0 to 3 i := j*32 - IF k[j] - dst[i+31:i] := Convert_FP32_To_UnsignedInt32(a[i+31:i]) - ELSE - dst[i+31:i] := src[i+31:i] - FI + k[j] := CheckFPClass_FP32(a[i+31:i], imm8[7:0]) ENDFOR -dst[MAX:512] := 0 +k[MAX:4] := 0 - +
immintrin.h
 - + Floating Point - Integer - AVX512F - Convert - - - - Convert packed single-precision (32-bit) floating-point elements in "a" to packed unsigned 32-bit integers, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -FOR j := 0 to 15 - i := 32*j - IF k[j] - dst[i+31:i] := Convert_FP32_To_UnsignedInt32(a[i+31:i]) + Mask + AVX512VL + AVX512DQ + Miscellaneous + + + + + Test packed single-precision (32-bit) floating-point elements in "a" for special categories specified by "imm8", and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). + [fpclass_note] + FOR j := 0 to 3 + i := j*32 + IF k1[j] + k[j] := CheckFPClass_FP32(a[i+31:i], imm8[7:0]) ELSE - dst[i+31:i] := src[i+31:i] + k[j] := 0 FI ENDFOR +k[MAX:4] := 0 + + +
immintrin.h
+ + + Floating Point + Mask + AVX512DQ + Miscellaneous + + + + Test the lower double-precision (64-bit) floating-point element in "a" for special categories specified by "imm8", and store the result in mask vector "k". + [fpclass_note] + k[0] := CheckFPClass_FP64(a[63:0], imm8[7:0]) +k[MAX:1] := 0 + + +
immintrin.h
+ + + Floating Point + Mask + AVX512DQ + Miscellaneous + + + + + Test the lower double-precision (64-bit) floating-point element in "a" for special categories specified by "imm8", and store the result in mask vector "k" using zeromask "k1" (the element is zeroed out when mask bit 0 is not set). + [fpclass_note] + IF k1[0] + k[0] := CheckFPClass_FP64(a[63:0], imm8[7:0]) +ELSE + k[0] := 0 +FI +k[MAX:1] := 0 + + +
immintrin.h
+ + + Floating Point + Mask + AVX512DQ + Miscellaneous + + + + Test the lower single-precision (32-bit) floating-point element in "a" for special categories specified by "imm8", and store the result in mask vector "k. + [fpclass_note] + k[0] := CheckFPClass_FP32(a[31:0], imm8[7:0]) +k[MAX:1] := 0 + + +
immintrin.h
+ + + Floating Point + Mask + AVX512DQ + Miscellaneous + + + + + Test the lower single-precision (32-bit) floating-point element in "a" for special categories specified by "imm8", and store the result in mask vector "k" using zeromask "k1" (the element is zeroed out when mask bit 0 is not set). + [fpclass_note] + IF k1[0] + k[0] := CheckFPClass_FP32(a[31:0], imm8[7:0]) +ELSE + k[0] := 0 +FI +k[MAX:1] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512DQ + Miscellaneous + + + + + Copy "a" to "dst", then insert 256 bits (composed of 8 packed single-precision (32-bit) floating-point elements) from "b" into "dst" at the location specified by "imm8". + +dst[511:0] := a[511:0] +CASE (imm8[0]) OF +0: dst[255:0] := b[255:0] +1: dst[511:256] := b[255:0] +ESAC dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - Integer - AVX512F - Convert - - - - Convert packed single-precision (32-bit) floating-point elements in "a" to packed unsigned 32-bit integers, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - [round_note] - + AVX512DQ + Miscellaneous + + + + + + + Copy "a" to "tmp", then insert 256 bits (composed of 8 packed single-precision (32-bit) floating-point elements) from "b" into "tmp" at the location specified by "imm8". Store "tmp" to "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). +tmp[511:0] := a[511:0] +CASE (imm8[0]) OF +0: tmp[255:0] := b[255:0] +1: tmp[511:256] := b[255:0] +ESAC FOR j := 0 to 15 - i := 32*j + i := j*32 IF k[j] - dst[i+31:i] := Convert_FP32_To_UnsignedInt32(a[i+31:i]) + dst[i+31:i] := tmp[i+31:i] ELSE - dst[i+31:i] := 0 + dst[i+31:i] := src[i+31:i] FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - Integer - AVX512F - Convert - - - Convert packed single-precision (32-bit) floating-point elements in "a" to packed unsigned 32-bit integers, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + AVX512DQ + Miscellaneous + + + + + + Copy "a" to "tmp", then insert 256 bits (composed of 8 packed single-precision (32-bit) floating-point elements) from "b" into "tmp" at the location specified by "imm8". Store "tmp" to "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). +tmp[511:0] := a[511:0] +CASE (imm8[0]) OF +0: tmp[255:0] := b[255:0] +1: tmp[511:256] := b[255:0] +ESAC FOR j := 0 to 15 - i := 32*j + i := j*32 IF k[j] - dst[i+31:i] := Convert_FP32_To_UnsignedInt32(a[i+31:i]) + dst[i+31:i] := tmp[i+31:i] ELSE dst[i+31:i] := 0 FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - AVX512F - Convert - - - Convert the lower double-precision (64-bit) floating-point element in "a" to a 32-bit integer, and store the result in "dst". - [round_note] - + AVX512VL + AVX512DQ + Miscellaneous + + + + + Copy "a" to "dst", then insert 128 bits (composed of 2 packed double-precision (64-bit) floating-point elements) from "b" into "dst" at the location specified by "imm8". -dst[31:0] := Convert_FP64_To_Int32(a[63:0]) +dst[255:0] := a[255:0] +CASE imm8[0] OF +0: dst[127:0] := b[127:0] +1: dst[255:128] := b[127:0] +ESAC +dst[MAX:256] := 0 - +
immintrin.h
 - + Floating Point - AVX512F - Convert - - - Convert the lower double-precision (64-bit) floating-point element in "a" to a 64-bit integer, and store the result in "dst". - [round_note] - + AVX512VL + AVX512DQ + Miscellaneous + + + + + + + Copy "a" to "tmp", then insert 128 bits (composed of 2 packed double-precision (64-bit) floating-point elements) from "b" into "tmp" at the location specified by "imm8". Store "tmp" to "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -dst[63:0] := Convert_FP64_To_Int64(a[63:0]) +tmp[255:0] := a[255:0] +CASE (imm8[0]) OF +0: tmp[127:0] := b[127:0] +1: tmp[255:128] := b[127:0] +ESAC +FOR j := 0 to 3 + i := j*64 + IF k[j] + dst[i+63:i] := tmp[i+63:i] + ELSE + dst[i+63:i] := src[i+63:i] + FI +ENDFOR +dst[MAX:256] := 0 - +
immintrin.h
 - + Floating Point - Integer - AVX512F - Convert - - - Convert the lower double-precision (64-bit) floating-point element in "a" to a 32-bit integer, and store the result in "dst". - [round_note] - + AVX512VL + AVX512DQ + Miscellaneous + + + + + + Copy "a" to "tmp", then insert 128 bits (composed of 2 packed double-precision (64-bit) floating-point elements) from "b" into "tmp" at the location specified by "imm8". Store "tmp" to "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -dst[31:0] := Convert_FP64_To_Int32(a[63:0]) +tmp[255:0] := a[255:0] +CASE (imm8[0]) OF +0: tmp[127:0] := b[127:0] +1: tmp[255:128] := b[127:0] +ESAC +FOR j := 0 to 3 + i := j*64 + IF k[j] + dst[i+63:i] := tmp[i+63:i] + ELSE + dst[i+63:i] := 0 + FI +ENDFOR +dst[MAX:256] := 0 - +
immintrin.h
 - + Floating Point - Integer - AVX512F - Convert - - - Convert the lower double-precision (64-bit) floating-point element in "a" to a 64-bit integer, and store the result in "dst". - [round_note] - + AVX512DQ + Miscellaneous + + + + + Copy "a" to "dst", then insert 128 bits (composed of 2 packed double-precision (64-bit) floating-point elements) from "b" into "dst" at the location specified by "imm8". -dst[63:0] := Convert_FP64_To_Int64(a[63:0]) +dst[511:0] := a[511:0] +CASE imm8[1:0] OF +0: dst[127:0] := b[127:0] +1: dst[255:128] := b[127:0] +2: dst[383:256] := b[127:0] +3: dst[511:384] := b[127:0] +ESAC +dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - AVX512F - Convert - - Convert the lower double-precision (64-bit) floating-point element in "a" to a 32-bit integer, and store the result in "dst". + AVX512DQ + Miscellaneous + + + + + + + Copy "a" to "tmp", then insert 128 bits (composed of 2 packed double-precision (64-bit) floating-point elements) from "b" into "tmp" at the location specified by "imm8". Store "tmp" to "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -dst[31:0] := Convert_FP64_To_Int32(a[63:0]) +tmp[511:0] := a[511:0] +CASE (imm8[1:0]) OF +0: tmp[127:0] := b[127:0] +1: tmp[255:128] := b[127:0] +2: tmp[383:256] := b[127:0] +3: tmp[511:384] := b[127:0] +ESAC +FOR j := 0 to 7 + i := j*64 + IF k[j] + dst[i+63:i] := tmp[i+63:i] + ELSE + dst[i+63:i] := src[i+63:i] + FI +ENDFOR +dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - AVX512F - Convert - - Convert the lower double-precision (64-bit) floating-point element in "a" to a 64-bit integer, and store the result in "dst". + AVX512DQ + Miscellaneous + + + + + + Copy "a" to "tmp", then insert 128 bits (composed of 2 packed double-precision (64-bit) floating-point elements) from "b" into "tmp" at the location specified by "imm8". Store "tmp" to "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -dst[63:0] := Convert_FP64_To_Int64(a[63:0]) +tmp[511:0] := a[511:0] +CASE (imm8[1:0]) OF +0: tmp[127:0] := b[127:0] +1: tmp[255:128] := b[127:0] +2: tmp[383:256] := b[127:0] +3: tmp[511:384] := b[127:0] +ESAC +FOR j := 0 to 7 + i := j*64 + IF k[j] + dst[i+63:i] := tmp[i+63:i] + ELSE + dst[i+63:i] := 0 + FI +ENDFOR +dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point - AVX512F - Convert - - - - Convert the lower double-precision (64-bit) floating-point element in "b" to a single-precision (32-bit) floating-point element, store the result in the lower element of "dst", and copy the upper 3 packed elements from "a" to the upper elements of "dst". - [round_note] - + + AVX512DQ + Miscellaneous + + + + + Copy "a" to "dst", then insert 256 bits (composed of 8 packed 32-bit integers) from "b" into "dst" at the location specified by "imm8". -dst[31:0] := Convert_FP64_To_FP32(b[63:0]) -dst[127:32] := a[127:31] -dst[MAX:64] := 0 +dst[511:0] := a[511:0] +CASE imm8[0] OF +0: dst[255:0] := b[255:0] +1: dst[511:256] := b[255:0] +ESAC +dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point - AVX512F - Convert - - - - - - Convert the lower double-precision (64-bit) floating-point element in "b" to a single-precision (32-bit) floating-point element, store the result in the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". - [round_note] - + + AVX512DQ + Miscellaneous + + + + + + + Copy "a" to "tmp", then insert 256 bits (composed of 8 packed 32-bit integers) from "b" into "tmp" at the location specified by "imm8". Store "tmp" to "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -IF k[0] - dst[31:0] := Convert_FP64_To_FP32(b[63:0]) -ELSE - dst[31:0] := src[31:0] -FI -dst[127:32] := a[127:31] -dst[MAX:64] := 0 +tmp[511:0] := a[511:0] +CASE (imm8[0]) OF +0: tmp[255:0] := b[255:0] +1: tmp[511:256] := b[255:0] +ESAC +FOR j := 0 to 15 + i := j*32 + IF k[j] + dst[i+31:i] := tmp[i+31:i] + ELSE + dst[i+31:i] := src[i+31:i] + FI +ENDFOR +dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point - AVX512F - Convert - - - - - Convert the lower double-precision (64-bit) floating-point element in "b" to a single-precision (32-bit) floating-point element, store the result in the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". - + + AVX512DQ + Miscellaneous + + + + + + Copy "a" to "tmp", then insert 256 bits (composed of 8 packed 32-bit integers) from "b" into "tmp" at the location specified by "imm8". Store "tmp" to "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -IF k[0] - dst[31:0] := Convert_FP64_To_FP32(b[63:0]) -ELSE - dst[31:0] := src[31:0] -FI -dst[127:32] := a[127:31] -dst[MAX:64] := 0 +tmp[511:0] := a[511:0] +CASE (imm8[0]) OF +0: tmp[255:0] := b[255:0] +1: tmp[511:256] := b[255:0] +ESAC +FOR j := 0 to 15 + i := j*32 + IF k[j] + dst[i+31:i] := tmp[i+31:i] + ELSE + dst[i+31:i] := 0 + FI +ENDFOR +dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point - AVX512F - Convert - - - - - Convert the lower double-precision (64-bit) floating-point element in "b" to a single-precision (32-bit) floating-point element, store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". - [round_note] - + + AVX512VL + AVX512DQ + Miscellaneous + + + + + Copy "a" to "dst", then insert 128 bits (composed of 2 packed 64-bit integers) from "b" into "dst" at the location specified by "imm8". -IF k[0] - dst[31:0] := Convert_FP64_To_FP32(b[63:0]) -ELSE - dst[31:0] := 0 -FI -dst[127:32] := a[127:31] -dst[MAX:64] := 0 +dst[255:0] := a[255:0] +CASE imm8[0] OF +0: dst[127:0] := b[127:0] +1: dst[255:128] := b[127:0] +ESAC +dst[MAX:256] := 0 - +
immintrin.h
 - - Floating Point - AVX512F - Convert - - - - Convert the lower double-precision (64-bit) floating-point element in "b" to a single-precision (32-bit) floating-point element, store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". - + + AVX512VL + AVX512DQ + Miscellaneous + + + + + + + Copy "a" to "tmp", then insert 128 bits (composed of 2 packed 64-bit integers) from "b" into "tmp" at the location specified by "imm8". Store "tmp" to "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -IF k[0] - dst[31:0] := Convert_FP64_To_FP32(b[63:0]) -ELSE - dst[31:0] := 0 -FI -dst[127:32] := a[127:31] -dst[MAX:64] := 0 +tmp[255:0] := a[255:0] +CASE (imm8[0]) OF +0: tmp[127:0] := b[127:0] +1: tmp[255:128] := b[127:0] +ESAC +FOR j := 0 to 3 + i := j*64 + IF k[j] + dst[i+63:i] := tmp[i+63:i] + ELSE + dst[i+63:i] := src[i+63:i] + FI +ENDFOR +dst[MAX:256] := 0 - +
immintrin.h
 - - Floating Point - Integer - AVX512F - Convert - - - Convert the lower double-precision (64-bit) floating-point element in "a" to an unsigned 32-bit integer, and store the result in "dst". - [round_note] - + + AVX512VL + AVX512DQ + Miscellaneous + + + + + + Copy "a" to "tmp", then insert 128 bits (composed of 2 packed 64-bit integers) from "b" into "tmp" at the location specified by "imm8". Store "tmp" to "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -dst[31:0] := Convert_FP64_To_UnsignedInt32(a[63:0]) +tmp[255:0] := a[255:0] +CASE (imm8[0]) OF +0: tmp[127:0] := b[127:0] +1: tmp[255:128] := b[127:0] +ESAC +FOR j := 0 to 3 + i := j*64 + IF k[j] + dst[i+63:i] := tmp[i+63:i] + ELSE + dst[i+63:i] := 0 + FI +ENDFOR +dst[MAX:256] := 0 - +
immintrin.h
 - - Floating Point - Integer - AVX512F - Convert - - - Convert the lower double-precision (64-bit) floating-point element in "a" to an unsigned 64-bit integer, and store the result in "dst". - [round_note] - + + AVX512DQ + Miscellaneous + + + + + Copy "a" to "dst", then insert 128 bits (composed of 2 packed 64-bit integers) from "b" into "dst" at the location specified by "imm8". -dst[63:0] := Convert_FP64_To_UnsignedInt64(a[63:0]) +dst[511:0] := a[511:0] +CASE imm8[1:0] OF +0: dst[127:0] := b[127:0] +1: dst[255:128] := b[127:0] +2: dst[383:256] := b[127:0] +3: dst[511:384] := b[127:0] +ESAC +dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point - Integer - AVX512F - Convert - - Convert the lower double-precision (64-bit) floating-point element in "a" to an unsigned 32-bit integer, and store the result in "dst". + + AVX512DQ + Miscellaneous + + + + + + + Copy "a" to "tmp", then insert 128 bits (composed of 2 packed 64-bit integers) from "b" into "tmp" at the location specified by "imm8". Store "tmp" to "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -dst[31:0] := Convert_FP64_To_UnsignedInt32(a[63:0]) +tmp[511:0] := a[511:0] +CASE (imm8[1:0]) OF +0: tmp[127:0] := b[127:0] +1: tmp[255:128] := b[127:0] +2: tmp[383:256] := b[127:0] +3: tmp[511:384] := b[127:0] +ESAC +FOR j := 0 to 7 + i := j*64 + IF k[j] + dst[i+63:i] := tmp[i+63:i] + ELSE + dst[i+63:i] := src[i+63:i] + FI +ENDFOR +dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point - Integer - AVX512F - Convert - - Convert the lower double-precision (64-bit) floating-point element in "a" to an unsigned 64-bit integer, and store the result in "dst". + + AVX512DQ + Miscellaneous + + + + + + Copy "a" to "tmp", then insert 128 bits (composed of 2 packed 64-bit integers) from "b" into "tmp" at the location specified by "imm8". Store "tmp" to "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -dst[63:0] := Convert_FP64_To_UnsignedInt64(a[63:0]) +tmp[511:0] := a[511:0] +CASE (imm8[1:0]) OF +0: tmp[127:0] := b[127:0] +1: tmp[255:128] := b[127:0] +2: tmp[383:256] := b[127:0] +3: tmp[511:384] := b[127:0] +ESAC +FOR j := 0 to 7 + i := j*64 + IF k[j] + dst[i+63:i] := tmp[i+63:i] + ELSE + dst[i+63:i] := 0 + FI +ENDFOR +dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - AVX512F - Convert - - - - Convert the 64-bit integer "b" to a double-precision (64-bit) floating-point element, store the result in the lower element of "dst", and copy the upper element from "a" to the upper element of "dst". - [round_note] - + AVX512VL + AVX512DQ + Logical + + + + + + Compute the bitwise OR of packed double-precision (64-bit) floating-point elements in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -dst[63:0] := Convert_Int64_To_FP64(b[63:0]) -dst[127:64] := a[127:64] -dst[MAX:128] := 0 +FOR j := 0 to 3 + i := j*64 + IF k[j] + dst[i+63:i] := a[i+63:i] OR b[i+63:i] + ELSE + dst[i+63:i] := src[i+63:i] + FI +ENDFOR +dst[MAX:256] := 0 - +
immintrin.h
 - + Floating Point - Integer - AVX512F - Convert - - - - Convert the 64-bit integer "b" to a double-precision (64-bit) floating-point element, store the result in the lower element of "dst", and copy the upper element from "a" to the upper element of "dst". - [round_note] - + AVX512VL + AVX512DQ + Logical + + + + + Compute the bitwise OR of packed double-precision (64-bit) floating-point elements in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -dst[63:0] := Convert_Int64_To_FP64(b[63:0]) -dst[127:64] := a[127:64] -dst[MAX:128] := 0 +FOR j := 0 to 3 + i := j*64 + IF k[j] + dst[i+63:i] := a[i+63:i] OR b[i+63:i] + ELSE + dst[i+63:i] := 0 + FI +ENDFOR +dst[MAX:256] := 0 - +
immintrin.h
 - + Floating Point - AVX512F - Convert - - - Convert the 32-bit integer "b" to a double-precision (64-bit) floating-point element, store the result in the lower element of "dst", and copy the upper element from "a" to the upper element of "dst". + AVX512DQ + Logical + + + + + + Compute the bitwise OR of packed double-precision (64-bit) floating-point elements in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -dst[63:0] := Convert_Int32_To_FP64(b[31:0]) -dst[127:64] := a[127:64] -dst[MAX:128] := 0 +FOR j := 0 to 7 + i := j*64 + IF k[j] + dst[i+63:i] := a[i+63:i] OR b[i+63:i] + ELSE + dst[i+63:i] := src[i+63:i] + FI +ENDFOR +dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - AVX512F - Convert - - - Convert the 64-bit integer "b" to a double-precision (64-bit) floating-point element, store the result in the lower element of "dst", and copy the upper element from "a" to the upper element of "dst". + AVX512DQ + Logical + + + + + Compute the bitwise OR of packed double-precision (64-bit) floating-point elements in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -dst[63:0] := Convert_Int64_To_FP64(b[63:0]) -dst[127:64] := a[127:64] -dst[MAX:128] := 0 +FOR j := 0 to 7 + i := j*64 + IF k[j] + dst[i+63:i] := a[i+63:i] OR b[i+63:i] + ELSE + dst[i+63:i] := 0 + FI +ENDFOR +dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - AVX512F - Convert - - - - Convert the 32-bit integer "b" to a single-precision (32-bit) floating-point element, store the result in the lower element of "dst", and copy the upper 3 packed elements from "a" to the upper elements of "dst". - [round_note] - + AVX512DQ + Logical + + + + Compute the bitwise OR of packed double-precision (64-bit) floating-point elements in "a" and "b", and store the results in "dst". -dst[31:0] := Convert_Int32_To_FP32(b[31:0]) -dst[127:32] := a[127:32] -dst[MAX:128] := 0 +FOR j := 0 to 7 + i := j*64 + dst[i+63:i] := a[i+63:i] OR b[i+63:i] +ENDFOR +dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - AVX512F - Convert - - - - Convert the 64-bit integer "b" to a single-precision (32-bit) floating-point element, store the result in the lower element of "dst", and copy the upper 3 packed elements from "a" to the upper elements of "dst". - [round_note] - + AVX512VL + AVX512DQ + Logical + + + + + + Compute the bitwise OR of packed double-precision (64-bit) floating-point elements in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -dst[31:0] := Convert_Int64_To_FP32(b[63:0]) -dst[127:32] := a[127:32] +FOR j := 0 to 1 + i := j*64 + IF k[j] + dst[i+63:i] := a[i+63:i] OR b[i+63:i] + ELSE + dst[i+63:i] := src[i+63:i] + FI +ENDFOR dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point - AVX512F - Convert - - - - Convert the 32-bit integer "b" to a single-precision (32-bit) floating-point element, store the result in the lower element of "dst", and copy the upper 3 packed elements from "a" to the upper elements of "dst". - [round_note] - + AVX512VL + AVX512DQ + Logical + + + + + Compute the bitwise OR of packed double-precision (64-bit) floating-point elements in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -dst[31:0] := Convert_Int32_To_FP32(b[31:0]) -dst[127:32] := a[127:32] +FOR j := 0 to 1 + i := j*64 + IF k[j] + dst[i+63:i] := a[i+63:i] OR b[i+63:i] + ELSE + dst[i+63:i] := 0 + FI +ENDFOR dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point - Integer - AVX512F - Convert - - - - Convert the 64-bit integer "b" to a single-precision (32-bit) floating-point element, store the result in the lower element of "dst", and copy the upper 3 packed elements from "a" to the upper elements of "dst". - [round_note] - + AVX512VL + AVX512DQ + Logical + + + + + + Compute the bitwise OR of packed single-precision (32-bit) floating-point elements in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -dst[31:0] := Convert_Int64_To_FP32(b[63:0]) -dst[127:32] := a[127:32] -dst[MAX:128] := 0 +FOR j := 0 to 7 + i := j*32 + IF k[j] + dst[i+31:i] := a[i+31:i] OR b[i+31:i] + ELSE + dst[i+31:i] := src[i+31:i] + FI +ENDFOR +dst[MAX:256] := 0 - +
immintrin.h
 - + Floating Point - AVX512F - Convert - - - Convert the 32-bit integer "b" to a single-precision (32-bit) floating-point element, store the result in the lower element of "dst", and copy the upper 3 packed elements from "a" to the upper elements of "dst". + AVX512VL + AVX512DQ + Logical + + + + + Compute the bitwise OR of packed single-precision (32-bit) floating-point elements in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -dst[31:0] := Convert_Int32_To_FP32(b[31:0]) -dst[127:32] := a[127:32] -dst[MAX:128] := 0 +FOR j := 0 to 7 + i := j*32 + IF k[j] + dst[i+31:i] := a[i+31:i] OR b[i+31:i] + ELSE + dst[i+31:i] := 0 + FI +ENDFOR +dst[MAX:256] := 0 - +
immintrin.h
 - + Floating Point - AVX512F - Convert - - - Convert the 64-bit integer "b" to a single-precision (32-bit) floating-point element, store the result in the lower element of "dst", and copy the upper 3 packed elements from "a" to the upper elements of "dst". + AVX512DQ + Logical + + + + + + Compute the bitwise OR of packed single-precision (32-bit) floating-point elements in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -dst[31:0] := Convert_Int64_To_FP32(b[63:0]) -dst[127:32] := a[127:32] -dst[MAX:128] := 0 +FOR j := 0 to 15 + i := j*32 + IF k[j] + dst[i+31:i] := a[i+31:i] OR b[i+31:i] + ELSE + dst[i+31:i] := src[i+31:i] + FI +ENDFOR +dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - AVX512F - Convert - - - - Convert the lower single-precision (32-bit) floating-point element in "b" to a double-precision (64-bit) floating-point element, store the result in the lower element of "dst", and copy the upper element from "a" to the upper element of "dst". - [round_note] - + AVX512DQ + Logical + + + + + Compute the bitwise OR of packed single-precision (32-bit) floating-point elements in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -dst[63:0] := Convert_FP32_To_FP64(b[31:0]) -dst[127:64] := a[127:64] -dst[MAX:64] := 0 +FOR j := 0 to 15 + i := j*32 + IF k[j] + dst[i+31:i] := a[i+31:i] OR b[i+31:i] + ELSE + dst[i+31:i] := 0 + FI +ENDFOR +dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - AVX512F - Convert - - - - - - Convert the lower single-precision (32-bit) floating-point element in "b" to a double-precision (64-bit) floating-point element, store the result in the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". - [round_note] - + AVX512DQ + Logical + + + + Compute the bitwise OR of packed single-precision (32-bit) floating-point elements in "a" and "b", and store the results in "dst". -IF k[0] - dst[63:0] := Convert_FP32_To_FP64(b[31:0]) -ELSE - dst[63:0] := src[63:0] -FI -dst[127:64] := a[127:64] -dst[MAX:64] := 0 +FOR j := 0 to 15 + i := j*32 + dst[i+31:i] := a[i+31:i] OR b[i+31:i] +ENDFOR +dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - AVX512F - Convert - - - - - Convert the lower single-precision (32-bit) floating-point element in "b" to a double-precision (64-bit) floating-point element, store the result in the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". - + AVX512VL + AVX512DQ + Logical + + + + + + Compute the bitwise OR of packed single-precision (32-bit) floating-point elements in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -IF k[0] - dst[63:0] := Convert_FP32_To_FP64(b[31:0]) -ELSE - dst[63:0] := src[63:0] -FI -dst[127:64] := a[127:64] -dst[MAX:64] := 0 - - -
immintrin.h
- - - Floating Point - AVX512F - Convert - - - - - Convert the lower single-precision (32-bit) floating-point element in "b" to a double-precision (64-bit) floating-point element, store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". - [round_note] - - -IF k[0] - dst[63:0] := Convert_FP32_To_FP64(b[31:0]) -ELSE - dst[63:0] := 0 -FI -dst[127:64] := a[127:64] -dst[MAX:64] := 0 - - -
immintrin.h
- - - Floating Point - AVX512F - Convert - - - - Convert the lower single-precision (32-bit) floating-point element in "b" to a double-precision (64-bit) floating-point element, store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". - -IF k[0] - dst[63:0] := Convert_FP32_To_FP64(b[31:0]) -ELSE - dst[63:0] := 0 -FI -dst[127:64] := a[127:64] -dst[MAX:64] := 0 - - -
immintrin.h
- - - Floating Point - AVX512F - Convert - - - Convert the lower single-precision (32-bit) floating-point element in "a" to a 32-bit integer, and store the result in "dst". - [round_note] - - -dst[31:0] := Convert_FP32_To_Int32(a[31:0]) - - -
immintrin.h
- - - Floating Point - AVX512F - Convert - - - Convert the lower single-precision (32-bit) floating-point element in "a" to a 64-bit integer, and store the result in "dst". - [round_note] - - -dst[63:0] := Convert_FP32_To_Int64(a[31:0]) - - -
immintrin.h
- - - Floating Point - Integer - AVX512F - Convert - - - Convert the lower single-precision (32-bit) floating-point element in "a" to a 32-bit integer, and store the result in "dst". - [round_note] - - -dst[31:0] := Convert_FP32_To_Int32(a[31:0]) - - -
immintrin.h
- - - Floating Point - Integer - AVX512F - Convert - - - Convert the lower single-precision (32-bit) floating-point element in "a" to a 64-bit integer, and store the result in "dst". - [round_note] - - -dst[63:0] := Convert_FP32_To_Int64(a[31:0]) - - -
immintrin.h
- - - Floating Point - AVX512F - Convert - - Convert the lower single-precision (32-bit) floating-point element in "a" to a 32-bit integer, and store the result in "dst". - -dst[31:0] := Convert_FP32_To_Int32(a[31:0]) - - -
immintrin.h
- - - Floating Point - AVX512F - Convert - - Convert the lower single-precision (32-bit) floating-point element in "a" to a 64-bit integer, and store the result in "dst". - -dst[63:0] := Convert_FP32_To_Int64(a[31:0]) - - -
immintrin.h
- - - Floating Point - Integer - AVX512F - Convert - - - Convert the lower single-precision (32-bit) floating-point element in "a" to an unsigned 32-bit integer, and store the result in "dst". - [round_note] - - -dst[31:0] := Convert_FP32_To_UnsignedInt32(a[31:0]) - - -
immintrin.h
- - - Floating Point - Integer - AVX512F - Convert - - - Convert the lower single-precision (32-bit) floating-point element in "a" to an unsigned 64-bit integer, and store the result in "dst". - [round_note] - - -dst[63:0] := Convert_FP32_To_UnsignedInt64(a[31:0]) +FOR j := 0 to 3 + i := j*32 + IF k[j] + dst[i+31:i] := a[i+31:i] OR b[i+31:i] + ELSE + dst[i+31:i] := src[i+31:i] + FI +ENDFOR +dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point - Integer - AVX512F - Convert - - Convert the lower single-precision (32-bit) floating-point element in "a" to an unsigned 32-bit integer, and store the result in "dst". + AVX512VL + AVX512DQ + Logical + + + + + Compute the bitwise OR of packed single-precision (32-bit) floating-point elements in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -dst[31:0] := Convert_FP32_To_UnsignedInt32(a[31:0]) +FOR j := 0 to 3 + i := j*32 + IF k[j] + dst[i+31:i] := a[i+31:i] OR b[i+31:i] + ELSE + dst[i+31:i] := 0 + FI +ENDFOR +dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point + Integer - AVX512F - Convert - - Convert the lower single-precision (32-bit) floating-point element in "a" to an unsigned 64-bit integer, and store the result in "dst". + Mask + AVX512VL + AVX512DQ + Miscellaneous + + + Set each bit of mask register "k" based on the most significant bit of the corresponding packed 32-bit integer in "a". -dst[63:0] := Convert_FP32_To_UnsignedInt64(a[31:0]) +FOR j := 0 to 7 + i := j*32 + IF a[i+31] + k[j] := 1 + ELSE + k[j] := 0 + FI +ENDFOR +k[MAX:8] := 0 - +
immintrin.h
 - - Floating Point + Integer - AVX512F - Convert - - - Convert packed double-precision (64-bit) floating-point elements in "a" to packed 32-bit integers with truncation, and store the results in "dst". - Pass __MM_FROUND_NO_EXC to "sae" to suppress all exceptions. + Mask + AVX512DQ + Miscellaneous + + + Set each bit of mask register "k" based on the most significant bit of the corresponding packed 32-bit integer in "a". -FOR j := 0 to 7 - i := 32*i - k := 64*j - dst[i+31:i] := Convert_FP64_To_IntegerTruncate(a[k+63:k]) +FOR j := 0 to 15 + i := j*32 + IF a[i+31] + k[j] := 1 + ELSE + k[j] := 0 + FI ENDFOR -dst[MAX:256] := 0 +k[MAX:16] := 0 - +
immintrin.h
 - - Floating Point + Integer - AVX512F - Convert - - Convert packed double-precision (64-bit) floating-point elements in "a" to packed 32-bit integers with truncation, and store the results in "dst". + Mask + AVX512VL + AVX512DQ + Miscellaneous + + + Set each bit of mask register "k" based on the most significant bit of the corresponding packed 32-bit integer in "a". -FOR j := 0 to 7 - i := 32*j - k := 64*j - dst[i+31:i] := Convert_FP64_To_Int32_Truncate(a[k+63:k]) +FOR j := 0 to 3 + i := j*32 + IF a[i+31] + k[j] := 1 + ELSE + k[j] := 0 + FI ENDFOR -dst[MAX:256] := 0 +k[MAX:4] := 0 - +
immintrin.h
 - - Floating Point + Integer - AVX512F - Convert - - - - - Convert packed double-precision (64-bit) floating-point elements in "a" to packed 32-bit integers with truncation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). Pass __MM_FROUND_NO_EXC to "sae" to suppress all exceptions. + AVX512VL + AVX512DQ + Miscellaneous + + + Set each packed 32-bit integer in "dst" to all ones or all zeros based on the value of the corresponding bit in "k". FOR j := 0 to 7 - i := 32*i - l := 64*j + i := j*32 IF k[j] - dst[i+31:i] := Convert_FP64_To_IntegerTruncate(a[l+63:l]) + dst[i+31:i] := 0xFFFFFFFF ELSE - dst[i+31:i] := src[i+31:i] + dst[i+31:i] := 0 FI ENDFOR dst[MAX:256] := 0 - +
immintrin.h
 - - Floating Point + Integer - AVX512F - Convert - - - - Convert packed double-precision (64-bit) floating-point elements in "a" to packed 32-bit integers with truncation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + AVX512DQ + Miscellaneous + + + Set each packed 32-bit integer in "dst" to all ones or all zeros based on the value of the corresponding bit in "k". -FOR j := 0 to 7 - i := 32*j - l := 64*j +FOR j := 0 to 15 + i := j*32 IF k[j] - dst[i+31:i] := Convert_FP64_To_Int32_Truncate(a[l+63:l]) + dst[i+31:i] := 0xFFFFFFFF ELSE - dst[i+31:i] := src[i+31:i] + dst[i+31:i] := 0 FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point + Integer - AVX512F - Convert - - - - Convert packed double-precision (64-bit) floating-point elements in "a" to packed 32-bit integers with truncation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - Pass __MM_FROUND_NO_EXC to "sae" to suppress all exceptions. + AVX512VL + AVX512DQ + Miscellaneous + + + Set each packed 32-bit integer in "dst" to all ones or all zeros based on the value of the corresponding bit in "k". -FOR j := 0 to 7 - i := 32*i - l := 64*j +FOR j := 0 to 3 + i := j*32 IF k[j] - dst[i+31:i] := Convert_FP64_To_IntegerTruncate(a[l+63:l]) + dst[i+31:i] := 0xFFFFFFFF ELSE dst[i+31:i] := 0 FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point + Integer - AVX512F - Convert - - - Convert packed double-precision (64-bit) floating-point elements in "a" to packed 32-bit integers with truncation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + AVX512VL + AVX512DQ + Miscellaneous + + + Set each packed 64-bit integer in "dst" to all ones or all zeros based on the value of the corresponding bit in "k". -FOR j := 0 to 7 - i := 32*j - l := 64*j +FOR j := 0 to 3 + i := j*64 IF k[j] - dst[i+31:i] := Convert_FP64_To_Int32_Truncate(a[l+63:l]) + dst[i+63:i] := 0xFFFFFFFFFFFFFFFF ELSE - dst[i+31:i] := 0 + dst[i+63:i] := 0 FI ENDFOR dst[MAX:256] := 0 - +
immintrin.h
 - - Floating Point + Integer - AVX512F - Convert - - - Convert packed double-precision (64-bit) floating-point elements in "a" to packed unsigned 32-bit integers with truncation, and store the results in "dst". - Pass __MM_FROUND_NO_EXC to "sae" to suppress all exceptions. + AVX512DQ + Miscellaneous + + + Set each packed 64-bit integer in "dst" to all ones or all zeros based on the value of the corresponding bit in "k". FOR j := 0 to 7 - i := 32*i - k := 64*j - dst[i+31:i] := Convert_FP64_To_UnsignedIntegerTruncate(a[k+63:k]) + i := j*64 + IF k[j] + dst[i+63:i] := 0xFFFFFFFFFFFFFFFF + ELSE + dst[i+63:i] := 0 + FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point + Integer - AVX512F - Convert - - Convert packed double-precision (64-bit) floating-point elements in "a" to packed unsigned 32-bit integers with truncation, and store the results in "dst". + AVX512VL + AVX512DQ + Miscellaneous + + + Set each packed 64-bit integer in "dst" to all ones or all zeros based on the value of the corresponding bit in "k". -FOR j := 0 to 7 - i := 32*j - k := 64*j - dst[i+31:i] := Convert_FP64_To_UnsignedInt32_Truncate(a[k+63:k]) +FOR j := 0 to 1 + i := j*64 + IF k[j] + dst[i+63:i] := 0xFFFFFFFFFFFFFFFF + ELSE + dst[i+63:i] := 0 + FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point + Integer - AVX512F - Convert - - - - - Convert packed double-precision (64-bit) floating-point elements in "a" to packed unsigned 32-bit integers with truncation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - Pass __MM_FROUND_NO_EXC to "sae" to suppress all exceptions. + Mask + AVX512VL + AVX512DQ + Miscellaneous + + + Set each bit of mask register "k" based on the most significant bit of the corresponding packed 64-bit integer in "a". -FOR j := 0 to 7 - i := 32*i - l := 64*j - IF k[j] - dst[i+31:i] := Convert_FP64_To_UnsignedIntegerTruncate(a[l+63:l]) +FOR j := 0 to 3 + i := j*64 + IF a[i+63] + k[j] := 1 ELSE - dst[i+31:i] := src[i+31:i] + k[j] := 0 FI ENDFOR -dst[MAX:256] := 0 +k[MAX:4] := 0 - +
immintrin.h
 - - Floating Point + Integer - AVX512F - Convert - - - - Convert packed double-precision (64-bit) floating-point elements in "a" to packed unsigned 32-bit integers with truncation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Mask + AVX512DQ + Miscellaneous + + + Set each bit of mask register "k" based on the most significant bit of the corresponding packed 64-bit integer in "a". FOR j := 0 to 7 - i := 32*j - l := 64*j - IF k[j] - dst[i+31:i] := Convert_FP64_To_UnsignedInt32_Truncate(a[l+63:l]) + i := j*64 + IF a[i+63] + k[j] := 1 ELSE - dst[i+31:i] := src[i+31:i] + k[j] := 0 FI ENDFOR -dst[MAX:256] := 0 +k[MAX:8] := 0 - +
immintrin.h
 - - Floating Point + Integer - AVX512F - Convert - - - - Convert packed double-precision (64-bit) floating-point elements in "a" to packed unsigned 32-bit integers with truncation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - Pass __MM_FROUND_NO_EXC to "sae" to suppress all exceptions. + Mask + AVX512VL + AVX512DQ + Miscellaneous + + + Set each bit of mask register "k" based on the most significant bit of the corresponding packed 64-bit integer in "a". -FOR j := 0 to 7 - i := 32*i - l := 64*j - IF k[j] - dst[i+31:i] := Convert_FP64_To_UnsignedIntegerTruncate(a[l+63:l]) +FOR j := 0 to 1 + i := j*64 + IF a[i+63] + k[j] := 1 ELSE - dst[i+31:i] := 0 + k[j] := 0 FI ENDFOR -dst[MAX:256] := 0 +k[MAX:2] := 0 - +
immintrin.h
 - - Floating Point + Integer - AVX512F - Convert - - - Convert packed double-precision (64-bit) floating-point elements in "a" to packed unsigned 32-bit integers with truncation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + AVX512VL + AVX512DQ + Arithmetic + + + + + + Multiply the packed 64-bit integers in "a" and "b", producing intermediate 128-bit integers, and store the low 64 bits of the intermediate integers in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := 32*j - l := 64*j +FOR j := 0 to 3 + i := j*64 IF k[j] - dst[i+31:i] := Convert_FP64_To_UnsignedInt32_Truncate(a[l+63:l]) + tmp[127:0] := a[i+63:i] * b[i+63:i] + dst[i+63:i] := tmp[63:0] ELSE - dst[i+31:i] := 0 + dst[i+63:i] := src[i+63:i] FI ENDFOR dst[MAX:256] := 0 - +
immintrin.h
 - - Floating Point + Integer - AVX512F - Convert - - - Convert packed single-precision (32-bit) floating-point elements in "a" to packed 32-bit integers with truncation, and store the results in "dst". - Pass __MM_FROUND_NO_EXC to "sae" to suppress all exceptions. + AVX512VL + AVX512DQ + Arithmetic + + + + + Multiply the packed 64-bit integers in "a" and "b", producing intermediate 128-bit integers, and store the low 64 bits of the intermediate integers in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := 32*i - dst[i+31:i] := Convert_FP32_To_IntegerTruncate(a[i+31:i]) +FOR j := 0 to 3 + i := j*64 + IF k[j] + tmp[127:0] := a[i+63:i] * b[i+63:i] + dst[i+63:i] := tmp[63:0] + ELSE + dst[i+63:i] := 0 + FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - - Floating Point + Integer - AVX512F - Convert - - Convert packed single-precision (32-bit) floating-point elements in "a" to packed 32-bit integers with truncation, and store the results in "dst". + AVX512VL + AVX512DQ + Arithmetic + + + + Multiply the packed 64-bit integers in "a" and "b", producing intermediate 128-bit integers, and store the low 64 bits of the intermediate integers in "dst". -FOR j := 0 to 15 - i := 32*j - dst[i+31:i] := Convert_FP32_To_Int32_Truncate(a[i+31:i]) +FOR j := 0 to 3 + i := j*64 + tmp[127:0] := a[i+63:i] * b[i+63:i] + dst[i+63:i] := tmp[63:0] ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - - Floating Point + Integer - AVX512F - Convert - - - - - Convert packed single-precision (32-bit) floating-point elements in "a" to packed 32-bit integers with truncation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - Pass __MM_FROUND_NO_EXC to "sae" to suppress all exceptions. + AVX512DQ + Arithmetic + + + + + + Multiply the packed 64-bit integers in "a" and "b", producing intermediate 128-bit integers, and store the low 64 bits of the intermediate integers in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := 32*i +FOR j := 0 to 7 + i := j*64 IF k[j] - dst[i+31:i] := Convert_FP32_To_IntegerTruncate(a[i+31:i]) + tmp[127:0] := a[i+63:i] * b[i+63:i] + dst[i+63:i] := tmp[63:0] ELSE - dst[i+31:i] := src[i+31:i] + dst[i+63:i] := src[i+63:i] FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point + Integer - AVX512F - Convert - - - - Convert packed single-precision (32-bit) floating-point elements in "a" to packed 32-bit integers with truncation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + AVX512DQ + Arithmetic + + + + + Multiply the packed 64-bit integers in "a" and "b", producing intermediate 128-bit integers, and store the low 64 bits of the intermediate integers in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := 32*j +FOR j := 0 to 7 + i := j*64 IF k[j] - dst[i+31:i] := Convert_FP32_To_Int32_Truncate(a[i+31:i]) + tmp[127:0] := a[i+63:i] * b[i+63:i] + dst[i+63:i] := tmp[63:0] ELSE - dst[i+31:i] := src[i+31:i] + dst[i+63:i] := 0 FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point + Integer - AVX512F - Convert - - - - Convert packed single-precision (32-bit) floating-point elements in "a" to packed 32-bit integers with truncation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - Pass __MM_FROUND_NO_EXC to "sae" to suppress all exceptions. + AVX512DQ + Arithmetic + + + + Multiply the packed 64-bit integers in "a" and "b", producing intermediate 128-bit integers, and store the low 64 bits of the intermediate integers in "dst". -FOR j := 0 to 15 - i := 32*i - IF k[j] - dst[i+31:i] := Convert_FP32_To_IntegerTruncate(a[i+31:i]) - ELSE - dst[i+31:i] := 0 - FI +FOR j := 0 to 7 + i := j*64 + tmp[127:0] := a[i+63:i] * b[i+63:i] + dst[i+63:i] := tmp[63:0] ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point + Integer - AVX512F - Convert - - - Convert packed single-precision (32-bit) floating-point elements in "a" to packed 32-bit integers with truncation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + AVX512VL + AVX512DQ + Arithmetic + + + + + + Multiply the packed 64-bit integers in "a" and "b", producing intermediate 128-bit integers, and store the low 64 bits of the intermediate integers in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := 32*j +FOR j := 0 to 1 + i := j*64 IF k[j] - dst[i+31:i] := Convert_FP32_To_Int32_Truncate(a[i+31:i]) + tmp[127:0] := a[i+63:i] * b[i+63:i] + dst[i+63:i] := tmp[63:0] ELSE - dst[i+31:i] := 0 + dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point + Integer - AVX512F - Convert - - - Convert packed single-precision (32-bit) floating-point elements in "a" to packed unsigned 32-bit integers with truncation, and store the results in "dst". - Pass __MM_FROUND_NO_EXC to "sae" to suppress all exceptions. + AVX512VL + AVX512DQ + Arithmetic + + + + + Multiply the packed 64-bit integers in "a" and "b", producing intermediate 128-bit integers, and store the low 64 bits of the intermediate integers in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := 32*i - dst[i+31:i] := Convert_FP32_To_UnsignedIntegerTruncate(a[i+31:i]) +FOR j := 0 to 1 + i := j*64 + IF k[j] + tmp[127:0] := a[i+63:i] * b[i+63:i] + dst[i+63:i] := tmp[63:0] + ELSE + dst[i+63:i] := 0 + FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point + Integer - AVX512F - Convert - - Convert packed single-precision (32-bit) floating-point elements in "a" to packed unsigned 32-bit integers with truncation, and store the results in "dst". + AVX512VL + AVX512DQ + Arithmetic + + + + Multiply the packed 64-bit integers in "a" and "b", producing intermediate 128-bit integers, and store the low 64 bits of the intermediate integers in "dst". -FOR j := 0 to 15 - i := 32*j - dst[i+31:i] := Convert_FP32_To_UnsignedInt32_Truncate(a[i+31:i]) +FOR j := 0 to 1 + i := j*64 + tmp[127:0] := a[i+63:i] * b[i+63:i] + dst[i+63:i] := tmp[63:0] ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point - Integer - AVX512F - Convert - - - - - Convert packed single-precision (32-bit) floating-point elements in "a" to packed unsigned 32-bit integers with truncation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - Pass __MM_FROUND_NO_EXC to "sae" to suppress all exceptions. + AVX512VL + AVX512DQ + Miscellaneous + + + + + + + Calculate the max, min, absolute max, or absolute min (depending on control in "imm8") for packed double-precision (64-bit) floating-point elements in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + imm8[1:0] specifies the operation control: 00 = min, 01 = max, 10 = absolute max, 11 = absolute min. + imm8[3:2] specifies the sign control: 00 = sign from a, 01 = sign from compare result, 10 = clear sign bit, 11 = set sign bit. -FOR j := 0 to 15 - i := 32*i +DEFINE RANGE(src1[63:0], src2[63:0], opCtl[1:0], signSelCtl[1:0]) { + CASE opCtl[1:0] OF + 0: tmp[63:0] := (src1[63:0] <= src2[63:0]) ? src1[63:0] : src2[63:0] + 1: tmp[63:0] := (src1[63:0] <= src2[63:0]) ? src2[63:0] : src1[63:0] + 2: tmp[63:0] := (ABS(src1[63:0]) <= ABS(src2[63:0])) ? src1[63:0] : src2[63:0] + 3: tmp[63:0] := (ABS(src1[63:0]) <= ABS(src2[63:0])) ? src2[63:0] : src1[63:0] + ESAC + + CASE signSelCtl[1:0] OF + 0: dst[63:0] := (src1[63] << 63) OR (tmp[62:0]) + 1: dst[63:0] := tmp[63:0] + 2: dst[63:0] := (0 << 63) OR (tmp[62:0]) + 3: dst[63:0] := (1 << 63) OR (tmp[62:0]) + ESAC + + RETURN dst +} +FOR j := 0 to 3 + i := j*64 IF k[j] - dst[i+31:i] := Convert_FP32_To_UnsignedIntegerTruncate(a[i+31:i]) + dst[i+63:i] := RANGE(a[i+63:i], b[i+63:i], imm8[1:0], imm8[3:2]) ELSE - dst[i+31:i] := src[i+31:i] + dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Floating Point - Integer - AVX512F - Convert - - - - Convert packed double-precision (32-bit) floating-point elements in "a" to packed unsigned 32-bit integers with truncation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + AVX512VL + AVX512DQ + Miscellaneous + + + + + + Calculate the max, min, absolute max, or absolute min (depending on control in "imm8") for packed double-precision (64-bit) floating-point elements in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + imm8[1:0] specifies the operation control: 00 = min, 01 = max, 10 = absolute max, 11 = absolute min. + imm8[3:2] specifies the sign control: 00 = sign from a, 01 = sign from compare result, 10 = clear sign bit, 11 = set sign bit. -FOR j := 0 to 15 - i := 32*j - IF k[j] - dst[i+31:i] := Convert_FP64_To_UnsignedInt32_Truncate(a[i+31:i]) - ELSE - dst[i+31:i] := src[i+31:i] +DEFINE RANGE(src1[63:0], src2[63:0], opCtl[1:0], signSelCtl[1:0]) { + CASE opCtl[1:0] OF + 0: tmp[63:0] := (src1[63:0] <= src2[63:0]) ? src1[63:0] : src2[63:0] + 1: tmp[63:0] := (src1[63:0] <= src2[63:0]) ? src2[63:0] : src1[63:0] + 2: tmp[63:0] := (ABS(src1[63:0]) <= ABS(src2[63:0])) ? src1[63:0] : src2[63:0] + 3: tmp[63:0] := (ABS(src1[63:0]) <= ABS(src2[63:0])) ? src2[63:0] : src1[63:0] + ESAC + + CASE signSelCtl[1:0] OF + 0: dst[63:0] := (src1[63] << 63) OR (tmp[62:0]) + 1: dst[63:0] := tmp[63:0] + 2: dst[63:0] := (0 << 63) OR (tmp[62:0]) + 3: dst[63:0] := (1 << 63) OR (tmp[62:0]) + ESAC + + RETURN dst +} +FOR j := 0 to 3 + i := j*64 + IF k[j] + dst[i+63:i] := RANGE(a[i+63:i], b[i+63:i], imm8[1:0], imm8[3:2]) + ELSE + dst[i+63:i] := 0 FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Floating Point - Integer - AVX512F - Convert - - - - Convert packed single-precision (32-bit) floating-point elements in "a" to packed unsigned 32-bit integers with truncation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - Pass __MM_FROUND_NO_EXC to "sae" to suppress all exceptions. + AVX512VL + AVX512DQ + Miscellaneous + + + + + Calculate the max, min, absolute max, or absolute min (depending on control in "imm8") for packed double-precision (64-bit) floating-point elements in "a" and "b", and store the results in "dst". + imm8[1:0] specifies the operation control: 00 = min, 01 = max, 10 = absolute max, 11 = absolute min. + imm8[3:2] specifies the sign control: 00 = sign from a, 01 = sign from compare result, 10 = clear sign bit, 11 = set sign bit. -FOR j := 0 to 15 - i := 32*i - IF k[j] - dst[i+31:i] := Convert_FP32_To_UnsignedIntegerTruncate(a[i+31:i]) - ELSE - dst[i+31:i] := 0 - FI +DEFINE RANGE(src1[63:0], src2[63:0], opCtl[1:0], signSelCtl[1:0]) { + CASE opCtl[1:0] OF + 0: tmp[63:0] := (src1[63:0] <= src2[63:0]) ? src1[63:0] : src2[63:0] + 1: tmp[63:0] := (src1[63:0] <= src2[63:0]) ? src2[63:0] : src1[63:0] + 2: tmp[63:0] := (ABS(src1[63:0]) <= ABS(src2[63:0])) ? src1[63:0] : src2[63:0] + 3: tmp[63:0] := (ABS(src1[63:0]) <= ABS(src2[63:0])) ? src2[63:0] : src1[63:0] + ESAC + + CASE signSelCtl[1:0] OF + 0: dst[63:0] := (src1[63] << 63) OR (tmp[62:0]) + 1: dst[63:0] := tmp[63:0] + 2: dst[63:0] := (0 << 63) OR (tmp[62:0]) + 3: dst[63:0] := (1 << 63) OR (tmp[62:0]) + ESAC + + RETURN dst +} +FOR j := 0 to 3 + i := j*64 + dst[i+63:i] := RANGE(a[i+63:i], b[i+63:i], imm8[1:0], imm8[3:2]) ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Floating Point - Integer - AVX512F - Convert - - - Convert packed double-precision (32-bit) floating-point elements in "a" to packed unsigned 32-bit integers with truncation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + AVX512DQ + Miscellaneous + + + + + + + Calculate the max, min, absolute max, or absolute min (depending on control in "imm8") for packed double-precision (64-bit) floating-point elements in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + imm8[1:0] specifies the operation control: 00 = min, 01 = max, 10 = absolute max, 11 = absolute min. + imm8[3:2] specifies the sign control: 00 = sign from a, 01 = sign from compare result, 10 = clear sign bit, 11 = set sign bit. -FOR j := 0 to 15 - i := 32*j +DEFINE RANGE(src1[63:0], src2[63:0], opCtl[1:0], signSelCtl[1:0]) { + CASE opCtl[1:0] OF + 0: tmp[63:0] := (src1[63:0] <= src2[63:0]) ? src1[63:0] : src2[63:0] + 1: tmp[63:0] := (src1[63:0] <= src2[63:0]) ? src2[63:0] : src1[63:0] + 2: tmp[63:0] := (ABS(src1[63:0]) <= ABS(src2[63:0])) ? src1[63:0] : src2[63:0] + 3: tmp[63:0] := (ABS(src1[63:0]) <= ABS(src2[63:0])) ? src2[63:0] : src1[63:0] + ESAC + + CASE signSelCtl[1:0] OF + 0: dst[63:0] := (src1[63] << 63) OR (tmp[62:0]) + 1: dst[63:0] := tmp[63:0] + 2: dst[63:0] := (0 << 63) OR (tmp[62:0]) + 3: dst[63:0] := (1 << 63) OR (tmp[62:0]) + ESAC + + RETURN dst +} +FOR j := 0 to 7 + i := j*64 IF k[j] - dst[i+31:i] := Convert_FP64_To_UnsignedInt32_Truncate(a[i+31:i]) + dst[i+63:i] := RANGE(a[i+63:i], b[i+63:i], imm8[1:0], imm8[3:2]) ELSE - dst[i+31:i] := 0 + dst[i+63:i] := src[i+63:i] FI ENDFOR dst[MAX:512] := 0 - -
immintrin.h
- - - Floating Point - AVX512F - Convert - - - Convert the lower double-precision (64-bit) floating-point element in "a" to a 32-bit integer with truncation, and store the result in "dst". - [round_note] - - -dst[31:0] := Convert_FP64_To_Int32_Truncate(a[63:0]) - - -
immintrin.h
- - - Floating Point - AVX512F - Convert - - - Convert the lower double-precision (64-bit) floating-point element in "a" to a 64-bit integer with truncation, and store the result in "dst". - [round_note] - - -dst[63:0] := Convert_FP64_To_Int64_Truncate(a[63:0]) - - -
immintrin.h
- - - Floating Point - Integer - AVX512F - Convert - - - Convert the lower double-precision (64-bit) floating-point element in "a" to a 32-bit integer with truncation, and store the result in "dst". - [round_note] - - -dst[31:0] := Convert_FP64_To_Int32_Truncate(a[63:0]) - - -
immintrin.h
- - - Floating Point - Integer - AVX512F - Convert - - - Convert the lower double-precision (64-bit) floating-point element in "a" to a 64-bit integer with truncation, and store the result in "dst". - [round_note] - - -dst[63:0] := Convert_FP64_To_Int64_Truncate(a[63:0]) - - -
immintrin.h
- - - Floating Point - AVX512F - Convert - - Convert the lower double-precision (64-bit) floating-point element in "a" to a 32-bit integer with truncation, and store the result in "dst". - -dst[31:0] := Convert_FP64_To_Int32_Truncate(a[63:0]) - - -
immintrin.h
- - - Floating Point - AVX512F - Convert - - Convert the lower double-precision (64-bit) floating-point element in "a" to a 64-bit integer with truncation, and store the result in "dst". - -dst[63:0] := Convert_FP64_To_Int64_Truncate(a[63:0]) - - -
immintrin.h
- - - Floating Point - Integer - AVX512F - Convert - - - Convert the lower double-precision (64-bit) floating-point element in "a" to an unsigned 32-bit integer with truncation, and store the result in "dst". - [round_note] - - -dst[31:0] := Convert_FP64_To_UnsignedInt32_Truncate(a[63:0]) - - -
immintrin.h
- - - Floating Point - Integer - AVX512F - Convert - - - Convert the lower double-precision (64-bit) floating-point element in "a" to an unsigned 64-bit integer with truncation, and store the result in "dst". - [round_note] - - -dst[63:0] := Convert_FP64_To_UnsignedInt64_Truncate(a[63:0]) - - -
immintrin.h
- - - Floating Point - Integer - AVX512F - Convert - - Convert the lower double-precision (64-bit) floating-point element in "a" to an unsigned 32-bit integer with truncation, and store the result in "dst". - -dst[31:0] := Convert_FP64_To_UnsignedInt32_Truncate(a[63:0]) - - -
immintrin.h
- - - Floating Point - Integer - AVX512F - Convert - - Convert the lower double-precision (64-bit) floating-point element in "a" to an unsigned 64-bit integer with truncation, and store the result in "dst". - -dst[63:0] := Convert_FP64_To_UnsignedInt64_Truncate(a[63:0]) - - -
immintrin.h
- - - Floating Point - AVX512F - Convert - - - Convert the lower single-precision (32-bit) floating-point element in "a" to a 32-bit integer with truncation, and store the result in "dst". - [round_note] - - -dst[31:0] := Convert_FP32_To_Int32_Truncate(a[31:0]) - - -
immintrin.h
- - - Floating Point - AVX512F - Convert - - - Convert the lower single-precision (32-bit) floating-point element in "a" to a 64-bit integer with truncation, and store the result in "dst". - [round_note] - - -dst[63:0] := Convert_FP32_To_Int64_Truncate(a[31:0]) - - -
immintrin.h
- - - Floating Point - Integer - AVX512F - Convert - - - Convert the lower single-precision (32-bit) floating-point element in "a" to a 32-bit integer with truncation, and store the result in "dst". - [round_note] - - -dst[31:0] := Convert_FP32_To_Int32_Truncate(a[31:0]) - - -
immintrin.h
- - - Floating Point - Integer - AVX512F - Convert - - - Convert the lower single-precision (32-bit) floating-point element in "a" to a 64-bit integer with truncation, and store the result in "dst". - [round_note] - - -dst[63:0] := Convert_FP32_To_Int64_Truncate(a[31:0]) - - -
immintrin.h
- - - Floating Point - AVX512F - Convert - - Convert the lower single-precision (32-bit) floating-point element in "a" to a 32-bit integer with truncation, and store the result in "dst". - -dst[31:0] := Convert_FP32_To_Int32_Truncate(a[31:0]) - - -
immintrin.h
- - - Floating Point - AVX512F - Convert - - Convert the lower single-precision (32-bit) floating-point element in "a" to a 64-bit integer with truncation, and store the result in "dst". - -dst[63:0] := Convert_FP32_To_Int64_Truncate(a[31:0]) - - -
immintrin.h
- - - Floating Point - Integer - AVX512F - Convert - - - Convert the lower single-precision (32-bit) floating-point element in "a" to an unsigned 32-bit integer with truncation, and store the result in "dst". - [round_note] - - -dst[31:0] := Convert_FP32_To_UnsignedInt32_Truncate(a[31:0]) - - -
immintrin.h
- - - Floating Point - Integer - AVX512F - Convert - - - Convert the lower single-precision (32-bit) floating-point element in "a" to an unsigned 64-bit integer with truncation, and store the result in "dst". - [round_note] - - -dst[63:0] := Convert_FP32_To_UnsignedInt64_Truncate(a[31:0]) - - -
immintrin.h
- - - Floating Point - Integer - AVX512F - Convert - - Convert the lower single-precision (32-bit) floating-point element in "a" to an unsigned 32-bit integer with truncation, and store the result in "dst". - -dst[31:0] := Convert_FP32_To_UnsignedInt32_Truncate(a[31:0]) - - -
immintrin.h
- - - Floating Point - Integer - AVX512F - Convert - - Convert the lower single-precision (32-bit) floating-point element in "a" to an unsigned 64-bit integer with truncation, and store the result in "dst". - -dst[63:0] := Convert_FP32_To_UnsignedInt64_Truncate(a[31:0]) - - +
immintrin.h
 - + Floating Point - Integer - AVX512F - Convert - - Convert packed unsigned 32-bit integers in "a" to packed double-precision (64-bit) floating-point elements, and store the results in "dst". + AVX512DQ + Miscellaneous + + + + + + + + Calculate the max, min, absolute max, or absolute min (depending on control in "imm8") for packed double-precision (64-bit) floating-point elements in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + imm8[1:0] specifies the operation control: 00 = min, 01 = max, 10 = absolute max, 11 = absolute min. + imm8[3:2] specifies the sign control: 00 = sign from a, 01 = sign from compare result, 10 = clear sign bit, 11 = set sign bit. [sae_note] +DEFINE RANGE(src1[63:0], src2[63:0], opCtl[1:0], signSelCtl[1:0]) { + CASE opCtl[1:0] OF + 0: tmp[63:0] := (src1[63:0] <= src2[63:0]) ? src1[63:0] : src2[63:0] + 1: tmp[63:0] := (src1[63:0] <= src2[63:0]) ? src2[63:0] : src1[63:0] + 2: tmp[63:0] := (ABS(src1[63:0]) <= ABS(src2[63:0])) ? src1[63:0] : src2[63:0] + 3: tmp[63:0] := (ABS(src1[63:0]) <= ABS(src2[63:0])) ? src2[63:0] : src1[63:0] + ESAC + + CASE signSelCtl[1:0] OF + 0: dst[63:0] := (src1[63] << 63) OR (tmp[62:0]) + 1: dst[63:0] := tmp[63:0] + 2: dst[63:0] := (0 << 63) OR (tmp[62:0]) + 3: dst[63:0] := (1 << 63) OR (tmp[62:0]) + ESAC + + RETURN dst +} FOR j := 0 to 7 i := j*64 - l := j*32 - dst[i+63:i] := ConvertUnsignedIntegerTo_FP64(a[l+31:l]) + IF k[j] + dst[i+63:i] := RANGE(a[i+63:i], b[i+63:i], imm8[1:0], imm8[3:2]) + ELSE + dst[i+63:i] := src[i+63:i] + FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - Integer - AVX512F - Convert - - - - Convert packed unsigned 32-bit integers in "a" to packed double-precision (64-bit) floating-point elements, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - + AVX512DQ + Miscellaneous + + + + + + Calculate the max, min, absolute max, or absolute min (depending on control in "imm8") for packed double-precision (64-bit) floating-point elements in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + imm8[1:0] specifies the operation control: 00 = min, 01 = max, 10 = absolute max, 11 = absolute min. + imm8[3:2] specifies the sign control: 00 = sign from a, 01 = sign from compare result, 10 = clear sign bit, 11 = set sign bit. +DEFINE RANGE(src1[63:0], src2[63:0], opCtl[1:0], signSelCtl[1:0]) { + CASE opCtl[1:0] OF + 0: tmp[63:0] := (src1[63:0] <= src2[63:0]) ? src1[63:0] : src2[63:0] + 1: tmp[63:0] := (src1[63:0] <= src2[63:0]) ? src2[63:0] : src1[63:0] + 2: tmp[63:0] := (ABS(src1[63:0]) <= ABS(src2[63:0])) ? src1[63:0] : src2[63:0] + 3: tmp[63:0] := (ABS(src1[63:0]) <= ABS(src2[63:0])) ? src2[63:0] : src1[63:0] + ESAC + + CASE signSelCtl[1:0] OF + 0: dst[63:0] := (src1[63] << 63) OR (tmp[62:0]) + 1: dst[63:0] := tmp[63:0] + 2: dst[63:0] := (0 << 63) OR (tmp[62:0]) + 3: dst[63:0] := (1 << 63) OR (tmp[62:0]) + ESAC + + RETURN dst +} FOR j := 0 to 7 i := j*64 - l := j*32 IF k[j] - dst[i+63:i] := ConvertUnsignedIntegerTo_FP64(a[l+31:l]) + dst[i+63:i] := RANGE(a[i+63:i], b[i+63:i], imm8[1:0], imm8[3:2]) ELSE - dst[i+63:i] := src[i+63:i] - FI + dst[i+63:i] := 0 + FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - Integer - AVX512F - Convert - - - Convert packed unsigned 32-bit integers in "a" to packed double-precision (64-bit) floating-point elements, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + AVX512DQ + Miscellaneous + + + + + + + Calculate the max, min, absolute max, or absolute min (depending on control in "imm8") for packed double-precision (64-bit) floating-point elements in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + imm8[1:0] specifies the operation control: 00 = min, 01 = max, 10 = absolute max, 11 = absolute min. + imm8[3:2] specifies the sign control: 00 = sign from a, 01 = sign from compare result, 10 = clear sign bit, 11 = set sign bit. [sae_note] +DEFINE RANGE(src1[63:0], src2[63:0], opCtl[1:0], signSelCtl[1:0]) { + CASE opCtl[1:0] OF + 0: tmp[63:0] := (src1[63:0] <= src2[63:0]) ? src1[63:0] : src2[63:0] + 1: tmp[63:0] := (src1[63:0] <= src2[63:0]) ? src2[63:0] : src1[63:0] + 2: tmp[63:0] := (ABS(src1[63:0]) <= ABS(src2[63:0])) ? src1[63:0] : src2[63:0] + 3: tmp[63:0] := (ABS(src1[63:0]) <= ABS(src2[63:0])) ? src2[63:0] : src1[63:0] + ESAC + + CASE signSelCtl[1:0] OF + 0: dst[63:0] := (src1[63] << 63) OR (tmp[62:0]) + 1: dst[63:0] := tmp[63:0] + 2: dst[63:0] := (0 << 63) OR (tmp[62:0]) + 3: dst[63:0] := (1 << 63) OR (tmp[62:0]) + ESAC + + RETURN dst +} FOR j := 0 to 7 i := j*64 - l := j*32 IF k[j] - dst[i+63:i] := ConvertUnsignedIntegerTo_FP64(a[l+31:l]) + dst[i+63:i] := RANGE(a[i+63:i], b[i+63:i], imm8[1:0], imm8[3:2]) ELSE dst[i+63:i] := 0 - FI + FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - Integer - AVX512F - Convert - - - Convert packed unsigned 32-bit integers in "a" to packed single-precision (32-bit) floating-point elements, and store the results in "dst". - [round_note] - + AVX512DQ + Miscellaneous + + + + + Calculate the max, min, absolute max, or absolute min (depending on control in "imm8") for packed double-precision (64-bit) floating-point elements in "a" and "b", and store the results in "dst". + imm8[1:0] specifies the operation control: 00 = min, 01 = max, 10 = absolute max, 11 = absolute min. + imm8[3:2] specifies the sign control: 00 = sign from a, 01 = sign from compare result, 10 = clear sign bit, 11 = set sign bit. -FOR j := 0 to 15 - i := 32*j - dst[i+31:i] := ConvertUnsignedInt32_To_FP32(a[i+31:i]) +DEFINE RANGE(src1[63:0], src2[63:0], opCtl[1:0], signSelCtl[1:0]) { + CASE opCtl[1:0] OF + 0: tmp[63:0] := (src1[63:0] <= src2[63:0]) ? src1[63:0] : src2[63:0] + 1: tmp[63:0] := (src1[63:0] <= src2[63:0]) ? src2[63:0] : src1[63:0] + 2: tmp[63:0] := (ABS(src1[63:0]) <= ABS(src2[63:0])) ? src1[63:0] : src2[63:0] + 3: tmp[63:0] := (ABS(src1[63:0]) <= ABS(src2[63:0])) ? src2[63:0] : src1[63:0] + ESAC + + CASE signSelCtl[1:0] OF + 0: dst[63:0] := (src1[63] << 63) OR (tmp[62:0]) + 1: dst[63:0] := tmp[63:0] + 2: dst[63:0] := (0 << 63) OR (tmp[62:0]) + 3: dst[63:0] := (1 << 63) OR (tmp[62:0]) + ESAC + + RETURN dst +} +FOR j := 0 to 7 + i := j*64 + dst[i+63:i] := RANGE(a[i+63:i], b[i+63:i], imm8[1:0], imm8[3:2]) ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - Integer - AVX512F - Convert - - Convert packed unsigned 32-bit integers in "a" to packed single-precision (32-bit) floating-point elements, and store the results in "dst". + AVX512DQ + Miscellaneous + + + + + + Calculate the max, min, absolute max, or absolute min (depending on control in "imm8") for packed double-precision (64-bit) floating-point elements in "a" and "b", and store the results in "dst". + imm8[1:0] specifies the operation control: 00 = min, 01 = max, 10 = absolute max, 11 = absolute min. + imm8[3:2] specifies the sign control: 00 = sign from a, 01 = sign from compare result, 10 = clear sign bit, 11 = set sign bit. [sae_note] -FOR j := 0 to 15 - i := 32*j - dst[i+31:i] := ConvertUnsignedInt32_To_FP32(a[i+31:i]) +DEFINE RANGE(src1[63:0], src2[63:0], opCtl[1:0], signSelCtl[1:0]) { + CASE opCtl[1:0] OF + 0: tmp[63:0] := (src1[63:0] <= src2[63:0]) ? src1[63:0] : src2[63:0] + 1: tmp[63:0] := (src1[63:0] <= src2[63:0]) ? src2[63:0] : src1[63:0] + 2: tmp[63:0] := (ABS(src1[63:0]) <= ABS(src2[63:0])) ? src1[63:0] : src2[63:0] + 3: tmp[63:0] := (ABS(src1[63:0]) <= ABS(src2[63:0])) ? src2[63:0] : src1[63:0] + ESAC + + CASE signSelCtl[1:0] OF + 0: dst[63:0] := (src1[63] << 63) OR (tmp[62:0]) + 1: dst[63:0] := tmp[63:0] + 2: dst[63:0] := (0 << 63) OR (tmp[62:0]) + 3: dst[63:0] := (1 << 63) OR (tmp[62:0]) + ESAC + + RETURN dst +} +FOR j := 0 to 7 + i := j*64 + dst[i+63:i] := RANGE(a[i+63:i], b[i+63:i], imm8[1:0], imm8[3:2]) ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - Integer - AVX512F - Convert - - - - - Convert packed unsigned 32-bit integers in "a" to packed single-precision (32-bit) floating-point elements, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - [round_note] - + AVX512VL + AVX512DQ + Miscellaneous + + + + + + + Calculate the max, min, absolute max, or absolute min (depending on control in "imm8") for packed double-precision (64-bit) floating-point elements in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + imm8[1:0] specifies the operation control: 00 = min, 01 = max, 10 = absolute max, 11 = absolute min. + imm8[3:2] specifies the sign control: 00 = sign from a, 01 = sign from compare result, 10 = clear sign bit, 11 = set sign bit. -FOR j := 0 to 15 - i := j*32 +DEFINE RANGE(src1[63:0], src2[63:0], opCtl[1:0], signSelCtl[1:0]) { + CASE opCtl[1:0] OF + 0: tmp[63:0] := (src1[63:0] <= src2[63:0]) ? src1[63:0] : src2[63:0] + 1: tmp[63:0] := (src1[63:0] <= src2[63:0]) ? src2[63:0] : src1[63:0] + 2: tmp[63:0] := (ABS(src1[63:0]) <= ABS(src2[63:0])) ? src1[63:0] : src2[63:0] + 3: tmp[63:0] := (ABS(src1[63:0]) <= ABS(src2[63:0])) ? src2[63:0] : src1[63:0] + ESAC + + CASE signSelCtl[1:0] OF + 0: dst[63:0] := (src1[63] << 63) OR (tmp[62:0]) + 1: dst[63:0] := tmp[63:0] + 2: dst[63:0] := (0 << 63) OR (tmp[62:0]) + 3: dst[63:0] := (1 << 63) OR (tmp[62:0]) + ESAC + + RETURN dst +} +FOR j := 0 to 1 + i := j*64 IF k[j] - dst[i+31:i] := ConvertUnsignedInt32_To_FP32(a[i+31:i]) + dst[i+63:i] := RANGE(a[i+63:i], b[i+63:i], imm8[1:0], imm8[3:2]) ELSE - dst[i+31:i] := src[i+31:i] + dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point - Integer - AVX512F - Convert - - - - Convert packed unsigned 32-bit integers in "a" to packed single-precision (32-bit) floating-point elements, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - + AVX512VL + AVX512DQ + Miscellaneous + + + + + + Calculate the max, min, absolute max, or absolute min (depending on control in "imm8") for packed double-precision (64-bit) floating-point elements in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + imm8[1:0] specifies the operation control: 00 = min, 01 = max, 10 = absolute max, 11 = absolute min. + imm8[3:2] specifies the sign control: 00 = sign from a, 01 = sign from compare result, 10 = clear sign bit, 11 = set sign bit. -FOR j := 0 to 15 - i := j*32 +DEFINE RANGE(src1[63:0], src2[63:0], opCtl[1:0], signSelCtl[1:0]) { + CASE opCtl[1:0] OF + 0: tmp[63:0] := (src1[63:0] <= src2[63:0]) ? src1[63:0] : src2[63:0] + 1: tmp[63:0] := (src1[63:0] <= src2[63:0]) ? src2[63:0] : src1[63:0] + 2: tmp[63:0] := (ABS(src1[63:0]) <= ABS(src2[63:0])) ? src1[63:0] : src2[63:0] + 3: tmp[63:0] := (ABS(src1[63:0]) <= ABS(src2[63:0])) ? src2[63:0] : src1[63:0] + ESAC + + CASE signSelCtl[1:0] OF + 0: dst[63:0] := (src1[63] << 63) OR (tmp[62:0]) + 1: dst[63:0] := tmp[63:0] + 2: dst[63:0] := (0 << 63) OR (tmp[62:0]) + 3: dst[63:0] := (1 << 63) OR (tmp[62:0]) + ESAC + + RETURN dst +} +FOR j := 0 to 1 + i := j*64 IF k[j] - dst[i+31:i] := ConvertUnsignedInt32_To_FP32(a[i+31:i]) + dst[i+63:i] := RANGE(a[i+63:i], b[i+63:i], imm8[1:0], imm8[3:2]) ELSE - dst[i+31:i] := src[i+31:i] + dst[i+63:i] := 0 FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point - Integer - AVX512F - Convert - - - - Convert packed unsigned 32-bit integers in "a" to packed single-precision (32-bit) floating-point elements, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - [round_note] - + AVX512VL + AVX512DQ + Miscellaneous + + + + + Calculate the max, min, absolute max, or absolute min (depending on control in "imm8") for packed double-precision (64-bit) floating-point elements in "a" and "b", and store the results in "dst". + imm8[1:0] specifies the operation control: 00 = min, 01 = max, 10 = absolute max, 11 = absolute min. + imm8[3:2] specifies the sign control: 00 = sign from a, 01 = sign from compare result, 10 = clear sign bit, 11 = set sign bit. -FOR j := 0 to 15 - i := 32*j - IF k[j] - dst[i+31:i] := ConvertUnsignedInt32_To_FP32(a[i+31:i]) - ELSE - dst[i+31:i] := 0 - FI +DEFINE RANGE(src1[63:0], src2[63:0], opCtl[1:0], signSelCtl[1:0]) { + CASE opCtl[1:0] OF + 0: tmp[63:0] := (src1[63:0] <= src2[63:0]) ? src1[63:0] : src2[63:0] + 1: tmp[63:0] := (src1[63:0] <= src2[63:0]) ? src2[63:0] : src1[63:0] + 2: tmp[63:0] := (ABS(src1[63:0]) <= ABS(src2[63:0])) ? src1[63:0] : src2[63:0] + 3: tmp[63:0] := (ABS(src1[63:0]) <= ABS(src2[63:0])) ? src2[63:0] : src1[63:0] + ESAC + + CASE signSelCtl[1:0] OF + 0: dst[63:0] := (src1[63] << 63) OR (tmp[62:0]) + 1: dst[63:0] := tmp[63:0] + 2: dst[63:0] := (0 << 63) OR (tmp[62:0]) + 3: dst[63:0] := (1 << 63) OR (tmp[62:0]) + ESAC + + RETURN dst +} +FOR j := 0 to 1 + i := j*64 + dst[i+63:i] := RANGE(a[i+63:i], b[i+63:i], imm8[1:0], imm8[3:2]) ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point - Integer - AVX512F - Convert - - - Convert packed unsigned 32-bit integers in "a" to packed single-precision (32-bit) floating-point elements, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - + AVX512VL + AVX512DQ + Miscellaneous + + + + + + + Calculate the max, min, absolute max, or absolute min (depending on control in "imm8") for packed single-precision (32-bit) floating-point elements in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + imm8[1:0] specifies the operation control: 00 = min, 01 = max, 10 = absolute max, 11 = absolute min. + imm8[3:2] specifies the sign control: 00 = sign from a, 01 = sign from compare result, 10 = clear sign bit, 11 = set sign bit. -FOR j := 0 to 15 - i := 32*j +DEFINE RANGE(src1[31:0], src2[31:0], opCtl[1:0], signSelCtl[1:0]) { + CASE opCtl[1:0] OF + 0: tmp[31:0] := (src1[31:0] <= src2[31:0]) ? src1[31:0] : src2[31:0] + 1: tmp[31:0] := (src1[31:0] <= src2[31:0]) ? src2[31:0] : src1[31:0] + 2: tmp[31:0] := (ABS(src1[31:0]) <= ABS(src2[31:0])) ? src1[31:0] : src2[31:0] + 3: tmp[31:0] := (ABS(src1[31:0]) <= ABS(src2[31:0])) ? src2[31:0] : src1[31:0] + ESAC + + CASE signSelCtl[1:0] OF + 0: dst[31:0] := (src1[31] << 31) OR (tmp[30:0]) + 1: dst[31:0] := tmp[63:0] + 2: dst[31:0] := (0 << 31) OR (tmp[30:0]) + 3: dst[31:0] := (1 << 31) OR (tmp[30:0]) + ESAC + + RETURN dst +} +FOR j := 0 to 7 + i := j*32 IF k[j] - dst[i+31:i] := ConvertUnsignedInt32_To_FP32(a[i+31:i]) + dst[i+31:i] := RANGE(a[i+31:i], b[i+31:i], imm8[1:0], imm8[3:2]) ELSE - dst[i+31:i] := 0 + dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:512] := 0 - - -
immintrin.h
- - - Floating Point - Integer - AVX512F - Convert - - - - Convert the unsigned 64-bit integer "b" to a double-precision (64-bit) floating-point element, store the result in the lower element of "dst", and copy the upper element from "a" to the upper element of "dst". - [round_note] - - -dst[63:0] := Convert_UnsignedInt64_To_FP64(b[63:0]) -dst[127:64] := a[127:64] -dst[MAX:128] := 0 - - -
immintrin.h
- - - Floating Point - Integer - AVX512F - Convert - - - Convert the unsigned 32-bit integer "b" to a double-precision (64-bit) floating-point element, store the result in the lower element of "dst", and copy the upper element from "a" to the upper element of "dst". - -dst[63:0] := Convert_UnsignedInt32_To_FP64(b[31:0]) -dst[127:64] := a[127:64] -dst[MAX:128] := 0 - - -
immintrin.h
- - - Floating Point - Integer - AVX512F - Convert - - - Convert the unsigned 64-bit integer "b" to a double-precision (64-bit) floating-point element, store the result in the lower element of "dst", and copy the upper element from "a" to the upper element of "dst". - -dst[63:0] := Convert_UnsignedInt64_To_FP64(b[63:0]) -dst[127:64] := a[127:64] -dst[MAX:128] := 0 - - -
immintrin.h
- - - Floating Point - Integer - AVX512F - Convert - - - - Convert the unsigned 32-bit integer "b" to a single-precision (32-bit) floating-point element, store the result in the lower element of "dst", and copy the upper 3 packed elements from "a" to the upper elements of "dst". - [round_note] - - -dst[31:0] := Convert_UnsignedInt32_To_FP32(b[31:0]) -dst[127:32] := a[127:32] -dst[MAX:128] := 0 - - -
immintrin.h
- - - Floating Point - Integer - AVX512F - Convert - - - - Convert the unsigned 64-bit integer "b" to a single-precision (32-bit) floating-point element, store the result in the lower element of "dst", and copy the upper 3 packed elements from "a" to the upper elements of "dst". - [round_note] - - -dst[31:0] := Convert_UnsignedInt64_To_FP32(b[63:0]) -dst[127:32] := a[127:32] -dst[MAX:128] := 0 - - -
immintrin.h
- - - Floating Point - Integer - AVX512F - Convert - - - Convert the unsigned 32-bit integer "b" to a single-precision (32-bit) floating-point element, store the result in the lower element of "dst", and copy the upper 3 packed elements from "a" to the upper elements of "dst". - -dst[31:0] := Convert_UnsignedInt32_To_FP32(b[31:0]) -dst[127:32] := a[127:32] -dst[MAX:128] := 0 - - -
immintrin.h
- - - Floating Point - Integer - AVX512F - Convert - - - Convert the unsigned 64-bit integer "b" to a single-precision (32-bit) floating-point element, store the result in the lower element of "dst", and copy the upper 3 packed elements from "a" to the upper elements of "dst". - -dst[31:0] := Convert_UnsignedInt64_To_FP32(b[63:0]) -dst[127:32] := a[127:32] -dst[MAX:128] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Floating Point - AVX512F - Arithmetic - - - Divide packed double-precision (64-bit) floating-point elements in "a" by packed elements in "b", and store the results in "dst". + AVX512VL + AVX512DQ + Miscellaneous + + + + + + Calculate the max, min, absolute max, or absolute min (depending on control in "imm8") for packed single-precision (32-bit) floating-point elements in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + imm8[1:0] specifies the operation control: 00 = min, 01 = max, 10 = absolute max, 11 = absolute min. + imm8[3:2] specifies the sign control: 00 = sign from a, 01 = sign from compare result, 10 = clear sign bit, 11 = set sign bit. +DEFINE RANGE(src1[31:0], src2[31:0], opCtl[1:0], signSelCtl[1:0]) { + CASE opCtl[1:0] OF + 0: tmp[31:0] := (src1[31:0] <= src2[31:0]) ? src1[31:0] : src2[31:0] + 1: tmp[31:0] := (src1[31:0] <= src2[31:0]) ? src2[31:0] : src1[31:0] + 2: tmp[31:0] := (ABS(src1[31:0]) <= ABS(src2[31:0])) ? src1[31:0] : src2[31:0] + 3: tmp[31:0] := (ABS(src1[31:0]) <= ABS(src2[31:0])) ? src2[31:0] : src1[31:0] + ESAC + + CASE signSelCtl[1:0] OF + 0: dst[31:0] := (src1[31] << 31) OR (tmp[30:0]) + 1: dst[31:0] := tmp[63:0] + 2: dst[31:0] := (0 << 31) OR (tmp[30:0]) + 3: dst[31:0] := (1 << 31) OR (tmp[30:0]) + ESAC + + RETURN dst +} FOR j := 0 to 7 - i := 64*j - dst[i+63:i] := a[i+63:i] / b[i+63:i] + i := j*32 + IF k[j] + dst[i+31:i] := RANGE(a[i+31:i], b[i+31:i], imm8[1:0], imm8[3:2]) + ELSE + dst[i+31:i] := 0 + FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Floating Point - AVX512F - Arithmetic - - - - Divide packed double-precision (64-bit) floating-point elements in "a" by packed elements in "b", =and store the results in "dst". - [round_note] - + AVX512VL + AVX512DQ + Miscellaneous + + + + + Calculate the max, min, absolute max, or absolute min (depending on control in "imm8") for packed single-precision (32-bit) floating-point elements in "a" and "b", and store the results in "dst". + imm8[1:0] specifies the operation control: 00 = min, 01 = max, 10 = absolute max, 11 = absolute min. + imm8[3:2] specifies the sign control: 00 = sign from a, 01 = sign from compare result, 10 = clear sign bit, 11 = set sign bit. +DEFINE RANGE(src1[31:0], src2[31:0], opCtl[1:0], signSelCtl[1:0]) { + CASE opCtl[1:0] OF + 0: tmp[31:0] := (src1[31:0] <= src2[31:0]) ? src1[31:0] : src2[31:0] + 1: tmp[31:0] := (src1[31:0] <= src2[31:0]) ? src2[31:0] : src1[31:0] + 2: tmp[31:0] := (ABS(src1[31:0]) <= ABS(src2[31:0])) ? src1[31:0] : src2[31:0] + 3: tmp[31:0] := (ABS(src1[31:0]) <= ABS(src2[31:0])) ? src2[31:0] : src1[31:0] + ESAC + + CASE signSelCtl[1:0] OF + 0: dst[31:0] := (src1[31] << 31) OR (tmp[30:0]) + 1: dst[31:0] := tmp[63:0] + 2: dst[31:0] := (0 << 31) OR (tmp[30:0]) + 3: dst[31:0] := (1 << 31) OR (tmp[30:0]) + ESAC + + RETURN dst +} FOR j := 0 to 7 - i := 64*j - dst[i+63:i] := a[i+63:i] / b[i+63:i] + i := j*32 + dst[i+31:i] := RANGE(a[i+31:i], b[i+31:i], imm8[1:0], imm8[3:2]) ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Floating Point - AVX512F - Arithmetic - - - - - Divide packed double-precision (64-bit) floating-point elements in "a" by packed elements in "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + AVX512DQ + Miscellaneous + + + + + + + Calculate the max, min, absolute max, or absolute min (depending on control in "imm8") for packed single-precision (32-bit) floating-point elements in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + imm8[1:0] specifies the operation control: 00 = min, 01 = max, 10 = absolute max, 11 = absolute min. + imm8[3:2] specifies the sign control: 00 = sign from a, 01 = sign from compare result, 10 = clear sign bit, 11 = set sign bit. -FOR j := 0 to 7 - i := 64*j +DEFINE RANGE(src1[31:0], src2[31:0], opCtl[1:0], signSelCtl[1:0]) { + CASE opCtl[1:0] OF + 0: tmp[31:0] := (src1[31:0] <= src2[31:0]) ? src1[31:0] : src2[31:0] + 1: tmp[31:0] := (src1[31:0] <= src2[31:0]) ? src2[31:0] : src1[31:0] + 2: tmp[31:0] := (ABS(src1[31:0]) <= ABS(src2[31:0])) ? src1[31:0] : src2[31:0] + 3: tmp[31:0] := (ABS(src1[31:0]) <= ABS(src2[31:0])) ? src2[31:0] : src1[31:0] + ESAC + + CASE signSelCtl[1:0] OF + 0: dst[31:0] := (src1[31] << 31) OR (tmp[30:0]) + 1: dst[31:0] := tmp[63:0] + 2: dst[31:0] := (0 << 31) OR (tmp[30:0]) + 3: dst[31:0] := (1 << 31) OR (tmp[30:0]) + ESAC + + RETURN dst +} +FOR j := 0 to 15 + i := j*32 IF k[j] - dst[i+63:i] := a[i+63:i] / b[i+63:i] + dst[i+31:i] := RANGE(a[i+31:i], b[i+31:i], imm8[1:0], imm8[3:2]) ELSE - dst[i+63:i] := src[i+63:i] + dst[i+31:i] := src[i+31:i] FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - AVX512F - Arithmetic - - - - - - - Divide packed double-precision (64-bit) floating-point elements in "a" by packed elements in "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - [round_note] - + AVX512DQ + Miscellaneous + + + + + + + + Calculate the max, min, absolute max, or absolute min (depending on control in "imm8") for packed single-precision (32-bit) floating-point elements in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + imm8[1:0] specifies the operation control: 00 = min, 01 = max, 10 = absolute max, 11 = absolute min. + imm8[3:2] specifies the sign control: 00 = sign from a, 01 = sign from compare result, 10 = clear sign bit, 11 = set sign bit. [sae_note] -FOR j := 0 to 7 - i := 64*j +DEFINE RANGE(src1[31:0], src2[31:0], opCtl[1:0], signSelCtl[1:0]) { + CASE opCtl[1:0] OF + 0: tmp[31:0] := (src1[31:0] <= src2[31:0]) ? src1[31:0] : src2[31:0] + 1: tmp[31:0] := (src1[31:0] <= src2[31:0]) ? src2[31:0] : src1[31:0] + 2: tmp[31:0] := (ABS(src1[31:0]) <= ABS(src2[31:0])) ? src1[31:0] : src2[31:0] + 3: tmp[31:0] := (ABS(src1[31:0]) <= ABS(src2[31:0])) ? src2[31:0] : src1[31:0] + ESAC + + CASE signSelCtl[1:0] OF + 0: dst[31:0] := (src1[31] << 31) OR (tmp[30:0]) + 1: dst[31:0] := tmp[63:0] + 2: dst[31:0] := (0 << 31) OR (tmp[30:0]) + 3: dst[31:0] := (1 << 31) OR (tmp[30:0]) + ESAC + + RETURN dst +} +FOR j := 0 to 15 + i := j*32 IF k[j] - dst[i+63:i] := a[i+63:i] / b[i+63:i] + dst[i+31:i] := RANGE(a[i+31:i], b[i+31:i], imm8[1:0], imm8[3:2]) ELSE - dst[i+63:i] := src[i+63:i] + dst[i+31:i] := src[i+31:i] FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - AVX512F - Arithmetic - - - - Divide packed double-precision (64-bit) floating-point elements in "a" by packed elements in "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + AVX512DQ + Miscellaneous + + + + + + Calculate the max, min, absolute max, or absolute min (depending on control in "imm8") for packed single-precision (32-bit) floating-point elements in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + imm8[1:0] specifies the operation control: 00 = min, 01 = max, 10 = absolute max, 11 = absolute min. + imm8[3:2] specifies the sign control: 00 = sign from a, 01 = sign from compare result, 10 = clear sign bit, 11 = set sign bit. -FOR j := 0 to 7 - i := 64*j +DEFINE RANGE(src1[31:0], src2[31:0], opCtl[1:0], signSelCtl[1:0]) { + CASE opCtl[1:0] OF + 0: tmp[31:0] := (src1[31:0] <= src2[31:0]) ? src1[31:0] : src2[31:0] + 1: tmp[31:0] := (src1[31:0] <= src2[31:0]) ? src2[31:0] : src1[31:0] + 2: tmp[31:0] := (ABS(src1[31:0]) <= ABS(src2[31:0])) ? src1[31:0] : src2[31:0] + 3: tmp[31:0] := (ABS(src1[31:0]) <= ABS(src2[31:0])) ? src2[31:0] : src1[31:0] + ESAC + + CASE signSelCtl[1:0] OF + 0: dst[31:0] := (src1[31] << 31) OR (tmp[30:0]) + 1: dst[31:0] := tmp[63:0] + 2: dst[31:0] := (0 << 31) OR (tmp[30:0]) + 3: dst[31:0] := (1 << 31) OR (tmp[30:0]) + ESAC + + RETURN dst +} +FOR j := 0 to 15 + i := j*32 IF k[j] - dst[i+63:i] := a[i+63:i] / b[i+63:i] + dst[i+31:i] := RANGE(a[i+31:i], b[i+31:i], imm8[1:0], imm8[3:2]) ELSE - dst[i+63:i] := 0 + dst[i+31:i] := 0 FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - AVX512F - Arithmetic - - - - - Divide packed double-precision (64-bit) floating-point elements in "a" by packed elements in "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - [round_note] - + AVX512DQ + Miscellaneous + + + + + + + Calculate the max, min, absolute max, or absolute min (depending on control in "imm8") for packed single-precision (32-bit) floating-point elements in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + imm8[1:0] specifies the operation control: 00 = min, 01 = max, 10 = absolute max, 11 = absolute min. + imm8[3:2] specifies the sign control: 00 = sign from a, 01 = sign from compare result, 10 = clear sign bit, 11 = set sign bit. [sae_note] -FOR j := 0 to 7 - i := 64*j +DEFINE RANGE(src1[31:0], src2[31:0], opCtl[1:0], signSelCtl[1:0]) { + CASE opCtl[1:0] OF + 0: tmp[31:0] := (src1[31:0] <= src2[31:0]) ? src1[31:0] : src2[31:0] + 1: tmp[31:0] := (src1[31:0] <= src2[31:0]) ? src2[31:0] : src1[31:0] + 2: tmp[31:0] := (ABS(src1[31:0]) <= ABS(src2[31:0])) ? src1[31:0] : src2[31:0] + 3: tmp[31:0] := (ABS(src1[31:0]) <= ABS(src2[31:0])) ? src2[31:0] : src1[31:0] + ESAC + + CASE signSelCtl[1:0] OF + 0: dst[31:0] := (src1[31] << 31) OR (tmp[30:0]) + 1: dst[31:0] := tmp[63:0] + 2: dst[31:0] := (0 << 31) OR (tmp[30:0]) + 3: dst[31:0] := (1 << 31) OR (tmp[30:0]) + ESAC + + RETURN dst +} +FOR j := 0 to 15 + i := j*32 IF k[j] - dst[i+63:i] := a[i+63:i] / b[i+63:i] + dst[i+31:i] := RANGE(a[i+31:i], b[i+31:i], imm8[1:0], imm8[3:2]) ELSE - dst[i+63:i] := 0 + dst[i+31:i] := 0 FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - AVX512F - Arithmetic - - - Divide packed single-precision (32-bit) floating-point elements in "a" by packed elements in "b", and store the results in "dst". - -FOR j := 0 to 15 - i := 32*j - dst[i+31:i] := a[i+31:i] / b[i+31:i] -ENDFOR -dst[MAX:512] := 0 - - -
immintrin.h
- - - Floating Point - AVX512F - Arithmetic - - - - Divide packed single-precision (32-bit) floating-point elements in "a" by packed elements in "b", and store the results in "dst". - [round_note] - + AVX512DQ + Miscellaneous + + + + + Calculate the max, min, absolute max, or absolute min (depending on control in "imm8") for packed single-precision (32-bit) floating-point elements in "a" and "b", and store the results in "dst". + imm8[1:0] specifies the operation control: 00 = min, 01 = max, 10 = absolute max, 11 = absolute min. + imm8[3:2] specifies the sign control: 00 = sign from a, 01 = sign from compare result, 10 = clear sign bit, 11 = set sign bit. +DEFINE RANGE(src1[31:0], src2[31:0], opCtl[1:0], signSelCtl[1:0]) { + CASE opCtl[1:0] OF + 0: tmp[31:0] := (src1[31:0] <= src2[31:0]) ? src1[31:0] : src2[31:0] + 1: tmp[31:0] := (src1[31:0] <= src2[31:0]) ? src2[31:0] : src1[31:0] + 2: tmp[31:0] := (ABS(src1[31:0]) <= ABS(src2[31:0])) ? src1[31:0] : src2[31:0] + 3: tmp[31:0] := (ABS(src1[31:0]) <= ABS(src2[31:0])) ? src2[31:0] : src1[31:0] + ESAC + + CASE signSelCtl[1:0] OF + 0: dst[31:0] := (src1[31] << 31) OR (tmp[30:0]) + 1: dst[31:0] := tmp[63:0] + 2: dst[31:0] := (0 << 31) OR (tmp[30:0]) + 3: dst[31:0] := (1 << 31) OR (tmp[30:0]) + ESAC + + RETURN dst +} FOR j := 0 to 15 - i := 32*j - dst[i+31:i] := a[i+31:i] / b[i+31:i] + i := j*32 + dst[i+31:i] := RANGE(a[i+31:i], b[i+31:i], imm8[1:0], imm8[3:2]) ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - AVX512F - Arithmetic - - - - - Divide packed single-precision (32-bit) floating-point elements in "a" by packed elements in "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + AVX512DQ + Miscellaneous + + + + + + Calculate the max, min, absolute max, or absolute min (depending on control in "imm8") for packed single-precision (32-bit) floating-point elements in "a" and "b", and store the results in "dst". + imm8[1:0] specifies the operation control: 00 = min, 01 = max, 10 = absolute max, 11 = absolute min. + imm8[3:2] specifies the sign control: 00 = sign from a, 01 = sign from compare result, 10 = clear sign bit, 11 = set sign bit. [sae_note] +DEFINE RANGE(src1[31:0], src2[31:0], opCtl[1:0], signSelCtl[1:0]) { + CASE opCtl[1:0] OF + 0: tmp[31:0] := (src1[31:0] <= src2[31:0]) ? src1[31:0] : src2[31:0] + 1: tmp[31:0] := (src1[31:0] <= src2[31:0]) ? src2[31:0] : src1[31:0] + 2: tmp[31:0] := (ABS(src1[31:0]) <= ABS(src2[31:0])) ? src1[31:0] : src2[31:0] + 3: tmp[31:0] := (ABS(src1[31:0]) <= ABS(src2[31:0])) ? src2[31:0] : src1[31:0] + ESAC + + CASE signSelCtl[1:0] OF + 0: dst[31:0] := (src1[31] << 31) OR (tmp[30:0]) + 1: dst[31:0] := tmp[63:0] + 2: dst[31:0] := (0 << 31) OR (tmp[30:0]) + 3: dst[31:0] := (1 << 31) OR (tmp[30:0]) + ESAC + + RETURN dst +} FOR j := 0 to 15 - i := 32*j - IF k[j] - dst[i+31:i] := a[i+31:i] / b[i+31:i] - ELSE - dst[i+31:i] := src[i+31:i] - FI + i := j*32 + dst[i+31:i] := RANGE(a[i+31:i], b[i+31:i], imm8[1:0], imm8[3:2]) ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - AVX512F - Arithmetic - - - - - - - Divide packed single-precision (32-bit) floating-point elements in "a" by packed elements in "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - [round_note] - + AVX512VL + AVX512DQ + Miscellaneous + + + + + + + Calculate the max, min, absolute max, or absolute min (depending on control in "imm8") for packed single-precision (32-bit) floating-point elements in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + imm8[1:0] specifies the operation control: 00 = min, 01 = max, 10 = absolute max, 11 = absolute min. + imm8[3:2] specifies the sign control: 00 = sign from a, 01 = sign from compare result, 10 = clear sign bit, 11 = set sign bit. -FOR j := 0 to 15 - i := 32*j +DEFINE RANGE(src1[31:0], src2[31:0], opCtl[1:0], signSelCtl[1:0]) { + CASE opCtl[1:0] OF + 0: tmp[31:0] := (src1[31:0] <= src2[31:0]) ? src1[31:0] : src2[31:0] + 1: tmp[31:0] := (src1[31:0] <= src2[31:0]) ? src2[31:0] : src1[31:0] + 2: tmp[31:0] := (ABS(src1[31:0]) <= ABS(src2[31:0])) ? src1[31:0] : src2[31:0] + 3: tmp[31:0] := (ABS(src1[31:0]) <= ABS(src2[31:0])) ? src2[31:0] : src1[31:0] + ESAC + + CASE signSelCtl[1:0] OF + 0: dst[31:0] := (src1[31] << 31) OR (tmp[30:0]) + 1: dst[31:0] := tmp[63:0] + 2: dst[31:0] := (0 << 31) OR (tmp[30:0]) + 3: dst[31:0] := (1 << 31) OR (tmp[30:0]) + ESAC + + RETURN dst +} +FOR j := 0 to 3 + i := j*32 IF k[j] - dst[i+31:i] := a[i+31:i] / b[i+31:i] + dst[i+31:i] := RANGE(a[i+31:i], b[i+31:i], imm8[1:0], imm8[3:2]) ELSE dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point - AVX512F - Arithmetic - - - - Divide packed single-precision (32-bit) floating-point elements in "a" by packed elements in "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + AVX512VL + AVX512DQ + Miscellaneous + + + + + + Calculate the max, min, absolute max, or absolute min (depending on control in "imm8") for packed single-precision (32-bit) floating-point elements in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + imm8[1:0] specifies the operation control: 00 = min, 01 = max, 10 = absolute max, 11 = absolute min. + imm8[3:2] specifies the sign control: 00 = sign from a, 01 = sign from compare result, 10 = clear sign bit, 11 = set sign bit. -FOR j := 0 to 15 - i := 32*j +DEFINE RANGE(src1[31:0], src2[31:0], opCtl[1:0], signSelCtl[1:0]) { + CASE opCtl[1:0] OF + 0: tmp[31:0] := (src1[31:0] <= src2[31:0]) ? src1[31:0] : src2[31:0] + 1: tmp[31:0] := (src1[31:0] <= src2[31:0]) ? src2[31:0] : src1[31:0] + 2: tmp[31:0] := (ABS(src1[31:0]) <= ABS(src2[31:0])) ? src1[31:0] : src2[31:0] + 3: tmp[31:0] := (ABS(src1[31:0]) <= ABS(src2[31:0])) ? src2[31:0] : src1[31:0] + ESAC + + CASE signSelCtl[1:0] OF + 0: dst[31:0] := (src1[31] << 31) OR (tmp[30:0]) + 1: dst[31:0] := tmp[63:0] + 2: dst[31:0] := (0 << 31) OR (tmp[30:0]) + 3: dst[31:0] := (1 << 31) OR (tmp[30:0]) + ESAC + + RETURN dst +} +FOR j := 0 to 3 + i := j*32 IF k[j] - dst[i+31:i] := a[i+31:i] / b[i+31:i] + dst[i+31:i] := RANGE(a[i+31:i], b[i+31:i], imm8[1:0], imm8[3:2]) ELSE dst[i+31:i] := 0 FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point - AVX512F - Arithmetic - - - - - Divide packed single-precision (32-bit) floating-point elements in "a" by packed elements in "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - [round_note] - + AVX512VL + AVX512DQ + Miscellaneous + + + + + Calculate the max, min, absolute max, or absolute min (depending on control in "imm8") for packed single-precision (32-bit) floating-point elements in "a" and "b", and store the results in "dst". + imm8[1:0] specifies the operation control: 00 = min, 01 = max, 10 = absolute max, 11 = absolute min. + imm8[3:2] specifies the sign control: 00 = sign from a, 01 = sign from compare result, 10 = clear sign bit, 11 = set sign bit. -FOR j := 0 to 15 - i := 32*j - IF k[j] - dst[i+31:i] := a[i+31:i] / b[i+31:i] - ELSE - dst[i+31:i] := 0 - FI +DEFINE RANGE(src1[31:0], src2[31:0], opCtl[1:0], signSelCtl[1:0]) { + CASE opCtl[1:0] OF + 0: tmp[31:0] := (src1[31:0] <= src2[31:0]) ? src1[31:0] : src2[31:0] + 1: tmp[31:0] := (src1[31:0] <= src2[31:0]) ? src2[31:0] : src1[31:0] + 2: tmp[31:0] := (ABS(src1[31:0]) <= ABS(src2[31:0])) ? src1[31:0] : src2[31:0] + 3: tmp[31:0] := (ABS(src1[31:0]) <= ABS(src2[31:0])) ? src2[31:0] : src1[31:0] + ESAC + + CASE signSelCtl[1:0] OF + 0: dst[31:0] := (src1[31] << 31) OR (tmp[30:0]) + 1: dst[31:0] := tmp[63:0] + 2: dst[31:0] := (0 << 31) OR (tmp[30:0]) + 3: dst[31:0] := (1 << 31) OR (tmp[30:0]) + ESAC + + RETURN dst +} +FOR j := 0 to 3 + i := j*32 + dst[i+31:i] := RANGE(a[i+31:i], b[i+31:i], imm8[1:0], imm8[3:2]) ENDFOR -dst[MAX:512] := 0 - - -
immintrin.h
- - - Floating Point - AVX512F - Arithmetic - - - - Divide the lower double-precision (64-bit) floating-point element in "a" by the lower double-precision (64-bit) floating-point element in "b", store the result in the lower element of "dst", and copy the upper element from "a" to the upper element of "dst". - [round_note] - - -dst[63:0] := a[63:0] / b[63:0] -dst[127:64] := a[127:64] dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point - AVX512F - Arithmetic - - - - - - Divide the lower double-precision (64-bit) floating-point element in "a" by the lower double-precision (64-bit) floating-point element in "b", store the result in the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". - [round_note] - + AVX512DQ + Miscellaneous + + + + + + + + Calculate the max, min, absolute max, or absolute min (depending on control in "imm8") for the lower double-precision (64-bit) floating-point element in "a" and "b", store the result in the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". + imm8[1:0] specifies the operation control: 00 = min, 01 = max, 10 = absolute max, 11 = absolute min. + imm8[3:2] specifies the sign control: 00 = sign from a, 01 = sign from compare result, 10 = clear sign bit, 11 = set sign bit. [sae_note] +DEFINE RANGE(src1[63:0], src2[63:0], opCtl[1:0], signSelCtl[1:0]) { + CASE opCtl[1:0] OF + 0: tmp[63:0] := (src1[63:0] <= src2[63:0]) ? src1[63:0] : src2[63:0] + 1: tmp[63:0] := (src1[63:0] <= src2[63:0]) ? src2[63:0] : src1[63:0] + 2: tmp[63:0] := (ABS(src1[63:0]) <= ABS(src2[63:0])) ? src1[63:0] : src2[63:0] + 3: tmp[63:0] := (ABS(src1[63:0]) <= ABS(src2[63:0])) ? src2[63:0] : src1[63:0] + ESAC + + CASE signSelCtl[1:0] OF + 0: dst[63:0] := (src1[63] << 63) OR (tmp[62:0]) + 1: dst[63:0] := tmp[63:0] + 2: dst[63:0] := (0 << 63) OR (tmp[62:0]) + 3: dst[63:0] := (1 << 63) OR (tmp[62:0]) + ESAC + + RETURN dst +} IF k[0] - dst[63:0] := a[63:0] / b[63:0] + dst[63:0] := RANGE(a[63:0], b[63:0], imm8[1:0], imm8[3:2]) ELSE dst[63:0] := src[63:0] FI dst[127:64] := a[127:64] dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point - AVX512F - Arithmetic - - - - - Divide the lower double-precision (64-bit) floating-point element in "a" by the lower double-precision (64-bit) floating-point element in "b", store the result in the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". + AVX512DQ + Miscellaneous + + + + + + + Calculate the max, min, absolute max, or absolute min (depending on control in "imm8") for the lower double-precision (64-bit) floating-point element in "a" and "b", store the result in the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". + imm8[1:0] specifies the operation control: 00 = min, 01 = max, 10 = absolute max, 11 = absolute min. + imm8[3:2] specifies the sign control: 00 = sign from a, 01 = sign from compare result, 10 = clear sign bit, 11 = set sign bit. +DEFINE RANGE(src1[63:0], src2[63:0], opCtl[1:0], signSelCtl[1:0]) { + CASE opCtl[1:0] OF + 0: tmp[63:0] := (src1[63:0] <= src2[63:0]) ? src1[63:0] : src2[63:0] + 1: tmp[63:0] := (src1[63:0] <= src2[63:0]) ? src2[63:0] : src1[63:0] + 2: tmp[63:0] := (ABS(src1[63:0]) <= ABS(src2[63:0])) ? src1[63:0] : src2[63:0] + 3: tmp[63:0] := (ABS(src1[63:0]) <= ABS(src2[63:0])) ? src2[63:0] : src1[63:0] + ESAC + + CASE signSelCtl[1:0] OF + 0: dst[63:0] := (src1[63] << 63) OR (tmp[62:0]) + 1: dst[63:0] := tmp[63:0] + 2: dst[63:0] := (0 << 63) OR (tmp[62:0]) + 3: dst[63:0] := (1 << 63) OR (tmp[62:0]) + ESAC + + RETURN dst +} IF k[0] - dst[63:0] := a[63:0] / b[63:0] + dst[63:0] := RANGE(a[63:0], b[63:0], imm8[1:0], imm8[3:2]) ELSE dst[63:0] := src[63:0] FI dst[127:64] := a[127:64] dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point - AVX512F - Arithmetic - - - - - Divide the lower double-precision (64-bit) floating-point element in "a" by the lower double-precision (64-bit) floating-point element in "b", store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". - [round_note] - + AVX512DQ + Miscellaneous + + + + + + + Calculate the max, min, absolute max, or absolute min (depending on control in "imm8") for the lower double-precision (64-bit) floating-point element in "a" and "b", store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". + imm8[1:0] specifies the operation control: 00 = min, 01 = max, 10 = absolute max, 11 = absolute min. + imm8[3:2] specifies the sign control: 00 = sign from a, 01 = sign from compare result, 10 = clear sign bit, 11 = set sign bit. [sae_note] +DEFINE RANGE(src1[63:0], src2[63:0], opCtl[1:0], signSelCtl[1:0]) { + CASE opCtl[1:0] OF + 0: tmp[63:0] := (src1[63:0] <= src2[63:0]) ? src1[63:0] : src2[63:0] + 1: tmp[63:0] := (src1[63:0] <= src2[63:0]) ? src2[63:0] : src1[63:0] + 2: tmp[63:0] := (ABS(src1[63:0]) <= ABS(src2[63:0])) ? src1[63:0] : src2[63:0] + 3: tmp[63:0] := (ABS(src1[63:0]) <= ABS(src2[63:0])) ? src2[63:0] : src1[63:0] + ESAC + + CASE signSelCtl[1:0] OF + 0: dst[63:0] := (src1[63] << 63) OR (tmp[62:0]) + 1: dst[63:0] := tmp[63:0] + 2: dst[63:0] := (0 << 63) OR (tmp[62:0]) + 3: dst[63:0] := (1 << 63) OR (tmp[62:0]) + ESAC + + RETURN dst +} IF k[0] - dst[63:0] := a[63:0] / b[63:0] + dst[63:0] := RANGE(a[63:0], b[63:0], imm8[1:0], imm8[3:2]) ELSE dst[63:0] := 0 FI dst[127:64] := a[127:64] dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point - AVX512F - Arithmetic - - - - Divide the lower double-precision (64-bit) floating-point element in "a" by the lower double-precision (64-bit) floating-point element in "b", store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". + AVX512DQ + Miscellaneous + + + + + + Calculate the max, min, absolute max, or absolute min (depending on control in "imm8") for the lower double-precision (64-bit) floating-point element in "a" and "b", store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". + imm8[1:0] specifies the operation control: 00 = min, 01 = max, 10 = absolute max, 11 = absolute min. + imm8[3:2] specifies the sign control: 00 = sign from a, 01 = sign from compare result, 10 = clear sign bit, 11 = set sign bit. +DEFINE RANGE(src1[63:0], src2[63:0], opCtl[1:0], signSelCtl[1:0]) { + CASE opCtl[1:0] OF + 0: tmp[63:0] := (src1[63:0] <= src2[63:0]) ? src1[63:0] : src2[63:0] + 1: tmp[63:0] := (src1[63:0] <= src2[63:0]) ? src2[63:0] : src1[63:0] + 2: tmp[63:0] := (ABS(src1[63:0]) <= ABS(src2[63:0])) ? src1[63:0] : src2[63:0] + 3: tmp[63:0] := (ABS(src1[63:0]) <= ABS(src2[63:0])) ? src2[63:0] : src1[63:0] + ESAC + + CASE signSelCtl[1:0] OF + 0: dst[63:0] := (src1[63] << 63) OR (tmp[62:0]) + 1: dst[63:0] := tmp[63:0] + 2: dst[63:0] := (0 << 63) OR (tmp[62:0]) + 3: dst[63:0] := (1 << 63) OR (tmp[62:0]) + ESAC + + RETURN dst +} IF k[0] - dst[63:0] := a[63:0] / b[63:0] + dst[63:0] := RANGE(a[63:0], b[63:0], imm8[1:0], imm8[3:2]) ELSE dst[63:0] := 0 FI dst[127:64] := a[127:64] dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point - AVX512F - Arithmetic - - - - Divide the lower single-precision (32-bit) floating-point element in "a" by the lower single-precision (32-bit) floating-point element in "b", store the result in the lower element of "dst", and copy the upper 3 packed elements from "a" to the upper elements of "dst". - [round_note] - + AVX512DQ + Miscellaneous + + + + + + Calculate the max, min, absolute max, or absolute min (depending on control in "imm8") for the lower double-precision (64-bit) floating-point element in "a" and "b", store the result in the lower element of "dst", and copy the upper element from "a" to the upper element of "dst". + imm8[1:0] specifies the operation control: 00 = min, 01 = max, 10 = absolute max, 11 = absolute min. + imm8[3:2] specifies the sign control: 00 = sign from a, 01 = sign from compare result, 10 = clear sign bit, 11 = set sign bit. [sae_note] -dst[31:0] := a[31:0] / b[31:0] -dst[127:32] := a[127:32] +DEFINE RANGE(src1[63:0], src2[63:0], opCtl[1:0], signSelCtl[1:0]) { + CASE opCtl[1:0] OF + 0: tmp[63:0] := (src1[63:0] <= src2[63:0]) ? src1[63:0] : src2[63:0] + 1: tmp[63:0] := (src1[63:0] <= src2[63:0]) ? src2[63:0] : src1[63:0] + 2: tmp[63:0] := (ABS(src1[63:0]) <= ABS(src2[63:0])) ? src1[63:0] : src2[63:0] + 3: tmp[63:0] := (ABS(src1[63:0]) <= ABS(src2[63:0])) ? src2[63:0] : src1[63:0] + ESAC + + CASE signSelCtl[1:0] OF + 0: dst[63:0] := (src1[63] << 63) OR (tmp[62:0]) + 1: dst[63:0] := tmp[63:0] + 2: dst[63:0] := (0 << 63) OR (tmp[62:0]) + 3: dst[63:0] := (1 << 63) OR (tmp[62:0]) + ESAC + + RETURN dst +} +dst[63:0] := RANGE(a[63:0], b[63:0], imm8[1:0], imm8[3:2]) +dst[127:64] := a[127:64] dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point - AVX512F - Arithmetic - - - - - - Divide the lower single-precision (32-bit) floating-point element in "a" by the lower single-precision (32-bit) floating-point element in "b", store the result in the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". - [round_note] - + AVX512DQ + Miscellaneous + + + + + + + + Calculate the max, min, absolute max, or absolute min (depending on control in "imm8") for the lower single-precision (32-bit) floating-point element in "a" and "b", store the result in the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". + imm8[1:0] specifies the operation control: 00 = min, 01 = max, 10 = absolute max, 11 = absolute min. + imm8[3:2] specifies the sign control: 00 = sign from a, 01 = sign from compare result, 10 = clear sign bit, 11 = set sign bit. [sae_note] +DEFINE RANGE(src1[31:0], src2[31:0], opCtl[1:0], signSelCtl[1:0]) { + CASE opCtl[1:0] OF + 0: tmp[31:0] := (src1[31:0] <= src2[31:0]) ? src1[31:0] : src2[31:0] + 1: tmp[31:0] := (src1[31:0] <= src2[31:0]) ? src2[31:0] : src1[31:0] + 2: tmp[31:0] := (ABS(src1[31:0]) <= ABS(src2[31:0])) ? src1[31:0] : src2[31:0] + 3: tmp[31:0] := (ABS(src1[31:0]) <= ABS(src2[31:0])) ? src2[31:0] : src1[31:0] + ESAC + + CASE signSelCtl[1:0] OF + 0: dst[31:0] := (src1[31] << 31) OR (tmp[30:0]) + 1: dst[31:0] := tmp[31:0] + 2: dst[31:0] := (0 << 31) OR (tmp[30:0]) + 3: dst[31:0] := (1 << 31) OR (tmp[30:0]) + ESAC + + RETURN dst +} IF k[0] - dst[31:0] := a[31:0] / b[31:0] + dst[31:0] := RANGE(a[31:0], b[31:0], imm8[1:0], imm8[3:2]) ELSE dst[31:0] := src[31:0] FI dst[127:32] := a[127:32] dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point - AVX512F - Arithmetic - - - - - Divide the lower single-precision (32-bit) floating-point element in "a" by the lower single-precision (32-bit) floating-point element in "b", store the result in the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". + AVX512DQ + Miscellaneous + + + + + + + Calculate the max, min, absolute max, or absolute min (depending on control in "imm8") for the lower single-precision (32-bit) floating-point element in "a" and "b", store the result in the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". + imm8[1:0] specifies the operation control: 00 = min, 01 = max, 10 = absolute max, 11 = absolute min. + imm8[3:2] specifies the sign control: 00 = sign from a, 01 = sign from compare result, 10 = clear sign bit, 11 = set sign bit. +DEFINE RANGE(src1[31:0], src2[31:0], opCtl[1:0], signSelCtl[1:0]) { + CASE opCtl[1:0] OF + 0: tmp[31:0] := (src1[31:0] <= src2[31:0]) ? src1[31:0] : src2[31:0] + 1: tmp[31:0] := (src1[31:0] <= src2[31:0]) ? src2[31:0] : src1[31:0] + 2: tmp[31:0] := (ABS(src1[31:0]) <= ABS(src2[31:0])) ? src1[31:0] : src2[31:0] + 3: tmp[31:0] := (ABS(src1[31:0]) <= ABS(src2[31:0])) ? src2[31:0] : src1[31:0] + ESAC + + CASE signSelCtl[1:0] OF + 0: dst[31:0] := (src1[31] << 31) OR (tmp[30:0]) + 1: dst[31:0] := tmp[31:0] + 2: dst[31:0] := (0 << 31) OR (tmp[30:0]) + 3: dst[31:0] := (1 << 31) OR (tmp[30:0]) + ESAC + + RETURN dst +} IF k[0] - dst[31:0] := a[31:0] / b[31:0] + dst[31:0] := RANGE(a[31:0], b[31:0], imm8[1:0], imm8[3:2]) ELSE dst[31:0] := src[31:0] FI dst[127:32] := a[127:32] dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point - AVX512F - Arithmetic - - - - - Divide the lower single-precision (32-bit) floating-point element in "a" by the lower single-precision (32-bit) floating-point element in "b", store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". - [round_note] - + AVX512DQ + Miscellaneous + + + + + + + Calculate the max, min, absolute max, or absolute min (depending on control in "imm8") for the lower single-precision (32-bit) floating-point element in "a" and "b", store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". + imm8[1:0] specifies the operation control: 00 = min, 01 = max, 10 = absolute max, 11 = absolute min. + imm8[3:2] specifies the sign control: 00 = sign from a, 01 = sign from compare result, 10 = clear sign bit, 11 = set sign bit. [sae_note] +DEFINE RANGE(src1[31:0], src2[31:0], opCtl[1:0], signSelCtl[1:0]) { + CASE opCtl[1:0] OF + 0: tmp[31:0] := (src1[31:0] <= src2[31:0]) ? src1[31:0] : src2[31:0] + 1: tmp[31:0] := (src1[31:0] <= src2[31:0]) ? src2[31:0] : src1[31:0] + 2: tmp[31:0] := (ABS(src1[31:0]) <= ABS(src2[31:0])) ? src1[31:0] : src2[31:0] + 3: tmp[31:0] := (ABS(src1[31:0]) <= ABS(src2[31:0])) ? src2[31:0] : src1[31:0] + ESAC + + CASE signSelCtl[1:0] OF + 0: dst[31:0] := (src1[31] << 31) OR (tmp[30:0]) + 1: dst[31:0] := tmp[31:0] + 2: dst[31:0] := (0 << 31) OR (tmp[30:0]) + 3: dst[31:0] := (1 << 31) OR (tmp[30:0]) + ESAC + + RETURN dst +} IF k[0] - dst[31:0] := a[31:0] / b[31:0] + dst[31:0] := RANGE(a[31:0], b[31:0], imm8[1:0], imm8[3:2]) ELSE dst[31:0] := 0 FI dst[127:32] := a[127:32] dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point - AVX512F - Arithmetic - - - - Divide the lower single-precision (32-bit) floating-point element in "a" by the lower single-precision (32-bit) floating-point element in "b", store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". + AVX512DQ + Miscellaneous + + + + + + Calculate the max, min, absolute max, or absolute min (depending on control in "imm8") for the lower single-precision (32-bit) floating-point element in "a" and "b", store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". + imm8[1:0] specifies the operation control: 00 = min, 01 = max, 10 = absolute max, 11 = absolute min. + imm8[3:2] specifies the sign control: 00 = sign from a, 01 = sign from compare result, 10 = clear sign bit, 11 = set sign bit. +DEFINE RANGE(src1[31:0], src2[31:0], opCtl[1:0], signSelCtl[1:0]) { + CASE opCtl[1:0] OF + 0: tmp[31:0] := (src1[31:0] <= src2[31:0]) ? src1[31:0] : src2[31:0] + 1: tmp[31:0] := (src1[31:0] <= src2[31:0]) ? src2[31:0] : src1[31:0] + 2: tmp[31:0] := (ABS(src1[31:0]) <= ABS(src2[31:0])) ? src1[31:0] : src2[31:0] + 3: tmp[31:0] := (ABS(src1[31:0]) <= ABS(src2[31:0])) ? src2[31:0] : src1[31:0] + ESAC + + CASE signSelCtl[1:0] OF + 0: dst[31:0] := (src1[31] << 31) OR (tmp[30:0]) + 1: dst[31:0] := tmp[31:0] + 2: dst[31:0] := (0 << 31) OR (tmp[30:0]) + 3: dst[31:0] := (1 << 31) OR (tmp[30:0]) + ESAC + + RETURN dst +} IF k[0] - dst[31:0] := a[31:0] / b[31:0] + dst[31:0] := RANGE(a[31:0], b[31:0], imm8[1:0], imm8[3:2]) ELSE dst[31:0] := 0 FI dst[127:32] := a[127:32] dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point - AVX512F - Swizzle - - - - Load contiguous active double-precision (64-bit) floating-point elements from "a" (those with their respective bit set in mask "k"), and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + AVX512DQ + Miscellaneous + + + + + + Calculate the max, min, absolute max, or absolute min (depending on control in "imm8") for the lower single-precision (32-bit) floating-point element in "a" and "b", store the result in the lower element of "dst", and copy the upper 3 packed elements from "a" to the upper elements of "dst". + imm8[1:0] specifies the operation control: 00 = min, 01 = max, 10 = absolute max, 11 = absolute min. + imm8[3:2] specifies the sign control: 00 = sign from a, 01 = sign from compare result, 10 = clear sign bit, 11 = set sign bit. [sae_note] -m := 0 +DEFINE RANGE(src1[31:0], src2[31:0], opCtl[1:0], signSelCtl[1:0]) { + CASE opCtl[1:0] OF + 0: tmp[31:0] := (src1[31:0] <= src2[31:0]) ? src1[31:0] : src2[31:0] + 1: tmp[31:0] := (src1[31:0] <= src2[31:0]) ? src2[31:0] : src1[31:0] + 2: tmp[31:0] := (ABS(src1[31:0]) <= ABS(src2[31:0])) ? src1[31:0] : src2[31:0] + 3: tmp[31:0] := (ABS(src1[31:0]) <= ABS(src2[31:0])) ? src2[31:0] : src1[31:0] + ESAC + + CASE signSelCtl[1:0] OF + 0: dst[31:0] := (src1[31] << 31) OR (tmp[30:0]) + 1: dst[31:0] := tmp[31:0] + 2: dst[31:0] := (0 << 31) OR (tmp[30:0]) + 3: dst[31:0] := (1 << 31) OR (tmp[30:0]) + ESAC + + RETURN dst +} +dst[31:0] := RANGE(a[31:0], b[31:0], imm8[1:0], imm8[3:2]) +dst[127:32] := a[127:32] +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512VL + AVX512DQ + Miscellaneous + + + + + + Extract the reduced argument of packed double-precision (64-bit) floating-point elements in "a" by the number of bits specified by "imm8", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). [round_imm_note] + +DEFINE ReduceArgumentPD(src1[63:0], imm8[7:0]) { + m[63:0] := FP64(imm8[7:4]) // number of fraction bits after the binary point to be preserved + tmp[63:0] := POW(2.0, -m) * ROUND(POW(2.0, m) * src1[63:0], imm8[3:0]) + tmp[63:0] := src1[63:0] - tmp[63:0] + IF IsInf(tmp[63:0]) + tmp[63:0] := FP64(0.0) + FI + RETURN tmp[63:0] +} +FOR j := 0 to 3 + i := j*64 + IF k[j] + dst[i+63:i] := ReduceArgumentPD(a[i+63:i], imm8[7:0]) + ELSE + dst[i+63:i] := src[i+63:i] + FI +ENDFOR +dst[MAX:256] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512VL + AVX512DQ + Miscellaneous + + + + + Extract the reduced argument of packed double-precision (64-bit) floating-point elements in "a" by the number of bits specified by "imm8", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). [round_imm_note] + +DEFINE ReduceArgumentPD(src1[63:0], imm8[7:0]) { + m[63:0] := FP64(imm8[7:4]) // number of fraction bits after the binary point to be preserved + tmp[63:0] := POW(2.0, -m) * ROUND(POW(2.0, m) * src1[63:0], imm8[3:0]) + tmp[63:0] := src1[63:0] - tmp[63:0] + IF IsInf(tmp[63:0]) + tmp[63:0] := FP64(0.0) + FI + RETURN tmp[63:0] +} +FOR j := 0 to 3 + i := j*64 + IF k[j] + dst[i+63:i] := ReduceArgumentPD(a[i+63:i], imm8[7:0]) + ELSE + dst[i+63:i] := 0 + FI +ENDFOR +dst[MAX:256] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512VL + AVX512DQ + Miscellaneous + + + + Extract the reduced argument of packed double-precision (64-bit) floating-point elements in "a" by the number of bits specified by "imm8", and store the results in "dst". [round_imm_note] + +DEFINE ReduceArgumentPD(src1[63:0], imm8[7:0]) { + m[63:0] := FP64(imm8[7:4]) // number of fraction bits after the binary point to be preserved + tmp[63:0] := POW(2.0, -m) * ROUND(POW(2.0, m) * src1[63:0], imm8[3:0]) + tmp[63:0] := src1[63:0] - tmp[63:0] + IF IsInf(tmp[63:0]) + tmp[63:0] := FP64(0.0) + FI + RETURN tmp[63:0] +} +FOR j := 0 to 3 + i := j*64 + dst[i+63:i] := ReduceArgumentPD(a[i+63:i], imm8[7:0]) +ENDFOR +dst[MAX:256] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512DQ + Miscellaneous + + + + + + Extract the reduced argument of packed double-precision (64-bit) floating-point elements in "a" by the number of bits specified by "imm8", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). [round_imm_note] + +DEFINE ReduceArgumentPD(src1[63:0], imm8[7:0]) { + m[63:0] := FP64(imm8[7:4]) // number of fraction bits after the binary point to be preserved + tmp[63:0] := POW(2.0, -m) * ROUND(POW(2.0, m) * src1[63:0], imm8[3:0]) + tmp[63:0] := src1[63:0] - tmp[63:0] + IF IsInf(tmp[63:0]) + tmp[63:0] := FP64(0.0) + FI + RETURN tmp[63:0] +} FOR j := 0 to 7 i := j*64 IF k[j] - dst[i+63:i] := a[m+63:m] - m := m + 64 + dst[i+63:i] := ReduceArgumentPD(a[i+63:i], imm8[7:0]) ELSE dst[i+63:i] := src[i+63:i] FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - AVX512F - Load - Swizzle - - - - Load contiguous active double-precision (64-bit) floating-point elements from unaligned memory at "mem_addr" (those with their respective bit set in mask "k"), and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -m := 0 + AVX512DQ + Miscellaneous + + + + + + + Extract the reduced argument of packed double-precision (64-bit) floating-point elements in "a" by the number of bits specified by "imm8", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). [round_imm_note][sae_note] + +DEFINE ReduceArgumentPD(src1[63:0], imm8[7:0]) { + m[63:0] := FP64(imm8[7:4]) // number of fraction bits after the binary point to be preserved + tmp[63:0] := POW(2.0, -m) * ROUND(POW(2.0, m) * src1[63:0], imm8[3:0]) + tmp[63:0] := src1[63:0] - tmp[63:0] + IF IsInf(tmp[63:0]) + tmp[63:0] := FP64(0.0) + FI + RETURN tmp[63:0] +} FOR j := 0 to 7 i := j*64 IF k[j] - dst[i+63:i] := MEM[mem_addr+m+63:mem_addr+m] - m := m + 64 + dst[i+63:i] := ReduceArgumentPD(a[i+63:i], imm8[7:0]) ELSE dst[i+63:i] := src[i+63:i] FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - AVX512F - Swizzle - - - Load contiguous active double-precision (64-bit) floating-point elements from "a" (those with their respective bit set in mask "k"), and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + AVX512DQ + Miscellaneous + + + + + Extract the reduced argument of packed double-precision (64-bit) floating-point elements in "a" by the number of bits specified by "imm8", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). [round_imm_note] -m := 0 +DEFINE ReduceArgumentPD(src1[63:0], imm8[7:0]) { + m[63:0] := FP64(imm8[7:4]) // number of fraction bits after the binary point to be preserved + tmp[63:0] := POW(2.0, -m) * ROUND(POW(2.0, m) * src1[63:0], imm8[3:0]) + tmp[63:0] := src1[63:0] - tmp[63:0] + IF IsInf(tmp[63:0]) + tmp[63:0] := FP64(0.0) + FI + RETURN tmp[63:0] +} FOR j := 0 to 7 i := j*64 IF k[j] - dst[i+63:i] := a[m+63:m] - m := m + 64 + dst[i+63:i] := ReduceArgumentPD(a[i+63:i], imm8[7:0]) ELSE dst[i+63:i] := 0 FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - AVX512F - Load - Swizzle - - - Load contiguous active double-precision (64-bit) floating-point elements from unaligned memory at "mem_addr" (those with their respective bit set in mask "k"), and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + AVX512DQ + Miscellaneous + + + + + + Extract the reduced argument of packed double-precision (64-bit) floating-point elements in "a" by the number of bits specified by "imm8", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). [round_imm_note][sae_note] -m := 0 +DEFINE ReduceArgumentPD(src1[63:0], imm8[7:0]) { + m[63:0] := FP64(imm8[7:4]) // number of fraction bits after the binary point to be preserved + tmp[63:0] := POW(2.0, -m) * ROUND(POW(2.0, m) * src1[63:0], imm8[3:0]) + tmp[63:0] := src1[63:0] - tmp[63:0] + IF IsInf(tmp[63:0]) + tmp[63:0] := FP64(0.0) + FI + RETURN tmp[63:0] +} FOR j := 0 to 7 i := j*64 IF k[j] - dst[i+63:i] := MEM[mem_addr+m+63:mem_addr+m] - m := m + 64 + dst[i+63:i] := ReduceArgumentPD(a[i+63:i], imm8[7:0]) ELSE dst[i+63:i] := 0 FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - AVX512F - Swizzle - - - - Load contiguous active single-precision (32-bit) floating-point elements from "a" (those with their respective bit set in mask "k"), and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + AVX512DQ + Miscellaneous + + + + Extract the reduced argument of packed double-precision (64-bit) floating-point elements in "a" by the number of bits specified by "imm8", and store the results in "dst". [round_imm_note] -m := 0 -FOR j := 0 to 15 - i := j*32 - IF k[j] - dst[i+31:i] := a[m+31:m] - m := m + 32 - ELSE - dst[i+31:i] := src[i+31:i] +DEFINE ReduceArgumentPD(src1[63:0], imm8[7:0]) { + m[63:0] := FP64(imm8[7:4]) // number of fraction bits after the binary point to be preserved + tmp[63:0] := POW(2.0, -m) * ROUND(POW(2.0, m) * src1[63:0], imm8[3:0]) + tmp[63:0] := src1[63:0] - tmp[63:0] + IF IsInf(tmp[63:0]) + tmp[63:0] := FP64(0.0) FI + RETURN tmp[63:0] +} +FOR j := 0 to 7 + i := j*64 + dst[i+63:i] := ReduceArgumentPD(a[i+63:i], imm8[7:0]) ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - AVX512F - Load - Swizzle - - - - Load contiguous active single-precision (32-bit) floating-point elements from unaligned memory at "mem_addr" (those with their respective bit set in mask "k"), and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + AVX512DQ + Miscellaneous + + + + + Extract the reduced argument of packed double-precision (64-bit) floating-point elements in "a" by the number of bits specified by "imm8", and store the results in "dst". [round_imm_note][sae_note] -m := 0 -FOR j := 0 to 15 - i := j*32 - IF k[j] - dst[i+31:i] := MEM[mem_addr+m+31:mem_addr+m] - m := m + 32 - ELSE - dst[i+31:i] := src[i+31:i] +DEFINE ReduceArgumentPD(src1[63:0], imm8[7:0]) { + m[63:0] := FP64(imm8[7:4]) // number of fraction bits after the binary point to be preserved + tmp[63:0] := POW(2.0, -m) * ROUND(POW(2.0, m) * src1[63:0], imm8[3:0]) + tmp[63:0] := src1[63:0] - tmp[63:0] + IF IsInf(tmp[63:0]) + tmp[63:0] := FP64(0.0) FI + RETURN tmp[63:0] +} +FOR j := 0 to 7 + i := j*64 + dst[i+63:i] := ReduceArgumentPD(a[i+63:i], imm8[7:0]) ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - AVX512F - Swizzle - - - Load contiguous active single-precision (32-bit) floating-point elements from "a" (those with their respective bit set in mask "k"), and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + AVX512VL + AVX512DQ + Miscellaneous + + + + + + Extract the reduced argument of packed double-precision (64-bit) floating-point elements in "a" by the number of bits specified by "imm8", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). [round_imm_note] -m := 0 -FOR j := 0 to 15 - i := j*32 +DEFINE ReduceArgumentPD(src1[63:0], imm8[7:0]) { + m[63:0] := FP64(imm8[7:4]) // number of fraction bits after the binary point to be preserved + tmp[63:0] := POW(2.0, -m) * ROUND(POW(2.0, m) * src1[63:0], imm8[3:0]) + tmp[63:0] := src1[63:0] - tmp[63:0] + IF IsInf(tmp[63:0]) + tmp[63:0] := FP64(0.0) + FI + RETURN tmp[63:0] +} +FOR j := 0 to 1 + i := j*64 IF k[j] - dst[i+31:i] := a[m+31:m] - m := m + 32 + dst[i+63:i] := ReduceArgumentPD(a[i+63:i], imm8[7:0]) ELSE - dst[i+31:i] := 0 + dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point - AVX512F - Load - Swizzle - - - Load contiguous active single-precision (32-bit) floating-point elements from unaligned memory at "mem_addr" (those with their respective bit set in mask "k"), and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + AVX512VL + AVX512DQ + Miscellaneous + + + + + Extract the reduced argument of packed double-precision (64-bit) floating-point elements in "a" by the number of bits specified by "imm8", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). [round_imm_note] -m := 0 -FOR j := 0 to 15 - i := j*32 +DEFINE ReduceArgumentPD(src1[63:0], imm8[7:0]) { + m[63:0] := FP64(imm8[7:4]) // number of fraction bits after the binary point to be preserved + tmp[63:0] := POW(2.0, -m) * ROUND(POW(2.0, m) * src1[63:0], imm8[3:0]) + tmp[63:0] := src1[63:0] - tmp[63:0] + IF IsInf(tmp[63:0]) + tmp[63:0] := FP64(0.0) + FI + RETURN tmp[63:0] +} +FOR j := 0 to 1 + i := j*64 IF k[j] - dst[i+31:i] := MEM[mem_addr+m+31:mem_addr+m] - m := m + 32 + dst[i+63:i] := ReduceArgumentPD(a[i+63:i], imm8[7:0]) ELSE - dst[i+31:i] := 0 + dst[i+63:i] := 0 FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point - AVX512F - Swizzle - - - Extract 128 bits (composed of 4 packed single-precision (32-bit) floating-point elements) from "a", selected with "imm8", and store the result in "dst". + AVX512VL + AVX512DQ + Miscellaneous + + + + Extract the reduced argument of packed double-precision (64-bit) floating-point elements in "a" by the number of bits specified by "imm8", and store the results in "dst". [round_imm_note] -CASE imm8[7:0] of -0: dst[127:0] := a[127:0] -1: dst[127:0] := a[255:128] -2: dst[127:0] := a[383:256] -3: dst[127:0] := a[511:384] -ESAC +DEFINE ReduceArgumentPD(src1[63:0], imm8[7:0]) { + m[63:0] := FP64(imm8[7:4]) // number of fraction bits after the binary point to be preserved + tmp[63:0] := POW(2.0, -m) * ROUND(POW(2.0, m) * src1[63:0], imm8[3:0]) + tmp[63:0] := src1[63:0] - tmp[63:0] + IF IsInf(tmp[63:0]) + tmp[63:0] := FP64(0.0) + FI + RETURN tmp[63:0] +} +FOR j := 0 to 1 + i := j*64 + dst[i+63:i] := ReduceArgumentPD(a[i+63:i], imm8[7:0]) +ENDFOR dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point - AVX512F - Swizzle - - - - - Extract 128 bits (composed of 4 packed single-precision (32-bit) floating-point elements) from "a", selected with "imm8", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + AVX512VL + AVX512DQ + Miscellaneous + + + + + + Extract the reduced argument of packed single-precision (32-bit) floating-point elements in "a" by the number of bits specified by "imm8", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). [round_imm_note] -CASE imm8[7:0] of -0: tmp[127:0] := a[127:0] -1: tmp[127:0] := a[255:128] -2: tmp[127:0] := a[383:256] -3: tmp[127:0] := a[511:384] -ESAC -FOR j := 0 to 3 +DEFINE ReduceArgumentPS(src1[31:0], imm8[7:0]) { + m[31:0] := FP32(imm8[7:4]) // number of fraction bits after the binary point to be preserved + tmp[31:0] := POW(FP32(2.0), -m) * ROUND(POW(FP32(2.0), m) * src1[31:0], imm8[3:0]) + tmp[31:0] := src1[31:0] - tmp[31:0] + IF IsInf(tmp[31:0]) + tmp[31:0] := FP32(0.0) + FI + RETURN tmp[31:0] +} +FOR j := 0 to 7 i := j*32 IF k[j] - dst[i+31:i] := tmp[i+31:i] + dst[i+31:i] := ReduceArgumentPS(a[i+31:i], imm8[7:0]) ELSE dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Floating Point - AVX512F - Swizzle - - - - Extract 128 bits (composed of 4 packed single-precision (32-bit) floating-point elements) from "a", selected with "imm8", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + AVX512VL + AVX512DQ + Miscellaneous + + + + + Extract the reduced argument of packed single-precision (32-bit) floating-point elements in "a" by the number of bits specified by "imm8", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). [round_imm_note] -CASE imm8[7:0] of -0: tmp[127:0] := a[127:0] -1: tmp[127:0] := a[255:128] -2: tmp[127:0] := a[383:256] -3: tmp[127:0] := a[511:384] -ESAC -FOR j := 0 to 3 +DEFINE ReduceArgumentPS(src1[31:0], imm8[7:0]) { + m[31:0] := FP32(imm8[7:4]) // number of fraction bits after the binary point to be preserved + tmp[31:0] := POW(FP32(2.0), -m) * ROUND(POW(FP32(2.0), m) * src1[31:0], imm8[3:0]) + tmp[31:0] := src1[31:0] - tmp[31:0] + IF IsInf(tmp[31:0]) + tmp[31:0] := FP32(0.0) + FI + RETURN tmp[31:0] +} +FOR j := 0 to 7 i := j*32 IF k[j] - dst[i+31:i] := tmp[i+31:i] + dst[i+31:i] := ReduceArgumentPS(a[i+31:i], imm8[7:0]) ELSE dst[i+31:i] := 0 FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Floating Point - AVX512F - Swizzle - - - Extract 256 bits (composed of 4 packed double-precision (64-bit) floating-point elements) from "a", selected with "imm8", and store the result in "dst". + AVX512VL + AVX512DQ + Miscellaneous + + + + Extract the reduced argument of packed single-precision (32-bit) floating-point elements in "a" by the number of bits specified by "imm8", and store the results in "dst". [round_imm_note] -CASE imm8[7:0] of -0: dst[255:0] := a[255:0] -1: dst[255:0] := a[511:256] -ESAC +DEFINE ReduceArgumentPS(src1[31:0], imm8[7:0]) { + m[31:0] := FP32(imm8[7:4]) // number of fraction bits after the binary point to be preserved + tmp[31:0] := POW(FP32(2.0), -m) * ROUND(POW(FP32(2.0), m) * src1[31:0], imm8[3:0]) + tmp[31:0] := src1[31:0] - tmp[31:0] + RETURN tmp[31:0] + IF IsInf(tmp[31:0]) + tmp[31:0] := FP32(0.0) + FI +} +FOR j := 0 to 7 + i := j*32 + dst[i+31:i] := ReduceArgumentPS(a[i+31:i], imm8[7:0]) +ENDFOR dst[MAX:256] := 0 - +
immintrin.h
 - + Floating Point - AVX512F - Swizzle - - - - - Extract 256 bits (composed of 4 packed double-precision (64-bit) floating-point elements) from "a", selected with "imm8", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + AVX512DQ + Miscellaneous + + + + + + Extract the reduced argument of packed single-precision (32-bit) floating-point elements in "a" by the number of bits specified by "imm8", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). [round_imm_note] -CASE imm8[7:0] of -0: tmp[255:0] := a[255:0] -1: tmp[255:0] := a[511:256] -ESAC -FOR j := 0 to 3 - i := j*64 +DEFINE ReduceArgumentPS(src1[31:0], imm8[7:0]) { + m[31:0] := FP32(imm8[7:4]) // number of fraction bits after the binary point to be preserved + tmp[31:0] := POW(FP32(2.0), -m) * ROUND(POW(FP32(2.0), m) * src1[31:0], imm8[3:0]) + tmp[31:0] := src1[31:0] - tmp[31:0] + IF IsInf(tmp[31:0]) + tmp[31:0] := FP32(0.0) + FI + RETURN tmp[31:0] +} +FOR j := 0 to 15 + i := j*32 IF k[j] - dst[i+63:i] := tmp[i+63:i] + dst[i+31:i] := ReduceArgumentPS(a[i+31:i], imm8[7:0]) ELSE - dst[i+63:i] := src[i+63:i] + dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - AVX512F - Swizzle - - - - Extract 256 bits (composed of 4 packed double-precision (64-bit) floating-point elements) from "a", selected with "imm8", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - -CASE imm8[7:0] of -0: tmp[255:0] := a[255:0] -1: tmp[255:0] := a[511:256] -ESAC -FOR j := 0 to 3 - i := j*64 + AVX512DQ + Miscellaneous + + + + + + + Extract the reduced argument of packed single-precision (32-bit) floating-point elements in "a" by the number of bits specified by "imm8", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). [round_imm_note][sae_note] + +DEFINE ReduceArgumentPS(src1[31:0], imm8[7:0]) { + m[31:0] := FP32(imm8[7:4]) // number of fraction bits after the binary point to be preserved + tmp[31:0] := POW(FP32(2.0), -m) * ROUND(POW(FP32(2.0), m) * src1[31:0], imm8[3:0]) + tmp[31:0] := src1[31:0] - tmp[31:0] + IF IsInf(tmp[31:0]) + tmp[31:0] := FP32(0.0) + FI + RETURN tmp[31:0] +} +FOR j := 0 to 15 + i := j*32 IF k[j] - dst[i+63:i] := tmp[i+63:i] + dst[i+31:i] := ReduceArgumentPS(a[i+31:i], imm8[7:0]) ELSE - dst[i+63:i] := 0 + dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:256] := 0 - - -
immintrin.h
- - - Integer - AVX512F - Swizzle - - - Extract 128 bits (composed of 4 packed 32-bit integers) from "a", selected with "imm8", and store the result in "dst". - -CASE imm8[7:0] of -0: dst[127:0] := a[127:0] -1: dst[127:0] := a[255:128] -2: dst[127:0] := a[383:256] -3: dst[127:0] := a[511:384] -ESAC -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - AVX512F - Swizzle - - - - - Extract 128 bits (composed of 4 packed 32-bit integers) from "a", selected with "imm8", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Floating Point + AVX512DQ + Miscellaneous + + + + + Extract the reduced argument of packed single-precision (32-bit) floating-point elements in "a" by the number of bits specified by "imm8", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). [round_imm_note] -CASE imm8[7:0] of -0: dst[127:0] := a[127:0] -1: dst[127:0] := a[255:128] -2: dst[127:0] := a[383:256] -3: dst[127:0] := a[511:384] -ESAC -FOR j := 0 to 3 +DEFINE ReduceArgumentPS(src1[31:0], imm8[7:0]) { + m[31:0] := FP32(imm8[7:4]) // number of fraction bits after the binary point to be preserved + tmp[31:0] := POW(FP32(2.0), -m) * ROUND(POW(FP32(2.0), m) * src1[31:0], imm8[3:0]) + tmp[31:0] := src1[31:0] - tmp[31:0] + IF IsInf(tmp[31:0]) + tmp[31:0] := FP32(0.0) + FI + RETURN tmp[31:0] +} +FOR j := 0 to 15 i := j*32 IF k[j] - dst[i+31:i] := tmp[i+31:i] + dst[i+31:i] := ReduceArgumentPS(a[i+31:i], imm8[7:0]) ELSE - dst[i+31:i] := src[i+31:i] + dst[i+31:i] := 0 FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - AVX512F - Swizzle - - - - Extract 128 bits (composed of 4 packed 32-bit integers) from "a", selected with "imm8", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Floating Point + AVX512DQ + Miscellaneous + + + + + + Extract the reduced argument of packed single-precision (32-bit) floating-point elements in "a" by the number of bits specified by "imm8", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). [round_imm_note][sae_note] -CASE imm8[7:0] of -0: dst[127:0] := a[127:0] -1: dst[127:0] := a[255:128] -2: dst[127:0] := a[383:256] -3: dst[127:0] := a[511:384] -ESAC -FOR j := 0 to 3 +DEFINE ReduceArgumentPS(src1[31:0], imm8[7:0]) { + m[31:0] := FP32(imm8[7:4]) // number of fraction bits after the binary point to be preserved + tmp[31:0] := POW(FP32(2.0), -m) * ROUND(POW(FP32(2.0), m) * src1[31:0], imm8[3:0]) + tmp[31:0] := src1[31:0] - tmp[31:0] + IF IsInf(tmp[31:0]) + tmp[31:0] := FP32(0.0) + FI + RETURN tmp[31:0] +} +FOR j := 0 to 15 i := j*32 IF k[j] - dst[i+31:i] := tmp[i+31:i] + dst[i+31:i] := ReduceArgumentPS(a[i+31:i], imm8[7:0]) ELSE dst[i+31:i] := 0 FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - AVX512F - Swizzle - - - Extract 256 bits (composed of 4 packed 64-bit integers) from "a", selected with "imm8", and store the result in "dst". - -CASE imm8[7:0] of -0: dst[255:0] := a[255:0] -1: dst[255:0] := a[511:256] -ESAC -dst[MAX:256] := 0 + + Floating Point + AVX512DQ + Miscellaneous + + + + Extract the reduced argument of packed single-precision (32-bit) floating-point elements in "a" by the number of bits specified by "imm8", and store the results in "dst". [round_imm_note] + +DEFINE ReduceArgumentPS(src1[31:0], imm8[7:0]) { + m[31:0] := FP32(imm8[7:4]) // number of fraction bits after the binary point to be preserved + tmp[31:0] := POW(FP32(2.0), -m) * ROUND(POW(FP32(2.0), m) * src1[31:0], imm8[3:0]) + tmp[31:0] := src1[31:0] - tmp[31:0] + IF IsInf(tmp[31:0]) + tmp[31:0] := FP32(0.0) + FI + RETURN tmp[31:0] +} +FOR j := 0 to 15 + i := j*32 + dst[i+31:i] := ReduceArgumentPS(a[i+31:i], imm8[7:0]) +ENDFOR +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - AVX512F - Swizzle - - - - - Extract 256 bits (composed of 4 packed 64-bit integers) from "a", selected with "imm8", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Floating Point + AVX512DQ + Miscellaneous + + + + + Extract the reduced argument of packed single-precision (32-bit) floating-point elements in "a" by the number of bits specified by "imm8", and store the results in "dst". [round_imm_note][sae_note] -CASE imm8[7:0] of -0: dst[255:0] := a[255:0] -1: dst[255:0] := a[511:256] -ESAC +DEFINE ReduceArgumentPS(src1[31:0], imm8[7:0]) { + m[31:0] := FP32(imm8[7:4]) // number of fraction bits after the binary point to be preserved + tmp[31:0] := POW(FP32(2.0), -m) * ROUND(POW(FP32(2.0), m) * src1[31:0], imm8[3:0]) + tmp[31:0] := src1[31:0] - tmp[31:0] + IF IsInf(tmp[31:0]) + tmp[31:0] := FP32(0.0) + FI + RETURN tmp[31:0] +} +FOR j := 0 to 15 + i := j*32 + dst[i+31:i] := ReduceArgumentPS(a[i+31:i], imm8[7:0]) +ENDFOR +dst[MAX:512] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512VL + AVX512DQ + Miscellaneous + + + + + + Extract the reduced argument of packed single-precision (32-bit) floating-point elements in "a" by the number of bits specified by "imm8", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). [round_imm_note] + +DEFINE ReduceArgumentPS(src1[31:0], imm8[7:0]) { + m[31:0] := FP32(imm8[7:4]) // number of fraction bits after the binary point to be preserved + tmp[31:0] := POW(FP32(2.0), -m) * ROUND(POW(FP32(2.0), m) * src1[31:0], imm8[3:0]) + tmp[31:0] := src1[31:0] - tmp[31:0] + IF IsInf(tmp[31:0]) + tmp[31:0] := FP32(0.0) + FI + RETURN tmp[31:0] +} FOR j := 0 to 3 - i := j*64 + i := j*32 IF k[j] - dst[i+63:i] := tmp[i+63:i] + dst[i+31:i] := ReduceArgumentPS(a[i+31:i], imm8[7:0]) ELSE - dst[i+63:i] := src[i+63:i] + dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - - Integer - AVX512F - Swizzle - - - - Extract 256 bits (composed of 4 packed 64-bit integers) from "a", selected with "imm8", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Floating Point + AVX512VL + AVX512DQ + Miscellaneous + + + + + Extract the reduced argument of packed single-precision (32-bit) floating-point elements in "a" by the number of bits specified by "imm8", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). [round_imm_note] -CASE imm8[7:0] of -0: dst[255:0] := a[255:0] -1: dst[255:0] := a[511:256] -ESAC +DEFINE ReduceArgumentPS(src1[31:0], imm8[7:0]) { + m[31:0] := FP32(imm8[7:4]) // number of fraction bits after the binary point to be preserved + tmp[31:0] := POW(FP32(2.0), -m) * ROUND(POW(FP32(2.0), m) * src1[31:0], imm8[3:0]) + tmp[31:0] := src1[31:0] - tmp[31:0] + IF IsInf(tmp[31:0]) + tmp[31:0] := FP32(0.0) + FI + RETURN tmp[31:0] +} FOR j := 0 to 3 - i := j*64 + i := j*32 IF k[j] - dst[i+63:i] := tmp[i+63:i] + dst[i+31:i] := ReduceArgumentPS(a[i+31:i], imm8[7:0]) ELSE - dst[i+63:i] := 0 + dst[i+31:i] := 0 FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point - AVX512F + AVX512VL + AVX512DQ Miscellaneous - - - - - Fix up packed double-precision (64-bit) floating-point elements in "a" and "b" using packed 64-bit integers in "c", and store the results in "dst". "imm8" is used to set the required flags reporting. + + + + Extract the reduced argument of packed single-precision (32-bit) floating-point elements in "a" by the number of bits specified by "imm8", and store the results in "dst". [round_imm_note] -enum TOKEN_TYPE { - QNAN_TOKEN := 0, - SNAN_TOKEN := 1, - ZERO_VALUE_TOKEN := 2, - ONE_VALUE_TOKEN := 3, - NEG_INF_TOKEN := 4, - POS_INF_TOKEN := 5, - NEG_VALUE_TOKEN := 6, - POS_VALUE_TOKEN := 7 -} -FIXUPIMMPD(src1[63:0], src2[63:0], src3[63:0], imm8[7:0]){ - tsrc[63:0] := ((src2[62:52] == 0) AND (MXCSR.DAZ == 1)) ? 0.0 : src2[63:0] - CASE(tsrc[63:0] of TOKEN_TYPE) - QNAN_TOKEN:j := 0 - SNAN_TOKEN:j := 1 - ZERO_VALUE_TOKEN: j := 2 - ONE_VALUE_TOKEN: j := 3 - NEG_INF_TOKEN: j := 4 - POS_INF_TOKEN: j := 5 - NEG_VALUE_TOKEN: j := 6 - POS_VALUE_TOKEN: j := 7 - ESAC - - token_response[3:0] := src3[3+4*j:4*j] - - CASE(token_response[3:0]) of - 0 : dest[63:0] := src1[63:0] - 1 : dest[63:0] := tsrc[63:0] - 2 : dest[63:0] := QNaN(tsrc[63:0]) - 3 : dest[63:0] := QNAN_Indefinite - 4 : dest[63:0] := -INF - 5 : dest[63:0] := +INF - 6 : dest[63:0] := tsrc.sign? –INF : +INF - 7 : dest[63:0] := -0 - 8 : dest[63:0] := +0 - 9 : dest[63:0] := -1 - 10: dest[63:0] := +1 - 11: dest[63:0] := 1⁄2 - 12: dest[63:0] := 90.0 - 13: dest[63:0] := PI/2 - 14: dest[63:0] := MAX_FLOAT - 15: dest[63:0] := -MAX_FLOAT - ESAC - - CASE(tsrc[31:0] of TOKEN_TYPE) - ZERO_VALUE_TOKEN: if imm8[0] then set #ZE - ZERO_VALUE_TOKEN: if imm8[1] then set #IE - ONE_VALUE_TOKEN: if imm8[2] then set #ZE - ONE_VALUE_TOKEN: if imm8[3] then set #IE - SNAN_TOKEN: if imm8[4] then set #IE - NEG_INF_TOKEN: if imm8[5] then set #IE - NEG_VALUE_TOKEN: if imm8[6] then set #IE - POS_INF_TOKEN: if imm8[7] then set #IE - ESAC - RETURN dest[63:0] +DEFINE ReduceArgumentPS(src1[31:0], imm8[7:0]) { + m[31:0] := FP32(imm8[7:4]) // number of fraction bits after the binary point to be preserved + tmp[31:0] := POW(FP32(2.0), -m) * ROUND(POW(FP32(2.0), m) * src1[31:0], imm8[3:0]) + tmp[31:0] := src1[31:0] - tmp[31:0] + IF IsInf(tmp[31:0]) + tmp[31:0] := FP32(0.0) + FI + RETURN tmp[31:0] } - -FOR j := 0 to 7 - i := j*64 - dst[i+63:i] := FIXUPIMMPD(a[i+63:i], b[i+63:i], c[i+63:i], imm8[7:0]) +FOR j := 0 to 3 + i := j*32 + dst[i+31:i] := ReduceArgumentPS(a[i+31:i], imm8[7:0]) ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point - AVX512F + AVX512DQ Miscellaneous - - - - - - Fix up packed double-precision (64-bit) floating-point elements in "a" and "b" using packed 64-bit integers in "c", and store the results in "dst". "imm8" is used to set the required flags reporting. - [round_note] - - -enum TOKEN_TYPE { - QNAN_TOKEN := 0, - SNAN_TOKEN := 1, - ZERO_VALUE_TOKEN := 2, - ONE_VALUE_TOKEN := 3, - NEG_INF_TOKEN := 4, - POS_INF_TOKEN := 5, - NEG_VALUE_TOKEN := 6, - POS_VALUE_TOKEN := 7 + + + + + + + Extract the reduced argument of the lower double-precision (64-bit) floating-point element in "b" by the number of bits specified by "imm8", store the result in the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". [round_imm_note] + +DEFINE ReduceArgumentPD(src1[63:0], imm8[7:0]) { + m[63:0] := FP64(imm8[7:4]) // number of fraction bits after the binary point to be preserved + tmp[63:0] := POW(2.0, -m) * ROUND(POW(2.0, m) * src1[63:0], imm8[3:0]) + tmp[63:0] := src1[63:0] - tmp[63:0] + IF IsInf(tmp[63:0]) + tmp[63:0] := FP64(0.0) + FI + RETURN tmp[63:0] } -FIXUPIMMPD(src1[63:0], src2[63:0], src3[63:0], imm8[7:0]){ - tsrc[63:0] := ((src2[62:52] == 0) AND (MXCSR.DAZ == 1)) ? 0.0 : src2[63:0] - CASE(tsrc[63:0] of TOKEN_TYPE) - QNAN_TOKEN:j := 0 - SNAN_TOKEN:j := 1 - ZERO_VALUE_TOKEN: j := 2 - ONE_VALUE_TOKEN: j := 3 - NEG_INF_TOKEN: j := 4 - POS_INF_TOKEN: j := 5 - NEG_VALUE_TOKEN: j := 6 - POS_VALUE_TOKEN: j := 7 - ESAC - - token_response[3:0] := src3[3+4*j:4*j] - - CASE(token_response[3:0]) of - 0 : dest[63:0] := src1[63:0] - 1 : dest[63:0] := tsrc[63:0] - 2 : dest[63:0] := QNaN(tsrc[63:0]) - 3 : dest[63:0] := QNAN_Indefinite - 4 : dest[63:0] := -INF - 5 : dest[63:0] := +INF - 6 : dest[63:0] := tsrc.sign? –INF : +INF - 7 : dest[63:0] := -0 - 8 : dest[63:0] := +0 - 9 : dest[63:0] := -1 - 10: dest[63:0] := +1 - 11: dest[63:0] := 1⁄2 - 12: dest[63:0] := 90.0 - 13: dest[63:0] := PI/2 - 14: dest[63:0] := MAX_FLOAT - 15: dest[63:0] := -MAX_FLOAT - ESAC - - CASE(tsrc[31:0] of TOKEN_TYPE) - ZERO_VALUE_TOKEN: if imm8[0] then set #ZE - ZERO_VALUE_TOKEN: if imm8[1] then set #IE - ONE_VALUE_TOKEN: if imm8[2] then set #ZE - ONE_VALUE_TOKEN: if imm8[3] then set #IE - SNAN_TOKEN: if imm8[4] then set #IE - NEG_INF_TOKEN: if imm8[5] then set #IE - NEG_VALUE_TOKEN: if imm8[6] then set #IE - POS_INF_TOKEN: if imm8[7] then set #IE - ESAC - RETURN dest[63:0] +IF k[0] + dst[63:0] := ReduceArgumentPD(b[63:0], imm8[7:0]) +ELSE + dst[63:0] := src[63:0] +FI +dst[127:64] := a[127:64] +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512DQ + Miscellaneous + + + + + + + + Extract the reduced argument of the lower double-precision (64-bit) floating-point element in "b" by the number of bits specified by "imm8", store the result in the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". [round_imm_note][sae_note] + +DEFINE ReduceArgumentPD(src1[63:0], imm8[7:0]) { + m[63:0] := FP64(imm8[7:4]) // number of fraction bits after the binary point to be preserved + tmp[63:0] := POW(2.0, -m) * ROUND(POW(2.0, m) * src1[63:0], imm8[3:0]) + tmp[63:0] := src1[63:0] - tmp[63:0] + IF IsInf(tmp[63:0]) + tmp[63:0] := FP64(0.0) + FI + RETURN tmp[63:0] } - -FOR j := 0 to 7 - i := j*64 - dst[i+63:i] := FIXUPIMMPD(a[i+63:i], b[i+63:i], c[i+63:i], imm8[7:0]) -ENDFOR -dst[MAX:512] := 0 +IF k[0] + dst[63:0] := ReduceArgumentPD(b[63:0], imm8[7:0]) +ELSE + dst[63:0] := src[63:0] +FI +dst[127:64] := a[127:64] +dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point - AVX512F + AVX512DQ Miscellaneous - - - - - - Fix up packed double-precision (64-bit) floating-point elements in "a" and "b" using packed 64-bit integers in "c", and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). "imm8" is used to set the required flags reporting. + + + + + + Extract the reduced argument of the lower double-precision (64-bit) floating-point element in "b" by the number of bits specified by "imm8", store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". [round_imm_note] -enum TOKEN_TYPE { - QNAN_TOKEN := 0, - SNAN_TOKEN := 1, - ZERO_VALUE_TOKEN := 2, - ONE_VALUE_TOKEN := 3, - NEG_INF_TOKEN := 4, - POS_INF_TOKEN := 5, - NEG_VALUE_TOKEN := 6, - POS_VALUE_TOKEN := 7 +DEFINE ReduceArgumentPD(src1[63:0], imm8[7:0]) { + m[63:0] := FP64(imm8[7:4]) // number of fraction bits after the binary point to be preserved + tmp[63:0] := POW(2.0, -m) * ROUND(POW(2.0, m) * src1[63:0], imm8[3:0]) + tmp[63:0] := src1[63:0] - tmp[63:0] + IF IsInf(tmp[63:0]) + tmp[63:0] := FP64(0.0) + FI + RETURN tmp[63:0] } -FIXUPIMMPD(src1[63:0], src2[63:0], src3[63:0], imm8[7:0]){ - tsrc[63:0] := ((src2[62:52] == 0) AND (MXCSR.DAZ == 1)) ? 0.0 : src2[63:0] - CASE(tsrc[63:0] of TOKEN_TYPE) - QNAN_TOKEN:j := 0 - SNAN_TOKEN:j := 1 - ZERO_VALUE_TOKEN: j := 2 - ONE_VALUE_TOKEN: j := 3 - NEG_INF_TOKEN: j := 4 - POS_INF_TOKEN: j := 5 - NEG_VALUE_TOKEN: j := 6 - POS_VALUE_TOKEN: j := 7 - ESAC - - token_response[3:0] := src3[3+4*j:4*j] - - CASE(token_response[3:0]) of - 0 : dest[63:0] := src1[63:0] - 1 : dest[63:0] := tsrc[63:0] - 2 : dest[63:0] := QNaN(tsrc[63:0]) - 3 : dest[63:0] := QNAN_Indefinite - 4 : dest[63:0] := -INF - 5 : dest[63:0] := +INF - 6 : dest[63:0] := tsrc.sign? –INF : +INF - 7 : dest[63:0] := -0 - 8 : dest[63:0] := +0 - 9 : dest[63:0] := -1 - 10: dest[63:0] := +1 - 11: dest[63:0] := 1⁄2 - 12: dest[63:0] := 90.0 - 13: dest[63:0] := PI/2 - 14: dest[63:0] := MAX_FLOAT - 15: dest[63:0] := -MAX_FLOAT - ESAC - - CASE(tsrc[31:0] of TOKEN_TYPE) - ZERO_VALUE_TOKEN: if imm8[0] then set #ZE - ZERO_VALUE_TOKEN: if imm8[1] then set #IE - ONE_VALUE_TOKEN: if imm8[2] then set #ZE - ONE_VALUE_TOKEN: if imm8[3] then set #IE - SNAN_TOKEN: if imm8[4] then set #IE - NEG_INF_TOKEN: if imm8[5] then set #IE - NEG_VALUE_TOKEN: if imm8[6] then set #IE - POS_INF_TOKEN: if imm8[7] then set #IE - ESAC - RETURN dest[63:0] +IF k[0] + dst[63:0] := ReduceArgumentPD(b[63:0], imm8[7:0]) +ELSE + dst[63:0] := 0 +FI +dst[127:64] := a[127:64] +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512DQ + Miscellaneous + + + + + + + Extract the reduced argument of the lower double-precision (64-bit) floating-point element in "b" by the number of bits specified by "imm8", store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". [round_imm_note][sae_note] + +DEFINE ReduceArgumentPD(src1[63:0], imm8[7:0]) { + m[63:0] := FP64(imm8[7:4]) // number of fraction bits after the binary point to be preserved + tmp[63:0] := POW(2.0, -m) * ROUND(POW(2.0, m) * src1[63:0], imm8[3:0]) + tmp[63:0] := src1[63:0] - tmp[63:0] + IF IsInf(tmp[63:0]) + tmp[63:0] := FP64(0.0) + FI + RETURN tmp[63:0] } - -FOR j := 0 to 7 +IF k[0] + dst[63:0] := ReduceArgumentPD(b[63:0], imm8[7:0]) +ELSE + dst[63:0] := 0 +FI +dst[127:64] := a[127:64] +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512DQ + Miscellaneous + + + + + Extract the reduced argument of the lower double-precision (64-bit) floating-point element in "b" by the number of bits specified by "imm8", store the result in the lower element of "dst", and copy the upper element from "a" to the upper element of "dst". [round_imm_note] + +DEFINE ReduceArgumentPD(src1[63:0], imm8[7:0]) { + m[63:0] := FP64(imm8[7:4]) // number of fraction bits after the binary point to be preserved + tmp[63:0] := POW(2.0, -m) * ROUND(POW(2.0, m) * src1[63:0], imm8[3:0]) + tmp[63:0] := src1[63:0] - tmp[63:0] + IF IsInf(tmp[63:0]) + tmp[63:0] := FP64(0.0) + FI + RETURN tmp[63:0] +} +dst[63:0] := ReduceArgumentPD(b[63:0], imm8[7:0]) +dst[127:64] := a[127:64] +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512DQ + Miscellaneous + + + + + + Extract the reduced argument of the lower double-precision (64-bit) floating-point element in "b" by the number of bits specified by "imm8", store the result in the lower element of "dst", and copy the upper element from "a" to the upper element of "dst". [round_imm_note][sae_note] + +DEFINE ReduceArgumentPD(src1[63:0], imm8[7:0]) { + m[63:0] := FP64(imm8[7:4]) // number of fraction bits after the binary point to be preserved + tmp[63:0] := POW(2.0, -m) * ROUND(POW(2.0, m) * src1[63:0], imm8[3:0]) + tmp[63:0] := src1[63:0] - tmp[63:0] + IF IsInf(tmp[63:0]) + tmp[63:0] := FP64(0.0) + FI + RETURN tmp[63:0] +} +dst[63:0] := ReduceArgumentPD(b[63:0], imm8[7:0]) +dst[127:64] := a[127:64] +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512DQ + Miscellaneous + + + + + + + Extract the reduced argument of the lower single-precision (32-bit) floating-point element in "b" by the number of bits specified by "imm8", store the result in the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". [round_imm_note] + +DEFINE ReduceArgumentPS(src1[31:0], imm8[7:0]) { + m[31:0] := FP32(imm8[7:4]) // number of fraction bits after the binary point to be preserved + tmp[31:0] := POW(FP32(2.0), -m) * ROUND(POW(FP32(2.0), m) * src1[31:0], imm8[3:0]) + tmp[31:0] := src1[31:0] - tmp[31:0] + IF IsInf(tmp[31:0]) + tmp[31:0] := FP32(0.0) + FI + RETURN tmp[31:0] +} +IF k[0] + dst[31:0] := ReduceArgumentPS(b[31:0], imm8[7:0]) +ELSE + dst[31:0] := src[31:0] +FI +dst[127:32] := a[127:32] +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512DQ + Miscellaneous + + + + + + + + Extract the reduced argument of the lower single-precision (32-bit) floating-point element in "b" by the number of bits specified by "imm8", store the result in the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". [round_imm_note][sae_note] + +DEFINE ReduceArgumentPS(src1[31:0], imm8[7:0]) { + m[31:0] := FP32(imm8[7:4]) // number of fraction bits after the binary point to be preserved + tmp[31:0] := POW(FP32(2.0), -m) * ROUND(POW(FP32(2.0), m) * src1[31:0], imm8[3:0]) + tmp[31:0] := src1[31:0] - tmp[31:0] + IF IsInf(tmp[31:0]) + tmp[31:0] := FP32(0.0) + FI + RETURN tmp[31:0] +} +IF k[0] + dst[31:0] := ReduceArgumentPS(b[31:0], imm8[7:0]) +ELSE + dst[31:0] := src[31:0] +FI +dst[127:32] := a[127:32] +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512DQ + Miscellaneous + + + + + + Extract the reduced argument of the lower single-precision (32-bit) floating-point element in "b" by the number of bits specified by "imm8", store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". [round_imm_note] + +DEFINE ReduceArgumentPS(src1[31:0], imm8[7:0]) { + m[31:0] := FP32(imm8[7:4]) // number of fraction bits after the binary point to be preserved + tmp[31:0] := POW(FP32(2.0), -m) * ROUND(POW(FP32(2.0), m) * src1[31:0], imm8[3:0]) + tmp[31:0] := src1[31:0] - tmp[31:0] + IF IsInf(tmp[31:0]) + tmp[31:0] := FP32(0.0) + FI + RETURN tmp[31:0] +} +IF k[0] + dst[31:0] := ReduceArgumentPS(b[31:0], imm8[7:0]) +ELSE + dst[31:0] := 0 +FI +dst[127:32] := a[127:32] +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512DQ + Miscellaneous + + + + + + + Extract the reduced argument of the lower single-precision (32-bit) floating-point element in "b" by the number of bits specified by "imm8", store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". [round_imm_note][sae_note] + +DEFINE ReduceArgumentPS(src1[31:0], imm8[7:0]) { + m[31:0] := FP32(imm8[7:4]) // number of fraction bits after the binary point to be preserved + tmp[31:0] := POW(FP32(2.0), -m) * ROUND(POW(FP32(2.0), m) * src1[31:0], imm8[3:0]) + tmp[31:0] := src1[31:0] - tmp[31:0] + IF IsInf(tmp[31:0]) + tmp[31:0] := FP32(0.0) + FI + RETURN tmp[31:0] +} +IF k[0] + dst[31:0] := ReduceArgumentPS(b[31:0], imm8[7:0]) +ELSE + dst[31:0] := 0 +FI +dst[127:32] := a[127:32] +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512DQ + Miscellaneous + + + + + Extract the reduced argument of the lower single-precision (32-bit) floating-point element in "b" by the number of bits specified by "imm8", store the result in the lower element of "dst", and copy the upper 3 packed elements from "a" to the upper elements of "dst". [round_imm_note] + +DEFINE ReduceArgumentPS(src1[31:0], imm8[7:0]) { + m[31:0] := FP32(imm8[7:4]) // number of fraction bits after the binary point to be preserved + tmp[31:0] := POW(FP32(2.0), -m) * ROUND(POW(FP32(2.0), m) * src1[31:0], imm8[3:0]) + tmp[31:0] := src1[31:0] - tmp[31:0] + IF IsInf(tmp[31:0]) + tmp[31:0] := FP32(0.0) + FI + RETURN tmp[31:0] +} +dst[31:0] := ReduceArgumentPS(b[31:0], imm8[7:0]) +dst[127:32] := a[127:32] +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512DQ + Miscellaneous + + + + + + Extract the reduced argument of the lower single-precision (32-bit) floating-point element in "b" by the number of bits specified by "imm8", store the result in the lower element of "dst", and copy the upper 3 packed elements from "a" to the upper elements of "dst". [round_imm_note][sae_note] + +DEFINE ReduceArgumentPS(src1[31:0], imm8[7:0]) { + m[31:0] := FP32(imm8[7:4]) // number of fraction bits after the binary point to be preserved + tmp[31:0] := POW(FP32(2.0), -m) * ROUND(POW(FP32(2.0), m) * src1[31:0], imm8[3:0]) + tmp[31:0] := src1[31:0] - tmp[31:0] + IF IsInf(tmp[31:0]) + tmp[31:0] := FP32(0.0) + FI + RETURN tmp[31:0] +} +dst[31:0] := ReduceArgumentPS(b[31:0], imm8[7:0]) +dst[127:32] := a[127:32] +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512VL + AVX512DQ + Logical + + + + + + Compute the bitwise XOR of packed double-precision (64-bit) floating-point elements in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +FOR j := 0 to 3 i := j*64 IF k[j] - dst[i+63:i] := FIXUPIMMPD(a[i+63:i], b[i+63:i], c[i+63:i], imm8[7:0]) + dst[i+63:i] := a[i+63:i] XOR b[i+63:i] ELSE - dst[i+63:i] := a[i+63:i] + dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Floating Point - AVX512F - Miscellaneous - - - - - - - Fix up packed double-precision (64-bit) floating-point elements in "a" and "b" using packed 64-bit integers in "c", and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). "imm8" is used to set the required flags reporting. - [round_note] - - -enum TOKEN_TYPE { - QNAN_TOKEN := 0, - SNAN_TOKEN := 1, - ZERO_VALUE_TOKEN := 2, - ONE_VALUE_TOKEN := 3, - NEG_INF_TOKEN := 4, - POS_INF_TOKEN := 5, - NEG_VALUE_TOKEN := 6, - POS_VALUE_TOKEN := 7 -} -FIXUPIMMPD(src1[63:0], src2[63:0], src3[63:0], imm8[7:0]){ - tsrc[63:0] := ((src2[62:52] == 0) AND (MXCSR.DAZ == 1)) ? 0.0 : src2[63:0] - CASE(tsrc[63:0] of TOKEN_TYPE) - QNAN_TOKEN:j := 0 - SNAN_TOKEN:j := 1 - ZERO_VALUE_TOKEN: j := 2 - ONE_VALUE_TOKEN: j := 3 - NEG_INF_TOKEN: j := 4 - POS_INF_TOKEN: j := 5 - NEG_VALUE_TOKEN: j := 6 - POS_VALUE_TOKEN: j := 7 - ESAC - - token_response[3:0] := src3[3+4*j:4*j] - - CASE(token_response[3:0]) of - 0 : dest[63:0] := src1[63:0] - 1 : dest[63:0] := tsrc[63:0] - 2 : dest[63:0] := QNaN(tsrc[63:0]) - 3 : dest[63:0] := QNAN_Indefinite - 4 : dest[63:0] := -INF - 5 : dest[63:0] := +INF - 6 : dest[63:0] := tsrc.sign? –INF : +INF - 7 : dest[63:0] := -0 - 8 : dest[63:0] := +0 - 9 : dest[63:0] := -1 - 10: dest[63:0] := +1 - 11: dest[63:0] := 1⁄2 - 12: dest[63:0] := 90.0 - 13: dest[63:0] := PI/2 - 14: dest[63:0] := MAX_FLOAT - 15: dest[63:0] := -MAX_FLOAT - ESAC - - CASE(tsrc[31:0] of TOKEN_TYPE) - ZERO_VALUE_TOKEN: if imm8[0] then set #ZE - ZERO_VALUE_TOKEN: if imm8[1] then set #IE - ONE_VALUE_TOKEN: if imm8[2] then set #ZE - ONE_VALUE_TOKEN: if imm8[3] then set #IE - SNAN_TOKEN: if imm8[4] then set #IE - NEG_INF_TOKEN: if imm8[5] then set #IE - NEG_VALUE_TOKEN: if imm8[6] then set #IE - POS_INF_TOKEN: if imm8[7] then set #IE - ESAC - RETURN dest[63:0] -} - -FOR j := 0 to 7 + AVX512VL + AVX512DQ + Logical + + + + + Compute the bitwise XOR of packed double-precision (64-bit) floating-point elements in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + +FOR j := 0 to 3 i := j*64 IF k[j] - dst[i+63:i] := FIXUPIMMPD(a[i+63:i], b[i+63:i], c[i+63:i], imm8[7:0]) + dst[i+63:i] := a[i+63:i] XOR b[i+63:i] ELSE - dst[i+63:i] := a[i+63:i] + dst[i+63:i] := 0 FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Floating Point - AVX512F - Miscellaneous - - - - - - Fix up packed double-precision (64-bit) floating-point elements in "a" and "b" using packed 64-bit integers in "c", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). "imm8" is used to set the required flags reporting. + AVX512DQ + Logical + + + + + + Compute the bitwise XOR of packed double-precision (64-bit) floating-point elements in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -enum TOKEN_TYPE { - QNAN_TOKEN := 0, - SNAN_TOKEN := 1, - ZERO_VALUE_TOKEN := 2, - ONE_VALUE_TOKEN := 3, - NEG_INF_TOKEN := 4, - POS_INF_TOKEN := 5, - NEG_VALUE_TOKEN := 6, - POS_VALUE_TOKEN := 7 -} -FIXUPIMMPD(src1[63:0], src2[63:0], src3[63:0], imm8[7:0]){ - tsrc[63:0] := ((src2[62:52] == 0) AND (MXCSR.DAZ == 1)) ? 0.0 : src2[63:0] - CASE(tsrc[63:0] of TOKEN_TYPE) - QNAN_TOKEN:j := 0 - SNAN_TOKEN:j := 1 - ZERO_VALUE_TOKEN: j := 2 - ONE_VALUE_TOKEN: j := 3 - NEG_INF_TOKEN: j := 4 - POS_INF_TOKEN: j := 5 - NEG_VALUE_TOKEN: j := 6 - POS_VALUE_TOKEN: j := 7 - ESAC - - token_response[3:0] := src3[3+4*j:4*j] - - CASE(token_response[3:0]) of - 0 : dest[63:0] := src1[63:0] - 1 : dest[63:0] := tsrc[63:0] - 2 : dest[63:0] := QNaN(tsrc[63:0]) - 3 : dest[63:0] := QNAN_Indefinite - 4 : dest[63:0] := -INF - 5 : dest[63:0] := +INF - 6 : dest[63:0] := tsrc.sign? –INF : +INF - 7 : dest[63:0] := -0 - 8 : dest[63:0] := +0 - 9 : dest[63:0] := -1 - 10: dest[63:0] := +1 - 11: dest[63:0] := 1⁄2 - 12: dest[63:0] := 90.0 - 13: dest[63:0] := PI/2 - 14: dest[63:0] := MAX_FLOAT - 15: dest[63:0] := -MAX_FLOAT - ESAC - - CASE(tsrc[31:0] of TOKEN_TYPE) - ZERO_VALUE_TOKEN: if imm8[0] then set #ZE - ZERO_VALUE_TOKEN: if imm8[1] then set #IE - ONE_VALUE_TOKEN: if imm8[2] then set #ZE - ONE_VALUE_TOKEN: if imm8[3] then set #IE - SNAN_TOKEN: if imm8[4] then set #IE - NEG_INF_TOKEN: if imm8[5] then set #IE - NEG_VALUE_TOKEN: if imm8[6] then set #IE - POS_INF_TOKEN: if imm8[7] then set #IE - ESAC - RETURN dest[63:0] -} - FOR j := 0 to 7 i := j*64 IF k[j] - dst[i+63:i] := FIXUPIMMPD(a[i+63:i], b[i+63:i], c[i+63:i], imm8[7:0]) + dst[i+63:i] := a[i+63:i] XOR b[i+63:i] ELSE - dst[i+63:i] := 0 + dst[i+63:i] := src[i+63:i] FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - AVX512F - Miscellaneous - - - - - - - Fix up packed double-precision (64-bit) floating-point elements in "a" and "b" using packed 64-bit integers in "c", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). "imm8" is used to set the required flags reporting. - [round_note] - - -enum TOKEN_TYPE { - QNAN_TOKEN := 0, - SNAN_TOKEN := 1, - ZERO_VALUE_TOKEN := 2, - ONE_VALUE_TOKEN := 3, - NEG_INF_TOKEN := 4, - POS_INF_TOKEN := 5, - NEG_VALUE_TOKEN := 6, - POS_VALUE_TOKEN := 7 -} -FIXUPIMMPD(src1[63:0], src2[63:0], src3[63:0], imm8[7:0]){ - tsrc[63:0] := ((src2[62:52] == 0) AND (MXCSR.DAZ == 1)) ? 0.0 : src2[63:0] - CASE(tsrc[63:0] of TOKEN_TYPE) - QNAN_TOKEN:j := 0 - SNAN_TOKEN:j := 1 - ZERO_VALUE_TOKEN: j := 2 - ONE_VALUE_TOKEN: j := 3 - NEG_INF_TOKEN: j := 4 - POS_INF_TOKEN: j := 5 - NEG_VALUE_TOKEN: j := 6 - POS_VALUE_TOKEN: j := 7 - ESAC - - token_response[3:0] := src3[3+4*j:4*j] - - CASE(token_response[3:0]) of - 0 : dest[63:0] := src1[63:0] - 1 : dest[63:0] := tsrc[63:0] - 2 : dest[63:0] := QNaN(tsrc[63:0]) - 3 : dest[63:0] := QNAN_Indefinite - 4 : dest[63:0] := -INF - 5 : dest[63:0] := +INF - 6 : dest[63:0] := tsrc.sign? –INF : +INF - 7 : dest[63:0] := -0 - 8 : dest[63:0] := +0 - 9 : dest[63:0] := -1 - 10: dest[63:0] := +1 - 11: dest[63:0] := 1⁄2 - 12: dest[63:0] := 90.0 - 13: dest[63:0] := PI/2 - 14: dest[63:0] := MAX_FLOAT - 15: dest[63:0] := -MAX_FLOAT - ESAC - - CASE(tsrc[31:0] of TOKEN_TYPE) - ZERO_VALUE_TOKEN: if imm8[0] then set #ZE - ZERO_VALUE_TOKEN: if imm8[1] then set #IE - ONE_VALUE_TOKEN: if imm8[2] then set #ZE - ONE_VALUE_TOKEN: if imm8[3] then set #IE - SNAN_TOKEN: if imm8[4] then set #IE - NEG_INF_TOKEN: if imm8[5] then set #IE - NEG_VALUE_TOKEN: if imm8[6] then set #IE - POS_INF_TOKEN: if imm8[7] then set #IE - ESAC - RETURN dest[63:0] -} - + AVX512DQ + Logical + + + + + Compute the bitwise XOR of packed double-precision (64-bit) floating-point elements in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + FOR j := 0 to 7 i := j*64 IF k[j] - dst[i+63:i] := FIXUPIMMPD(a[i+63:i], b[i+63:i], c[i+63:i], imm8[7:0]) + dst[i+63:i] := a[i+63:i] XOR b[i+63:i] ELSE dst[i+63:i] := 0 FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - AVX512F - Miscellaneous - - - - - Fix up packed single-precision (32-bit) floating-point elements in "a" and "b" using packed 32-bit integers in "c", and store the results in "dst". "imm8" is used to set the required flags reporting. + AVX512DQ + Logical + + + + Compute the bitwise XOR of packed double-precision (64-bit) floating-point elements in "a" and "b", and store the results in "dst". -enum TOKEN_TYPE { - QNAN_TOKEN := 0, - SNAN_TOKEN L= 1, - ZERO_VALUE_TOKEN := 2, - ONE_VALUE_TOKEN := 3, - NEG_INF_TOKEN := 4, - POS_INF_TOKEN := 5, - NEG_VALUE_TOKEN := 6, - POS_VALUE_TOKEN := 7 -} -FIXUPIMMPD(src1[31:0], src2[31:0], src3[31:0], imm8[7:0]){ - tsrc[31:0] := ((src2[30:23] == 0) AND (MXCSR.DAZ == 1)) ? 0.0 : src2[31:0] - CASE(tsrc[31:0] of TOKEN_TYPE) - QNAN_TOKEN:j := 0 - SNAN_TOKEN:j := 1 - ZERO_VALUE_TOKEN: j := 2 - ONE_VALUE_TOKEN: j := 3 - NEG_INF_TOKEN: j := 4 - POS_INF_TOKEN: j := 5 - NEG_VALUE_TOKEN: j := 6 - POS_VALUE_TOKEN: j := 7 - ESAC - - token_response[3:0] := src3[3+4*j:4*j] - - CASE(token_response[3:0]) of - 0 : dest[31:0] := src1[31:0] - 1 : dest[31:0] := tsrc[31:0] - 2 : dest[31:0] := QNaN(tsrc[31:0]) - 3 : dest[31:0] := QNAN_Indefinite - 4 : dest[31:0] := -INF - 5 : dest[31:0] := +INF - 6 : dest[31:0] := tsrc.sign? –INF : +INF - 7 : dest[31:0] := -0 - 8 : dest[31:0] := +0 - 9 : dest[31:0] := -1 - 10: dest[31:0] := +1 - 11: dest[31:0] := 1⁄2 - 12: dest[31:0] := 90.0 - 13: dest[31:0] := PI/2 - 14: dest[31:0] := MAX_FLOAT - 15: dest[31:0] := -MAX_FLOAT - ESAC - - CASE(tsrc[31:0] of TOKEN_TYPE) - ZERO_VALUE_TOKEN: if imm8[0] then set #ZE - ZERO_VALUE_TOKEN: if imm8[1] then set #IE - ONE_VALUE_TOKEN: if imm8[2] then set #ZE - ONE_VALUE_TOKEN: if imm8[3] then set #IE - SNAN_TOKEN: if imm8[4] then set #IE - NEG_INF_TOKEN: if imm8[5] then set #IE - NEG_VALUE_TOKEN: if imm8[6] then set #IE - POS_INF_TOKEN: if imm8[7] then set #IE - ESAC - RETURN dest[31:0] -} - -FOR j := 0 to 15 - i := j*32 - dst[i+31:i] := FIXUPIMMPD(a[i+31:i], b[i+31:i], c[i+31:i], imm8[7:0]) +FOR j := 0 to 7 + i := j*64 + dst[i+63:i] := a[i+63:i] XOR b[i+63:i] ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - AVX512F - Miscellaneous - - - - - - Fix up packed single-precision (32-bit) floating-point elements in "a" and "b" using packed 32-bit integers in "c", and store the results in "dst". "imm8" is used to set the required flags reporting. - [round_note] - - -enum TOKEN_TYPE { - QNAN_TOKEN := 0, - SNAN_TOKEN L= 1, - ZERO_VALUE_TOKEN := 2, - ONE_VALUE_TOKEN := 3, - NEG_INF_TOKEN := 4, - POS_INF_TOKEN := 5, - NEG_VALUE_TOKEN := 6, - POS_VALUE_TOKEN := 7 -} -FIXUPIMMPD(src1[31:0], src2[31:0], src3[31:0], imm8[7:0]){ - tsrc[31:0] := ((src2[30:23] == 0) AND (MXCSR.DAZ == 1)) ? 0.0 : src2[31:0] - CASE(tsrc[31:0] of TOKEN_TYPE) - QNAN_TOKEN:j := 0 - SNAN_TOKEN:j := 1 - ZERO_VALUE_TOKEN: j := 2 - ONE_VALUE_TOKEN: j := 3 - NEG_INF_TOKEN: j := 4 - POS_INF_TOKEN: j := 5 - NEG_VALUE_TOKEN: j := 6 - POS_VALUE_TOKEN: j := 7 - ESAC - - token_response[3:0] := src3[3+4*j:4*j] - - CASE(token_response[3:0]) of - 0 : dest[31:0] := src1[31:0] - 1 : dest[31:0] := tsrc[31:0] - 2 : dest[31:0] := QNaN(tsrc[31:0]) - 3 : dest[31:0] := QNAN_Indefinite - 4 : dest[31:0] := -INF - 5 : dest[31:0] := +INF - 6 : dest[31:0] := tsrc.sign? –INF : +INF - 7 : dest[31:0] := -0 - 8 : dest[31:0] := +0 - 9 : dest[31:0] := -1 - 10: dest[31:0] := +1 - 11: dest[31:0] := 1⁄2 - 12: dest[31:0] := 90.0 - 13: dest[31:0] := PI/2 - 14: dest[31:0] := MAX_FLOAT - 15: dest[31:0] := -MAX_FLOAT - ESAC - - CASE(tsrc[31:0] of TOKEN_TYPE) - ZERO_VALUE_TOKEN: if imm8[0] then set #ZE - ZERO_VALUE_TOKEN: if imm8[1] then set #IE - ONE_VALUE_TOKEN: if imm8[2] then set #ZE - ONE_VALUE_TOKEN: if imm8[3] then set #IE - SNAN_TOKEN: if imm8[4] then set #IE - NEG_INF_TOKEN: if imm8[5] then set #IE - NEG_VALUE_TOKEN: if imm8[6] then set #IE - POS_INF_TOKEN: if imm8[7] then set #IE - ESAC - RETURN dest[31:0] -} - -FOR j := 0 to 15 - i := j*32 - dst[i+31:i] := FIXUPIMMPD(a[i+31:i], b[i+31:i], c[i+31:i], imm8[7:0]) + AVX512VL + AVX512DQ + Logical + + + + + + Compute the bitwise XOR of packed double-precision (64-bit) floating-point elements in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +FOR j := 0 to 1 + i := j*64 + IF k[j] + dst[i+63:i] := a[i+63:i] XOR b[i+63:i] + ELSE + dst[i+63:i] := src[i+63:i] + FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point - AVX512F - Miscellaneous - - - - - - Fix up packed single-precision (32-bit) floating-point elements in "a" and "b" using packed 32-bit integers in "c", and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). "imm8" is used to set the required flags reporting. + AVX512VL + AVX512DQ + Logical + + + + + Compute the bitwise XOR of packed double-precision (64-bit) floating-point elements in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -enum TOKEN_TYPE { - QNAN_TOKEN := 0, - SNAN_TOKEN L= 1, - ZERO_VALUE_TOKEN := 2, - ONE_VALUE_TOKEN := 3, - NEG_INF_TOKEN := 4, - POS_INF_TOKEN := 5, - NEG_VALUE_TOKEN := 6, - POS_VALUE_TOKEN := 7 -} -FIXUPIMMPD(src1[31:0], src2[31:0], src3[31:0], imm8[7:0]){ - tsrc[31:0] := ((src2[30:23] == 0) AND (MXCSR.DAZ == 1)) ? 0.0 : src2[31:0] - CASE(tsrc[31:0] of TOKEN_TYPE) - QNAN_TOKEN:j := 0 - SNAN_TOKEN:j := 1 - ZERO_VALUE_TOKEN: j := 2 - ONE_VALUE_TOKEN: j := 3 - NEG_INF_TOKEN: j := 4 - POS_INF_TOKEN: j := 5 - NEG_VALUE_TOKEN: j := 6 - POS_VALUE_TOKEN: j := 7 - ESAC - - token_response[3:0] := src3[3+4*j:4*j] - - CASE(token_response[3:0]) of - 0 : dest[31:0] := src1[31:0] - 1 : dest[31:0] := tsrc[31:0] - 2 : dest[31:0] := QNaN(tsrc[31:0]) - 3 : dest[31:0] := QNAN_Indefinite - 4 : dest[31:0] := -INF - 5 : dest[31:0] := +INF - 6 : dest[31:0] := tsrc.sign? –INF : +INF - 7 : dest[31:0] := -0 - 8 : dest[31:0] := +0 - 9 : dest[31:0] := -1 - 10: dest[31:0] := +1 - 11: dest[31:0] := 1⁄2 - 12: dest[31:0] := 90.0 - 13: dest[31:0] := PI/2 - 14: dest[31:0] := MAX_FLOAT - 15: dest[31:0] := -MAX_FLOAT - ESAC - - CASE(tsrc[31:0] of TOKEN_TYPE) - ZERO_VALUE_TOKEN: if imm8[0] then set #ZE - ZERO_VALUE_TOKEN: if imm8[1] then set #IE - ONE_VALUE_TOKEN: if imm8[2] then set #ZE - ONE_VALUE_TOKEN: if imm8[3] then set #IE - SNAN_TOKEN: if imm8[4] then set #IE - NEG_INF_TOKEN: if imm8[5] then set #IE - NEG_VALUE_TOKEN: if imm8[6] then set #IE - POS_INF_TOKEN: if imm8[7] then set #IE - ESAC - RETURN dest[31:0] -} - -FOR j := 0 to 15 +FOR j := 0 to 1 + i := j*64 + IF k[j] + dst[i+63:i] := a[i+63:i] XOR b[i+63:i] + ELSE + dst[i+63:i] := 0 + FI +ENDFOR +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512VL + AVX512DQ + Logical + + + + + + Compute the bitwise XOR of packed single-precision (32-bit) floating-point elements in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +FOR j := 0 to 7 i := j*32 IF k[j] - dst[i+31:i] := FIXUPIMMPD(a[i+31:i], b[i+31:i], c[i+31:i], imm8[7:0]) + dst[i+31:i] := a[i+31:i] XOR b[i+31:i] ELSE - dst[i+31:i] := a[i+31:i] + dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Floating Point - AVX512F - Miscellaneous - - - - - - - Fix up packed single-precision (32-bit) floating-point elements in "a" and "b" using packed 32-bit integers in "c", and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). "imm8" is used to set the required flags reporting. - [round_note] - - -enum TOKEN_TYPE { - QNAN_TOKEN := 0, - SNAN_TOKEN L= 1, - ZERO_VALUE_TOKEN := 2, - ONE_VALUE_TOKEN := 3, - NEG_INF_TOKEN := 4, - POS_INF_TOKEN := 5, - NEG_VALUE_TOKEN := 6, - POS_VALUE_TOKEN := 7 -} -FIXUPIMMPD(src1[31:0], src2[31:0], src3[31:0], imm8[7:0]){ - tsrc[31:0] := ((src2[30:23] == 0) AND (MXCSR.DAZ == 1)) ? 0.0 : src2[31:0] - CASE(tsrc[31:0] of TOKEN_TYPE) - QNAN_TOKEN:j := 0 - SNAN_TOKEN:j := 1 - ZERO_VALUE_TOKEN: j := 2 - ONE_VALUE_TOKEN: j := 3 - NEG_INF_TOKEN: j := 4 - POS_INF_TOKEN: j := 5 - NEG_VALUE_TOKEN: j := 6 - POS_VALUE_TOKEN: j := 7 - ESAC - - token_response[3:0] := src3[3+4*j:4*j] - - CASE(token_response[3:0]) of - 0 : dest[31:0] := src1[31:0] - 1 : dest[31:0] := tsrc[31:0] - 2 : dest[31:0] := QNaN(tsrc[31:0]) - 3 : dest[31:0] := QNAN_Indefinite - 4 : dest[31:0] := -INF - 5 : dest[31:0] := +INF - 6 : dest[31:0] := tsrc.sign? –INF : +INF - 7 : dest[31:0] := -0 - 8 : dest[31:0] := +0 - 9 : dest[31:0] := -1 - 10: dest[31:0] := +1 - 11: dest[31:0] := 1⁄2 - 12: dest[31:0] := 90.0 - 13: dest[31:0] := PI/2 - 14: dest[31:0] := MAX_FLOAT - 15: dest[31:0] := -MAX_FLOAT - ESAC - - CASE(tsrc[31:0] of TOKEN_TYPE) - ZERO_VALUE_TOKEN: if imm8[0] then set #ZE - ZERO_VALUE_TOKEN: if imm8[1] then set #IE - ONE_VALUE_TOKEN: if imm8[2] then set #ZE - ONE_VALUE_TOKEN: if imm8[3] then set #IE - SNAN_TOKEN: if imm8[4] then set #IE - NEG_INF_TOKEN: if imm8[5] then set #IE - NEG_VALUE_TOKEN: if imm8[6] then set #IE - POS_INF_TOKEN: if imm8[7] then set #IE - ESAC - RETURN dest[31:0] -} - -FOR j := 0 to 15 + AVX512VL + AVX512DQ + Logical + + + + + Compute the bitwise XOR of packed single-precision (32-bit) floating-point elements in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + +FOR j := 0 to 7 i := j*32 IF k[j] - dst[i+31:i] := FIXUPIMMPD(a[i+31:i], b[i+31:i], c[i+31:i], imm8[7:0]) + dst[i+31:i] := a[i+31:i] XOR b[i+31:i] ELSE - dst[i+31:i] := a[i+31:i] + dst[i+31:i] := 0 FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Floating Point - AVX512F - Miscellaneous - - - - - - Fix up packed single-precision (32-bit) floating-point elements in "a" and "b" using packed 32-bit integers in "c", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). "imm8" is used to set the required flags reporting. + AVX512DQ + Logical + + + + + + Compute the bitwise XOR of packed single-precision (32-bit) floating-point elements in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -enum TOKEN_TYPE { - QNAN_TOKEN := 0, - SNAN_TOKEN L= 1, - ZERO_VALUE_TOKEN := 2, - ONE_VALUE_TOKEN := 3, - NEG_INF_TOKEN := 4, - POS_INF_TOKEN := 5, - NEG_VALUE_TOKEN := 6, - POS_VALUE_TOKEN := 7 -} -FIXUPIMMPD(src1[31:0], src2[31:0], src3[31:0], imm8[7:0]){ - tsrc[31:0] := ((src2[30:23] == 0) AND (MXCSR.DAZ == 1)) ? 0.0 : src2[31:0] - CASE(tsrc[31:0] of TOKEN_TYPE) - QNAN_TOKEN:j := 0 - SNAN_TOKEN:j := 1 - ZERO_VALUE_TOKEN: j := 2 - ONE_VALUE_TOKEN: j := 3 - NEG_INF_TOKEN: j := 4 - POS_INF_TOKEN: j := 5 - NEG_VALUE_TOKEN: j := 6 - POS_VALUE_TOKEN: j := 7 - ESAC - - token_response[3:0] := src3[3+4*j:4*j] - - CASE(token_response[3:0]) of - 0 : dest[31:0] := src1[31:0] - 1 : dest[31:0] := tsrc[31:0] - 2 : dest[31:0] := QNaN(tsrc[31:0]) - 3 : dest[31:0] := QNAN_Indefinite - 4 : dest[31:0] := -INF - 5 : dest[31:0] := +INF - 6 : dest[31:0] := tsrc.sign? –INF : +INF - 7 : dest[31:0] := -0 - 8 : dest[31:0] := +0 - 9 : dest[31:0] := -1 - 10: dest[31:0] := +1 - 11: dest[31:0] := 1⁄2 - 12: dest[31:0] := 90.0 - 13: dest[31:0] := PI/2 - 14: dest[31:0] := MAX_FLOAT - 15: dest[31:0] := -MAX_FLOAT - ESAC - - CASE(tsrc[31:0] of TOKEN_TYPE) - ZERO_VALUE_TOKEN: if imm8[0] then set #ZE - ZERO_VALUE_TOKEN: if imm8[1] then set #IE - ONE_VALUE_TOKEN: if imm8[2] then set #ZE - ONE_VALUE_TOKEN: if imm8[3] then set #IE - SNAN_TOKEN: if imm8[4] then set #IE - NEG_INF_TOKEN: if imm8[5] then set #IE - NEG_VALUE_TOKEN: if imm8[6] then set #IE - POS_INF_TOKEN: if imm8[7] then set #IE - ESAC - RETURN dest[31:0] -} - FOR j := 0 to 15 i := j*32 IF k[j] - dst[i+31:i] := FIXUPIMMPD(a[i+31:i], b[i+31:i], c[i+31:i], imm8[7:0]) + dst[i+31:i] := a[i+31:i] XOR b[i+31:i] ELSE - dst[i+31:i] := 0 + dst[i+31:i] := src[i+31:i] FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - AVX512F - Miscellaneous - - - - - - - Fix up packed single-precision (32-bit) floating-point elements in "a" and "b" using packed 32-bit integers in "c", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). "imm8" is used to set the required flags reporting. - [round_note] - - -enum TOKEN_TYPE { - QNAN_TOKEN := 0, - SNAN_TOKEN L= 1, - ZERO_VALUE_TOKEN := 2, - ONE_VALUE_TOKEN := 3, - NEG_INF_TOKEN := 4, - POS_INF_TOKEN := 5, - NEG_VALUE_TOKEN := 6, - POS_VALUE_TOKEN := 7 -} -FIXUPIMMPD(src1[31:0], src2[31:0], src3[31:0], imm8[7:0]){ - tsrc[31:0] := ((src2[30:23] == 0) AND (MXCSR.DAZ == 1)) ? 0.0 : src2[31:0] - CASE(tsrc[31:0] of TOKEN_TYPE) - QNAN_TOKEN:j := 0 - SNAN_TOKEN:j := 1 - ZERO_VALUE_TOKEN: j := 2 - ONE_VALUE_TOKEN: j := 3 - NEG_INF_TOKEN: j := 4 - POS_INF_TOKEN: j := 5 - NEG_VALUE_TOKEN: j := 6 - POS_VALUE_TOKEN: j := 7 - ESAC - - token_response[3:0] := src3[3+4*j:4*j] - - CASE(token_response[3:0]) of - 0 : dest[31:0] := src1[31:0] - 1 : dest[31:0] := tsrc[31:0] - 2 : dest[31:0] := QNaN(tsrc[31:0]) - 3 : dest[31:0] := QNAN_Indefinite - 4 : dest[31:0] := -INF - 5 : dest[31:0] := +INF - 6 : dest[31:0] := tsrc.sign? –INF : +INF - 7 : dest[31:0] := -0 - 8 : dest[31:0] := +0 - 9 : dest[31:0] := -1 - 10: dest[31:0] := +1 - 11: dest[31:0] := 1⁄2 - 12: dest[31:0] := 90.0 - 13: dest[31:0] := PI/2 - 14: dest[31:0] := MAX_FLOAT - 15: dest[31:0] := -MAX_FLOAT - ESAC - - CASE(tsrc[31:0] of TOKEN_TYPE) - ZERO_VALUE_TOKEN: if imm8[0] then set #ZE - ZERO_VALUE_TOKEN: if imm8[1] then set #IE - ONE_VALUE_TOKEN: if imm8[2] then set #ZE - ONE_VALUE_TOKEN: if imm8[3] then set #IE - SNAN_TOKEN: if imm8[4] then set #IE - NEG_INF_TOKEN: if imm8[5] then set #IE - NEG_VALUE_TOKEN: if imm8[6] then set #IE - POS_INF_TOKEN: if imm8[7] then set #IE - ESAC - RETURN dest[31:0] -} - + AVX512DQ + Logical + + + + + Compute the bitwise XOR of packed single-precision (32-bit) floating-point elements in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + FOR j := 0 to 15 i := j*32 IF k[j] - dst[i+31:i] := FIXUPIMMPD(a[i+31:i], b[i+31:i], c[i+31:i], imm8[7:0]) + dst[i+31:i] := a[i+31:i] XOR b[i+31:i] ELSE dst[i+31:i] := 0 FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - AVX512F - Miscellaneous - - - - - - Fix up the lower double-precision (64-bit) floating-point elements in "a" and "b" using the lower 64-bit integer in "c", store the result in the lower element of "dst", and copy the upper element from "a" to the upper element of "dst". "imm8" is used to set the required flags reporting. - [round_note] - - -enum TOKEN_TYPE { - QNAN_TOKEN := 0, - SNAN_TOKEN := 1, - ZERO_VALUE_TOKEN := 2, - ONE_VALUE_TOKEN := 3, - NEG_INF_TOKEN := 4, - POS_INF_TOKEN := 5, - NEG_VALUE_TOKEN := 6, - POS_VALUE_TOKEN := 7 -} -FIXUPIMMPD(src1[63:0], src2[63:0], src3[63:0], imm8[7:0]){ - tsrc[63:0] := ((src2[62:52] == 0) AND (MXCSR.DAZ == 1)) ? 0.0 : src2[63:0] - CASE(tsrc[63:0] of TOKEN_TYPE) - QNAN_TOKEN:j := 0 - SNAN_TOKEN:j := 1 - ZERO_VALUE_TOKEN: j := 2 - ONE_VALUE_TOKEN: j := 3 - NEG_INF_TOKEN: j := 4 - POS_INF_TOKEN: j := 5 - NEG_VALUE_TOKEN: j := 6 - POS_VALUE_TOKEN: j := 7 - ESAC - - token_response[3:0] := src3[3+4*j:4*j] - - CASE(token_response[3:0]) of - 0 : dest[63:0] := src1[63:0] - 1 : dest[63:0] := tsrc[63:0] - 2 : dest[63:0] := QNaN(tsrc[63:0]) - 3 : dest[63:0] := QNAN_Indefinite - 4 : dest[63:0] := -INF - 5 : dest[63:0] := +INF - 6 : dest[63:0] := tsrc.sign? –INF : +INF - 7 : dest[63:0] := -0 - 8 : dest[63:0] := +0 - 9 : dest[63:0] := -1 - 10: dest[63:0] := +1 - 11: dest[63:0] := 1⁄2 - 12: dest[63:0] := 90.0 - 13: dest[63:0] := PI/2 - 14: dest[63:0] := MAX_FLOAT - 15: dest[63:0] := -MAX_FLOAT - ESAC - - CASE(tsrc[31:0] of TOKEN_TYPE) - ZERO_VALUE_TOKEN: if imm8[0] then set #ZE - ZERO_VALUE_TOKEN: if imm8[1] then set #IE - ONE_VALUE_TOKEN: if imm8[2] then set #ZE - ONE_VALUE_TOKEN: if imm8[3] then set #IE - SNAN_TOKEN: if imm8[4] then set #IE - NEG_INF_TOKEN: if imm8[5] then set #IE - NEG_VALUE_TOKEN: if imm8[6] then set #IE - POS_INF_TOKEN: if imm8[7] then set #IE - ESAC - RETURN dest[63:0] -} - -dst[63:0] := FIXUPIMMPD(a[63:0], b[63:0], c[63:0], imm8[7:0]) -dst[127:64] := a[127:64] -dst[MAX:128] := 0 + AVX512DQ + Logical + + + + Compute the bitwise XOR of packed single-precision (32-bit) floating-point elements in "a" and "b", and store the results in "dst". + +FOR j := 0 to 15 + i := j*32 + dst[i+31:i] := a[i+31:i] XOR b[i+31:i] +ENDFOR +dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - AVX512F - Miscellaneous - - - - - Fix up the lower double-precision (64-bit) floating-point elements in "a" and "b" using the lower 64-bit integer in "c", store the result in the lower element of "dst", and copy the upper element from "a" to the upper element of "dst". "imm8" is used to set the required flags reporting. + AVX512VL + AVX512DQ + Logical + + + + + + Compute the bitwise XOR of packed single-precision (32-bit) floating-point elements in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -enum TOKEN_TYPE { - QNAN_TOKEN := 0, - SNAN_TOKEN := 1, - ZERO_VALUE_TOKEN := 2, - ONE_VALUE_TOKEN := 3, - NEG_INF_TOKEN := 4, - POS_INF_TOKEN := 5, - NEG_VALUE_TOKEN := 6, - POS_VALUE_TOKEN := 7 -} -FIXUPIMMPD(src1[63:0], src2[63:0], src3[63:0], imm8[7:0]){ - tsrc[63:0] := ((src2[62:52] == 0) AND (MXCSR.DAZ == 1)) ? 0.0 : src2[63:0] - CASE(tsrc[63:0] of TOKEN_TYPE) - QNAN_TOKEN:j := 0 - SNAN_TOKEN:j := 1 - ZERO_VALUE_TOKEN: j := 2 - ONE_VALUE_TOKEN: j := 3 - NEG_INF_TOKEN: j := 4 - POS_INF_TOKEN: j := 5 - NEG_VALUE_TOKEN: j := 6 - POS_VALUE_TOKEN: j := 7 - ESAC - - token_response[3:0] := src3[3+4*j:4*j] - - CASE(token_response[3:0]) of - 0 : dest[63:0] := src1[63:0] - 1 : dest[63:0] := tsrc[63:0] - 2 : dest[63:0] := QNaN(tsrc[63:0]) - 3 : dest[63:0] := QNAN_Indefinite - 4 : dest[63:0] := -INF - 5 : dest[63:0] := +INF - 6 : dest[63:0] := tsrc.sign? –INF : +INF - 7 : dest[63:0] := -0 - 8 : dest[63:0] := +0 - 9 : dest[63:0] := -1 - 10: dest[63:0] := +1 - 11: dest[63:0] := 1⁄2 - 12: dest[63:0] := 90.0 - 13: dest[63:0] := PI/2 - 14: dest[63:0] := MAX_FLOAT - 15: dest[63:0] := -MAX_FLOAT - ESAC - - CASE(tsrc[31:0] of TOKEN_TYPE) - ZERO_VALUE_TOKEN: if imm8[0] then set #ZE - ZERO_VALUE_TOKEN: if imm8[1] then set #IE - ONE_VALUE_TOKEN: if imm8[2] then set #ZE - ONE_VALUE_TOKEN: if imm8[3] then set #IE - SNAN_TOKEN: if imm8[4] then set #IE - NEG_INF_TOKEN: if imm8[5] then set #IE - NEG_VALUE_TOKEN: if imm8[6] then set #IE - POS_INF_TOKEN: if imm8[7] then set #IE - ESAC - RETURN dest[63:0] -} - -dst[63:0] := FIXUPIMMPD(a[63:0], b[63:0], c[63:0], imm8[7:0]) -dst[127:64] := a[127:64] +FOR j := 0 to 3 + i := j*32 + IF k[j] + dst[i+31:i] := a[i+31:i] XOR b[i+31:i] + ELSE + dst[i+31:i] := src[i+31:i] + FI +ENDFOR dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point - AVX512F - Miscellaneous - - - - - - - Fix up the lower double-precision (64-bit) floating-point elements in "a" and "b" using the lower 64-bit integer in "c", store the result in the lower element of "dst" using writemask "k" (the element is copied from "a" when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". "imm8" is used to set the required flags reporting. - [round_note] - - -enum TOKEN_TYPE { - QNAN_TOKEN := 0, - SNAN_TOKEN := 1, - ZERO_VALUE_TOKEN := 2, - ONE_VALUE_TOKEN := 3, - NEG_INF_TOKEN := 4, - POS_INF_TOKEN := 5, - NEG_VALUE_TOKEN := 6, - POS_VALUE_TOKEN := 7 -} -FIXUPIMMPD(src1[63:0], src2[63:0], src3[63:0], imm8[7:0]){ - tsrc[63:0] := ((src2[62:52] == 0) AND (MXCSR.DAZ == 1)) ? 0.0 : src2[63:0] - CASE(tsrc[63:0] of TOKEN_TYPE) - QNAN_TOKEN:j := 0 - SNAN_TOKEN:j := 1 - ZERO_VALUE_TOKEN: j := 2 - ONE_VALUE_TOKEN: j := 3 - NEG_INF_TOKEN: j := 4 - POS_INF_TOKEN: j := 5 - NEG_VALUE_TOKEN: j := 6 - POS_VALUE_TOKEN: j := 7 - ESAC - - token_response[3:0] := src3[3+4*j:4*j] - - CASE(token_response[3:0]) of - 0 : dest[63:0] := src1[63:0] - 1 : dest[63:0] := tsrc[63:0] - 2 : dest[63:0] := QNaN(tsrc[63:0]) - 3 : dest[63:0] := QNAN_Indefinite - 4 : dest[63:0] := -INF - 5 : dest[63:0] := +INF - 6 : dest[63:0] := tsrc.sign? –INF : +INF - 7 : dest[63:0] := -0 - 8 : dest[63:0] := +0 - 9 : dest[63:0] := -1 - 10: dest[63:0] := +1 - 11: dest[63:0] := 1⁄2 - 12: dest[63:0] := 90.0 - 13: dest[63:0] := PI/2 - 14: dest[63:0] := MAX_FLOAT - 15: dest[63:0] := -MAX_FLOAT - ESAC - - CASE(tsrc[31:0] of TOKEN_TYPE) - ZERO_VALUE_TOKEN: if imm8[0] then set #ZE - ZERO_VALUE_TOKEN: if imm8[1] then set #IE - ONE_VALUE_TOKEN: if imm8[2] then set #ZE - ONE_VALUE_TOKEN: if imm8[3] then set #IE - SNAN_TOKEN: if imm8[4] then set #IE - NEG_INF_TOKEN: if imm8[5] then set #IE - NEG_VALUE_TOKEN: if imm8[6] then set #IE - POS_INF_TOKEN: if imm8[7] then set #IE - ESAC - RETURN dest[63:0] -} - -IF k[0] - dst[63:0] := FIXUPIMMPD(a[63:0], b[63:0], c[63:0], imm8[7:0]) -ELSE - dst[63:0] := a[63:0] -FI -dst[127:64] := a[127:64] + AVX512VL + AVX512DQ + Logical + + + + + Compute the bitwise XOR of packed single-precision (32-bit) floating-point elements in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + +FOR j := 0 to 3 + i := j*32 + IF k[j] + dst[i+31:i] := a[i+31:i] XOR b[i+31:i] + ELSE + dst[i+31:i] := 0 + FI +ENDFOR dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point - AVX512F - Miscellaneous - - - - - - Fix up the lower double-precision (64-bit) floating-point elements in "a" and "b" using the lower 64-bit integer in "c", store the result in the lower element of "dst" using writemask "k" (the element is copied from "a" when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". "imm8" is used to set the required flags reporting. + + Mask + AVX512DQ + Mask + + + + Add 8-bit masks in "a" and "b", and store the result in "k". -enum TOKEN_TYPE { - QNAN_TOKEN := 0, - SNAN_TOKEN := 1, - ZERO_VALUE_TOKEN := 2, - ONE_VALUE_TOKEN := 3, - NEG_INF_TOKEN := 4, - POS_INF_TOKEN := 5, - NEG_VALUE_TOKEN := 6, - POS_VALUE_TOKEN := 7 -} -FIXUPIMMPD(src1[63:0], src2[63:0], src3[63:0], imm8[7:0]){ - tsrc[63:0] := ((src2[62:52] == 0) AND (MXCSR.DAZ == 1)) ? 0.0 : src2[63:0] - CASE(tsrc[63:0] of TOKEN_TYPE) - QNAN_TOKEN:j := 0 - SNAN_TOKEN:j := 1 - ZERO_VALUE_TOKEN: j := 2 - ONE_VALUE_TOKEN: j := 3 - NEG_INF_TOKEN: j := 4 - POS_INF_TOKEN: j := 5 - NEG_VALUE_TOKEN: j := 6 - POS_VALUE_TOKEN: j := 7 - ESAC - - token_response[3:0] := src3[3+4*j:4*j] - - CASE(token_response[3:0]) of - 0 : dest[63:0] := src1[63:0] - 1 : dest[63:0] := tsrc[63:0] - 2 : dest[63:0] := QNaN(tsrc[63:0]) - 3 : dest[63:0] := QNAN_Indefinite - 4 : dest[63:0] := -INF - 5 : dest[63:0] := +INF - 6 : dest[63:0] := tsrc.sign? –INF : +INF - 7 : dest[63:0] := -0 - 8 : dest[63:0] := +0 - 9 : dest[63:0] := -1 - 10: dest[63:0] := +1 - 11: dest[63:0] := 1⁄2 - 12: dest[63:0] := 90.0 - 13: dest[63:0] := PI/2 - 14: dest[63:0] := MAX_FLOAT - 15: dest[63:0] := -MAX_FLOAT - ESAC - - CASE(tsrc[31:0] of TOKEN_TYPE) - ZERO_VALUE_TOKEN: if imm8[0] then set #ZE - ZERO_VALUE_TOKEN: if imm8[1] then set #IE - ONE_VALUE_TOKEN: if imm8[2] then set #ZE - ONE_VALUE_TOKEN: if imm8[3] then set #IE - SNAN_TOKEN: if imm8[4] then set #IE - NEG_INF_TOKEN: if imm8[5] then set #IE - NEG_VALUE_TOKEN: if imm8[6] then set #IE - POS_INF_TOKEN: if imm8[7] then set #IE - ESAC - RETURN dest[63:0] -} - -IF k[0] - dst[63:0] := FIXUPIMMPD(a[63:0], b[63:0], c[63:0], imm8[7:0]) -ELSE - dst[63:0] := a[63:0] -FI -dst[127:64] := a[127:64] -dst[MAX:128] := 0 +k[7:0] := a[7:0] + b[7:0] +k[MAX:8] := 0 - +
immintrin.h
 - - Floating Point - AVX512F - Miscellaneous - - - - - - - Fix up the lower double-precision (64-bit) floating-point elements in "a" and "b" using the lower 64-bit integer in "c", store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". "imm8" is used to set the required flags reporting. - [round_note] - - -enum TOKEN_TYPE { - QNAN_TOKEN := 0, - SNAN_TOKEN := 1, - ZERO_VALUE_TOKEN := 2, - ONE_VALUE_TOKEN := 3, - NEG_INF_TOKEN := 4, - POS_INF_TOKEN := 5, - NEG_VALUE_TOKEN := 6, - POS_VALUE_TOKEN := 7 -} -FIXUPIMMPD(src1[63:0], src2[63:0], src3[63:0], imm8[7:0]){ - tsrc[63:0] := ((src2[62:52] == 0) AND (MXCSR.DAZ == 1)) ? 0.0 : src2[63:0] - CASE(tsrc[63:0] of TOKEN_TYPE) - QNAN_TOKEN:j := 0 - SNAN_TOKEN:j := 1 - ZERO_VALUE_TOKEN: j := 2 - ONE_VALUE_TOKEN: j := 3 - NEG_INF_TOKEN: j := 4 - POS_INF_TOKEN: j := 5 - NEG_VALUE_TOKEN: j := 6 - POS_VALUE_TOKEN: j := 7 - ESAC - - token_response[3:0] := src3[3+4*j:4*j] - - CASE(token_response[3:0]) of - 0 : dest[63:0] := src1[63:0] - 1 : dest[63:0] := tsrc[63:0] - 2 : dest[63:0] := QNaN(tsrc[63:0]) - 3 : dest[63:0] := QNAN_Indefinite - 4 : dest[63:0] := -INF - 5 : dest[63:0] := +INF - 6 : dest[63:0] := tsrc.sign? –INF : +INF - 7 : dest[63:0] := -0 - 8 : dest[63:0] := +0 - 9 : dest[63:0] := -1 - 10: dest[63:0] := +1 - 11: dest[63:0] := 1⁄2 - 12: dest[63:0] := 90.0 - 13: dest[63:0] := PI/2 - 14: dest[63:0] := MAX_FLOAT - 15: dest[63:0] := -MAX_FLOAT - ESAC - - CASE(tsrc[31:0] of TOKEN_TYPE) - ZERO_VALUE_TOKEN: if imm8[0] then set #ZE - ZERO_VALUE_TOKEN: if imm8[1] then set #IE - ONE_VALUE_TOKEN: if imm8[2] then set #ZE - ONE_VALUE_TOKEN: if imm8[3] then set #IE - SNAN_TOKEN: if imm8[4] then set #IE - NEG_INF_TOKEN: if imm8[5] then set #IE - NEG_VALUE_TOKEN: if imm8[6] then set #IE - POS_INF_TOKEN: if imm8[7] then set #IE - ESAC - RETURN dest[63:0] -} - -IF k[0] - dst[63:0] := FIXUPIMMPD(a[63:0], b[63:0], c[63:0], imm8[7:0]) -ELSE - dst[63:0] := 0 -FI -dst[127:64] := a[127:64] -dst[MAX:128] := 0 + + Mask + AVX512DQ + Mask + + + + Add 16-bit masks in "a" and "b", and store the result in "k". + +k[15:0] := a[15:0] + b[15:0] +k[MAX:16] := 0 - +
immintrin.h
 - - Floating Point - AVX512F - Miscellaneous - - - - - - Fix up the lower double-precision (64-bit) floating-point elements in "a" and "b" using the lower 64-bit integer in "c", store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". "imm8" is used to set the required flags reporting. + + Mask + AVX512DQ + Mask + + + + Compute the bitwise AND of 8-bit masks "a" and "b", and store the result in "k". -enum TOKEN_TYPE { - QNAN_TOKEN := 0, - SNAN_TOKEN := 1, - ZERO_VALUE_TOKEN := 2, - ONE_VALUE_TOKEN := 3, - NEG_INF_TOKEN := 4, - POS_INF_TOKEN := 5, - NEG_VALUE_TOKEN := 6, - POS_VALUE_TOKEN := 7 -} -FIXUPIMMPD(src1[63:0], src2[63:0], src3[63:0], imm8[7:0]){ - tsrc[63:0] := ((src2[62:52] == 0) AND (MXCSR.DAZ == 1)) ? 0.0 : src2[63:0] - CASE(tsrc[63:0] of TOKEN_TYPE) - QNAN_TOKEN:j := 0 - SNAN_TOKEN:j := 1 - ZERO_VALUE_TOKEN: j := 2 - ONE_VALUE_TOKEN: j := 3 - NEG_INF_TOKEN: j := 4 - POS_INF_TOKEN: j := 5 - NEG_VALUE_TOKEN: j := 6 - POS_VALUE_TOKEN: j := 7 - ESAC - - token_response[3:0] := src3[3+4*j:4*j] - - CASE(token_response[3:0]) of - 0 : dest[63:0] := src1[63:0] - 1 : dest[63:0] := tsrc[63:0] - 2 : dest[63:0] := QNaN(tsrc[63:0]) - 3 : dest[63:0] := QNAN_Indefinite - 4 : dest[63:0] := -INF - 5 : dest[63:0] := +INF - 6 : dest[63:0] := tsrc.sign? –INF : +INF - 7 : dest[63:0] := -0 - 8 : dest[63:0] := +0 - 9 : dest[63:0] := -1 - 10: dest[63:0] := +1 - 11: dest[63:0] := 1⁄2 - 12: dest[63:0] := 90.0 - 13: dest[63:0] := PI/2 - 14: dest[63:0] := MAX_FLOAT - 15: dest[63:0] := -MAX_FLOAT - ESAC - - CASE(tsrc[31:0] of TOKEN_TYPE) - ZERO_VALUE_TOKEN: if imm8[0] then set #ZE - ZERO_VALUE_TOKEN: if imm8[1] then set #IE - ONE_VALUE_TOKEN: if imm8[2] then set #ZE - ONE_VALUE_TOKEN: if imm8[3] then set #IE - SNAN_TOKEN: if imm8[4] then set #IE - NEG_INF_TOKEN: if imm8[5] then set #IE - NEG_VALUE_TOKEN: if imm8[6] then set #IE - POS_INF_TOKEN: if imm8[7] then set #IE - ESAC - RETURN dest[63:0] -} - -IF k[0] - dst[63:0] := FIXUPIMMPD(a[63:0], b[63:0], c[63:0], imm8[7:0]) -ELSE - dst[63:0] := 0 -FI -dst[127:64] := a[127:64] -dst[MAX:128] := 0 +k[7:0] := a[7:0] AND b[7:0] +k[MAX:8] := 0 - +
immintrin.h
 - - Floating Point - AVX512F - Miscellaneous - - - - - - Fix up the lower single-precision (32-bit) floating-point elements in "a" and "b" using the lower 32-bit integer in "c", store the result in the lower element of "dst", and copy the upper 3 packed elements from "a" to the upper elements of "dst". "imm8" is used to set the required flags reporting. - [round_note] - - -enum TOKEN_TYPE { - QNAN_TOKEN := 0, - SNAN_TOKEN L= 1, - ZERO_VALUE_TOKEN := 2, - ONE_VALUE_TOKEN := 3, - NEG_INF_TOKEN := 4, - POS_INF_TOKEN := 5, - NEG_VALUE_TOKEN := 6, - POS_VALUE_TOKEN := 7 -} -FIXUPIMMPD(src1[31:0], src2[31:0], src3[31:0], imm8[7:0]){ - tsrc[31:0] := ((src2[30:23] == 0) AND (MXCSR.DAZ == 1)) ? 0.0 : src2[31:0] - CASE(tsrc[31:0] of TOKEN_TYPE) - QNAN_TOKEN:j := 0 - SNAN_TOKEN:j := 1 - ZERO_VALUE_TOKEN: j := 2 - ONE_VALUE_TOKEN: j := 3 - NEG_INF_TOKEN: j := 4 - POS_INF_TOKEN: j := 5 - NEG_VALUE_TOKEN: j := 6 - POS_VALUE_TOKEN: j := 7 - ESAC - - token_response[3:0] := src3[3+4*j:4*j] - - CASE(token_response[3:0]) of - 0 : dest[31:0] := src1[31:0] - 1 : dest[31:0] := tsrc[31:0] - 2 : dest[31:0] := QNaN(tsrc[31:0]) - 3 : dest[31:0] := QNAN_Indefinite - 4 : dest[31:0] := -INF - 5 : dest[31:0] := +INF - 6 : dest[31:0] := tsrc.sign? –INF : +INF - 7 : dest[31:0] := -0 - 8 : dest[31:0] := +0 - 9 : dest[31:0] := -1 - 10: dest[31:0] := +1 - 11: dest[31:0] := 1⁄2 - 12: dest[31:0] := 90.0 - 13: dest[31:0] := PI/2 - 14: dest[31:0] := MAX_FLOAT - 15: dest[31:0] := -MAX_FLOAT - ESAC - - CASE(tsrc[31:0] of TOKEN_TYPE) - ZERO_VALUE_TOKEN: if imm8[0] then set #ZE - ZERO_VALUE_TOKEN: if imm8[1] then set #IE - ONE_VALUE_TOKEN: if imm8[2] then set #ZE - ONE_VALUE_TOKEN: if imm8[3] then set #IE - SNAN_TOKEN: if imm8[4] then set #IE - NEG_INF_TOKEN: if imm8[5] then set #IE - NEG_VALUE_TOKEN: if imm8[6] then set #IE - POS_INF_TOKEN: if imm8[7] then set #IE - ESAC - RETURN dest[31:0] -} - -dst[31:0] := FIXUPIMMPD(a[31:0], b[31:0], c[31:0], imm8[7:0]) -dst[127:32] := a[127:32] -dst[MAX:128] := 0 + + Mask + AVX512DQ + Mask + + + + Compute the bitwise NOT of 8-bit masks "a" and then AND with "b", and store the result in "k". + +k[7:0] := (NOT a[7:0]) AND b[7:0] +k[MAX:8] := 0 - +
immintrin.h
 - - Floating Point - AVX512F - Miscellaneous - - - - - Fix up the lower single-precision (32-bit) floating-point elements in "a" and "b" using the lower 32-bit integer in "c", store the result in the lower element of "dst", and copy the upper 3 packed elements from "a" to the upper elements of "dst". "imm8" is used to set the required flags reporting. + + Mask + AVX512DQ + Mask + + + Compute the bitwise NOT of 8-bit mask "a", and store the result in "k". -enum TOKEN_TYPE { - QNAN_TOKEN := 0, - SNAN_TOKEN L= 1, - ZERO_VALUE_TOKEN := 2, - ONE_VALUE_TOKEN := 3, - NEG_INF_TOKEN := 4, - POS_INF_TOKEN := 5, - NEG_VALUE_TOKEN := 6, - POS_VALUE_TOKEN := 7 -} -FIXUPIMMPD(src1[31:0], src2[31:0], src3[31:0], imm8[7:0]){ - tsrc[31:0] := ((src2[30:23] == 0) AND (MXCSR.DAZ == 1)) ? 0.0 : src2[31:0] - CASE(tsrc[31:0] of TOKEN_TYPE) - QNAN_TOKEN:j := 0 - SNAN_TOKEN:j := 1 - ZERO_VALUE_TOKEN: j := 2 - ONE_VALUE_TOKEN: j := 3 - NEG_INF_TOKEN: j := 4 - POS_INF_TOKEN: j := 5 - NEG_VALUE_TOKEN: j := 6 - POS_VALUE_TOKEN: j := 7 - ESAC - - token_response[3:0] := src3[3+4*j:4*j] - - CASE(token_response[3:0]) of - 0 : dest[31:0] := src1[31:0] - 1 : dest[31:0] := tsrc[31:0] - 2 : dest[31:0] := QNaN(tsrc[31:0]) - 3 : dest[31:0] := QNAN_Indefinite - 4 : dest[31:0] := -INF - 5 : dest[31:0] := +INF - 6 : dest[31:0] := tsrc.sign? –INF : +INF - 7 : dest[31:0] := -0 - 8 : dest[31:0] := +0 - 9 : dest[31:0] := -1 - 10: dest[31:0] := +1 - 11: dest[31:0] := 1⁄2 - 12: dest[31:0] := 90.0 - 13: dest[31:0] := PI/2 - 14: dest[31:0] := MAX_FLOAT - 15: dest[31:0] := -MAX_FLOAT - ESAC - - CASE(tsrc[31:0] of TOKEN_TYPE) - ZERO_VALUE_TOKEN: if imm8[0] then set #ZE - ZERO_VALUE_TOKEN: if imm8[1] then set #IE - ONE_VALUE_TOKEN: if imm8[2] then set #ZE - ONE_VALUE_TOKEN: if imm8[3] then set #IE - SNAN_TOKEN: if imm8[4] then set #IE - NEG_INF_TOKEN: if imm8[5] then set #IE - NEG_VALUE_TOKEN: if imm8[6] then set #IE - POS_INF_TOKEN: if imm8[7] then set #IE - ESAC - RETURN dest[31:0] -} - -dst[31:0] := FIXUPIMMPD(a[31:0], b[31:0], c[31:0], imm8[7:0]) -dst[127:32] := a[127:32] -dst[MAX:128] := 0 +k[7:0] := NOT a[7:0] +k[MAX:8] := 0 - +
immintrin.h
 - - Floating Point - AVX512F - Miscellaneous - - - - - - - Fix up the lower single-precision (32-bit) floating-point elements in "a" and "b" using the lower 32-bit integer in "c", store the result in the lower element of "dst" using writemask "k" (the element is copied from "a" when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". "imm8" is used to set the required flags reporting. - [round_note] - - -enum TOKEN_TYPE { - QNAN_TOKEN := 0, - SNAN_TOKEN L= 1, - ZERO_VALUE_TOKEN := 2, - ONE_VALUE_TOKEN := 3, - NEG_INF_TOKEN := 4, - POS_INF_TOKEN := 5, - NEG_VALUE_TOKEN := 6, - POS_VALUE_TOKEN := 7 -} -FIXUPIMMPD(src1[31:0], src2[31:0], src3[31:0], imm8[7:0]){ - tsrc[31:0] := ((src2[30:23] == 0) AND (MXCSR.DAZ == 1)) ? 0.0 : src2[31:0] - CASE(tsrc[31:0] of TOKEN_TYPE) - QNAN_TOKEN:j := 0 - SNAN_TOKEN:j := 1 - ZERO_VALUE_TOKEN: j := 2 - ONE_VALUE_TOKEN: j := 3 - NEG_INF_TOKEN: j := 4 - POS_INF_TOKEN: j := 5 - NEG_VALUE_TOKEN: j := 6 - POS_VALUE_TOKEN: j := 7 - ESAC - - token_response[3:0] := src3[3+4*j:4*j] - - CASE(token_response[3:0]) of - 0 : dest[31:0] := src1[31:0] - 1 : dest[31:0] := tsrc[31:0] - 2 : dest[31:0] := QNaN(tsrc[31:0]) - 3 : dest[31:0] := QNAN_Indefinite - 4 : dest[31:0] := -INF - 5 : dest[31:0] := +INF - 6 : dest[31:0] := tsrc.sign? –INF : +INF - 7 : dest[31:0] := -0 - 8 : dest[31:0] := +0 - 9 : dest[31:0] := -1 - 10: dest[31:0] := +1 - 11: dest[31:0] := 1⁄2 - 12: dest[31:0] := 90.0 - 13: dest[31:0] := PI/2 - 14: dest[31:0] := MAX_FLOAT - 15: dest[31:0] := -MAX_FLOAT - ESAC - - CASE(tsrc[31:0] of TOKEN_TYPE) - ZERO_VALUE_TOKEN: if imm8[0] then set #ZE - ZERO_VALUE_TOKEN: if imm8[1] then set #IE - ONE_VALUE_TOKEN: if imm8[2] then set #ZE - ONE_VALUE_TOKEN: if imm8[3] then set #IE - SNAN_TOKEN: if imm8[4] then set #IE - NEG_INF_TOKEN: if imm8[5] then set #IE - NEG_VALUE_TOKEN: if imm8[6] then set #IE - POS_INF_TOKEN: if imm8[7] then set #IE - ESAC - RETURN dest[31:0] -} - -IF k[0] - dst[31:0] := FIXUPIMMPD(a[31:0], b[31:0], c[31:0], imm8[7:0]) -ELSE - dst[31:0] := a[31:0] -FI -dst[127:32] := a[127:32] -dst[MAX:128] := 0 + + Mask + AVX512DQ + Mask + + + + Compute the bitwise OR of 8-bit masks "a" and "b", and store the result in "k". + +k[7:0] := a[7:0] OR b[7:0] +k[MAX:8] := 0 - +
immintrin.h
 - - Floating Point - AVX512F - Miscellaneous - - - - - - Fix up the lower single-precision (32-bit) floating-point elements in "a" and "b" using the lower 32-bit integer in "c", store the result in the lower element of "dst" using writemask "k" (the element is copied from "a" when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". "imm8" is used to set the required flags reporting. + + Mask + AVX512DQ + Mask + + + + Compute the bitwise XNOR of 8-bit masks "a" and "b", and store the result in "k". -enum TOKEN_TYPE { - QNAN_TOKEN := 0, - SNAN_TOKEN L= 1, - ZERO_VALUE_TOKEN := 2, - ONE_VALUE_TOKEN := 3, - NEG_INF_TOKEN := 4, - POS_INF_TOKEN := 5, - NEG_VALUE_TOKEN := 6, - POS_VALUE_TOKEN := 7 -} -FIXUPIMMPD(src1[31:0], src2[31:0], src3[31:0], imm8[7:0]){ - tsrc[31:0] := ((src2[30:23] == 0) AND (MXCSR.DAZ == 1)) ? 0.0 : src2[31:0] - CASE(tsrc[31:0] of TOKEN_TYPE) - QNAN_TOKEN:j := 0 - SNAN_TOKEN:j := 1 - ZERO_VALUE_TOKEN: j := 2 - ONE_VALUE_TOKEN: j := 3 - NEG_INF_TOKEN: j := 4 - POS_INF_TOKEN: j := 5 - NEG_VALUE_TOKEN: j := 6 - POS_VALUE_TOKEN: j := 7 - ESAC - - token_response[3:0] := src3[3+4*j:4*j] - - CASE(token_response[3:0]) of - 0 : dest[31:0] := src1[31:0] - 1 : dest[31:0] := tsrc[31:0] - 2 : dest[31:0] := QNaN(tsrc[31:0]) - 3 : dest[31:0] := QNAN_Indefinite - 4 : dest[31:0] := -INF - 5 : dest[31:0] := +INF - 6 : dest[31:0] := tsrc.sign? –INF : +INF - 7 : dest[31:0] := -0 - 8 : dest[31:0] := +0 - 9 : dest[31:0] := -1 - 10: dest[31:0] := +1 - 11: dest[31:0] := 1⁄2 - 12: dest[31:0] := 90.0 - 13: dest[31:0] := PI/2 - 14: dest[31:0] := MAX_FLOAT - 15: dest[31:0] := -MAX_FLOAT - ESAC - - CASE(tsrc[31:0] of TOKEN_TYPE) - ZERO_VALUE_TOKEN: if imm8[0] then set #ZE - ZERO_VALUE_TOKEN: if imm8[1] then set #IE - ONE_VALUE_TOKEN: if imm8[2] then set #ZE - ONE_VALUE_TOKEN: if imm8[3] then set #IE - SNAN_TOKEN: if imm8[4] then set #IE - NEG_INF_TOKEN: if imm8[5] then set #IE - NEG_VALUE_TOKEN: if imm8[6] then set #IE - POS_INF_TOKEN: if imm8[7] then set #IE - ESAC - RETURN dest[31:0] -} - -IF k[0] - dst[31:0] := FIXUPIMMPD(a[31:0], b[31:0], c[31:0], imm8[7:0]) -ELSE - dst[31:0] := a[31:0] -FI -dst[127:32] := a[127:32] -dst[MAX:128] := 0 +k[7:0] := NOT (a[7:0] XOR b[7:0]) +k[MAX:8] := 0 - +
immintrin.h
 - - Floating Point - AVX512F - Miscellaneous - - - - - - - Fix up the lower single-precision (32-bit) floating-point elements in "a" and "b" using the lower 32-bit integer in "c", store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". "imm8" is used to set the required flags reporting. - [round_note] - - -enum TOKEN_TYPE { - QNAN_TOKEN := 0, - SNAN_TOKEN L= 1, - ZERO_VALUE_TOKEN := 2, - ONE_VALUE_TOKEN := 3, - NEG_INF_TOKEN := 4, - POS_INF_TOKEN := 5, - NEG_VALUE_TOKEN := 6, - POS_VALUE_TOKEN := 7 -} -FIXUPIMMPD(src1[31:0], src2[31:0], src3[31:0], imm8[7:0]){ - tsrc[31:0] := ((src2[30:23] == 0) AND (MXCSR.DAZ == 1)) ? 0.0 : src2[31:0] - CASE(tsrc[31:0] of TOKEN_TYPE) - QNAN_TOKEN:j := 0 - SNAN_TOKEN:j := 1 - ZERO_VALUE_TOKEN: j := 2 - ONE_VALUE_TOKEN: j := 3 - NEG_INF_TOKEN: j := 4 - POS_INF_TOKEN: j := 5 - NEG_VALUE_TOKEN: j := 6 - POS_VALUE_TOKEN: j := 7 - ESAC - - token_response[3:0] := src3[3+4*j:4*j] - - CASE(token_response[3:0]) of - 0 : dest[31:0] := src1[31:0] - 1 : dest[31:0] := tsrc[31:0] - 2 : dest[31:0] := QNaN(tsrc[31:0]) - 3 : dest[31:0] := QNAN_Indefinite - 4 : dest[31:0] := -INF - 5 : dest[31:0] := +INF - 6 : dest[31:0] := tsrc.sign? –INF : +INF - 7 : dest[31:0] := -0 - 8 : dest[31:0] := +0 - 9 : dest[31:0] := -1 - 10: dest[31:0] := +1 - 11: dest[31:0] := 1⁄2 - 12: dest[31:0] := 90.0 - 13: dest[31:0] := PI/2 - 14: dest[31:0] := MAX_FLOAT - 15: dest[31:0] := -MAX_FLOAT - ESAC - - CASE(tsrc[31:0] of TOKEN_TYPE) - ZERO_VALUE_TOKEN: if imm8[0] then set #ZE - ZERO_VALUE_TOKEN: if imm8[1] then set #IE - ONE_VALUE_TOKEN: if imm8[2] then set #ZE - ONE_VALUE_TOKEN: if imm8[3] then set #IE - SNAN_TOKEN: if imm8[4] then set #IE - NEG_INF_TOKEN: if imm8[5] then set #IE - NEG_VALUE_TOKEN: if imm8[6] then set #IE - POS_INF_TOKEN: if imm8[7] then set #IE - ESAC - RETURN dest[31:0] -} - -IF k[0] - dst[31:0] := FIXUPIMMPD(a[31:0], b[31:0], c[31:0], imm8[7:0]) -ELSE - dst[31:0] := 0 + + Mask + AVX512DQ + Mask + + + + Compute the bitwise XOR of 8-bit masks "a" and "b", and store the result in "k". + +k[7:0] := a[7:0] XOR b[7:0] +k[MAX:8] := 0 + + +
immintrin.h
+ + + Mask + AVX512DQ + Mask + + + + Shift the bits of 8-bit mask "a" left by "count" while shifting in zeros, and store the least significant 8 bits of the result in "k". + +k[MAX:0] := 0 +IF count[7:0] <= 7 + k[7:0] := a[7:0] << count[7:0] FI -dst[127:32] := a[127:32] -dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point - AVX512F - Miscellaneous - - - - - - Fix up the lower single-precision (32-bit) floating-point elements in "a" and "b" using the lower 32-bit integer in "c", store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". "imm8" is used to set the required flags reporting. - -enum TOKEN_TYPE { - QNAN_TOKEN := 0, - SNAN_TOKEN L= 1, - ZERO_VALUE_TOKEN := 2, - ONE_VALUE_TOKEN := 3, - NEG_INF_TOKEN := 4, - POS_INF_TOKEN := 5, - NEG_VALUE_TOKEN := 6, - POS_VALUE_TOKEN := 7 -} -FIXUPIMMPD(src1[31:0], src2[31:0], src3[31:0], imm8[7:0]){ - tsrc[31:0] := ((src2[30:23] == 0) AND (MXCSR.DAZ == 1)) ? 0.0 : src2[31:0] - CASE(tsrc[31:0] of TOKEN_TYPE) - QNAN_TOKEN:j := 0 - SNAN_TOKEN:j := 1 - ZERO_VALUE_TOKEN: j := 2 - ONE_VALUE_TOKEN: j := 3 - NEG_INF_TOKEN: j := 4 - POS_INF_TOKEN: j := 5 - NEG_VALUE_TOKEN: j := 6 - POS_VALUE_TOKEN: j := 7 - ESAC - - token_response[3:0] := src3[3+4*j:4*j] - - CASE(token_response[3:0]) of - 0 : dest[31:0] := src1[31:0] - 1 : dest[31:0] := tsrc[31:0] - 2 : dest[31:0] := QNaN(tsrc[31:0]) - 3 : dest[31:0] := QNAN_Indefinite - 4 : dest[31:0] := -INF - 5 : dest[31:0] := +INF - 6 : dest[31:0] := tsrc.sign? –INF : +INF - 7 : dest[31:0] := -0 - 8 : dest[31:0] := +0 - 9 : dest[31:0] := -1 - 10: dest[31:0] := +1 - 11: dest[31:0] := 1⁄2 - 12: dest[31:0] := 90.0 - 13: dest[31:0] := PI/2 - 14: dest[31:0] := MAX_FLOAT - 15: dest[31:0] := -MAX_FLOAT - ESAC - - CASE(tsrc[31:0] of TOKEN_TYPE) - ZERO_VALUE_TOKEN: if imm8[0] then set #ZE - ZERO_VALUE_TOKEN: if imm8[1] then set #IE - ONE_VALUE_TOKEN: if imm8[2] then set #ZE - ONE_VALUE_TOKEN: if imm8[3] then set #IE - SNAN_TOKEN: if imm8[4] then set #IE - NEG_INF_TOKEN: if imm8[5] then set #IE - NEG_VALUE_TOKEN: if imm8[6] then set #IE - POS_INF_TOKEN: if imm8[7] then set #IE - ESAC - RETURN dest[31:0] -} - -IF k[0] - dst[31:0] := FIXUPIMMPD(a[31:0], b[31:0], c[31:0], imm8[7:0]) -ELSE - dst[31:0] := 0 + + Mask + AVX512DQ + Mask + + + + Shift the bits of 8-bit mask "a" right by "count" while shifting in zeros, and store the least significant 8 bits of the result in "k". + +k[MAX:0] := 0 +IF count[7:0] <= 7 + k[7:0] := a[7:0] >> count[7:0] FI -dst[127:32] := a[127:32] -dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point - AVX512F/KNCNI - Arithmetic - - - - Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", add the intermediate result to packed elements in "c", and store the results in "dst". - + + Mask + AVX512DQ + Load + + + Load 8-bit mask from memory into "k". -FOR j := 0 to 7 - i := j*64 - dst[i+63:i] := (a[i+63:i] * b[i+63:i]) + c[i+63:i] -ENDFOR -dst[MAX:512] := 0 +k[7:0] := MEM[mem_addr+7:mem_addr] - - - +
immintrin.h
 - - Floating Point - AVX512F/KNCNI - Arithmetic - - - - - Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", add the intermediate result to packed elements in "c", and store the results in "dst". - [round_note] - + + Mask + AVX512DQ + Store + + + + Store 8-bit mask from "a" into memory. -FOR j := 0 to 7 - i := j*64 - dst[i+63:i] := (a[i+63:i] * b[i+63:i]) + c[i+63:i] -ENDFOR -dst[MAX:512] := 0 +MEM[mem_addr+7:mem_addr] := a[7:0] - - - +
immintrin.h
 - - Floating Point - AVX512F/KNCNI - Arithmetic - - - - - Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", add the intermediate result to packed elements in "c", and store the results in "dst" using writemask "k" (elements are copied from "c" when the corresponding mask bit is not set). - - -FOR j := 0 to 7 - i := j*64 - IF k[j] - dst[i+63:i] := (a[i+63:i] * b[i+63:i]) + c[i+63:i] - ELSE - dst[i+63:i] := c[i+63:i] - FI -ENDFOR -dst[MAX:512] := 0 + + Mask + AVX512DQ + Mask + + + + + Compute the bitwise OR of 8-bit masks "a" and "b". If the result is all zeros, store 1 in "dst", otherwise store 0 in "dst". If the result is all ones, store 1 in "all_ones", otherwise store 0 in "all_ones". + +tmp[7:0] := a[7:0] OR b[7:0] +IF tmp[7:0] == 0x0 + dst := 1 +ELSE + dst := 0 +FI +IF tmp[7:0] == 0xFF + MEM[all_ones+7:all_ones] := 1 +ELSE + MEM[all_ones+7:all_ones] := 0 +FI - - - +
immintrin.h
 - - Floating Point - AVX512F/KNCNI - Arithmetic - - - - - - Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", add the intermediate result to packed elements in "c", and store the results in "dst" using writemask "k" (elements are copied from "c" when the corresponding mask bit is not set). - [round_note] - + + Mask + AVX512DQ + Mask + + + + Compute the bitwise OR of 8-bit masks "a" and "b". If the result is all zeroes, store 1 in "dst", otherwise store 0 in "dst". -FOR j := 0 to 7 - i := j*64 - IF k[j] - dst[i+63:i] := (a[i+63:i] * b[i+63:i]) + c[i+63:i] - ELSE - dst[i+63:i] := c[i+63:i] - FI -ENDFOR -dst[MAX:512] := 0 +tmp[7:0] := a[7:0] OR b[7:0] +IF tmp[7:0] == 0x0 + dst := 1 +ELSE + dst := 0 +FI - - - +
immintrin.h
 - - Floating Point - AVX512F/KNCNI - Arithmetic - - - - - Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", add the intermediate result to packed elements in "c", and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). - + + Mask + AVX512DQ + Mask + + + + Compute the bitwise OR of 8-bit masks "a" and "b". If the result is all ones, store 1 in "dst", otherwise store 0 in "dst". -FOR j := 0 to 7 - i := j*64 - IF k[j] - dst[i+63:i] := (a[i+63:i] * b[i+63:i]) + c[i+63:i] - ELSE - dst[i+63:i] := a[i+63:i] - FI -ENDFOR -dst[MAX:512] := 0 +tmp[7:0] := a[7:0] OR b[7:0] +IF tmp[7:0] == 0xFF + dst := 1 +ELSE + dst := 0 +FI - - - +
immintrin.h
 - - Floating Point - AVX512F/KNCNI - Arithmetic - - - - - - Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", add the intermediate result to packed elements in "c", and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). - [round_note] - -FOR j := 0 to 7 - i := j*64 - IF k[j] - dst[i+63:i] := (a[i+63:i] * b[i+63:i]) + c[i+63:i] - ELSE - dst[i+63:i] := a[i+63:i] - FI -ENDFOR -dst[MAX:512] := 0 + + Mask + AVX512DQ + Mask + + + + + Compute the bitwise AND of 8-bit masks "a" and "b", and if the result is all zeros, store 1 in "dst", otherwise store 0 in "dst". Compute the bitwise NOT of "a" and then AND with "b", if the result is all zeros, store 1 in "and_not", otherwise store 0 in "and_not". + +tmp1[7:0] := a[7:0] AND b[7:0] +IF tmp1[7:0] == 0x0 + dst := 1 +ELSE + dst := 0 +FI +tmp2[7:0] := (NOT a[7:0]) AND b[7:0] +IF tmp2[7:0] == 0x0 + MEM[and_not+7:and_not] := 1 +ELSE + MEM[and_not+7:and_not] := 0 +FI - - - +
immintrin.h
 - - Floating Point - AVX512F - Arithmetic - - - - - Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", add the intermediate result to packed elements in "c", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - + + Mask + AVX512DQ + Mask + + + + Compute the bitwise AND of 8-bit masks "a" and "b", and if the result is all zeros, store 1 in "dst", otherwise store 0 in "dst". -FOR j := 0 to 7 - i := j*64 - IF k[j] - dst[i+63:i] := (a[i+63:i] * b[i+63:i]) + c[i+63:i] - ELSE - dst[i+63:i] := 0 - FI -ENDFOR -dst[MAX:512] := 0 +tmp[7:0] := a[7:0] AND b[7:0] +IF tmp[7:0] == 0x0 + dst := 1 +ELSE + dst := 0 +FI - - - +
immintrin.h
 - - Floating Point - AVX512F - Arithmetic - - - - - - Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", add the intermediate result to packed elements in "c", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - [round_note] + + Mask + AVX512DQ + Mask + + + + Compute the bitwise NOT of 8-bit mask "a" and then AND with "b", if the result is all zeroes, store 1 in "dst", otherwise store 0 in "dst". -FOR j := 0 to 7 - i := j*64 - IF k[j] - dst[i+63:i] := (a[i+63:i] * b[i+63:i]) + c[i+63:i] - ELSE - dst[i+63:i] := 0 - FI -ENDFOR -dst[MAX:512] := 0 +tmp[7:0] := (NOT a[7:0]) AND b[7:0] +IF tmp[7:0] == 0x0 + dst := 1 +ELSE + dst := 0 +FI + + +
immintrin.h
+ + + Mask + AVX512DQ + Mask + + + + + Compute the bitwise AND of 16-bit masks "a" and "b", and if the result is all zeros, store 1 in "dst", otherwise store 0 in "dst". Compute the bitwise NOT of "a" and then AND with "b", if the result is all zeros, store 1 in "and_not", otherwise store 0 in "and_not". + +tmp1[15:0] := a[15:0] AND b[15:0] +IF tmp1[15:0] == 0x0 + dst := 1 +ELSE + dst := 0 +FI +tmp2[15:0] := (NOT a[15:0]) AND b[15:0] +IF tmp2[15:0] == 0x0 + MEM[and_not+7:and_not] := 1 +ELSE + MEM[and_not+7:and_not] := 0 +FI + + +
immintrin.h
+ + + Mask + AVX512DQ + Mask + + + + Compute the bitwise AND of 16-bit masks "a" and "b", and if the result is all zeros, store 1 in "dst", otherwise store 0 in "dst". + +tmp[15:0] := a[15:0] AND b[15:0] +IF tmp[15:0] == 0x0 + dst := 1 +ELSE + dst := 0 +FI + + +
immintrin.h
+ + + Mask + AVX512DQ + Mask + + + + Compute the bitwise NOT of 16-bit mask "a" and then AND with "b", if the result is all zeroes, store 1 in "dst", otherwise store 0 in "dst". + +tmp[15:0] := (NOT a[15:0]) AND b[15:0] +IF tmp[15:0] == 0x0 + dst := 1 +ELSE + dst := 0 +FI + + +
immintrin.h
+ + + AVX512DQ + Mask + + + Convert 8-bit mask "a" into an integer value, and store the result in "dst". + +dst := ZeroExtend32(a[7:0]) + + +
immintrin.h
+ + + AVX512DQ + Mask + + + Convert integer value "a" into an 8-bit mask, and store the result in "k". + +k := a[7:0] - - - +
immintrin.h
 - + Floating Point - AVX512F/KNCNI - Arithmetic - - - - Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", add the intermediate result to packed elements in "c", and store the results in "dst". - + AVX512ER + Elementary Math Functions + + + + Compute the approximate exponential value of 2 raised to the power of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst". The maximum relative error for this approximation is less than 2^-23. [sae_note] FOR j := 0 to 15 i := j*32 - dst[i+31:i] := (a[i+31:i] * b[i+31:i]) + c[i+31:i] + dst[i+31:i] := POW(FP32(2.0), a[i+31:i]) ENDFOR dst[MAX:512] := 0 - - - +
immintrin.h
 - + Floating Point - AVX512F/KNCNI - Arithmetic - - - - - Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", add the intermediate result to packed elements in "c", and store the results in "dst". - [round_note] + AVX512ER + Elementary Math Functions + + + Compute the approximate exponential value of 2 raised to the power of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst". The maximum relative error for this approximation is less than 2^-23. FOR j := 0 to 15 i := j*32 - dst[i+31:i] := (a[i+31:i] * b[i+31:i]) + c[i+31:i] + dst[i+31:i] := POW(FP32(2.0), a[i+31:i]) ENDFOR dst[MAX:512] := 0 - - - +
immintrin.h
 - + Floating Point - AVX512F/KNCNI - Arithmetic - - - - - Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", add the intermediate result to packed elements in "c", and store the results in "dst" using writemask "k" (elements are copied from "c" when the corresponding mask bit is not set). - + AVX512ER + Elementary Math Functions + + + + + + Compute the approximate exponential value of 2 raised to the power of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). The maximum relative error for this approximation is less than 2^-23. [sae_note] FOR j := 0 to 15 i := j*32 IF k[j] - dst[i+31:i] := (a[i+31:i] * b[i+31:i]) + c[i+31:i] + dst[i+31:i] := POW(FP32(2.0), a[i+31:i]) ELSE - dst[i+31:i] := c[i+31:i] + dst[i+31:i] := src[i+31:i] FI ENDFOR dst[MAX:512] := 0 - - - +
immintrin.h
 - + Floating Point - AVX512F/KNCNI - Arithmetic - - - - - - Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", add the intermediate result to packed elements in "c", and store the results in "dst" using writemask "k" (elements are copied from "c" when the corresponding mask bit is not set). - [round_note] + AVX512ER + Elementary Math Functions + + + + + Compute the approximate exponential value of 2 raised to the power of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). The maximum relative error for this approximation is less than 2^-23. FOR j := 0 to 15 i := j*32 IF k[j] - dst[i+31:i] := (a[i+31:i] * b[i+31:i]) + c[i+31:i] + dst[i+31:i] := POW(FP32(2.0), a[i+31:i]) ELSE - dst[i+31:i] := c[i+31:i] + dst[i+31:i] := src[i+31:i] FI ENDFOR dst[MAX:512] := 0 - - - +
immintrin.h
 - + Floating Point - AVX512F/KNCNI - Arithmetic - - - - - Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", add the intermediate result to packed elements in "c", and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). - + AVX512ER + Elementary Math Functions + + + + + Compute the approximate exponential value of 2 raised to the power of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). The maximum relative error for this approximation is less than 2^-23. [sae_note] FOR j := 0 to 15 i := j*32 IF k[j] - dst[i+31:i] := (a[i+31:i] * b[i+31:i]) + c[i+31:i] + dst[i+31:i] := POW(FP32(2.0), a[i+31:i]) ELSE - dst[i+31:i] := a[i+31:i] + dst[i+31:i] := 0 FI ENDFOR dst[MAX:512] := 0 - - - +
immintrin.h
 - + Floating Point - AVX512F/KNCNI - Arithmetic - - - - - - Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", add the intermediate result to packed elements in "c", and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). - [round_note] + AVX512ER + Elementary Math Functions + + + + Compute the approximate exponential value of 2 raised to the power of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). The maximum relative error for this approximation is less than 2^-23. FOR j := 0 to 15 i := j*32 IF k[j] - dst[i+31:i] := (a[i+31:i] * b[i+31:i]) + c[i+31:i] + dst[i+31:i] := POW(FP32(2.0), a[i+31:i]) ELSE - dst[i+31:i] := a[i+31:i] + dst[i+31:i] := 0 FI ENDFOR dst[MAX:512] := 0 - - - +
immintrin.h
 - + Floating Point - AVX512F - Arithmetic - - - - - Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", add the intermediate result to packed elements in "c", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - + AVX512ER + Elementary Math Functions + + + + Compute the approximate exponential value of 2 raised to the power of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst". The maximum relative error for this approximation is less than 2^-23. [sae_note] -FOR j := 0 to 15 - i := j*32 +FOR j := 0 to 7 + i := j*64 + dst[i+63:i] := POW(2.0, a[i+63:i]) +ENDFOR +dst[MAX:512] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512ER + Elementary Math Functions + + + Compute the approximate exponential value of 2 raised to the power of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst". The maximum relative error for this approximation is less than 2^-23. + +FOR j := 0 to 7 + i := j*64 + dst[i+63:i] := POW(2.0, a[i+63:i]) +ENDFOR +dst[MAX:512] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512ER + Elementary Math Functions + + + + + + Compute the approximate exponential value of 2 raised to the power of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). The maximum relative error for this approximation is less than 2^-23. [sae_note] + +FOR j := 0 to 7 + i := j*64 IF k[j] - dst[i+31:i] := (a[i+31:i] * b[i+31:i]) + c[i+31:i] + dst[i+63:i] := POW(2.0, a[i+63:i]) ELSE - dst[i+31:i] := 0 + dst[i+63:i] := src[i+63:i] FI ENDFOR dst[MAX:512] := 0 - - - +
immintrin.h
 - + Floating Point - AVX512F - Arithmetic - - - - - - Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", add the intermediate result to packed elements in "c", and store the results in "a" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - [round_note] + AVX512ER + Elementary Math Functions + + + + + Compute the approximate exponential value of 2 raised to the power of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). The maximum relative error for this approximation is less than 2^-23. -FOR j := 0 to 15 - i := j*32 +FOR j := 0 to 7 + i := j*64 IF k[j] - a[i+31:i] := (a[i+31:i] * b[i+31:i]) + c[i+31:i] + dst[i+63:i] := POW(2.0, a[i+63:i]) ELSE - dst[i+31:i] := 0 + dst[i+63:i] := src[i+63:i] FI ENDFOR dst[MAX:512] := 0 - - - +
immintrin.h
 - + Floating Point - AVX512F - Arithmetic - - - - - Multiply the lower double-precision (64-bit) floating-point elements in "a" and "b", and add the intermediate result to the lower element in "c". Store the result in the lower element of "dst", and copy the upper element from "a" to the upper element of "dst". - [round_note] + AVX512ER + Elementary Math Functions + + + + + Compute the approximate exponential value of 2 raised to the power of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). The maximum relative error for this approximation is less than 2^-23. [sae_note] -dst[63:0] := (a[63:0] * b[63:0]) + c[63:0] -dst[127:64] := a[127:64] -dst[MAX:128] := 0 +FOR j := 0 to 7 + i := j*64 + IF k[j] + dst[i+63:i] := POW(2.0, a[i+63:i]) + ELSE + dst[i+63:i] := 0 + FI +ENDFOR +dst[MAX:512] := 0 - - - +
immintrin.h
 - + Floating Point - AVX512F - Arithmetic - - - - - - Multiply the lower double-precision (64-bit) floating-point elements in "a" and "b", and add the intermediate result to the lower element in "c". Store the result in the lower element of "dst" using writemask "k" (the element is copied from "c" when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". - [round_note] + AVX512ER + Elementary Math Functions + + + + Compute the approximate exponential value of 2 raised to the power of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). The maximum relative error for this approximation is less than 2^-23. -IF k[0] - dst[63:0] := (a[63:0] * b[63:0]) + c[63:0] -ELSE - dst[63:0] := c[63:0] -FI -dst[127:64] := a[127:64] -dst[MAX:128] := 0 +FOR j := 0 to 7 + i := j*64 + IF k[j] + dst[i+63:i] := POW(2.0, a[i+63:i]) + ELSE + dst[i+63:i] := 0 + FI +ENDFOR +dst[MAX:512] := 0 - - - +
immintrin.h
 - + Floating Point - AVX512F - Arithmetic - - - - - Multiply the lower double-precision (64-bit) floating-point elements in "a" and "b", and add the intermediate result to the lower element in "c". Store the result in the lower element of "dst" using writemask "k" (the element is copied from "c" when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". - + AVX512ER + Elementary Math Functions + + + + + Compute the approximate reciprocal of the lower double-precision (64-bit) floating-point element in "b", store the result in the lower element of "dst", and copy the upper element from "a" to the upper element of "dst". The maximum relative error for this approximation is less than 2^-28. [sae_note] -IF k[0] - dst[63:0] := (a[63:0] * b[63:0]) + c[63:0] -ELSE - dst[63:0] := c[63:0] -FI +dst[63:0] := (1.0 / b[63:0]) dst[127:64] := a[127:64] dst[MAX:128] := 0 - - - +
immintrin.h
 - + Floating Point - AVX512F - Arithmetic - - - - - - Multiply the lower double-precision (64-bit) floating-point elements in "a" and "b", and add the intermediate result to the lower element in "c". Store the result in the lower element of "dst" using writemask "k" (the element is copied from "a" when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". - [round_note] + AVX512ER + Elementary Math Functions + + + + Compute the approximate reciprocal of the lower double-precision (64-bit) floating-point element in "b", store the result in the lower element of "dst", and copy the upper element from "a" to the upper element of "dst". The maximum relative error for this approximation is less than 2^-28. -IF k[0] - dst[63:0] := (a[63:0] * b[63:0]) + c[63:0] -ELSE - dst[63:0] := a[63:0] -FI +dst[63:0] := (1.0 / b[63:0]) dst[127:64] := a[127:64] dst[MAX:128] := 0 - - - +
immintrin.h
 - + Floating Point - AVX512F - Arithmetic - - - - - Multiply the lower double-precision (64-bit) floating-point elements in "a" and "b", and add the intermediate result to the lower element in "c". Store the result in the lower element of "dst" using writemask "k" (the element is copied from "a" when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". - + AVX512ER + Elementary Math Functions + + + + + + + Compute the approximate reciprocal of the lower double-precision (64-bit) floating-point element in "b", store the result in the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". The maximum relative error for this approximation is less than 2^-28. [sae_note] IF k[0] - dst[63:0] := (a[63:0] * b[63:0]) + c[63:0] + dst[63:0] := (1.0 / b[63:0]) ELSE - dst[63:0] := a[63:0] + dst[63:0] := src[63:0] FI dst[127:64] := a[127:64] dst[MAX:128] := 0 - - - +
immintrin.h
 - + Floating Point - AVX512F - Arithmetic - - - - - - Multiply the lower double-precision (64-bit) floating-point elements in "a" and "b", and add the intermediate result to the lower element in "c". Store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". - [round_note] + AVX512ER + Elementary Math Functions + + + + + + Compute the approximate reciprocal of the lower double-precision (64-bit) floating-point element in "b", store the result in the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". The maximum relative error for this approximation is less than 2^-28. IF k[0] - dst[63:0] := (a[63:0] * b[63:0]) + c[63:0] + dst[63:0] := (1.0 / b[63:0]) ELSE - dst[63:0] := 0 + dst[63:0] := src[63:0] FI dst[127:64] := a[127:64] dst[MAX:128] := 0 - - - +
immintrin.h
 - + Floating Point - AVX512F - Arithmetic - - - - - Multiply the lower double-precision (64-bit) floating-point elements in "a" and "b", and add the intermediate result to the lower element in "c". Store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". - + AVX512ER + Elementary Math Functions + + + + + + Compute the approximate reciprocal of the lower double-precision (64-bit) floating-point element in "b", store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". The maximum relative error for this approximation is less than 2^-28. [sae_note] IF k[0] - dst[63:0] := (a[63:0] * b[63:0]) + c[63:0] + dst[63:0] := (1.0 / b[63:0]) ELSE dst[63:0] := 0 FI dst[127:64] := a[127:64] dst[MAX:128] := 0 - - - +
immintrin.h
 - + Floating Point - AVX512F - Arithmetic - - - - - - Multiply the lower single-precision (32-bit) floating-point elements in "a" and "b", and add the intermediate result to the lower element in "c". Store the result in the lower element of "dst" using writemask "k" (the element is copied from "c" when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". - [round_note] + AVX512ER + Elementary Math Functions + + + + + Compute the approximate reciprocal of the lower double-precision (64-bit) floating-point element in "b", store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". The maximum relative error for this approximation is less than 2^-28. IF k[0] - dst[31:0] := (a[31:0] * b[31:0]) + c[31:0] + dst[63:0] := (1.0 / b[63:0]) ELSE - dst[31:0] := c[31:0] + dst[63:0] := 0 FI -dst[127:32] := a[127:32] +dst[127:64] := a[127:64] dst[MAX:128] := 0 - - - +
immintrin.h
 - + Floating Point - AVX512F - Arithmetic - - - - - Multiply the lower single-precision (32-bit) floating-point elements in "a" and "b", and add the intermediate result to the lower element in "c". Store the result in the lower element of "dst" using writemask "k" (the element is copied from "c" when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". - + AVX512ER + Elementary Math Functions + + + + + Compute the approximate reciprocal of the lower single-precision (32-bit) floating-point element in "b", store the result in the lower element of "dst". The maximum relative error for this approximation is less than 2^-28, and copy the upper 3 packed elements from "a" to the upper elements of "dst". [sae_note] -IF k[0] - dst[31:0] := (a[31:0] * b[31:0]) + c[31:0] -ELSE - dst[31:0] := c[31:0] -FI +dst[31:0] := (1.0 / b[31:0]) dst[127:32] := a[127:32] dst[MAX:128] := 0 - - - +
immintrin.h
 - + Floating Point - AVX512F - Arithmetic - - - - - Multiply the lower single-precision (32-bit) floating-point elements in "a" and "b", and add the intermediate result to the lower element in "c". Store the result in the lower element of "dst", and copy the upper 3 packed elements from "a" to the upper elements of "dst". - [round_note] + AVX512ER + Elementary Math Functions + + + + Compute the approximate reciprocal of the lower single-precision (32-bit) floating-point element in "b", store the result in the lower element of "dst", and copy the upper 3 packed elements from "a" to the upper elements of "dst". The maximum relative error for this approximation is less than 2^-28. -dst[31:0] := (a[31:0] * b[31:0]) + c[31:0] +dst[31:0] := (1.0 / b[31:0]) dst[127:32] := a[127:32] dst[MAX:128] := 0 - - - +
immintrin.h
 - + Floating Point - AVX512F - Arithmetic - - - - - - Multiply the lower single-precision (32-bit) floating-point elements in "a" and "b", and add the intermediate result to the lower element in "c". Store the result in the lower element of "dst" using writemask "k" (the element is copied from "a" when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". - [round_note] + AVX512ER + Elementary Math Functions + + + + + + + Compute the approximate reciprocal of the lower single-precision (32-bit) floating-point element in "b", store the result in the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". The maximum relative error for this approximation is less than 2^-28. [sae_note] IF k[0] - dst[31:0] := (a[31:0] * b[31:0]) + c[31:0] + dst[31:0] := (1.0 / b[31:0]) ELSE - dst[31:0] := a[31:0] + dst[31:0] := src[31:0] FI dst[127:32] := a[127:32] dst[MAX:128] := 0 - - - +
immintrin.h
 - + Floating Point - AVX512F - Arithmetic - - - - - Multiply the lower single-precision (32-bit) floating-point elements in "a" and "b", and add the intermediate result to the lower element in "c". Store the result in the lower element of "dst" using writemask "k" (the element is copied from "a" when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". - + AVX512ER + Elementary Math Functions + + + + + + Compute the approximate reciprocal of the lower single-precision (32-bit) floating-point element in "b", store the result in the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". The maximum relative error for this approximation is less than 2^-28. IF k[0] - dst[31:0] := (a[31:0] * b[31:0]) + c[31:0] + dst[31:0] := (1.0 / b[31:0]) ELSE - dst[31:0] := a[31:0] + dst[31:0] := src[31:0] FI dst[127:32] := a[127:32] dst[MAX:128] := 0 - - - +
immintrin.h
 - + Floating Point - AVX512F - Arithmetic - - - - - - Multiply the lower single-precision (32-bit) floating-point elements in "a" and "b", and add the intermediate result to the lower element in "c". Store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". - [round_note] + AVX512ER + Elementary Math Functions + + + + + + Compute the approximate reciprocal of the lower single-precision (32-bit) floating-point element in "b", store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". The maximum relative error for this approximation is less than 2^-28. [sae_note] IF k[0] - dst[31:0] := (a[31:0] * b[31:0]) + c[31:0] + dst[31:0] := (1.0 / b[31:0]) ELSE dst[31:0] := 0 FI dst[127:32] := a[127:32] dst[MAX:128] := 0 - - - +
immintrin.h
 - + Floating Point - AVX512F - Arithmetic - - - - - Multiply the lower single-precision (32-bit) floating-point elements in "a" and "b", and add the intermediate result to the lower element in "c". Store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". - + AVX512ER + Elementary Math Functions + + + + + Compute the approximate reciprocal of the lower single-precision (32-bit) floating-point element in "b", store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". The maximum relative error for this approximation is less than 2^-28. IF k[0] - dst[31:0] := (a[31:0] * b[31:0]) + c[31:0] + dst[31:0] := (1.0 / b[31:0]) ELSE dst[31:0] := 0 FI dst[127:32] := a[127:32] dst[MAX:128] := 0 - - - +
immintrin.h
 - + Floating Point - AVX512F - Arithmetic - - - - Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", alternatively add and subtract packed elements in "c" to/from the intermediate result, and store the results in "dst". - + AVX512ER + Elementary Math Functions + + + + Compute the approximate reciprocal of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst". The maximum relative error for this approximation is less than 2^-28. [sae_note] -FOR j := 0 to 7 - i := j*64 - IF (j is even) - dst[i+63:i] := (a[i+63:i] * b[i+63:i]) - c[i+63:i] - ELSE - dst[i+63:i] := (a[i+63:i] * b[i+63:i]) + c[i+63:i] - FI +FOR j := 0 to 15 + i := j*32 + dst[i+31:i] := (1.0 / a[i+31:i]) ENDFOR -dst[MAX:512] := 0 - - - +
immintrin.h
 - + Floating Point - AVX512F - Arithmetic - - - - - Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", alternatively add and subtract packed elements in "c" to/from the intermediate result, and store the results in "dst". - [round_note] + AVX512ER + Elementary Math Functions + + + Compute the approximate reciprocal of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst". The maximum relative error for this approximation is less than 2^-28. -FOR j := 0 to 7 - i := j*64 - IF (j is even) - dst[i+63:i] := (a[i+63:i] * b[i+63:i]) - c[i+63:i] - ELSE - dst[i+63:i] := (a[i+63:i] * b[i+63:i]) + c[i+63:i] - FI +FOR j := 0 to 15 + i := j*32 + dst[i+31:i] := (1.0 / a[i+31:i]) ENDFOR -dst[MAX:512] := 0 - - - +
immintrin.h
 - + Floating Point - AVX512F - Arithmetic - - - - - Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", alternatively add and subtract packed elements in "c" to/from the intermediate result, and store the results in "dst" using writemask "k" (elements are copied from "c" when the corresponding mask bit is not set). + AVX512ER + Elementary Math Functions + + + + + + Compute the approximate reciprocal of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). The maximum relative error for this approximation is less than 2^-28. [sae_note] -FOR j := 0 to 7 - i := j*64 +FOR j := 0 to 15 + i := j*32 IF k[j] - IF (j is even) - dst[i+63:i] := (a[i+63:i] * b[i+63:i]) - c[i+63:i] - ELSE - dst[i+63:i] := (a[i+63:i] * b[i+63:i]) + c[i+63:i] - FI + dst[i+31:i] := (1.0 / a[i+31:i]) ELSE - dst[i+63:i] := c[i+63:i] + dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:512] := 0 - - - +
immintrin.h
 - + Floating Point - AVX512F - Arithmetic - - - - - - Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", alternatively add and subtract packed elements in "c" to/from the intermediate result, and store the results in "dst" using writemask "k" (elements are copied from "c" when the corresponding mask bit is not set). [round_note] + AVX512ER + Elementary Math Functions + + + + + Compute the approximate reciprocal of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). The maximum relative error for this approximation is less than 2^-28. -FOR j := 0 to 7 - i := j*64 +FOR j := 0 to 15 + i := j*32 IF k[j] - IF (j is even) - dst[i+63:i] := (a[i+63:i] * b[i+63:i]) - c[i+63:i] - ELSE - dst[i+63:i] := (a[i+63:i] * b[i+63:i]) + c[i+63:i] - FI - ELSE - dst[i+63:i] := c[i+63:i] + dst[i+31:i] := (1.0 / a[i+31:i]) + ELSE + dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:512] := 0 - - - +
immintrin.h
 - + Floating Point - AVX512F - Arithmetic - - - - - Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", alternatively add and subtract packed elements in "c" to/from the intermediate result, and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). + AVX512ER + Elementary Math Functions + + + + + Compute the approximate reciprocal of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). The maximum relative error for this approximation is less than 2^-28. [sae_note] -FOR j := 0 to 7 - i := j*64 +FOR j := 0 to 15 + i := j*32 IF k[j] - IF (j is even) - dst[i+63:i] := (a[i+63:i] * b[i+63:i]) - c[i+63:i] - ELSE - dst[i+63:i] := (a[i+63:i] * b[i+63:i]) + c[i+63:i] - FI + dst[i+31:i] := (1.0 / a[i+31:i]) ELSE - dst[i+63:i] := a[i+63:i] + dst[i+31:i] := 0 FI ENDFOR -dst[MAX:512] := 0 - - - +
immintrin.h
 - + Floating Point - AVX512F - Arithmetic - - - - - - Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", alternatively add and subtract packed elements in "c" to/from the intermediate result, and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). [round_note] + AVX512ER + Elementary Math Functions + + + + Compute the approximate reciprocal of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). The maximum relative error for this approximation is less than 2^-28. -FOR j := 0 to 7 - i := j*64 +FOR j := 0 to 15 + i := j*32 IF k[j] - IF (j is even) - dst[i+63:i] := (a[i+63:i] * b[i+63:i]) - c[i+63:i] - ELSE - dst[i+63:i] := (a[i+63:i] * b[i+63:i]) + c[i+63:i] - FI + dst[i+31:i] := (1.0 / a[i+31:i]) ELSE - dst[i+63:i] := a[i+63:i] + dst[i+31:i] := 0 FI ENDFOR -dst[MAX:512] := 0 - - - +
immintrin.h
 - + Floating Point - AVX512F - Arithmetic - - - - - Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", alternatively add and subtract packed elements in "c" to/from the intermediate result, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - + AVX512ER + Elementary Math Functions + + + + Compute the approximate reciprocal of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst". The maximum relative error for this approximation is less than 2^-28. [sae_note] FOR j := 0 to 7 i := j*64 - IF k[j] - IF (j is even) - dst[i+63:i] := (a[i+63:i] * b[i+63:i]) - c[i+63:i] - ELSE - dst[i+63:i] := (a[i+63:i] * b[i+63:i]) + c[i+63:i] - FI - ELSE - dst[i+63:i] := 0 - FI + dst[i+63:i] := (1.0 / a[i+63:i]) ENDFOR -dst[MAX:512] := 0 - - - +
immintrin.h
 - + Floating Point - AVX512F - Arithmetic - - - - - - Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", alternatively add and subtract packed elements in "c" to/from the intermediate result, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - [round_note] + AVX512ER + Elementary Math Functions + + + Compute the approximate reciprocal of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst". The maximum relative error for this approximation is less than 2^-28. FOR j := 0 to 7 i := j*64 - IF k[j] - IF (j is even) - dst[i+63:i] := (a[i+63:i] * b[i+63:i]) - c[i+63:i] - ELSE - dst[i+63:i] := (a[i+63:i] * b[i+63:i]) + c[i+63:i] - FI - ELSE - dst[i+63:i] := 0 - FI -ENDFOR -dst[MAX:512] := 0 - - - - -
immintrin.h
- - - Floating Point - AVX512F - Arithmetic - - - - Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", alternatively add and subtract packed elements in "c" to/from the intermediate result, and store the results in "dst". - - -FOR j := 0 to 15 - i := j*32 - IF (j is even) - dst[i+31:i] := (a[i+31:i] * b[i+31:i]) - c[i+31:i] - ELSE - dst[i+31:i] := (a[i+31:i] * b[i+31:i]) + c[i+31:i] - FI + dst[i+63:i] := (1.0 / a[i+63:i]) ENDFOR -dst[MAX:512] := 0 - - - +
immintrin.h
 - + Floating Point - AVX512F - Arithmetic - - - - - Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", alternatively add and subtract packed elements in "c" to/from the intermediate result, and store the results in "dst". - [round_note] + AVX512ER + Elementary Math Functions + + + + + + Compute the approximate reciprocal of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). The maximum relative error for this approximation is less than 2^-28. [sae_note] -FOR j := 0 to 15 - i := j*32 - IF (j is even) - dst[i+31:i] := (a[i+31:i] * b[i+31:i]) - c[i+31:i] +FOR j := 0 to 7 + i := j*64 + IF k[j] + dst[i+63:i] := (1.0 / a[i+63:i]) ELSE - dst[i+31:i] := (a[i+31:i] * b[i+31:i]) + c[i+31:i] + dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:512] := 0 - - - +
immintrin.h
 - + Floating Point - AVX512F - Arithmetic - - - - - Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", alternatively add and subtract packed elements in "c" to/from the intermediate result, and store the results in "dst" using writemask "k" (elements are copied from "c" when the corresponding mask bit is not set). + AVX512ER + Elementary Math Functions + + + + + Compute the approximate reciprocal of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). The maximum relative error for this approximation is less than 2^-28. -FOR j := 0 to 15 - i := j*32 +FOR j := 0 to 7 + i := j*64 IF k[j] - IF (j is even) - dst[i+31:i] := (a[i+31:i] * b[i+31:i]) - c[i+31:i] - ELSE - dst[i+31:i] := (a[i+31:i] * b[i+31:i]) + c[i+31:i] - FI + dst[i+63:i] := (1.0 / a[i+63:i]) ELSE - dst[i+31:i] := c[i+31:i] + dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:512] := 0 - - - +
immintrin.h
 - + Floating Point - AVX512F - Arithmetic - - - - - - Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", alternatively add and subtract packed elements in "c" to/from the intermediate result, and store the results in "dst" using writemask "k" (elements are copied from "c" when the corresponding mask bit is not set). [round_note] + AVX512ER + Elementary Math Functions + + + + + Compute the approximate reciprocal of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). The maximum relative error for this approximation is less than 2^-28. [sae_note] -FOR j := 0 to 15 - i := j*32 +FOR j := 0 to 7 + i := j*64 IF k[j] - IF (j is even) - dst[i+31:i] := (a[i+31:i] * b[i+31:i]) - c[i+31:i] - ELSE - dst[i+31:i] := (a[i+31:i] * b[i+31:i]) + c[i+31:i] - FI + dst[i+63:i] := (1.0 / a[i+63:i]) ELSE - dst[i+31:i] := c[i+31:i] + dst[i+63:i] := 0 FI ENDFOR -dst[MAX:512] := 0 - - - +
immintrin.h
 - + Floating Point - AVX512F - Arithmetic - - - - - Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", alternatively add and subtract packed elements in "c" to/from the intermediate result, and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). + AVX512ER + Elementary Math Functions + + + + Compute the approximate reciprocal of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). The maximum relative error for this approximation is less than 2^-28. -FOR j := 0 to 15 - i := j*32 +FOR j := 0 to 7 + i := j*64 IF k[j] - IF (j is even) - dst[i+31:i] := (a[i+31:i] * b[i+31:i]) - c[i+31:i] - ELSE - dst[i+31:i] := (a[i+31:i] * b[i+31:i]) + c[i+31:i] - FI + dst[i+63:i] := (1.0 / a[i+63:i]) ELSE - dst[i+31:i] := a[i+31:i] + dst[i+63:i] := 0 FI ENDFOR -dst[MAX:512] := 0 - - - +
immintrin.h
 - + Floating Point - AVX512F - Arithmetic - - - - - - Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", alternatively add and subtract packed elements in "c" to/from the intermediate result, and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). [round_note] + AVX512ER + Elementary Math Functions + + + + + Compute the approximate reciprocal square root of the lower double-precision (64-bit) floating-point element in "b", store the result in the lower element of "dst", and copy the upper element from "a" to the upper element of "dst". The maximum relative error for this approximation is less than 2^-28. [sae_note] + +dst[63:0] := (1.0 / SQRT(b[63:0])) +dst[127:64] := a[127:64] +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512ER + Elementary Math Functions + + + + Compute the approximate reciprocal square root of the lower double-precision (64-bit) floating-point element in "b", store the result in the lower element of "dst", and copy the upper element from "a" to the upper element of "dst". The maximum relative error for this approximation is less than 2^-28. + +dst[63:0] := (1.0 / SQRT(b[63:0])) +dst[127:64] := a[127:64] +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512ER + Elementary Math Functions + + + + + + + Compute the approximate reciprocal square root of the lower double-precision (64-bit) floating-point element in "b", store the result in the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". The maximum relative error for this approximation is less than 2^-28. [sae_note] + +IF k[0] + dst[63:0] := (1.0 / SQRT(b[63:0])) +ELSE + dst[63:0] := src[63:0] +FI +dst[127:64] := a[127:64] +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512ER + Elementary Math Functions + + + + + + Compute the approximate reciprocal square root of the lower double-precision (64-bit) floating-point element in "b", store the result in the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". The maximum relative error for this approximation is less than 2^-28. + +IF k[0] + dst[63:0] := (1.0 / SQRT(b[63:0])) +ELSE + dst[63:0] := src[63:0] +FI +dst[127:64] := a[127:64] +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512ER + Elementary Math Functions + + + + + + Compute the approximate reciprocal square root of the lower double-precision (64-bit) floating-point element in "b", store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". The maximum relative error for this approximation is less than 2^-28. [sae_note] + +IF k[0] + dst[63:0] := (1.0 / SQRT(b[63:0])) +ELSE + dst[63:0] := 0 +FI +dst[127:64] := a[127:64] +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512ER + Elementary Math Functions + + + + + Compute the approximate reciprocal square root of the lower double-precision (64-bit) floating-point element in "b", store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". The maximum relative error for this approximation is less than 2^-28. + +IF k[0] + dst[63:0] := (1.0 / SQRT(b[63:0])) +ELSE + dst[63:0] := 0 +FI +dst[127:64] := a[127:64] +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512ER + Elementary Math Functions + + + + + Compute the approximate reciprocal square root of the lower single-precision (32-bit) floating-point element in "b", store the result in the lower element of "dst", and copy the upper 3 packed elements from "a" to the upper elements of "dst". The maximum relative error for this approximation is less than 2^-28. [sae_note] + +dst[31:0] := (1.0 / SQRT(b[31:0])) +dst[127:32] := a[127:32] +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512ER + Elementary Math Functions + + + + Compute the approximate reciprocal square root of the lower single-precision (32-bit) floating-point element in "b", store the result in the lower element of "dst", and copy the upper 3 packed elements from "a" to the upper elements of "dst". The maximum relative error for this approximation is less than 2^-28. + +dst[31:0] := (1.0 / SQRT(b[31:0])) +dst[127:32] := a[127:32] +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512ER + Elementary Math Functions + + + + + + + Compute the approximate reciprocal square root of the lower single-precision (32-bit) floating-point element in "b", store the result in the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". The maximum relative error for this approximation is less than 2^-28. [sae_note] + +IF k[0] + dst[31:0] := (1.0 / SQRT(b[31:0])) +ELSE + dst[31:0] := src[31:0] +FI +dst[127:32] := a[127:32] +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512ER + Elementary Math Functions + + + + + + Compute the approximate reciprocal square root of the lower single-precision (32-bit) floating-point element in "b", store the result in the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". The maximum relative error for this approximation is less than 2^-28. + +IF k[0] + dst[31:0] := (1.0 / SQRT(b[31:0])) +ELSE + dst[31:0] := src[31:0] +FI +dst[127:32] := a[127:32] +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512ER + Elementary Math Functions + + + + + + Compute the approximate reciprocal square root of the lower single-precision (32-bit) floating-point element in "b", store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". The maximum relative error for this approximation is less than 2^-28. [sae_note] + +IF k[0] + dst[31:0] := (1.0 / SQRT(b[31:0])) +ELSE + dst[31:0] := 0 +FI +dst[127:32] := a[127:32] +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512ER + Elementary Math Functions + + + + + Compute the approximate reciprocal square root of the lower single-precision (32-bit) floating-point element in "b", store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". The maximum relative error for this approximation is less than 2^-28. + +IF k[0] + dst[31:0] := (1.0 / SQRT(b[31:0])) +ELSE + dst[31:0] := 0 +FI +dst[127:32] := a[127:32] +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512ER + Elementary Math Functions + + + + Compute the approximate reciprocal square root of packed single-precision (32-bit) floating-point elements in "a", store the results in "dst". The maximum relative error for this approximation is less than 2^-28. [sae_note] + +FOR j := 0 to 15 + i := j*32 + dst[i+31:i] := (1.0 / SQRT(a[i+31:i])) +ENDFOR + + +
immintrin.h
+ + + Floating Point + AVX512ER + Elementary Math Functions + + + Compute the approximate reciprocal square root of packed single-precision (32-bit) floating-point elements in "a", store the results in "dst". The maximum relative error for this approximation is less than 2^-28. + +FOR j := 0 to 15 + i := j*32 + dst[i+31:i] := (1.0 / SQRT(a[i+31:i])) +ENDFOR + + +
immintrin.h
+ + + Floating Point + AVX512ER + Elementary Math Functions + + + + + + Compute the approximate reciprocal square root of packed single-precision (32-bit) floating-point elements in "a", store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). The maximum relative error for this approximation is less than 2^-28. [sae_note] FOR j := 0 to 15 i := j*32 IF k[j] - IF (j is even) - dst[i+31:i] := (a[i+31:i] * b[i+31:i]) - c[i+31:i] - ELSE - dst[i+31:i] := (a[i+31:i] * b[i+31:i]) + c[i+31:i] - FI + dst[i+31:i] := (1.0 / SQRT(a[i+31:i])) ELSE - dst[i+31:i] := a[i+31:i] + dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:512] := 0 - - - +
immintrin.h
 - + Floating Point - AVX512F - Arithmetic - - - - - Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", alternatively add and subtract packed elements in "c" to/from the intermediate result, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - + AVX512ER + Elementary Math Functions + + + + + Compute the approximate reciprocal square root of packed single-precision (32-bit) floating-point elements in "a", store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). The maximum relative error for this approximation is less than 2^-28. FOR j := 0 to 15 i := j*32 IF k[j] - IF (j is even) - dst[i+31:i] := (a[i+31:i] * b[i+31:i]) - c[i+31:i] - ELSE - dst[i+31:i] := (a[i+31:i] * b[i+31:i]) + c[i+31:i] - FI + dst[i+31:i] := (1.0 / SQRT(a[i+31:i])) + ELSE + dst[i+31:i] := src[i+31:i] + FI +ENDFOR + + +
immintrin.h
+ + + Floating Point + AVX512ER + Elementary Math Functions + + + + + Compute the approximate reciprocal square root of packed single-precision (32-bit) floating-point elements in "a", store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). The maximum relative error for this approximation is less than 2^-28. [sae_note] + +FOR j := 0 to 15 + i := j*32 + IF k[j] + dst[i+31:i] := (1.0 / SQRT(a[i+31:i])) ELSE dst[i+31:i] := 0 FI ENDFOR -dst[MAX:512] := 0 - - - +
immintrin.h
 - + Floating Point - AVX512F - Arithmetic - - - - - - Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", alternatively add and subtract packed elements in "c" to/from the intermediate result, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - [round_note] + AVX512ER + Elementary Math Functions + + + + Compute the approximate reciprocal square root of packed single-precision (32-bit) floating-point elements in "a", store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). The maximum relative error for this approximation is less than 2^-28. FOR j := 0 to 15 i := j*32 IF k[j] - IF (j is even) - dst[i+31:i] := (a[i+31:i] * b[i+31:i]) - c[i+31:i] - ELSE - dst[i+31:i] := (a[i+31:i] * b[i+31:i]) + c[i+31:i] - FI + dst[i+31:i] := (1.0 / SQRT(a[i+31:i])) ELSE dst[i+31:i] := 0 FI ENDFOR -dst[MAX:512] := 0 - - - +
immintrin.h
 - + Floating Point - AVX512F/KNCNI - Arithmetic - - - - Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", subtract packed elements in "c" from the intermediate result, and store the results in "dst". + AVX512ER + Elementary Math Functions + + + + Compute the approximate reciprocal square root of packed double-precision (64-bit) floating-point elements in "a", store the results in "dst". The maximum relative error for this approximation is less than 2^-28. [sae_note] FOR j := 0 to 7 i := j*64 - dst[i+63:i] := (a[i+63:i] * b[i+63:i]) - c[i+63:i] + dst[i+63:i] := (1.0 / SQRT(a[i+63:i])) ENDFOR -dst[MAX:512] := 0 - - - +
immintrin.h
 - + Floating Point - AVX512F/KNCNI - Arithmetic - - - - - Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", subtract packed elements in "c" from the intermediate result, and store the results in "dst". - [round_note] + AVX512ER + Elementary Math Functions + + + Compute the approximate reciprocal square root of packed double-precision (64-bit) floating-point elements in "a", store the results in "dst". The maximum relative error for this approximation is less than 2^-28. FOR j := 0 to 7 i := j*64 - dst[i+63:i] := (a[i+63:i] * b[i+63:i]) - c[i+63:i] + dst[i+63:i] := (1.0 / SQRT(a[i+63:i])) ENDFOR -dst[MAX:512] := 0 - - - +
immintrin.h
 - + Floating Point - AVX512F/KNCNI - Arithmetic - - - - - Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", subtract packed elements in "c" from the intermediate result, and store the results in "dst" using writemask "k" (elements are copied from "c" when the corresponding mask bit is not set). + AVX512ER + Elementary Math Functions + + + + + + Compute the approximate reciprocal square root of packed double-precision (64-bit) floating-point elements in "a", store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). The maximum relative error for this approximation is less than 2^-28. [sae_note] FOR j := 0 to 7 i := j*64 IF k[j] - dst[i+63:i] := (a[i+63:i] * b[i+63:i]) - c[i+63:i] + dst[i+63:i] := (1.0 / SQRT(a[i+63:i])) ELSE - dst[i+63:i] := c[i+63:i] + dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:512] := 0 - - - +
immintrin.h
 - + Floating Point - AVX512F/KNCNI - Arithmetic - - - - - - Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", subtract packed elements in "c" from the intermediate result, and store the results in "dst" using writemask "k" (elements are copied from "c" when the corresponding mask bit is not set). [round_note] + AVX512ER + Elementary Math Functions + + + + + Compute the approximate reciprocal square root of packed double-precision (64-bit) floating-point elements in "a", store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). The maximum relative error for this approximation is less than 2^-28. FOR j := 0 to 7 i := j*64 IF k[j] - dst[i+63:i] := (a[i+63:i] * b[i+63:i]) - c[i+63:i] + dst[i+63:i] := (1.0 / SQRT(a[i+63:i])) ELSE - dst[i+63:i] := c[i+63:i] + dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:512] := 0 - - - +
immintrin.h
 - + Floating Point - AVX512F/KNCNI - Arithmetic - - - - - Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", subtract packed elements in "c" from the intermediate result, and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). + AVX512ER + Elementary Math Functions + + + + + Compute the approximate reciprocal square root of packed double-precision (64-bit) floating-point elements in "a", store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). The maximum relative error for this approximation is less than 2^-28. [sae_note] FOR j := 0 to 7 i := j*64 IF k[j] - dst[i+63:i] := (a[i+63:i] * b[i+63:i]) - c[i+63:i] + dst[i+63:i] := (1.0 / SQRT(a[i+63:i])) ELSE - dst[i+63:i] := a[i+63:i] + dst[i+63:i] := 0 FI ENDFOR -dst[MAX:512] := 0 - - - +
immintrin.h
 - + Floating Point - AVX512F/KNCNI - Arithmetic - - - - - - Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", subtract packed elements in "c" from the intermediate result, and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). [round_note] + AVX512ER + Elementary Math Functions + + + + Compute the approximate reciprocal square root of packed double-precision (64-bit) floating-point elements in "a", store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). The maximum relative error for this approximation is less than 2^-28. FOR j := 0 to 7 i := j*64 IF k[j] - dst[i+63:i] := (a[i+63:i] * b[i+63:i]) - c[i+63:i] + dst[i+63:i] := (1.0 / SQRT(a[i+63:i])) ELSE - dst[i+63:i] := a[i+63:i] + dst[i+63:i] := 0 FI ENDFOR -dst[MAX:512] := 0 - - - +
immintrin.h
 - + Floating Point AVX512F + AVX512VL Arithmetic - - - - - Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", subtract packed elements in "c" from the intermediate result, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - + + + + + + Add packed double-precision (64-bit) floating-point elements in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 7 +FOR j := 0 to 3 i := j*64 IF k[j] - dst[i+63:i] := (a[i+63:i] * b[i+63:i]) - c[i+63:i] + dst[i+63:i] := a[i+63:i] + b[i+63:i] ELSE - dst[i+63:i] := 0 + dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - - - +
immintrin.h
 - + Floating Point AVX512F + AVX512VL Arithmetic - - - - - - Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", subtract packed elements in "c" from the intermediate result, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - [round_note] + + + + + Add packed double-precision (64-bit) floating-point elements in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 +FOR j := 0 to 3 i := j*64 IF k[j] - dst[i+63:i] := (a[i+63:i] * b[i+63:i]) - c[i+63:i] + dst[i+63:i] := a[i+63:i] + b[i+63:i] ELSE dst[i+63:i] := 0 FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - - - +
immintrin.h
 - + Floating Point - AVX512F/KNCNI + AVX512F + AVX512VL Arithmetic - - - - Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", subtract packed elements in "c" from the intermediate result, and store the results in "dst". - + + + + + + Add packed double-precision (64-bit) floating-point elements in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*32 - dst[i+31:i] := (a[i+31:i] * b[i+31:i]) - c[i+31:i] +FOR j := 0 to 1 + i := j*64 + IF k[j] + dst[i+63:i] := a[i+63:i] + b[i+63:i] + ELSE + dst[i+63:i] := src[i+63:i] + FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - - - +
immintrin.h
 - + Floating Point - AVX512F/KNCNI + AVX512F + AVX512VL Arithmetic - - - - - Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", subtract packed elements in "c" from the intermediate result, and store the results in "dst". - [round_note] + + + + + Add packed double-precision (64-bit) floating-point elements in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*32 - dst[i+31:i] := (a[i+31:i] * b[i+31:i]) - c[i+31:i] +FOR j := 0 to 1 + i := j*64 + IF k[j] + dst[i+63:i] := a[i+63:i] + b[i+63:i] + ELSE + dst[i+63:i] := 0 + FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - - - +
immintrin.h
 - + Floating Point - AVX512F/KNCNI + AVX512F + AVX512VL Arithmetic - - - - - Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", subtract packed elements in "c" from the intermediate result, and store the results in "dst" using writemask "k" (elements are copied from "c" when the corresponding mask bit is not set). + + + + + + Add packed single-precision (32-bit) floating-point elements in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 15 +FOR j := 0 to 7 i := j*32 IF k[j] - dst[i+31:i] := (a[i+31:i] * b[i+31:i]) - c[i+31:i] + dst[i+31:i] := a[i+31:i] + b[i+31:i] ELSE - dst[i+31:i] := c[i+31:i] + dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - - - +
immintrin.h
 - + Floating Point - AVX512F/KNCNI + AVX512F + AVX512VL Arithmetic - - - - - - Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", subtract packed elements in "c" from the intermediate result, and store the results in "dst" using writemask "k" (elements are copied from "c" when the corresponding mask bit is not set). [round_note] + + + + + Add packed single-precision (32-bit) floating-point elements in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 15 +FOR j := 0 to 7 i := j*32 IF k[j] - dst[i+31:i] := (a[i+31:i] * b[i+31:i]) - c[i+31:i] + dst[i+31:i] := a[i+31:i] + b[i+31:i] ELSE - dst[i+31:i] := c[i+31:i] + dst[i+31:i] := 0 FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - - - +
immintrin.h
 - + Floating Point - AVX512F/KNCNI + AVX512F + AVX512VL Arithmetic - - - - - Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", subtract packed elements in "c" from the intermediate result, and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). + + + + + + Add packed single-precision (32-bit) floating-point elements in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 15 +FOR j := 0 to 3 i := j*32 IF k[j] - dst[i+31:i] := (a[i+31:i] * b[i+31:i]) - c[i+31:i] + dst[i+31:i] := a[i+31:i] + b[i+31:i] ELSE - dst[i+31:i] := a[i+31:i] + dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - - - +
immintrin.h
 - + Floating Point - AVX512F/KNCNI + AVX512F + AVX512VL Arithmetic - - - - - - Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", subtract packed elements in "c" from the intermediate result, and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). [round_note] + + + + + Add packed single-precision (32-bit) floating-point elements in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 15 +FOR j := 0 to 3 i := j*32 IF k[j] - dst[i+31:i] := (a[i+31:i] * b[i+31:i]) - c[i+31:i] + dst[i+31:i] := a[i+31:i] + b[i+31:i] ELSE - dst[i+31:i] := a[i+31:i] + dst[i+31:i] := 0 FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - - - +
immintrin.h
 - - Floating Point + + Integer AVX512F - Arithmetic - - - - - Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", subtract packed elements in "c" from the intermediate result, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - + AVX512VL + Miscellaneous + + + + + Concatenate "a" and "b" into a 64-byte immediate result, shift the result right by "imm8" 32-bit elements, and store the low 32 bytes (8 elements) in "dst". -FOR j := 0 to 15 +temp[511:256] := a[255:0] +temp[255:0] := b[255:0] +temp[511:0] := temp[511:0] >> (32*imm8[2:0]) +dst[255:0] := temp[255:0] +dst[MAX:256] := 0 + + +
immintrin.h
+ + + Integer + AVX512F + AVX512VL + Miscellaneous + + + + + + + Concatenate "a" and "b" into a 64-byte immediate result, shift the result right by "imm8" 32-bit elements, and store the low 32 bytes (8 elements) in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +temp[511:256] := a[255:0] +temp[255:0] := b[255:0] +temp[511:0] := temp[511:0] >> (32*imm8[2:0]) +FOR j := 0 to 7 i := j*32 IF k[j] - dst[i+31:i] := (a[i+31:i] * b[i+31:i]) - c[i+31:i] + dst[i+31:i] := temp[i+31:i] ELSE - dst[i+31:i] := 0 + dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - - - +
immintrin.h
 - - Floating Point + + Integer AVX512F - Arithmetic - - - - - - Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", subtract packed elements in "c" from the intermediate result, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - [round_note] + AVX512VL + Miscellaneous + + + + + + Concatenate "a" and "b" into a 64-byte immediate result, shift the result right by "imm8" 32-bit elements, and store the low 32 bytes (8 elements) in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 15 +temp[511:256] := a[255:0] +temp[255:0] := b[255:0] +temp[511:0] := temp[511:0] >> (32*imm8[2:0]) +FOR j := 0 to 7 i := j*32 IF k[j] - dst[i+31:i] := (a[i+31:i] * b[i+31:i]) - c[i+31:i] + dst[i+31:i] := temp[i+31:i] ELSE dst[i+31:i] := 0 FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - - - +
immintrin.h
 - - Floating Point + + Integer AVX512F - Arithmetic - - - - - Multiply the lower double-precision (64-bit) floating-point elements in "a" and "b", and subtract the lower element in "c" from the intermediate result. Store the result in the lower element of "dst", and copy the upper element from "a" to the upper element of "dst". - [round_note] + AVX512VL + Miscellaneous + + + + + Concatenate "a" and "b" into a 32-byte immediate result, shift the result right by "imm8" 32-bit elements, and store the low 16 bytes (4 elements) in "dst". -dst[63:0] := (a[63:0] * b[63:0]) - c[63:0] -dst[127:64] := a[127:64] +temp[255:128] := a[127:0] +temp[127:0] := b[127:0] +temp[255:0] := temp[255:0] >> (32*imm8[1:0]) +dst[127:0] := temp[127:0] dst[MAX:128] := 0 - - - +
immintrin.h
 - - - Floating Point + + Integer AVX512F - Arithmetic - - - - - - Multiply the lower double-precision (64-bit) floating-point elements in "a" and "b", and subtract the lower element in "c" from the intermediate result. Store the result in the lower element of "dst" using writemask "k" (the element is copied from "c" when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". - [round_note] + AVX512VL + Miscellaneous + + + + + + + Concatenate "a" and "b" into a 32-byte immediate result, shift the result right by "imm8" 32-bit elements, and store the low 16 bytes (4 elements) in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -IF k[0] - dst[63:0] := (a[63:0] * b[63:0]) - c[63:0] -ELSE - dst[63:0] := c[63:0] -FI -dst[127:64] := a[127:64] +temp[255:128] := a[127:0] +temp[127:0] := b[127:0] +temp[255:0] := temp[255:0] >> (32*imm8[1:0]) +FOR j := 0 to 3 + i := j*32 + IF k[j] + dst[i+31:i] := temp[i+31:i] + ELSE + dst[i+31:i] := src[i+31:i] + FI +ENDFOR dst[MAX:128] := 0 - - - +
immintrin.h
 - - Floating Point + + Integer AVX512F - Arithmetic - - - - - Multiply the lower double-precision (64-bit) floating-point elements in "a" and "b", and subtract the lower element in "c" from the intermediate result. Store the result in the lower element of "dst" using writemask "k" (the element is copied from "c" when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". + AVX512VL + Miscellaneous + + + + + + Concatenate "a" and "b" into a 32-byte immediate result, shift the result right by "imm8" 32-bit elements, and store the low 16 bytes (4 elements) in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -IF k[0] - dst[63:0] := (a[63:0] * b[63:0]) - c[63:0] -ELSE - dst[63:0] := c[63:0] -FI -dst[127:64] := a[127:64] +temp[255:128] := a[127:0] +temp[127:0] := b[127:0] +temp[255:0] := temp[255:0] >> (32*imm8[1:0]) +FOR j := 0 to 3 + i := j*32 + IF k[j] + dst[i+31:i] := temp[i+31:i] + ELSE + dst[i+31:i] := 0 + FI +ENDFOR dst[MAX:128] := 0 - - - +
immintrin.h
 - - Floating Point + + Integer AVX512F - Arithmetic - - - - - - Multiply the lower double-precision (64-bit) floating-point elements in "a" and "b", and subtract the lower element in "c" from the intermediate result. Store the result in the lower element of "dst" using writemask "k" (the element is copied from "a" when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". - [round_note] + AVX512VL + Miscellaneous + + + + + Concatenate "a" and "b" into a 64-byte immediate result, shift the result right by "imm8" 64-bit elements, and store the low 32 bytes (4 elements) in "dst". -IF k[0] - dst[63:0] := (a[63:0] * b[63:0]) - c[63:0] -ELSE - dst[63:0] := a[63:0] -FI -dst[127:64] := a[127:64] -dst[MAX:128] := 0 +temp[511:256] := a[255:0] +temp[255:0] := b[255:0] +temp[511:0] := temp[511:0] >> (64*imm8[1:0]) +dst[255:0] := temp[255:0] +dst[MAX:256] := 0 - - - +
immintrin.h
 - - Floating Point + + Integer AVX512F - Arithmetic - - - - - Multiply the lower double-precision (64-bit) floating-point elements in "a" and "b", and subtract the lower element in "c" from the intermediate result. Store the result in the lower element of "dst" using writemask "k" (the element is copied from "a" when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". + AVX512VL + Miscellaneous + + + + + + + Concatenate "a" and "b" into a 64-byte immediate result, shift the result right by "imm8" 64-bit elements, and store the low 32 bytes (4 elements) in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -IF k[0] - dst[63:0] := (a[63:0] * b[63:0]) - c[63:0] -ELSE - dst[63:0] := a[63:0] -FI -dst[127:64] := a[127:64] -dst[MAX:128] := 0 +temp[511:256] := a[255:0] +temp[255:0] := b[255:0] +temp[511:0] := temp[511:0] >> (64*imm8[1:0]) +FOR j := 0 to 3 + i := j*64 + IF k[j] + dst[i+63:i] := temp[i+63:i] + ELSE + dst[i+63:i] := src[i+63:i] + FI +ENDFOR +dst[MAX:256] := 0 - - - +
immintrin.h
 - - Floating Point + + Integer AVX512F - Arithmetic - - - - - - Multiply the lower double-precision (64-bit) floating-point elements in "a" and "b", and subtract the lower element in "c" from the intermediate result. Store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". - [round_note] - -IF k[0] - dst[63:0] := (a[63:0] * b[63:0]) - c[63:0] -ELSE - dst[63:0] := 0 -FI -dst[127:64] := a[127:64] -dst[MAX:128] := 0 - - - - -
immintrin.h
- - - Floating Point - AVX512F - Arithmetic - - - - - Multiply the lower double-precision (64-bit) floating-point elements in "a" and "b", and subtract the lower element in "c" from the intermediate result. Store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". - -IF k[0] - dst[63:0] := (a[63:0] * b[63:0]) - c[63:0] -ELSE - dst[63:0] := 0 -FI -dst[127:64] := a[127:64] -dst[MAX:128] := 0 - - - - -
immintrin.h
- - - Floating Point - AVX512F - Arithmetic - - - - - Multiply the lower single-precision (32-bit) floating-point elements in "a" and "b", and subtract the lower element in "c" from the intermediate result. Store the result in the lower element of "dst", and copy the upper 3 packed elements from "a" to the upper elements of "dst". - [round_note] - -dst[31:0] := (a[31:0] * b[31:0]) - c[31:0] -dst[127:32] := a[127:32] -dst[MAX:128] := 0 - - - - -
immintrin.h
- - - Floating Point - AVX512F - Arithmetic - - - - - - Multiply the lower single-precision (32-bit) floating-point elements in "a" and "b", and subtract the lower element in "c" from the intermediate result. Store the result in the lower element of "dst" using writemask "k" (the element is copied from "c" when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". - [round_note] - -IF k[0] - dst[31:0] := (a[31:0] * b[31:0]) - c[31:0] -ELSE - dst[31:0] := c[31:0] -FI -dst[127:32] := a[127:32] -dst[MAX:128] := 0 - - - - -
immintrin.h
- - - Floating Point - AVX512F - Arithmetic - - - - - Multiply the lower single-precision (32-bit) floating-point elements in "a" and "b", and subtract the lower element in "c" from the intermediate result. Store the result in the lower element of "dst" using writemask "k" (the element is copied from "c" when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". - - -IF k[0] - dst[31:0] := (a[31:0] * b[31:0]) - c[31:0] -ELSE - dst[31:0] := c[31:0] -FI -dst[127:32] := a[127:32] -dst[MAX:128] := 0 - - - - -
immintrin.h
- - - Floating Point - AVX512F - Arithmetic - - - - - - Multiply the lower single-precision (32-bit) floating-point elements in "a" and "b", and subtract the lower element in "c" from the intermediate result. Store the result in the lower element of "dst" using writemask "k" (the element is copied from "a" when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". - [round_note] - -IF k[0] - dst[31:0] := (a[31:0] * b[31:0]) - c[31:0] -ELSE - dst[31:0] := a[31:0] -FI -dst[127:32] := a[127:32] -dst[MAX:128] := 0 - - - - -
immintrin.h
- - - Floating Point - AVX512F - Arithmetic - - - - - Multiply the lower single-precision (32-bit) floating-point elements in "a" and "b", and subtract the lower element in "c" from the intermediate result. Store the result in the lower element of "dst" using writemask "k" (the element is copied from "a" when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". - -IF k[0] - dst[31:0] := (a[31:0] * b[31:0]) - c[31:0] -ELSE - dst[31:0] := a[31:0] -FI -dst[127:32] := a[127:32] -dst[MAX:128] := 0 - - - - -
immintrin.h
- - - Floating Point - AVX512F - Arithmetic - - - - - - Multiply the lower single-precision (32-bit) floating-point elements in "a" and "b", and subtract the lower element in "c" from the intermediate result. Store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". - [round_note] - -IF k[0] - dst[31:0] := (a[31:0] * b[31:0]) - c[31:0] -ELSE - dst[31:0] := 0 -FI -dst[127:32] := a[127:32] -dst[MAX:128] := 0 - - - - -
immintrin.h
- - - Floating Point - AVX512F - Arithmetic - - - - - Multiply the lower single-precision (32-bit) floating-point elements in "a" and "b", and subtract the lower element in "c" from the intermediate result. Store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". - -IF k[0] - dst[31:0] := (a[31:0] * b[31:0]) - c[31:0] -ELSE - dst[31:0] := 0 -FI -dst[127:32] := a[127:32] -dst[MAX:128] := 0 - - - - -
immintrin.h
- - - Floating Point - AVX512F - Arithmetic - - - - Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", alternatively subtract and add packed elements in "c" from/to the intermediate result, and store the results in "dst". - + AVX512VL + Miscellaneous + + + + + + Concatenate "a" and "b" into a 64-byte immediate result, shift the result right by "imm8" 64-bit elements, and store the low 32 bytes (4 elements) in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 +temp[511:256] := a[255:0] +temp[255:0] := b[255:0] +temp[511:0] := temp[511:0] >> (64*imm8[1:0]) +FOR j := 0 to 3 i := j*64 - IF (j is even) - dst[i+63:i] := (a[i+63:i] * b[i+63:i]) + c[i+63:i] + IF k[j] + dst[i+63:i] := temp[i+63:i] ELSE - dst[i+63:i] := (a[i+63:i] * b[i+63:i]) - c[i+63:i] + dst[i+63:i] := 0 FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - - - +
immintrin.h
 - - Floating Point + + Integer AVX512F - Arithmetic - - - - - Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", alternatively subtract and add packed elements in "c" from/to the intermediate result, and store the results in "dst". - [round_note] + AVX512VL + Miscellaneous + + + + + Concatenate "a" and "b" into a 32-byte immediate result, shift the result right by "imm8" 64-bit elements, and store the low 16 bytes (2 elements) in "dst". -FOR j := 0 to 7 - i := j*64 - IF (j is even) - dst[i+63:i] := (a[i+63:i] * b[i+63:i]) + c[i+63:i] - ELSE - dst[i+63:i] := (a[i+63:i] * b[i+63:i]) - c[i+63:i] - FI -ENDFOR -dst[MAX:512] := 0 +temp[255:128] := a[127:0] +temp[127:0] := b[127:0] +temp[255:0] := temp[255:0] >> (64*imm8[0]) +dst[127:0] := temp[127:0] +dst[MAX:128] := 0 - - - +
immintrin.h
 - - Floating Point + + Integer AVX512F - Arithmetic - - - - - Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", alternatively subtract and add packed elements in "c" from/to the intermediate result, and store the results in "dst" using writemask "k" (elements are copied from "c" when the corresponding mask bit is not set). + AVX512VL + Miscellaneous + + + + + + + Concatenate "a" and "b" into a 32-byte immediate result, shift the result right by "imm8" 64-bit elements, and store the low 16 bytes (2 elements) in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 7 +temp[255:128] := a[127:0] +temp[127:0] := b[127:0] +temp[255:0] := temp[255:0] >> (64*imm8[0]) +FOR j := 0 to 1 i := j*64 IF k[j] - IF (j is even) - dst[i+63:i] := (a[i+63:i] * b[i+63:i]) + c[i+63:i] - ELSE - dst[i+63:i] := (a[i+63:i] * b[i+63:i]) - c[i+63:i] - FI + dst[i+63:i] := temp[i+63:i] ELSE - dst[i+63:i] := c[i+63:i] + dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - - - +
immintrin.h
 - - Floating Point + + Integer AVX512F - Arithmetic - - - - - - Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", alternatively subtract and add packed elements in "c" from/to the intermediate result, and store the results in "dst" using writemask "k" (elements are copied from "c" when the corresponding mask bit is not set). [round_note] + AVX512VL + Miscellaneous + + + + + + Concatenate "a" and "b" into a 32-byte immediate result, shift the result right by "imm8" 64-bit elements, and store the low 16 bytes (2 elements) in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 +temp[255:128] := a[127:0] +temp[127:0] := b[127:0] +temp[255:0] := temp[255:0] >> (64*imm8[0]) +FOR j := 0 to 1 i := j*64 IF k[j] - IF (j is even) - dst[i+63:i] := (a[i+63:i] * b[i+63:i]) + c[i+63:i] - ELSE - dst[i+63:i] := (a[i+63:i] * b[i+63:i]) - c[i+63:i] - FI + dst[i+63:i] := temp[i+63:i] ELSE - dst[i+63:i] := c[i+63:i] + dst[i+63:i] := 0 FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - - - +
immintrin.h
 - + Floating Point + AVX512VL AVX512F - Arithmetic - - - - - Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", alternatively subtract and add packed elements in "c" from/to the intermediate result, and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). + Miscellaneous + + + + + Blend packed double-precision (64-bit) floating-point elements from "a" and "b" using control mask "k", and store the results in "dst". -FOR j := 0 to 7 +FOR j := 0 to 3 i := j*64 IF k[j] - IF (j is even) - dst[i+63:i] := (a[i+63:i] * b[i+63:i]) + c[i+63:i] - ELSE - dst[i+63:i] := (a[i+63:i] * b[i+63:i]) - c[i+63:i] - FI + dst[i+63:i] := b[i+63:i] ELSE dst[i+63:i] := a[i+63:i] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - - - +
immintrin.h
 - + Floating Point + AVX512VL AVX512F - Arithmetic - - - - - - Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", alternatively subtract and add packed elements in "c" from/to the intermediate result, and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). [round_note] + Miscellaneous + + + + + Blend packed double-precision (64-bit) floating-point elements from "a" and "b" using control mask "k", and store the results in "dst". -FOR j := 0 to 7 +FOR j := 0 to 1 i := j*64 IF k[j] - IF (j is even) - dst[i+63:i] := (a[i+63:i] * b[i+63:i]) + c[i+63:i] - ELSE - dst[i+63:i] := (a[i+63:i] * b[i+63:i]) - c[i+63:i] - FI + dst[i+63:i] := b[i+63:i] ELSE dst[i+63:i] := a[i+63:i] FI ENDFOR -dst[MAX:512] := 0 - - - - -
immintrin.h
- - - Floating Point - AVX512F - Arithmetic - - - - - Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", alternatively subtract and add packed elements in "c" from/to the intermediate result, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - - -FOR j := 0 to 7 - i := j*64 - IF k[j] - IF (j is even) - dst[i+63:i] := (a[i+63:i] * b[i+63:i]) + c[i+63:i] - ELSE - dst[i+63:i] := (a[i+63:i] * b[i+63:i]) - c[i+63:i] - FI - ELSE - dst[i+63:i] := 0 - FI -ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - - - +
immintrin.h
 - + Floating Point + AVX512VL AVX512F - Arithmetic - - - - - - Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", alternatively subtract and add packed elements in "c" from/to the intermediate result, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - [round_note] + Miscellaneous + + + + + Blend packed single-precision (32-bit) floating-point elements from "a" and "b" using control mask "k", and store the results in "dst". FOR j := 0 to 7 - i := j*64 + i := j*32 IF k[j] - IF (j is even) - dst[i+63:i] := (a[i+63:i] * b[i+63:i]) + c[i+63:i] - ELSE - dst[i+63:i] := (a[i+63:i] * b[i+63:i]) - c[i+63:i] - FI + dst[i+31:i] := b[i+31:i] ELSE - dst[i+63:i] := 0 + dst[i+31:i] := a[i+31:i] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - - - +
immintrin.h
 - + Floating Point + AVX512VL AVX512F - Arithmetic - - - - Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", alternatively subtract and add packed elements in "c" from/to the intermediate result, and store the results in "dst". - + Miscellaneous + + + + + Blend packed single-precision (32-bit) floating-point elements from "a" and "b" using control mask "k", and store the results in "dst". -FOR j := 0 to 15 +FOR j := 0 to 3 i := j*32 - IF (j is even) - dst[i+31:i] := (a[i+31:i] * b[i+31:i]) + c[i+31:i] + IF k[j] + dst[i+31:i] := b[i+31:i] ELSE - dst[i+31:i] := (a[i+31:i] * b[i+31:i]) - c[i+31:i] + dst[i+31:i] := a[i+31:i] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - - - +
immintrin.h
 - + Floating Point + AVX512VL AVX512F - Arithmetic - - - - - Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", alternatively subtract and add packed elements in "c" from/to the intermediate result, and store the results in "dst". - [round_note] + Miscellaneous + + + Broadcast the 4 packed single-precision (32-bit) floating-point elements from "a" to all elements of "dst". -FOR j := 0 to 15 +FOR j := 0 to 7 i := j*32 - IF (j is even) - dst[i+31:i] := (a[i+31:i] * b[i+31:i]) + c[i+31:i] - ELSE - dst[i+31:i] := (a[i+31:i] * b[i+31:i]) - c[i+31:i] - FI + n := (j % 4)*32 + dst[i+31:i] := a[n+31:n] ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - - - +
immintrin.h
 - + Floating Point + AVX512VL AVX512F - Arithmetic - - - - - Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", alternatively subtract and add packed elements in "c" from/to the intermediate result, and store the results in "dst" using writemask "k" (elements are copied from "c" when the corresponding mask bit is not set). + Miscellaneous + + + + + Broadcast the 4 packed single-precision (32-bit) floating-point elements from "a" to all elements of "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 15 +FOR j := 0 to 7 i := j*32 + n := (j % 4)*32 IF k[j] - IF (j is even) - dst[i+31:i] := (a[i+31:i] * b[i+31:i]) + c[i+31:i] - ELSE - dst[i+31:i] := (a[i+31:i] * b[i+31:i]) - c[i+31:i] - FI + dst[i+31:i] := a[n+31:n] ELSE - dst[i+31:i] := c[i+31:i] + dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - - - +
immintrin.h
 - + Floating Point + AVX512VL AVX512F - Arithmetic - - - - - - Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", alternatively subtract and add packed elements in "c" from/to the intermediate result, and store the results in "dst" using writemask "k" (elements are copied from "c" when the corresponding mask bit is not set). [round_note] + Miscellaneous + + + + Broadcast the 4 packed single-precision (32-bit) floating-point elements from "a" to all elements of "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 15 +FOR j := 0 to 7 i := j*32 + n := (j % 4)*32 IF k[j] - IF (j is even) - dst[i+31:i] := (a[i+31:i] * b[i+31:i]) + c[i+31:i] - ELSE - dst[i+31:i] := (a[i+31:i] * b[i+31:i]) - c[i+31:i] - FI + dst[i+31:i] := a[n+31:n] ELSE - dst[i+31:i] := c[i+31:i] + dst[i+31:i] := 0 FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - - - +
immintrin.h
 - - Floating Point + + AVX512VL AVX512F - Arithmetic - - - - - Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", alternatively subtract and add packed elements in "c" from/to the intermediate result, and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). + Miscellaneous + + + Broadcast the 4 packed 32-bit integers from "a" to all elements of "dst". -FOR j := 0 to 15 +FOR j := 0 to 7 i := j*32 - IF k[j] - IF (j is even) - dst[i+31:i] := (a[i+31:i] * b[i+31:i]) + c[i+31:i] - ELSE - dst[i+31:i] := (a[i+31:i] * b[i+31:i]) - c[i+31:i] - FI - ELSE - dst[i+31:i] := a[i+31:i] - FI + n := (j % 4)*32 + dst[i+31:i] := a[n+31:n] ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - - - +
immintrin.h
 - - Floating Point + + AVX512VL AVX512F - Arithmetic - - - - - - Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", alternatively subtract and add packed elements in "c" from/to the intermediate result, and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). - [round_note] + Miscellaneous + + + + + Broadcast the 4 packed 32-bit integers from "a" to all elements of "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 15 +FOR j := 0 to 7 i := j*32 + n := (j % 4)*32 IF k[j] - IF (j is even) - dst[i+31:i] := (a[i+31:i] * b[i+31:i]) + c[i+31:i] - ELSE - dst[i+31:i] := (a[i+31:i] * b[i+31:i]) - c[i+31:i] - FI + dst[i+31:i] := a[n+31:n] ELSE - dst[i+31:i] := a[i+31:i] + dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - - - +
immintrin.h
 - - Floating Point + + AVX512VL AVX512F - Arithmetic - - - - - Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", alternatively subtract and add packed elements in "c" from/to the intermediate result, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - + Miscellaneous + + + + Broadcast the 4 packed 32-bit integers from "a" to all elements of "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 15 +FOR j := 0 to 7 i := j*32 + n := (j % 4)*32 IF k[j] - IF (j is even) - dst[i+31:i] := (a[i+31:i] * b[i+31:i]) + c[i+31:i] - ELSE - dst[i+31:i] := (a[i+31:i] * b[i+31:i]) - c[i+31:i] - FI + dst[i+31:i] := a[n+31:n] ELSE dst[i+31:i] := 0 FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - - - +
immintrin.h
 - + Floating Point + AVX512VL AVX512F - Arithmetic - - - - - - Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", alternatively subtract and add packed elements in "c" from/to the intermediate result, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - [round_note] + Miscellaneous + + + + + Broadcast the low double-precision (64-bit) floating-point element from "a" to all elements of "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*32 +FOR j := 0 to 3 + i := j*64 IF k[j] - IF (j is even) - dst[i+31:i] := (a[i+31:i] * b[i+31:i]) + c[i+31:i] - ELSE - dst[i+31:i] := (a[i+31:i] * b[i+31:i]) - c[i+31:i] - FI + dst[i+63:i] := a[63:0] ELSE - dst[i+31:i] := 0 + dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:512] := 0 - - - - -
immintrin.h
- - - Floating Point - AVX512F/KNCNI - Arithmetic - - - - Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", add the negated intermediate result to packed elements in "c", and store the results in "dst". - - -FOR j := 0 to 7 - i := j*64 - dst[i+63:i] := -(a[i+63:i] * b[i+63:i]) + c[i+63:i] -ENDFOR -dst[MAX:512] := 0 - - - - -
immintrin.h
- - - Floating Point - AVX512F/KNCNI - Arithmetic - - - - - Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", add the negated intermediate result to packed elements in "c", and store the results in "dst". - [round_note] - -FOR j := 0 to 7 - i := j*64 - dst[i+63:i] := -(a[i+63:i] * b[i+63:i]) + c[i+63:i] -ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - - - +
immintrin.h
 - + Floating Point - AVX512F/KNCNI - Arithmetic - - - - - Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", add the negated intermediate result to packed elements in "c", and store the results in "dst" using writemask "k" (elements are copied from "c" when the corresponding mask bit is not set). + AVX512VL + AVX512F + Miscellaneous + + + + Broadcast the low double-precision (64-bit) floating-point element from "a" to all elements of "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 +FOR j := 0 to 3 i := j*64 IF k[j] - dst[i+63:i] := -(a[i+63:i] * b[i+63:i]) + c[i+63:i] + dst[i+63:i] := a[63:0] ELSE - dst[i+63:i] := c[i+63:i] + dst[i+63:i] := 0 FI -ENDFOR -dst[MAX:512] := 0 +ENDFOR +dst[MAX:256] := 0 - - - +
immintrin.h
 - + Floating Point - AVX512F/KNCNI - Arithmetic - - - - - - Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", add the negated intermediate result to packed elements in "c", and store the results in "dst" using writemask "k" (elements are copied from "c" when the corresponding mask bit is not set). [round_note] + AVX512VL + AVX512F + Miscellaneous + + + + + Broadcast the low single-precision (32-bit) floating-point element from "a" to all elements of "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). FOR j := 0 to 7 - i := j*64 + i := j*32 IF k[j] - dst[i+63:i] := -(a[i+63:i] * b[i+63:i]) + c[i+63:i] + dst[i+31:i] := a[31:0] ELSE - dst[i+63:i] := c[i+63:i] + dst[i+31:i] := src[i+31:i] FI -ENDFOR -dst[MAX:512] := 0 +ENDFOR +dst[MAX:256] := 0 - - - +
immintrin.h
 - + Floating Point - AVX512F/KNCNI - Arithmetic - - - - - Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", add the negated intermediate result to packed elements in "c", and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). + AVX512VL + AVX512F + Miscellaneous + + + + Broadcast the low single-precision (32-bit) floating-point element from "a" to all elements of "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). FOR j := 0 to 7 - i := j*64 + i := j*32 IF k[j] - dst[i+63:i] := -(a[i+63:i] * b[i+63:i]) + c[i+63:i] + dst[i+31:i] := a[31:0] ELSE - dst[i+63:i] := a[i+63:i] + dst[i+31:i] := 0 FI -ENDFOR -dst[MAX:512] := 0 +ENDFOR +dst[MAX:256] := 0 - - - +
immintrin.h
 - + Floating Point - AVX512F/KNCNI - Arithmetic - - - - - - Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", add the negated intermediate result to packed elements in "c", and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). [round_note] + AVX512VL + AVX512F + Miscellaneous + + + + + Broadcast the low single-precision (32-bit) floating-point element from "a" to all elements of "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := j*64 +FOR j := 0 to 3 + i := j*32 IF k[j] - dst[i+63:i] := -(a[i+63:i] * b[i+63:i]) + c[i+63:i] + dst[i+31:i] := a[31:0] ELSE - dst[i+63:i] := a[i+63:i] + dst[i+31:i] := src[i+31:i] FI -ENDFOR -dst[MAX:512] := 0 +ENDFOR +dst[MAX:128] := 0 - - - +
immintrin.h
 - + Floating Point + AVX512VL AVX512F - Arithmetic - - - - - Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", add the negated intermediate result to packed elements in "c", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Miscellaneous + + + + Broadcast the low single-precision (32-bit) floating-point element from "a" to all elements of "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := j*64 +FOR j := 0 to 3 + i := j*32 IF k[j] - dst[i+63:i] := -(a[i+63:i] * b[i+63:i]) + c[i+63:i] + dst[i+31:i] := a[31:0] ELSE - dst[i+63:i] := 0 + dst[i+31:i] := 0 FI -ENDFOR -dst[MAX:512] := 0 +ENDFOR +dst[MAX:128] := 0 - - - +
immintrin.h
 - + Floating Point + Mask + AVX512VL AVX512F - Arithmetic - - - - - - Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", add the negated intermediate result to packed elements in "c", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). [round_note] - -FOR j := 0 to 7 - i := j*64 - IF k[j] - dst[i+63:i] := -(a[i+63:i] * b[i+63:i]) + c[i+63:i] - ELSE - dst[i+63:i] := 0 - FI -ENDFOR -dst[MAX:512] := 0 + Compare + + + + + Compare packed double-precision (64-bit) floating-point elements in "a" and "b" based on the comparison operand specified by "imm8", and store the results in mask vector "k". + CASE (imm8[4:0]) OF +0: OP := _CMP_EQ_OQ +1: OP := _CMP_LT_OS +2: OP := _CMP_LE_OS +3: OP := _CMP_UNORD_Q +4: OP := _CMP_NEQ_UQ +5: OP := _CMP_NLT_US +6: OP := _CMP_NLE_US +7: OP := _CMP_ORD_Q +8: OP := _CMP_EQ_UQ +9: OP := _CMP_NGE_US +10: OP := _CMP_NGT_US +11: OP := _CMP_FALSE_OQ +12: OP := _CMP_NEQ_OQ +13: OP := _CMP_GE_OS +14: OP := _CMP_GT_OS +15: OP := _CMP_TRUE_UQ +16: OP := _CMP_EQ_OS +17: OP := _CMP_LT_OQ +18: OP := _CMP_LE_OQ +19: OP := _CMP_UNORD_S +20: OP := _CMP_NEQ_US +21: OP := _CMP_NLT_UQ +22: OP := _CMP_NLE_UQ +23: OP := _CMP_ORD_S +24: OP := _CMP_EQ_US +25: OP := _CMP_NGE_UQ +26: OP := _CMP_NGT_UQ +27: OP := _CMP_FALSE_OS +28: OP := _CMP_NEQ_OS +29: OP := _CMP_GE_OQ +30: OP := _CMP_GT_OQ +31: OP := _CMP_TRUE_US +ESAC +FOR j := 0 to 3 + i := j*64 + k[j] := (a[i+63:i] OP b[i+63:i]) ? 1 : 0 +ENDFOR +k[MAX:4] := 0 - - - +
immintrin.h
 - + Floating Point - AVX512F/KNCNI - Arithmetic - - - - Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", add the negated intermediate result to packed elements in "c", and store the results in "dst". - - -FOR j := 0 to 15 - i := j*32 - a[i+31:i] := -(a[i+31:i] * b[i+31:i]) + c[i+31:i] -ENDFOR -dst[MAX:512] := 0 + Mask + AVX512VL + AVX512F + Compare + + + + + + Compare packed double-precision (64-bit) floating-point elements in "a" and "b" based on the comparison operand specified by "imm8", and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). + CASE (imm8[4:0]) OF +0: OP := _CMP_EQ_OQ +1: OP := _CMP_LT_OS +2: OP := _CMP_LE_OS +3: OP := _CMP_UNORD_Q +4: OP := _CMP_NEQ_UQ +5: OP := _CMP_NLT_US +6: OP := _CMP_NLE_US +7: OP := _CMP_ORD_Q +8: OP := _CMP_EQ_UQ +9: OP := _CMP_NGE_US +10: OP := _CMP_NGT_US +11: OP := _CMP_FALSE_OQ +12: OP := _CMP_NEQ_OQ +13: OP := _CMP_GE_OS +14: OP := _CMP_GT_OS +15: OP := _CMP_TRUE_UQ +16: OP := _CMP_EQ_OS +17: OP := _CMP_LT_OQ +18: OP := _CMP_LE_OQ +19: OP := _CMP_UNORD_S +20: OP := _CMP_NEQ_US +21: OP := _CMP_NLT_UQ +22: OP := _CMP_NLE_UQ +23: OP := _CMP_ORD_S +24: OP := _CMP_EQ_US +25: OP := _CMP_NGE_UQ +26: OP := _CMP_NGT_UQ +27: OP := _CMP_FALSE_OS +28: OP := _CMP_NEQ_OS +29: OP := _CMP_GE_OQ +30: OP := _CMP_GT_OQ +31: OP := _CMP_TRUE_US +ESAC +FOR j := 0 to 3 + i := j*64 + IF k1[j] + k[j] := ( a[i+63:i] OP b[i+63:i] ) ? 1 : 0 + ELSE + k[j] := 0 + FI +ENDFOR +k[MAX:4] := 0 - - - +
immintrin.h
 - + Floating Point - AVX512F/KNCNI - Arithmetic - - - - - Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", add the negated intermediate result to packed elements in "c", and store the results in "dst". - [round_note] - -FOR j := 0 to 15 - i := j*32 - dst[i+31:i] := -(a[i+31:i] * b[i+31:i]) + c[i+31:i] -ENDFOR -dst[MAX:512] := 0 + Mask + AVX512VL + AVX512F + Compare + + + + + Compare packed double-precision (64-bit) floating-point elements in "a" and "b" based on the comparison operand specified by "imm8", and store the results in mask vector "k". + CASE (imm8[4:0]) OF +0: OP := _CMP_EQ_OQ +1: OP := _CMP_LT_OS +2: OP := _CMP_LE_OS +3: OP := _CMP_UNORD_Q +4: OP := _CMP_NEQ_UQ +5: OP := _CMP_NLT_US +6: OP := _CMP_NLE_US +7: OP := _CMP_ORD_Q +8: OP := _CMP_EQ_UQ +9: OP := _CMP_NGE_US +10: OP := _CMP_NGT_US +11: OP := _CMP_FALSE_OQ +12: OP := _CMP_NEQ_OQ +13: OP := _CMP_GE_OS +14: OP := _CMP_GT_OS +15: OP := _CMP_TRUE_UQ +16: OP := _CMP_EQ_OS +17: OP := _CMP_LT_OQ +18: OP := _CMP_LE_OQ +19: OP := _CMP_UNORD_S +20: OP := _CMP_NEQ_US +21: OP := _CMP_NLT_UQ +22: OP := _CMP_NLE_UQ +23: OP := _CMP_ORD_S +24: OP := _CMP_EQ_US +25: OP := _CMP_NGE_UQ +26: OP := _CMP_NGT_UQ +27: OP := _CMP_FALSE_OS +28: OP := _CMP_NEQ_OS +29: OP := _CMP_GE_OQ +30: OP := _CMP_GT_OQ +31: OP := _CMP_TRUE_US +ESAC +FOR j := 0 to 1 + i := j*64 + k[j] := ( a[i+63:i] OP b[i+63:i] ) ? 1 : 0 +ENDFOR +k[MAX:2] := 0 - - - +
immintrin.h
 - + Floating Point - AVX512F/KNCNI - Arithmetic - - - - - Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", add the negated intermediate result to packed elements in "c", and store the results in "dst" using writemask "k" (elements are copied from "c" when the corresponding mask bit is not set). - -FOR j := 0 to 15 - i := j*32 - IF k[j] - dst[i+31:i] := -(a[i+31:i] * b[i+31:i]) + c[i+31:i] - ELSE - dst[i+31:i] := c[i+31:i] + Mask + AVX512VL + AVX512F + Compare + + + + + + Compare packed double-precision (64-bit) floating-point elements in "a" and "b" based on the comparison operand specified by "imm8", and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). + CASE (imm8[4:0]) OF +0: OP := _CMP_EQ_OQ +1: OP := _CMP_LT_OS +2: OP := _CMP_LE_OS +3: OP := _CMP_UNORD_Q +4: OP := _CMP_NEQ_UQ +5: OP := _CMP_NLT_US +6: OP := _CMP_NLE_US +7: OP := _CMP_ORD_Q +8: OP := _CMP_EQ_UQ +9: OP := _CMP_NGE_US +10: OP := _CMP_NGT_US +11: OP := _CMP_FALSE_OQ +12: OP := _CMP_NEQ_OQ +13: OP := _CMP_GE_OS +14: OP := _CMP_GT_OS +15: OP := _CMP_TRUE_UQ +16: OP := _CMP_EQ_OS +17: OP := _CMP_LT_OQ +18: OP := _CMP_LE_OQ +19: OP := _CMP_UNORD_S +20: OP := _CMP_NEQ_US +21: OP := _CMP_NLT_UQ +22: OP := _CMP_NLE_UQ +23: OP := _CMP_ORD_S +24: OP := _CMP_EQ_US +25: OP := _CMP_NGE_UQ +26: OP := _CMP_NGT_UQ +27: OP := _CMP_FALSE_OS +28: OP := _CMP_NEQ_OS +29: OP := _CMP_GE_OQ +30: OP := _CMP_GT_OQ +31: OP := _CMP_TRUE_US +ESAC +FOR j := 0 to 1 + i := j*64 + IF k1[j] + k[j] := ( a[i+63:i] OP b[i+63:i] ) ? 1 : 0 + ELSE + k[j] := 0 FI -ENDFOR -dst[MAX:512] := 0 +ENDFOR +k[MAX:2] := 0 - - - +
immintrin.h
 - + Floating Point - AVX512F/KNCNI - Arithmetic - - - - - - Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", add the negated intermediate result to packed elements in "c", and store the results in "dst" using writemask "k" (elements are copied from "c" when the corresponding mask bit is not set). [round_note] - -FOR j := 0 to 15 + Mask + AVX512VL + AVX512F + Compare + + + + + Compare packed single-precision (32-bit) floating-point elements in "a" and "b" based on the comparison operand specified by "imm8", and store the results in mask vector "k". + CASE (imm8[4:0]) OF +0: OP := _CMP_EQ_OQ +1: OP := _CMP_LT_OS +2: OP := _CMP_LE_OS +3: OP := _CMP_UNORD_Q +4: OP := _CMP_NEQ_UQ +5: OP := _CMP_NLT_US +6: OP := _CMP_NLE_US +7: OP := _CMP_ORD_Q +8: OP := _CMP_EQ_UQ +9: OP := _CMP_NGE_US +10: OP := _CMP_NGT_US +11: OP := _CMP_FALSE_OQ +12: OP := _CMP_NEQ_OQ +13: OP := _CMP_GE_OS +14: OP := _CMP_GT_OS +15: OP := _CMP_TRUE_UQ +16: OP := _CMP_EQ_OS +17: OP := _CMP_LT_OQ +18: OP := _CMP_LE_OQ +19: OP := _CMP_UNORD_S +20: OP := _CMP_NEQ_US +21: OP := _CMP_NLT_UQ +22: OP := _CMP_NLE_UQ +23: OP := _CMP_ORD_S +24: OP := _CMP_EQ_US +25: OP := _CMP_NGE_UQ +26: OP := _CMP_NGT_UQ +27: OP := _CMP_FALSE_OS +28: OP := _CMP_NEQ_OS +29: OP := _CMP_GE_OQ +30: OP := _CMP_GT_OQ +31: OP := _CMP_TRUE_US +ESAC +FOR j := 0 to 7 i := j*32 - IF k[j] - dst[i+31:i] := -(a[i+31:i] * b[i+31:i]) + c[i+31:i] - ELSE - dst[i+31:i] := c[i+31:i] - FI -ENDFOR -dst[MAX:512] := 0 + k[j] := (a[i+31:i] OP b[i+31:i]) ? 1 : 0 +ENDFOR +k[MAX:8] := 0 - - - +
immintrin.h
 - + Floating Point - AVX512F/KNCNI - Arithmetic - - - - - Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", add the negated intermediate result to packed elements in "c", and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). - -FOR j := 0 to 15 + Mask + AVX512VL + AVX512F + Compare + + + + + + Compare packed single-precision (32-bit) floating-point elements in "a" and "b" based on the comparison operand specified by "imm8", and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). + CASE (imm8[4:0]) OF +0: OP := _CMP_EQ_OQ +1: OP := _CMP_LT_OS +2: OP := _CMP_LE_OS +3: OP := _CMP_UNORD_Q +4: OP := _CMP_NEQ_UQ +5: OP := _CMP_NLT_US +6: OP := _CMP_NLE_US +7: OP := _CMP_ORD_Q +8: OP := _CMP_EQ_UQ +9: OP := _CMP_NGE_US +10: OP := _CMP_NGT_US +11: OP := _CMP_FALSE_OQ +12: OP := _CMP_NEQ_OQ +13: OP := _CMP_GE_OS +14: OP := _CMP_GT_OS +15: OP := _CMP_TRUE_UQ +16: OP := _CMP_EQ_OS +17: OP := _CMP_LT_OQ +18: OP := _CMP_LE_OQ +19: OP := _CMP_UNORD_S +20: OP := _CMP_NEQ_US +21: OP := _CMP_NLT_UQ +22: OP := _CMP_NLE_UQ +23: OP := _CMP_ORD_S +24: OP := _CMP_EQ_US +25: OP := _CMP_NGE_UQ +26: OP := _CMP_NGT_UQ +27: OP := _CMP_FALSE_OS +28: OP := _CMP_NEQ_OS +29: OP := _CMP_GE_OQ +30: OP := _CMP_GT_OQ +31: OP := _CMP_TRUE_US +ESAC +FOR j := 0 to 7 i := j*32 - IF k[j] - dst[i+31:i] := -(a[i+31:i] * b[i+31:i]) + c[i+31:i] - ELSE - dst[i+31:i] := a[i+31:i] + IF k1[j] + k[j] := ( a[i+31:i] OP b[i+31:i] ) ? 1 : 0 + ELSE + k[j] := 0 FI -ENDFOR -dst[MAX:512] := 0 +ENDFOR +k[MAX:8] := 0 - - - +
immintrin.h
 - + Floating Point - AVX512F/KNCNI - Arithmetic - - - - - - Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", add the negated intermediate result to packed elements in "c", and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). [round_note] - -FOR j := 0 to 15 + Mask + AVX512VL + AVX512F + Compare + + + + + Compare packed single-precision (32-bit) floating-point elements in "a" and "b" based on the comparison operand specified by "imm8", and store the results in mask vector "k". + CASE (imm8[4:0]) OF +0: OP := _CMP_EQ_OQ +1: OP := _CMP_LT_OS +2: OP := _CMP_LE_OS +3: OP := _CMP_UNORD_Q +4: OP := _CMP_NEQ_UQ +5: OP := _CMP_NLT_US +6: OP := _CMP_NLE_US +7: OP := _CMP_ORD_Q +8: OP := _CMP_EQ_UQ +9: OP := _CMP_NGE_US +10: OP := _CMP_NGT_US +11: OP := _CMP_FALSE_OQ +12: OP := _CMP_NEQ_OQ +13: OP := _CMP_GE_OS +14: OP := _CMP_GT_OS +15: OP := _CMP_TRUE_UQ +16: OP := _CMP_EQ_OS +17: OP := _CMP_LT_OQ +18: OP := _CMP_LE_OQ +19: OP := _CMP_UNORD_S +20: OP := _CMP_NEQ_US +21: OP := _CMP_NLT_UQ +22: OP := _CMP_NLE_UQ +23: OP := _CMP_ORD_S +24: OP := _CMP_EQ_US +25: OP := _CMP_NGE_UQ +26: OP := _CMP_NGT_UQ +27: OP := _CMP_FALSE_OS +28: OP := _CMP_NEQ_OS +29: OP := _CMP_GE_OQ +30: OP := _CMP_GT_OQ +31: OP := _CMP_TRUE_US +ESAC +FOR j := 0 to 3 i := j*32 - IF k[j] - dst[i+31:i] := -(a[i+31:i] * b[i+31:i]) + c[i+31:i] - ELSE - dst[i+31:i] := a[i+31:i] - FI -ENDFOR -dst[MAX:512] := 0 + k[j] := ( a[i+31:i] OP b[i+31:i] ) ? 1 : 0 +ENDFOR +k[MAX:4] := 0 - - - +
immintrin.h
 - + Floating Point + Mask + AVX512VL AVX512F - Arithmetic - - - - - Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", add the negated intermediate result to packed elements in "c", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - -FOR j := 0 to 15 + Compare + + + + + + Compare packed single-precision (32-bit) floating-point elements in "a" and "b" based on the comparison operand specified by "imm8", and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). + CASE (imm8[4:0]) OF +0: OP := _CMP_EQ_OQ +1: OP := _CMP_LT_OS +2: OP := _CMP_LE_OS +3: OP := _CMP_UNORD_Q +4: OP := _CMP_NEQ_UQ +5: OP := _CMP_NLT_US +6: OP := _CMP_NLE_US +7: OP := _CMP_ORD_Q +8: OP := _CMP_EQ_UQ +9: OP := _CMP_NGE_US +10: OP := _CMP_NGT_US +11: OP := _CMP_FALSE_OQ +12: OP := _CMP_NEQ_OQ +13: OP := _CMP_GE_OS +14: OP := _CMP_GT_OS +15: OP := _CMP_TRUE_UQ +16: OP := _CMP_EQ_OS +17: OP := _CMP_LT_OQ +18: OP := _CMP_LE_OQ +19: OP := _CMP_UNORD_S +20: OP := _CMP_NEQ_US +21: OP := _CMP_NLT_UQ +22: OP := _CMP_NLE_UQ +23: OP := _CMP_ORD_S +24: OP := _CMP_EQ_US +25: OP := _CMP_NGE_UQ +26: OP := _CMP_NGT_UQ +27: OP := _CMP_FALSE_OS +28: OP := _CMP_NEQ_OS +29: OP := _CMP_GE_OQ +30: OP := _CMP_GT_OQ +31: OP := _CMP_TRUE_US +ESAC +FOR j := 0 to 3 i := j*32 - IF k[j] - dst[i+31:i] := -(a[i+31:i] * b[i+31:i]) + c[i+31:i] - ELSE - dst[i+31:i] := 0 + IF k1[j] + k[j] := ( a[i+31:i] OP b[i+31:i] ) ? 1 : 0 + ELSE + k[j] := 0 FI -ENDFOR -dst[MAX:512] := 0 +ENDFOR +k[MAX:4] := 0 - - - +
immintrin.h
 - + Floating Point + AVX512VL AVX512F - Arithmetic - - - - - - Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", add the negated intermediate result to packed elements in "c", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). [round_note] + Miscellaneous + + + + + Contiguously store the active double-precision (64-bit) floating-point elements in "a" (those with their respective bit set in writemask "k") to "dst", and pass through the remaining elements from "src". -FOR j := 0 to 15 - i := j*32 +size := 64 +m := 0 +FOR j := 0 to 3 + i := j*64 IF k[j] - dst[i+31:i] := -(a[i+31:i] * b[i+31:i]) + c[i+31:i] - ELSE - dst[i+31:i] := 0 + dst[m+size-1:m] := a[i+63:i] + m := m + size FI -ENDFOR -dst[MAX:512] := 0 +ENDFOR +dst[255:m] := src[255:m] +dst[MAX:256] := 0 - - - +
immintrin.h
 - + Floating Point + AVX512VL AVX512F - Arithmetic - - - - - Multiply the lower double-precision (64-bit) floating-point elements in "a" and "b", and add the negated intermediate result to the lower element in "c". Store the result in the lower element of "dst", and copy the upper element from "a" to the upper element of "dst". - [round_note] + Store + + + + + Contiguously store the active double-precision (64-bit) floating-point elements in "a" (those with their respective bit set in writemask "k") to unaligned memory at "base_addr". -dst[63:0] := -(a[63:0] * b[63:0]) + c[63:0] -dst[127:64] := a[127:64] -dst[MAX:128] := 0 +size := 64 +m := base_addr +FOR j := 0 to 3 + i := j*64 + IF k[j] + MEM[m+size-1:m] := a[i+63:i] + m := m + size + FI +ENDFOR - - - +
immintrin.h
 - + Floating Point + AVX512VL AVX512F - Arithmetic - - - - - - Multiply the lower double-precision (64-bit) floating-point elements in "a" and "b", and add the negated intermediate result to the lower element in "c". Store the result in the lower element of "dst" using writemask "k" (the element is copied from "c" when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". - [round_note] + Miscellaneous + + + + Contiguously store the active double-precision (64-bit) floating-point elements in "a" (those with their respective bit set in zeromask "k") to "dst", and set the remaining elements to zero. -IF k[0] - dst[63:0] := -(a[63:0] * b[63:0]) + c[63:0] -ELSE - dst[63:0] := c[63:0] -FI -dst[127:64] := a[127:64] -dst[MAX:128] := 0 +size := 64 +m := 0 +FOR j := 0 to 3 + i := j*64 + IF k[j] + dst[m+size-1:m] := a[i+63:i] + m := m + size + FI +ENDFOR +dst[255:m] := 0 +dst[MAX:256] := 0 - - - +
immintrin.h
 - + Floating Point + AVX512VL AVX512F - Arithmetic - - - - - Multiply the lower double-precision (64-bit) floating-point elements in "a" and "b", and add the negated intermediate result to the lower element in "c". Store the result in the lower element of "dst" using writemask "k" (the element is copied from "c" when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". + Miscellaneous + + + + + Contiguously store the active double-precision (64-bit) floating-point elements in "a" (those with their respective bit set in writemask "k") to "dst", and pass through the remaining elements from "src". -IF k[0] - dst[63:0] := -(a[63:0] * b[63:0]) + c[63:0] -ELSE - dst[63:0] := c[63:0] -FI -dst[127:64] := a[127:64] +size := 64 +m := 0 +FOR j := 0 to 1 + i := j*64 + IF k[j] + dst[m+size-1:m] := a[i+63:i] + m := m + size + FI +ENDFOR +dst[127:m] := src[127:m] dst[MAX:128] := 0 - - - +
immintrin.h
 - + Floating Point + AVX512VL AVX512F - Arithmetic - - - - - - Multiply the lower double-precision (64-bit) floating-point elements in "a" and "b", and add the negated intermediate result to the lower element in "c". Store the result in the lower element of "dst" using writemask "k" (the element is copied from "a" when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". - [round_note] + Store + + + + + Contiguously store the active double-precision (64-bit) floating-point elements in "a" (those with their respective bit set in writemask "k") to unaligned memory at "base_addr". -IF k[0] - dst[63:0] := -(a[63:0] * b[63:0]) + c[63:0] -ELSE - dst[63:0] := a[63:0] -FI -dst[127:64] := a[127:64] -dst[MAX:128] := 0 +size := 64 +m := base_addr +FOR j := 0 to 1 + i := j*64 + IF k[j] + MEM[m+size-1:m] := a[i+63:i] + m := m + size + FI +ENDFOR - - - +
immintrin.h
 - + Floating Point + AVX512VL AVX512F - Arithmetic - - - - - Multiply the lower double-precision (64-bit) floating-point elements in "a" and "b", and add the negated intermediate result to the lower element in "c". Store the result in the lower element of "dst" using writemask "k" (the element is copied from "a" when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". + Miscellaneous + + + + Contiguously store the active double-precision (64-bit) floating-point elements in "a" (those with their respective bit set in zeromask "k") to "dst", and set the remaining elements to zero. -IF k[0] - dst[63:0] := -(a[63:0] * b[63:0]) + c[63:0] -ELSE - dst[63:0] := a[63:0] -FI -dst[127:64] := a[127:64] +size := 64 +m := 0 +FOR j := 0 to 1 + i := j*64 + IF k[j] + dst[m+size-1:m] := a[i+63:i] + m := m + size + FI +ENDFOR +dst[127:m] := 0 dst[MAX:128] := 0 - - - +
immintrin.h
 - + Floating Point + AVX512VL AVX512F - Arithmetic - - - - - - Multiply the lower double-precision (64-bit) floating-point elements in "a" and "b", and add the negated intermediate result to the lower element in "c". Store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". - [round_note] + Miscellaneous + + + + + Contiguously store the active single-precision (32-bit) floating-point elements in "a" (those with their respective bit set in writemask "k") to "dst", and pass through the remaining elements from "src". -IF k[0] - dst[63:0] := -(a[63:0] * b[63:0]) + c[63:0] -ELSE - dst[63:0] := 0 -FI -dst[127:64] := a[127:64] -dst[MAX:128] := 0 +size := 32 +m := 0 +FOR j := 0 to 7 + i := j*32 + IF k[j] + dst[m+size-1:m] := a[i+31:i] + m := m + size + FI +ENDFOR +dst[255:m] := src[255:m] +dst[MAX:256] := 0 - - - +
immintrin.h
 - + Floating Point + AVX512VL AVX512F - Arithmetic - - - - - Multiply the lower double-precision (64-bit) floating-point elements in "a" and "b", and add the negated intermediate result to the lower element in "c". Store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". + Store + + + + + Contiguously store the active single-precision (32-bit) floating-point elements in "a" (those with their respective bit set in writemask "k") to unaligned memory at "base_addr". -IF k[0] - dst[63:0] := -(a[63:0] * b[63:0]) + c[63:0] -ELSE - dst[63:0] := 0 -FI -dst[127:64] := a[127:64] -dst[MAX:128] := 0 +size := 32 +m := base_addr +FOR j := 0 to 7 + i := j*32 + IF k[j] + MEM[m+size-1:m] := a[i+31:i] + m := m + size + FI +ENDFOR - - - +
immintrin.h
 - + Floating Point + AVX512VL AVX512F - Arithmetic - - - - - Multiply the lower single-precision (32-bit) floating-point elements in "a" and "b", and add the negated intermediate result to the lower element in "c". Store the result in the lower element of "dst", and copy the upper 3 packed elements from "a" to the upper elements of "dst". - [round_note] + Miscellaneous + + + + Contiguously store the active single-precision (32-bit) floating-point elements in "a" (those with their respective bit set in zeromask "k") to "dst", and set the remaining elements to zero. -dst[31:0] := -(a[31:0] * b[31:0]) + c[31:0] -dst[127:32] := a[127:32] -dst[MAX:128] := 0 +size := 32 +m := 0 +FOR j := 0 to 7 + i := j*32 + IF k[j] + dst[m+size-1:m] := a[i+31:i] + m := m + size + FI +ENDFOR +dst[255:m] := 0 +dst[MAX:256] := 0 - - - +
immintrin.h
 - + Floating Point + AVX512VL AVX512F - Arithmetic - - - - - - Multiply the lower single-precision (32-bit) floating-point elements in "a" and "b", and add the negated intermediate result to the lower element in "c". Store the result in the lower element of "dst" using writemask "k" (the element is copied from "c" when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". - [round_note] + Miscellaneous + + + + + Contiguously store the active single-precision (32-bit) floating-point elements in "a" (those with their respective bit set in writemask "k") to "dst", and pass through the remaining elements from "src". -IF k[0] - dst[31:0] := -(a[31:0] * b[31:0]) + c[31:0] -ELSE - dst[31:0] := c[31:0] -FI -dst[127:32] := a[127:32] +size := 32 +m := 0 +FOR j := 0 to 3 + i := j*32 + IF k[j] + dst[m+size-1:m] := a[i+31:i] + m := m + size + FI +ENDFOR +dst[127:m] := src[127:m] dst[MAX:128] := 0 - - - +
immintrin.h
 - + Floating Point + AVX512VL AVX512F - Arithmetic - - - - - Multiply the lower single-precision (32-bit) floating-point elements in "a" and "b", and add the negated intermediate result to the lower element in "c". Store the result in the lower element of "dst" using writemask "k" (the element is copied from "c" when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". + Store + + + + + Contiguously store the active single-precision (32-bit) floating-point elements in "a" (those with their respective bit set in writemask "k") to unaligned memory at "base_addr". -IF k[0] - dst[31:0] := -(a[31:0] * b[31:0]) + c[31:0] -ELSE - dst[31:0] := c[31:0] -FI -dst[127:32] := a[127:32] -dst[MAX:128] := 0 +size := 32 +m := base_addr +FOR j := 0 to 3 + i := j*32 + IF k[j] + MEM[m+size-1:m] := a[i+31:i] + m := m + size + FI +ENDFOR - - - +
immintrin.h
 - + Floating Point + AVX512VL AVX512F - Arithmetic - - - - - - Multiply the lower single-precision (32-bit) floating-point elements in "a" and "b", and add the negated intermediate result to the lower element in "c". Store the result in the lower element of "dst" using writemask "k" (the element is copied from "a" when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". - [round_note] + Miscellaneous + + + + Contiguously store the active single-precision (32-bit) floating-point elements in "a" (those with their respective bit set in zeromask "k") to "dst", and set the remaining elements to zero. -IF k[0] - dst[31:0] := -(a[31:0] * b[31:0]) + c[31:0] -ELSE - dst[31:0] := a[31:0] -FI -dst[127:32] := a[127:32] +size := 32 +m := 0 +FOR j := 0 to 3 + i := j*32 + IF k[j] + dst[m+size-1:m] := a[i+31:i] + m := m + size + FI +ENDFOR +dst[127:m] := 0 dst[MAX:128] := 0 - - - +
immintrin.h
 - + Floating Point + Integer + AVX512VL AVX512F - Arithmetic - - - - - Multiply the lower single-precision (32-bit) floating-point elements in "a" and "b", and add the negated intermediate result to the lower element in "c". Store the result in the lower element of "dst" using writemask "k" (the element is copied from "a" when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". + Convert + + + + + Convert packed signed 32-bit integers in "a" to packed double-precision (64-bit) floating-point elements, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -IF k[0] - dst[31:0] := -(a[31:0] * b[31:0]) + c[31:0] -ELSE - dst[31:0] := a[31:0] -FI -dst[127:32] := a[127:32] -dst[MAX:128] := 0 +FOR j := 0 to 3 + i := j*32 + m := j*64 + IF k[j] + dst[m+63:m] := Convert_Int32_To_FP64(a[i+31:i]) + ELSE + dst[m+63:m] := src[m+63:m] + FI +ENDFOR +dst[MAX:256] := 0 - - - +
immintrin.h
 - + Floating Point + Integer + AVX512VL AVX512F - Arithmetic - - - - - - Multiply the lower single-precision (32-bit) floating-point elements in "a" and "b", and add the negated intermediate result to the lower element in "c". Store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". - [round_note] + Convert + + + + Convert packed signed 32-bit integers in "a" to packed double-precision (64-bit) floating-point elements, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -IF k[0] - dst[31:0] := -(a[31:0] * b[31:0]) + c[31:0] -ELSE - dst[31:0] := 0 -FI -dst[127:32] := a[127:32] -dst[MAX:128] := 0 +FOR j := 0 to 3 + i := j*32 + m := j*64 + IF k[j] + dst[m+63:m] := Convert_Int32_To_FP64(a[i+31:i]) + ELSE + dst[m+63:m] := 0 + FI +ENDFOR +dst[MAX:256] := 0 - - - +
immintrin.h
 - + Floating Point + Integer + AVX512VL AVX512F - Arithmetic - - - - - Multiply the lower single-precision (32-bit) floating-point elements in "a" and "b", and add the negated intermediate result to the lower element in "c". Store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". - -IF k[0] - dst[31:0] := -(a[31:0] * b[31:0]) + c[31:0] -ELSE - dst[31:0] := 0 -FI -dst[127:32] := a[127:32] -dst[MAX:128] := 0 - - - - -
immintrin.h
- - - Floating Point - AVX512F/KNCNI - Arithmetic - - - - Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", subtract packed elements in "c" from the negated intermediate result, and store the results in "dst". - -FOR j := 0 to 7 - i := j*64 - dst[i+63:i] := -(a[i+63:i] * b[i+63:i]) - c[i+63:i] -ENDFOR -dst[MAX:512] := 0 - - - - -
immintrin.h
- - - Floating Point - AVX512F/KNCNI - Arithmetic - - - - - Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", subtract packed elements in "c" from the negated intermediate result, and store the results in "dst". - [round_note] - -FOR j := 0 to 7 - i := j*64 - dst[i+63:i] := -(a[i+63:i] * b[i+63:i]) - c[i+63:i] -ENDFOR -dst[MAX:512] := 0 - - - - -
immintrin.h
- - - Floating Point - AVX512F/KNCNI - Arithmetic - - - - - Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", subtract packed elements in "c" from the negated intermediate result, and store the results in "dst" using writemask "k" (elements are copied from "c" when the corresponding mask bit is not set). + Convert + + + + + Convert packed signed 32-bit integers in "a" to packed double-precision (64-bit) floating-point elements, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := j*64 +FOR j := 0 to 1 + i := j*32 + m := j*64 IF k[j] - dst[i+63:i] := -(a[i+63:i] * b[i+63:i]) - c[i+63:i] + dst[m+63:m] := Convert_Int32_To_FP64(a[i+31:i]) ELSE - dst[i+63:i] := c[i+63:i] + dst[m+63:m] := src[m+63:m] FI -ENDFOR -dst[MAX:512] := 0 +ENDFOR +dst[MAX:128] := 0 - - - +
immintrin.h
 - + Floating Point - AVX512F/KNCNI - Arithmetic - - - - - - Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", subtract packed elements in "c" from the negated intermediate result, and store the results in "dst" using writemask "k" (elements are copied from "c" when the corresponding mask bit is not set). [round_note] + Integer + AVX512VL + AVX512F + Convert + + + + Convert packed signed 32-bit integers in "a" to packed double-precision (64-bit) floating-point elements, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := j*64 +FOR j := 0 to 1 + i := j*32 + m := j*64 IF k[j] - dst[i+63:i] := -(a[i+63:i] * b[i+63:i]) - c[i+63:i] + dst[m+63:m] := Convert_Int32_To_FP64(a[i+31:i]) ELSE - dst[i+63:i] := c[i+63:i] + dst[m+63:m] := 0 FI -ENDFOR -dst[MAX:512] := 0 +ENDFOR +dst[MAX:128] := 0 - - - +
immintrin.h
 - + Floating Point - AVX512F/KNCNI - Arithmetic - - - - - Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", subtract packed elements in "c" from the negated intermediate result, and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). + Integer + AVX512VL + AVX512F + Convert + + + + + Convert packed signed 32-bit integers in "a" to packed single-precision (32-bit) floating-point elements, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). FOR j := 0 to 7 - i := j*64 + i := j*32 IF k[j] - dst[i+63:i] := -(a[i+63:i] * b[i+63:i]) - c[i+63:i] + dst[i+31:i] := Convert_Int32_To_FP32(a[i+31:i]) ELSE - dst[i+63:i] := a[i+63:i] + dst[i+31:i] := src[i+31:i] FI -ENDFOR -dst[MAX:512] := 0 +ENDFOR +dst[MAX:256] := 0 - - - +
immintrin.h
 - + Floating Point - AVX512F/KNCNI - Arithmetic - - - - - - Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", subtract packed elements in "c" from the negated intermediate result, and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). - [round_note] + Integer + AVX512VL + AVX512F + Convert + + + + Convert packed signed 32-bit integers in "a" to packed single-precision (32-bit) floating-point elements, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). FOR j := 0 to 7 - i := j*64 + i := 32*j IF k[j] - dst[i+63:i] := -(a[i+63:i] * b[i+63:i]) - c[i+63:i] + dst[i+31:i] := Convert_Int32_To_FP32(a[i+31:i]) ELSE - dst[i+63:i] := a[i+63:i] + dst[i+31:i] := 0 FI -ENDFOR -dst[MAX:512] := 0 +ENDFOR +dst[MAX:256] := 0 - - - +
immintrin.h
 - + Floating Point + Integer + AVX512VL AVX512F - Arithmetic - - - - - Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", subtract packed elements in "c" from the negated intermediate result, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Convert + + + + + Convert packed signed 32-bit integers in "a" to packed single-precision (32-bit) floating-point elements, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := j*64 +FOR j := 0 to 3 + i := j*32 IF k[j] - dst[i+63:i] := -(a[i+63:i] * b[i+63:i]) - c[i+63:i] + dst[i+31:i] := Convert_Int32_To_FP32(a[i+31:i]) ELSE - dst[i+63:i] := 0 + dst[i+31:i] := src[i+31:i] FI -ENDFOR -dst[MAX:512] := 0 +ENDFOR +dst[MAX:128] := 0 - - - +
immintrin.h
 - + Floating Point + Integer + AVX512VL AVX512F - Arithmetic - - - - - - Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", subtract packed elements in "c" from the negated intermediate result, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). [round_note] + Convert + + + + Convert packed signed 32-bit integers in "a" to packed single-precision (32-bit) floating-point elements, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := j*64 +FOR j := 0 to 3 + i := 32*j IF k[j] - dst[i+63:i] := -(a[i+63:i] * b[i+63:i]) - c[i+63:i] + dst[i+31:i] := Convert_Int32_To_FP32(a[i+31:i]) ELSE - dst[i+63:i] := 0 + dst[i+31:i] := 0 FI -ENDFOR -dst[MAX:512] := 0 +ENDFOR +dst[MAX:128] := 0 - - - +
immintrin.h
 - + Floating Point - AVX512F/KNCNI - Arithmetic - - - - Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", subtract packed elements in "c" from the negated intermediate result, and store the results in "dst". + Integer + AVX512VL + AVX512F + Convert + + + + + Convert packed double-precision (64-bit) floating-point elements in "a" to packed 32-bit integers, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 15 +FOR j := 0 to 3 i := j*32 - dst[i+31:i] := -(a[i+31:i] * b[i+31:i]) - c[i+31:i] -ENDFOR -dst[MAX:512] := 0 + l := j*64 + IF k[j] + dst[i+31:i] := Convert_FP64_To_Int32(a[l+63:l]) + ELSE + dst[i+31:i] := src[i+31:i] + FI +ENDFOR +dst[MAX:128] := 0 - - - +
immintrin.h
 - + Floating Point - AVX512F/KNCNI - Arithmetic - - - - - Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", subtract packed elements in "c" from the negated intermediate result, and store the results in "dst". - [round_note] + Integer + AVX512VL + AVX512F + Convert + + + + Convert packed double-precision (64-bit) floating-point elements in "a" to packed 32-bit integers, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*32 - dst[i+31:i] := -(a[i+31:i] * b[i+31:i]) - c[i+31:i] -ENDFOR -dst[MAX:512] := 0 +FOR j := 0 to 3 + i := 32*j + l := 64*j + IF k[j] + dst[i+31:i] := Convert_FP64_To_Int32(a[l+63:l]) + ELSE + dst[i+31:i] := 0 + FI +ENDFOR +dst[MAX:128] := 0 - - - +
immintrin.h
 - + Floating Point - AVX512F/KNCNI - Arithmetic - - - - - Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", subtract packed elements in "c" from the negated intermediate result, and store the results in "dst" using writemask "k" (elements are copied from "c" when the corresponding mask bit is not set). + Integer + AVX512VL + AVX512F + Convert + + + + + Convert packed double-precision (64-bit) floating-point elements in "a" to packed 32-bit integers, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 15 +FOR j := 0 to 1 i := j*32 + l := j*64 IF k[j] - dst[i+31:i] := -(a[i+31:i] * b[i+31:i]) - c[i+31:i] + dst[i+31:i] := Convert_FP64_To_Int32(a[l+63:l]) ELSE - dst[i+31:i] := c[i+31:i] + dst[i+31:i] := src[i+31:i] FI -ENDFOR -dst[MAX:512] := 0 +ENDFOR +dst[MAX:64] := 0 - - - +
immintrin.h
 - + Floating Point - AVX512F/KNCNI - Arithmetic - - - - - - Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", subtract packed elements in "c" from the negated intermediate result, and store the results in "dst" using writemask "k" (elements are copied from "c" when the corresponding mask bit is not set). [round_note] + Integer + AVX512VL + AVX512F + Convert + + + + Convert packed double-precision (64-bit) floating-point elements in "a" to packed 32-bit integers, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*32 +FOR j := 0 to 1 + i := 32*j + l := 64*j IF k[j] - dst[i+31:i] := -(a[i+31:i] * b[i+31:i]) - c[i+31:i] + dst[i+31:i] := Convert_FP64_To_Int32(a[l+63:l]) ELSE - dst[i+31:i] := c[i+31:i] + dst[i+31:i] := 0 FI -ENDFOR -dst[MAX:512] := 0 +ENDFOR +dst[MAX:64] := 0 - - - +
immintrin.h
 - + Floating Point - AVX512F/KNCNI - Arithmetic - - - - - Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", subtract packed elements in "c" from the negated intermediate result, and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). + AVX512VL + AVX512F + Convert + + + + + Convert packed double-precision (64-bit) floating-point elements in "a" to packed single-precision (32-bit) floating-point elements, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*32 +FOR j := 0 to 3 + i := 32*j + l := 64*j IF k[j] - dst[i+31:i] := -(a[i+31:i] * b[i+31:i]) - c[i+31:i] + dst[i+31:i] := Convert_FP64_To_FP32(a[l+63:l]) ELSE - dst[i+31:i] := a[i+31:i] + dst[i+31:i] := src[i+31:i] FI -ENDFOR -dst[MAX:512] := 0 +ENDFOR +dst[MAX:128] := 0 - - - +
immintrin.h
 - + Floating Point - AVX512F/KNCNI - Arithmetic - - - - - - Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", subtract packed elements in "c" from the negated intermediate result, and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). - [round_note] - + AVX512VL + AVX512F + Convert + + + + Convert packed double-precision (64-bit) floating-point elements in "a" to packed single-precision (32-bit) floating-point elements, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 15 +FOR j := 0 to 3 i := j*32 + l := j*64 IF k[j] - dst[i+31:i] := -(a[i+31:i] * b[i+31:i]) - c[i+31:i] + dst[i+31:i] := Convert_FP64_To_FP32(a[l+63:l]) ELSE - dst[i+31:i] := a[i+31:i] + dst[i+31:i] := 0 FI -ENDFOR -dst[MAX:512] := 0 +ENDFOR +dst[MAX:128] := 0 - - - +
immintrin.h
 - + Floating Point + AVX512VL AVX512F - Arithmetic - - - - - Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", subtract packed elements in "c" from the negated intermediate result, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Convert + + + + + Convert packed double-precision (64-bit) floating-point elements in "a" to packed single-precision (32-bit) floating-point elements, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*32 +FOR j := 0 to 1 + i := 32*j + l := 64*j IF k[j] - dst[i+31:i] := -(a[i+31:i] * b[i+31:i]) - c[i+31:i] + dst[i+31:i] := Convert_FP64_To_FP32(a[l+63:l]) ELSE - dst[i+31:i] := 0 + dst[i+31:i] := src[i+31:i] FI -ENDFOR -dst[MAX:512] := 0 +ENDFOR +dst[MAX:64] := 0 - - - +
immintrin.h
 - + Floating Point + AVX512VL AVX512F - Arithmetic - - - - - - Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", subtract packed elements in "c" from the negated intermediate result, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - [round_note] - + Convert + + + + Convert packed double-precision (64-bit) floating-point elements in "a" to packed single-precision (32-bit) floating-point elements, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 15 +FOR j := 0 to 1 i := j*32 + l := j*64 IF k[j] - dst[i+31:i] := -(a[i+31:i] * b[i+31:i]) - c[i+31:i] + dst[i+31:i] := Convert_FP64_To_FP32(a[l+63:l]) ELSE dst[i+31:i] := 0 FI -ENDFOR -dst[MAX:512] := 0 +ENDFOR +dst[MAX:64] := 0 - - - +
immintrin.h
 - + Floating Point + Integer + AVX512VL AVX512F - Arithmetic - - - - - Multiply the lower double-precision (64-bit) floating-point elements in "a" and "b", and subtract the lower element in "c" from the negated intermediate result. Store the result in the lower element of "dst", and copy the upper element from "a" to the upper element of "dst". - [round_note] + Convert + + + Convert packed double-precision (64-bit) floating-point elements in "a" to packed unsigned 32-bit integers, and store the results in "dst". -dst[63:0] := -(a[63:0] * b[63:0]) - c[63:0] -dst[127:64] := a[127:64] -dst[MAX:128] := 0 - - - - +FOR j := 0 to 3 + i := 32*j + k := 64*j + dst[i+31:i] := Convert_FP64_To_UInt32(a[k+63:k]) +ENDFOR +dst[MAX:128] := 0 + +
immintrin.h
 - + Floating Point + Integer + AVX512VL AVX512F - Arithmetic - - - - - - Multiply the lower double-precision (64-bit) floating-point elements in "a" and "b", and subtract the lower element in "c" from the negated intermediate result. Store the result in the lower element of "dst" using writemask "k" (the element is copied from "c" when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". - [round_note] + Convert + + + + + Convert packed double-precision (64-bit) floating-point elements in "a" to packed unsigned 32-bit integers, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -IF k[0] - dst[63:0] := -(a[63:0] * b[63:0]) - c[63:0] -ELSE - dst[63:0] := c[63:0] -FI -dst[127:64] := a[127:64] +FOR j := 0 to 3 + i := j*32 + l := j*64 + IF k[j] + dst[i+31:i] := Convert_FP64_To_UInt32(a[l+63:l]) + ELSE + dst[i+31:i] := src[i+31:i] + FI +ENDFOR dst[MAX:128] := 0 - - - +
immintrin.h
 - + Floating Point + Integer + AVX512VL AVX512F - Arithmetic - - - - - Multiply the lower double-precision (64-bit) floating-point elements in "a" and "b", and subtract the lower element in "c" from the negated intermediate result. Store the result in the lower element of "dst" using writemask "k" (the element is copied from "c" when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". + Convert + + + + Convert packed double-precision (64-bit) floating-point elements in "a" to packed unsigned 32-bit integers, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -IF k[0] - dst[63:0] := -(a[63:0] * b[63:0]) - c[63:0] -ELSE - dst[63:0] := c[63:0] -FI -dst[127:64] := a[127:64] +FOR j := 0 to 3 + i := 32*j + l := 64*j + IF k[j] + dst[i+31:i] := Convert_FP64_To_UInt32(a[l+63:l]) + ELSE + dst[i+31:i] := 0 + FI +ENDFOR dst[MAX:128] := 0 - - - +
immintrin.h
 - + Floating Point + Integer + AVX512VL AVX512F - Arithmetic - - - - - - Multiply the lower double-precision (64-bit) floating-point elements in "a" and "b", and subtract the lower element in "c" from the negated intermediate result. Store the result in the lower element of "dst" using writemask "k" (the element is copied from "c" when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". - [round_note] + Convert + + + Convert packed double-precision (64-bit) floating-point elements in "a" to packed unsigned 32-bit integers, and store the results in "dst". -IF k[0] - dst[63:0] := -(a[63:0] * b[63:0]) - c[63:0] -ELSE - dst[63:0] := a[63:0] -FI -dst[127:64] := a[127:64] -dst[MAX:128] := 0 +FOR j := 0 to 1 + i := 32*j + k := 64*j + dst[i+31:i] := Convert_FP64_To_UInt32(a[k+63:k]) +ENDFOR +dst[MAX:64] := 0 - - - +
immintrin.h
 - + Floating Point + Integer + AVX512VL AVX512F - Arithmetic - - - - - Multiply the lower double-precision (64-bit) floating-point elements in "a" and "b", and subtract the lower element in "c" from the negated intermediate result. Store the result in the lower element of "dst" using writemask "k" (the element is copied from "c" when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". + Convert + + + + + Convert packed double-precision (64-bit) floating-point elements in "a" to packed unsigned 32-bit integers, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -IF k[0] - dst[63:0] := -(a[63:0] * b[63:0]) - c[63:0] -ELSE - dst[63:0] := a[63:0] -FI -dst[127:64] := a[127:64] -dst[MAX:128] := 0 +FOR j := 0 to 1 + i := j*32 + l := j*64 + IF k[j] + dst[i+31:i] := Convert_FP64_To_UInt32(a[l+63:l]) + ELSE + dst[i+31:i] := src[i+31:i] + FI +ENDFOR +dst[MAX:64] := 0 - - - +
immintrin.h
 - + Floating Point + Integer + AVX512VL AVX512F - Arithmetic - - - - - - Multiply the lower double-precision (64-bit) floating-point elements in "a" and "b", and subtract the lower element in "c" from the negated intermediate result. Store the result in "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". - [round_note] + Convert + + + + Convert packed double-precision (64-bit) floating-point elements in "a" to packed unsigned 32-bit integers, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -IF k[0] - dst[63:0] := -(a[63:0] * b[63:0]) - c[63:0] -ELSE - dst[63:0] := 0 -FI -dst[127:64] := a[127:64] -dst[MAX:128] := 0 +FOR j := 0 to 1 + i := 32*j + l := 64*j + IF k[j] + dst[i+31:i] := Convert_FP64_To_UInt32(a[l+63:l]) + ELSE + dst[i+31:i] := 0 + FI +ENDFOR +dst[MAX:64] := 0 - - - +
immintrin.h
 - + Floating Point + AVX512VL AVX512F - Arithmetic - - - - - Multiply the lower double-precision (64-bit) floating-point elements in "a" and "b", and subtract the lower element in "c" from the negated intermediate result. Store the result in "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". + Convert + + + + + Convert packed half-precision (16-bit) floating-point elements in "a" to packed single-precision (32-bit) floating-point elements, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -IF k[0] - dst[63:0] := -(a[63:0] * b[63:0]) - c[63:0] -ELSE - dst[63:0] := 0 -FI -dst[127:64] := a[127:64] -dst[MAX:128] := 0 +FOR j := 0 to 7 + i := j*32 + m := j*16 + IF k[j] + dst[i+31:i] := Convert_FP16_To_FP32(a[m+15:m]) + ELSE + dst[i+31:i] := src[i+31:i] + FI +ENDFOR +dst[MAX:256] := 0 - - - +
immintrin.h
 - + Floating Point + AVX512VL AVX512F - Arithmetic - - - - - Multiply the lower single-precision (32-bit) floating-point elements in "a" and "b", subtract the lower element in "c" from the negated intermediate result, store the result in the lower element of "dst", and copy the upper element from "a" to the upper element of "dst". - [round_note] + Convert + + + + Convert packed half-precision (16-bit) floating-point elements in "a" to packed single-precision (32-bit) floating-point elements, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -dst[31:0] := -(a[31:0] * b[31:0]) - c[31:0] -dst[127:32] := a[127:32] -dst[MAX:128] := 0 +FOR j := 0 to 7 + i := j*32 + m := j*16 + IF k[j] + dst[i+31:i] := Convert_FP16_To_FP32(a[m+15:m]) + ELSE + dst[i+31:i] := 0 + FI +ENDFOR +dst[MAX:256] := 0 - - - +
immintrin.h
 - + Floating Point + AVX512VL AVX512F - Arithmetic - - - - - - Multiply the lower single-precision (32-bit) floating-point elements in "a" and "b", subtract the lower element in "c" from the negated intermediate result, store the result in the lower element of "dst", and copy the upper element from "a" to the upper element of "dst" using writemask "k" (elements are copied from "c" when the corresponding mask bit is not set). - [round_note] + Convert + + + + + Convert packed half-precision (16-bit) floating-point elements in "a" to packed single-precision (32-bit) floating-point elements, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -IF k[0] - dst[31:0] := -(a[31:0] * b[31:0]) - c[31:0] -ELSE - dst[31:0] := c[31:0] -FI -dst[127:32] := a[127:32] +FOR j := 0 to 3 + i := j*32 + m := j*16 + IF k[j] + dst[i+31:i] := Convert_FP16_To_FP32(a[m+15:m]) + ELSE + dst[i+31:i] := src[i+31:i] + FI +ENDFOR dst[MAX:128] := 0 - - - +
immintrin.h
 - + Floating Point + AVX512VL AVX512F - Arithmetic - - - - - Multiply the lower single-precision (32-bit) floating-point elements in "a" and "b", and subtract the lower element in "c" from the negated intermediate result. Store the result in the lower element of "dst", and copy the upper element from "a" to the upper element of "dst" using writemask "k" (elements are copied from "c" when the corresponding mask bit is not set). + Convert + + + + Convert packed half-precision (16-bit) floating-point elements in "a" to packed single-precision (32-bit) floating-point elements, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -IF k[0] - dst[31:0] := -(a[31:0] * b[31:0]) - c[31:0] -ELSE - dst[31:0] := c[31:0] -FI -dst[127:32] := a[127:32] +FOR j := 0 to 3 + i := j*32 + m := j*16 + IF k[j] + dst[i+31:i] := Convert_FP16_To_FP32(a[m+15:m]) + ELSE + dst[i+31:i] := 0 + FI +ENDFOR dst[MAX:128] := 0 - - - +
immintrin.h
 - + Floating Point + Integer + AVX512VL AVX512F - Arithmetic - - - - - - Multiply the lower single-precision (32-bit) floating-point elements in "a" and "b", and subtract the lower element in "c" from the negated intermediate result. Store the result in the lower element of "dst" using writemask "k" (the element is copied from "c" when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". - [round_note] + Convert + + + + + Convert packed single-precision (32-bit) floating-point elements in "a" to packed 32-bit integers, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -IF k[0] - dst[31:0] := -(a[31:0] * b[31:0]) - c[31:0] -ELSE - dst[31:0] := a[31:0] -FI -dst[127:32] := a[127:32] -dst[MAX:128] := 0 +FOR j := 0 to 7 + i := j*32 + IF k[j] + dst[i+31:i] := Convert_FP32_To_Int32(a[i+31:i]) + ELSE + dst[i+31:i] := src[i+31:i] + FI +ENDFOR +dst[MAX:256] := 0 - - - +
immintrin.h
 - + Floating Point + Integer + AVX512VL AVX512F - Arithmetic - - - - - Multiply the lower single-precision (32-bit) floating-point elements in "a" and "b", and subtract the lower element in "c" from the negated intermediate result. Store the result in the lower element of "dst" using writemask "k" (the element is copied from "c" when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". + Convert + + + + Convert packed single-precision (32-bit) floating-point elements in "a" to packed 32-bit integers, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -IF k[0] - dst[31:0] := -(a[31:0] * b[31:0]) - c[31:0] -ELSE - dst[31:0] := a[31:0] -FI -dst[127:32] := a[127:32] -dst[MAX:128] := 0 +FOR j := 0 to 7 + i := 32*j + IF k[j] + dst[i+31:i] := Convert_FP32_To_Int32(a[i+31:i]) + ELSE + dst[i+31:i] := 0 + FI +ENDFOR +dst[MAX:256] := 0 - - - +
immintrin.h
 - + Floating Point + Integer + AVX512VL AVX512F - Arithmetic - - - - - - Multiply the lower single-precision (32-bit) floating-point elements in "a" and "b", and subtract the lower element in "c" from the negated intermediate result. Store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". - [round_note] + Convert + + + + + Convert packed single-precision (32-bit) floating-point elements in "a" to packed 32-bit integers, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -IF k[0] - dst[31:0] := -(a[31:0] * b[31:0]) - c[31:0] -ELSE - dst[31:0] := 0 -FI -dst[127:32] := a[127:32] +FOR j := 0 to 3 + i := j*32 + IF k[j] + dst[i+31:i] := Convert_FP32_To_Int32(a[i+31:i]) + ELSE + dst[i+31:i] := src[i+31:i] + FI +ENDFOR dst[MAX:128] := 0 - - - +
immintrin.h
 - + Floating Point + Integer + AVX512VL AVX512F - Arithmetic - - - - - Multiply the lower single-precision (32-bit) floating-point elements in "a" and "b", and subtract the lower element in "c" from the negated intermediate result. Store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". + Convert + + + + Convert packed single-precision (32-bit) floating-point elements in "a" to packed 32-bit integers, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -IF k[0] - dst[31:0] := -(a[31:0] * b[31:0]) - c[31:0] -ELSE - dst[31:0] := 0 -FI -dst[127:32] := a[127:32] +FOR j := 0 to 3 + i := 32*j + IF k[j] + dst[i+31:i] := Convert_FP32_To_Int32(a[i+31:i]) + ELSE + dst[i+31:i] := 0 + FI +ENDFOR dst[MAX:128] := 0 - - - +
immintrin.h
 - + Floating Point + AVX512VL AVX512F - Load - - - - - Gather double-precision (64-bit) floating-point elements from memory using 32-bit indices. 64-bit elements are loaded from addresses starting at "base_addr" and offset by each 32-bit element in "vindex" (each index is scaled by the factor in "scale"). Gathered elements are merged into "dst". "scale" should be 1, 2, 4 or 8. - + Convert + + + + + + Convert packed single-precision (32-bit) floating-point elements in "a" to packed half-precision (16-bit) floating-point elements, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). [round_imm_note] FOR j := 0 to 7 - i := j*64 - m := j*32 - dst[i+63:i] := MEM[base_addr + SignExtend(vindex[m+31:m])*scale] + i := 16*j + l := 32*j + IF k[j] + dst[i+15:i] := Convert_FP32_To_FP16(a[l+31:l]) + ELSE + dst[i+15:i] := src[i+15:i] + FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point + AVX512VL AVX512F - Load - - - - - - - Gather double-precision (64-bit) floating-point elements from memory using 32-bit indices. 64-bit elements are loaded from addresses starting at "base_addr" and offset by each 32-bit element in "vindex" (each index is scaled by the factor in "scale"). Gathered elements are merged into "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). "scale" should be 1, 2, 4 or 8. - + Convert + + + + + + Convert packed single-precision (32-bit) floating-point elements in "a" to packed half-precision (16-bit) floating-point elements, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). [round_imm_note] FOR j := 0 to 7 - i := j*64 - m := j*32 + i := 16*j + l := 32*j IF k[j] - dst[i+63:i] := MEM[base_addr + SignExtend(vindex[m+31:m])*scale] - k[j] := 0 + dst[i+15:i] := Convert_FP32_To_FP16(a[l+31:l]) ELSE - dst[i+63:i] := src[i+63:i] + dst[i+15:i] := src[i+15:i] FI ENDFOR -k[MAX:8] := 0 -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point - AVX512F/KNCNI - Load - - - - Gather single-precision (32-bit) floating-point elements from memory using 32-bit indices. 32-bit elements are loaded from addresses starting at "base_addr" and offset by each 32-bit element in "vindex" (each index is scaled by the factor in "scale"). Gathered elements are merged into "dst". "scale" should be 1, 2, 4 or 8. + AVX512VL + AVX512F + Convert + + + + + Convert packed single-precision (32-bit) floating-point elements in "a" to packed half-precision (16-bit) floating-point elements, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). [round_imm_note] -FOR j := 0 to 15 - i := j*32 - dst[i+31:i] := MEM[base_addr + SignExtend(vindex[i+31:i])*scale] +FOR j := 0 to 7 + i := 16*j + l := 32*j + IF k[j] + dst[i+15:i] := Convert_FP32_To_FP16(a[l+31:l]) + ELSE + dst[i+15:i] := 0 + FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point - AVX512F/KNCNI - Load - - - - - - - Gather single-precision (32-bit) floating-point elements from memory using 32-bit indices. 32-bit elements are loaded from addresses starting at "base_addr" and offset by each 32-bit element in "vindex" (each index is scaled by the factor in "scale"). Gathered elements are merged into "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). "scale" should be 1, 2, 4 or 8. - + AVX512VL + AVX512F + Convert + + + + + Convert packed single-precision (32-bit) floating-point elements in "a" to packed half-precision (16-bit) floating-point elements, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). [round_imm_note] -FOR j := 0 to 15 - i := j*32 +FOR j := 0 to 7 + i := 16*j + l := 32*j IF k[j] - dst[i+31:i] := MEM[base_addr + SignExtend(vindex[i+31:i])*scale] - k[j] := 0 + dst[i+15:i] := Convert_FP32_To_FP16(a[l+31:l]) ELSE - dst[i+31:i] := src[i+31:i] + dst[i+15:i] := 0 FI ENDFOR -k[MAX:16] := 0 -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point + AVX512VL AVX512F - Load - - - - - Gather double-precision (64-bit) floating-point elements from memory using 64-bit indices. 64-bit elements are loaded from addresses starting at "base_addr" and offset by each 64-bit element in "vindex" (each index is scaled by the factor in "scale"). Gathered elements are merged into "dst". "scale" should be 1, 2, 4 or 8. - + Convert + + + + + + Convert packed single-precision (32-bit) floating-point elements in "a" to packed half-precision (16-bit) floating-point elements, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). [round_imm_note] -FOR j := 0 to 7 - i := j*64 - dst[i+63:i] := MEM[base_addr + SignExtend(vindex[i+63:i])*scale] +FOR j := 0 to 3 + i := 16*j + l := 32*j + IF k[j] + dst[i+15:i] := Convert_FP32_To_FP16(a[l+31:l]) + ELSE + dst[i+15:i] := src[i+15:i] + FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:64] := 0 - +
immintrin.h
 - + Floating Point + AVX512VL AVX512F - Load - - - - - - - Gather double-precision (64-bit) floating-point elements from memory using 64-bit indices. 64-bit elements are loaded from addresses starting at "base_addr" and offset by each 64-bit element in "vindex" (each index is scaled by the factor in "scale"). Gathered elements are merged into "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). "scale" should be 1, 2, 4 or 8. - + Convert + + + + + + Convert packed single-precision (32-bit) floating-point elements in "a" to packed half-precision (16-bit) floating-point elements, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). [round_imm_note] -FOR j := 0 to 7 - i := j*64 +FOR j := 0 to 3 + i := 16*j + l := 32*j IF k[j] - dst[i+63:i] := MEM[base_addr + SignExtend(vindex[i+63:i])*scale] - k[j] := 0 + dst[i+15:i] := Convert_FP32_To_FP16(a[l+31:l]) ELSE - dst[i+63:i] := src[i+63:i] + dst[i+15:i] := src[i+15:i] FI ENDFOR -k[MAX:8] := 0 -dst[MAX:512] := 0 +dst[MAX:64] := 0 - +
immintrin.h
 - + Floating Point + AVX512VL AVX512F - Load - - - - - Gather single-precision (32-bit) floating-point elements from memory using 64-bit indices. 32-bit elements are loaded from addresses starting at "base_addr" and offset by each 64-bit element in "vindex" (each index is scaled by the factor in "scale"). Gathered elements are merged into "dst". "scale" should be 1, 2, 4 or 8. - + Convert + + + + + Convert packed single-precision (32-bit) floating-point elements in "a" to packed half-precision (16-bit) floating-point elements, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). [round_imm_note] -FOR j := 0 to 7 - i := j*32 - m := j*64 - dst[i+31:i] := MEM[base_addr + SignExtend(vindex[i+63:i])*scale] +FOR j := 0 to 3 + i := 16*j + l := 32*j + IF k[j] + dst[i+15:i] := Convert_FP32_To_FP16(a[l+31:l]) + ELSE + dst[i+15:i] := 0 + FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:64] := 0 - +
immintrin.h
 - + Floating Point + AVX512VL AVX512F - Load - - - - - - - Gather single-precision (32-bit) floating-point elements from memory using 64-bit indices. 32-bit elements are loaded from addresses starting at "base_addr" and offset by each 64-bit element in "vindex" (each index is scaled by the factor in "scale"). Gathered elements are merged into "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). "scale" should be 1, 2, 4 or 8. - + Convert + + + + + Convert packed single-precision (32-bit) floating-point elements in "a" to packed half-precision (16-bit) floating-point elements, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). [round_imm_note] -FOR j := 0 to 7 - i := j*32 - m := j*64 +FOR j := 0 to 3 + i := 16*j + l := 32*j IF k[j] - dst[i+31:i] := MEM[base_addr + SignExtend(vindex[i+63:i])*scale] - k[j] := 0 + dst[i+15:i] := Convert_FP32_To_FP16(a[l+31:l]) ELSE - dst[i+31:i] := src[i+31:i] + dst[i+15:i] := 0 FI ENDFOR -k[MAX:8] := 0 -dst[MAX:256] := 0 +dst[MAX:64] := 0 - +
immintrin.h
 - + Floating Point - AVX512F/KNCNI - Miscellaneous - - Convert the exponent of each packed double-precision (64-bit) floating-point element in "a" to a double-precision (64-bit) floating-point number representing the integer exponent, and store the results in "dst". This intrinsic essentially calculates "floor(log2(x))" for each element. + Integer + AVX512VL + AVX512F + Convert + + + Convert packed single-precision (32-bit) floating-point elements in "a" to packed unsigned 32-bit integers, and store the results in "dst". FOR j := 0 to 7 - i := j*64 - dst[i+63:i] := ConvertExpFP64(a[i+63:i]) + i := 32*j + dst[i+31:i] := Convert_FP32_To_UInt32(a[i+31:i]) ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Floating Point - AVX512F/KNCNI - Miscellaneous - - - Convert the exponent of each packed double-precision (64-bit) floating-point element in "a" to a double-precision (64-bit) floating-point number representing the integer exponent, and store the results in "dst". This intrinsic essentially calculates "floor(log2(x))" for each element. - [round_note] - + Integer + AVX512VL + AVX512F + Convert + + + + + Convert packed single-precision (32-bit) floating-point elements in "a" to packed unsigned 32-bit integers, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). FOR j := 0 to 7 - i := j*64 - dst[i+63:i] := ConvertExpFP64(a[i+63:i]) + i := 32*j + IF k[j] + dst[i+31:i] := Convert_FP32_To_UInt32(a[i+31:i]) + ELSE + dst[i+31:i] := src[i+31:i] + FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Floating Point - AVX512F/KNCNI - Miscellaneous - - - - Convert the exponent of each packed double-precision (64-bit) floating-point element in "a" to a double-precision (64-bit) floating-point number representing the integer exponent, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). This intrinsic essentially calculates "floor(log2(x))" for each element. + Integer + AVX512VL + AVX512F + Convert + + + + Convert packed single-precision (32-bit) floating-point elements in "a" to packed unsigned 32-bit integers, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). FOR j := 0 to 7 - i := j*64 + i := 32*j IF k[j] - dst[i+63:i] := ConvertExpFP64(a[i+63:i]) + dst[i+31:i] := Convert_FP32_To_UInt32(a[i+31:i]) ELSE - dst[i+63:i] := src[i+63:i] + dst[i+31:i] := 0 FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Floating Point - AVX512F/KNCNI - Miscellaneous - - - - - Convert the exponent of each packed double-precision (64-bit) floating-point element in "a" to a double-precision (64-bit) floating-point number representing the integer exponent, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). This intrinsic essentially calculates "floor(log2(x))" for each element. - [round_note] - + Integer + AVX512VL + AVX512F + Convert + + + Convert packed single-precision (32-bit) floating-point elements in "a" to packed unsigned 32-bit integers, and store the results in "dst". -FOR j := 0 to 7 - i := j*64 +FOR j := 0 to 3 + i := 32*j + dst[i+31:i] := Convert_FP32_To_UInt32(a[i+31:i]) +ENDFOR +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Floating Point + Integer + AVX512VL + AVX512F + Convert + + + + + Convert packed single-precision (32-bit) floating-point elements in "a" to packed unsigned 32-bit integers, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +FOR j := 0 to 3 + i := 32*j IF k[j] - dst[i+63:i] := ConvertExpFP64(a[i+63:i]) + dst[i+31:i] := Convert_FP32_To_UInt32(a[i+31:i]) ELSE - dst[i+63:i] := src[i+63:i] + dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point + Integer + AVX512VL AVX512F - Miscellaneous - - - Convert the exponent of each packed double-precision (64-bit) floating-point element in "a" to a double-precision (64-bit) floating-point number representing the integer exponent, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). This intrinsic essentially calculates "floor(log2(x))" for each element. + Convert + + + + Convert packed single-precision (32-bit) floating-point elements in "a" to packed unsigned 32-bit integers, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := j*64 +FOR j := 0 to 3 + i := 32*j IF k[j] - dst[i+63:i] := ConvertExpFP64(a[i+63:i]) + dst[i+31:i] := Convert_FP32_To_UInt32(a[i+31:i]) ELSE - dst[i+63:i] := 0 + dst[i+31:i] := 0 FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point + Integer + AVX512VL AVX512F - Miscellaneous - - - - Convert the exponent of each packed double-precision (64-bit) floating-point element in "a" to a double-precision (64-bit) floating-point number representing the integer exponent, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). This intrinsic essentially calculates "floor(log2(x))" for each element. - [round_note] - + Convert + + + + + Convert packed double-precision (64-bit) floating-point elements in "a" to packed 32-bit integers with truncation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := j*64 +FOR j := 0 to 3 + i := 32*j + l := 64*j IF k[j] - dst[i+63:i] := ConvertExpFP64(a[i+63:i]) + dst[i+31:i] := Convert_FP64_To_Int32_Truncate(a[l+63:l]) ELSE - dst[i+63:i] := 0 + dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point - AVX512F/KNCNI - Miscellaneous - - Convert the exponent of each packed single-precision (32-bit) floating-point element in "a" to a single-precision (32-bit) floating-point number representing the integer exponent, and store the results in "dst". This intrinsic essentially calculates "floor(log2(x))" for each element. + Integer + AVX512VL + AVX512F + Convert + + + + Convert packed double-precision (64-bit) floating-point elements in "a" to packed 32-bit integers with truncation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*32 - dst[i+31:i] := ConvertExpFP32(a[i+31:i]) +FOR j := 0 to 3 + i := 32*j + l := 64*j + IF k[j] + dst[i+31:i] := Convert_FP64_To_Int32_Truncate(a[l+63:l]) + ELSE + dst[i+31:i] := 0 + FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point - AVX512F/KNCNI - Miscellaneous - - - Convert the exponent of each packed single-precision (32-bit) floating-point element in "a" to a single-precision (32-bit) floating-point number representing the integer exponent, and store the results in "dst". This intrinsic essentially calculates "floor(log2(x))" for each element. - [round_note] - + Integer + AVX512VL + AVX512F + Convert + + + + + Convert packed double-precision (64-bit) floating-point elements in "a" to packed 32-bit integers with truncation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*32 - dst[i+31:i] := ConvertExpFP32(a[i+31:i]) +FOR j := 0 to 1 + i := 32*j + l := 64*j + IF k[j] + dst[i+31:i] := Convert_FP64_To_Int32_Truncate(a[l+63:l]) + ELSE + dst[i+31:i] := src[i+31:i] + FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:64] := 0 - +
immintrin.h
 - + Floating Point - AVX512F/KNCNI - Miscellaneous - - - - Convert the exponent of each packed single-precision (32-bit) floating-point element in "a" to a single-precision (32-bit) floating-point number representing the integer exponent, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). This intrinsic essentially calculates "floor(log2(x))" for each element. + Integer + AVX512VL + AVX512F + Convert + + + + Convert packed double-precision (64-bit) floating-point elements in "a" to packed 32-bit integers with truncation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*32 +FOR j := 0 to 1 + i := 32*j + l := 64*j IF k[j] - dst[i+31:i] := ConvertExpFP32(a[i+31:i]) + dst[i+31:i] := Convert_FP64_To_Int32_Truncate(a[l+63:l]) ELSE - dst[i+31:i] := src[i+31:i] + dst[i+31:i] := 0 FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:64] := 0 - +
immintrin.h
 - + Floating Point - AVX512F/KNCNI - Miscellaneous - - - - - Convert the exponent of each packed single-precision (32-bit) floating-point element in "a" to a single-precision (32-bit) floating-point number representing the integer exponent, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). This intrinsic essentially calculates "floor(log2(x))" for each element. - [round_note] - + Integer + AVX512VL + AVX512F + Convert + + + Convert packed double-precision (64-bit) floating-point elements in "a" to packed unsigned 32-bit integers with truncation, and store the results in "dst". -FOR j := 0 to 15 - i := j*32 - IF k[j] - dst[i+31:i] := ConvertExpFP32(a[i+31:i]) - ELSE - dst[i+31:i] := src[i+31:i] - FI +FOR j := 0 to 3 + i := 32*j + k := 64*j + dst[i+31:i] := Convert_FP64_To_UInt32_Truncate(a[k+63:k]) ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point + Integer + AVX512VL AVX512F - Miscellaneous - - - Convert the exponent of each packed single-precision (32-bit) floating-point element in "a" to a single-precision (32-bit) floating-point number representing the integer exponent, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). This intrinsic essentially calculates "floor(log2(x))" for each element. + Convert + + + + + Convert packed double-precision (64-bit) floating-point elements in "a" to packed unsigned 32-bit integers with truncation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*32 +FOR j := 0 to 3 + i := 32*j + l := 64*j IF k[j] - dst[i+31:i] := ConvertExpFP32(a[i+31:i]) + dst[i+31:i] := Convert_FP64_To_UInt32_Truncate(a[l+63:l]) ELSE - dst[i+31:i] := 0 + dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point + Integer + AVX512VL AVX512F - Miscellaneous - - - - Convert the exponent of each packed single-precision (32-bit) floating-point element in "a" to a single-precision (32-bit) floating-point number representing the integer exponent, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). This intrinsic essentially calculates "floor(log2(x))" for each element. - [round_note] - + Convert + + + + Convert packed double-precision (64-bit) floating-point elements in "a" to packed unsigned 32-bit integers with truncation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*32 +FOR j := 0 to 3 + i := 32*j + l := 64*j IF k[j] - dst[i+31:i] := ConvertExpFP32(a[i+31:i]) + dst[i+31:i] := Convert_FP64_To_UInt32_Truncate(a[l+63:l]) ELSE dst[i+31:i] := 0 FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point + Integer + AVX512VL AVX512F - Miscellaneous - - - - Convert the exponent of the lower double-precision (64-bit) floating-point element in "b" to a double-precision (64-bit) floating-point number representing the integer exponent, store the result in the lower element of "dst", and copy the upper element from "a" to the upper element of "dst". This intrinsic essentially calculates "floor(log2(x))" for the lower element. - [round_note] - + Convert + + + Convert packed double-precision (64-bit) floating-point elements in "a" to packed unsigned 32-bit integers with truncation, and store the results in "dst". -dst[63:0] := ConvertExpFP64(b[63:0]) -dst[127:64] := a[127:64] -dst[MAX:128] := 0 +FOR j := 0 to 1 + i := 32*j + k := 64*j + dst[i+31:i] := Convert_FP64_To_UInt32_Truncate(a[k+63:k]) +ENDFOR +dst[MAX:64] := 0 - +
immintrin.h
 - + Floating Point + Integer + AVX512VL AVX512F - Miscellaneous - - - Convert the exponent of the lower double-precision (64-bit) floating-point element in "b" to a double-precision (64-bit) floating-point number representing the integer exponent, store the result in the lower element of "dst", and copy the upper element from "a" to the upper element of "dst". This intrinsic essentially calculates "floor(log2(x))" for the lower element. + Convert + + + + + Convert packed double-precision (64-bit) floating-point elements in "a" to packed unsigned 32-bit integers with truncation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -dst[63:0] := ConvertExpFP64(b[63:0]) -dst[127:64] := a[127:64] -dst[MAX:128] := 0 +FOR j := 0 to 1 + i := 32*j + l := 64*j + IF k[j] + dst[i+31:i] := Convert_FP64_To_UInt32_Truncate(a[l+63:l]) + ELSE + dst[i+31:i] := src[i+31:i] + FI +ENDFOR +dst[MAX:64] := 0 - +
immintrin.h
 - + Floating Point + Integer + AVX512VL AVX512F - Miscellaneous - - - - - - Convert the exponent of the lower double-precision (64-bit) floating-point element in "b" to a double-precision (64-bit) floating-point number representing the integer exponent, store the result in the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". This intrinsic essentially calculates "floor(log2(x))" for the lower element. - [round_note] - + Convert + + + + Convert packed double-precision (64-bit) floating-point elements in "a" to packed unsigned 32-bit integers with truncation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -IF k[0] - dst[63:0] := ConvertExpFP64(b[63:0]) -ELSE - dst[63:0] := src[63:0] -FI -dst[127:64] := a[127:64] -dst[MAX:128] := 0 +FOR j := 0 to 1 + i := 32*j + l := 64*j + IF k[j] + dst[i+31:i] := Convert_FP64_To_UInt32_Truncate(a[l+63:l]) + ELSE + dst[i+31:i] := 0 + FI +ENDFOR +dst[MAX:64] := 0 - +
immintrin.h
 - + Floating Point + Integer + AVX512VL AVX512F - Miscellaneous - - - - - Convert the exponent of the lower double-precision (64-bit) floating-point element in "b" to a double-precision (64-bit) floating-point number representing the integer exponent, store the result in the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". This intrinsic essentially calculates "floor(log2(x))" for the lower element. + Convert + + + + + Convert packed single-precision (32-bit) floating-point elements in "a" to packed 32-bit integers with truncation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -IF k[0] - dst[63:0] := ConvertExpFP64(b[63:0]) -ELSE - dst[63:0] := src[63:0] -FI -dst[127:64] := a[127:64] -dst[MAX:128] := 0 +FOR j := 0 to 7 + i := 32*j + IF k[j] + dst[i+31:i] := Convert_FP32_To_Int32_Truncate(a[i+31:i]) + ELSE + dst[i+31:i] := src[i+31:i] + FI +ENDFOR +dst[MAX:256] := 0 - +
immintrin.h
 - + Floating Point + Integer + AVX512VL AVX512F - Miscellaneous - - - - - Convert the exponent of the lower double-precision (64-bit) floating-point element in "b" to a double-precision (64-bit) floating-point number representing the integer exponent, store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". This intrinsic essentially calculates "floor(log2(x))" for the lower element. - [round_note] - + Convert + + + + Convert packed single-precision (32-bit) floating-point elements in "a" to packed 32-bit integers with truncation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -IF k[0] - dst[63:0] := ConvertExpFP64(b[63:0]) -ELSE - dst[63:0] := 0 -FI -dst[127:64] := a[127:64] -dst[MAX:128] := 0 +FOR j := 0 to 7 + i := 32*j + IF k[j] + dst[i+31:i] := Convert_FP32_To_Int32_Truncate(a[i+31:i]) + ELSE + dst[i+31:i] := 0 + FI +ENDFOR +dst[MAX:256] := 0 - +
immintrin.h
 - + Floating Point + Integer + AVX512VL AVX512F - Miscellaneous - - - - Convert the exponent of the lower double-precision (64-bit) floating-point element in "b" to a double-precision (64-bit) floating-point number representing the integer exponent, store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". This intrinsic essentially calculates "floor(log2(x))" for the lower element. + Convert + + + + + Convert packed single-precision (32-bit) floating-point elements in "a" to packed 32-bit integers with truncation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -IF k[0] - dst[63:0] := ConvertExpFP64(b[63:0]) -ELSE - dst[63:0] := 0 -FI -dst[127:64] := a[127:64] +FOR j := 0 to 3 + i := 32*j + IF k[j] + dst[i+31:i] := Convert_FP32_To_Int32_Truncate(a[i+31:i]) + ELSE + dst[i+31:i] := src[i+31:i] + FI +ENDFOR dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point + Integer + AVX512VL AVX512F - Miscellaneous - - - - Convert the exponent of the lower single-precision (32-bit) floating-point element in "b" to a single-precision (32-bit) floating-point number representing the integer exponent, store the result in the lower element of "dst", and copy the upper 3 packed elements from "a" to the upper elements of "dst". This intrinsic essentially calculates "floor(log2(x))" for the lower element. - [round_note] - + Convert + + + + Convert packed single-precision (32-bit) floating-point elements in "a" to packed 32-bit integers with truncation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -dst[31:0] := ConvertExpFP32(b[31:0]) -dst[127:32] := a[127:32] +FOR j := 0 to 3 + i := 32*j + IF k[j] + dst[i+31:i] := Convert_FP32_To_Int32_Truncate(a[i+31:i]) + ELSE + dst[i+31:i] := 0 + FI +ENDFOR dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point + Integer + AVX512VL AVX512F - Miscellaneous - - - Convert the exponent of the lower single-precision (32-bit) floating-point element in "b" to a single-precision (32-bit) floating-point number representing the integer exponent, store the result in the lower element of "dst", and copy the upper 3 packed elements from "a" to the upper elements of "dst". This intrinsic essentially calculates "floor(log2(x))" for the lower element. + Convert + + + Convert packed single-precision (32-bit) floating-point elements in "a" to packed unsigned 32-bit integers with truncation, and store the results in "dst". -dst[31:0] := ConvertExpFP32(b[31:0]) -dst[127:32] := a[127:32] -dst[MAX:128] := 0 +FOR j := 0 to 7 + i := 32*j + dst[i+31:i] := Convert_FP32_To_UInt32_Truncate(a[i+31:i]) +ENDFOR +dst[MAX:256] := 0 - +
immintrin.h
 - + Floating Point + Integer + AVX512VL AVX512F - Miscellaneous - - - - - - Convert the exponent of the lower single-precision (32-bit) floating-point element in "b" to a single-precision (32-bit) floating-point number representing the integer exponent, store the result in the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". This intrinsic essentially calculates "floor(log2(x))" for the lower element. - [round_note] - + Convert + + + + + Convert packed double-precision (32-bit) floating-point elements in "a" to packed unsigned 32-bit integers with truncation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -IF k[0] - dst[31:0] := ConvertExpFP32(b[31:0]) -ELSE - dst[31:0] := src[31:0] -FI -dst[127:32] := a[127:32] -dst[MAX:128] := 0 +FOR j := 0 to 7 + i := 32*j + IF k[j] + dst[i+31:i] := Convert_FP64_To_UInt32_Truncate(a[i+31:i]) + ELSE + dst[i+31:i] := src[i+31:i] + FI +ENDFOR +dst[MAX:256] := 0 - +
immintrin.h
 - + Floating Point + Integer + AVX512VL AVX512F - Miscellaneous - - - - - Convert the exponent of the lower single-precision (32-bit) floating-point element in "b" to a single-precision (32-bit) floating-point number representing the integer exponent, store the result in the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". This intrinsic essentially calculates "floor(log2(x))" for the lower element. + Convert + + + + Convert packed double-precision (32-bit) floating-point elements in "a" to packed unsigned 32-bit integers with truncation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -IF k[0] - dst[31:0] := ConvertExpFP32(b[31:0]) -ELSE - dst[31:0] := src[31:0] -FI -dst[127:32] := a[127:32] -dst[MAX:128] := 0 +FOR j := 0 to 7 + i := 32*j + IF k[j] + dst[i+31:i] := Convert_FP64_To_UInt32_Truncate(a[i+31:i]) + ELSE + dst[i+31:i] := 0 + FI +ENDFOR +dst[MAX:256] := 0 - +
immintrin.h
 - + Floating Point + Integer + AVX512VL AVX512F - Miscellaneous - - - - - Convert the exponent of the lower single-precision (32-bit) floating-point element in "b" to a single-precision (32-bit) floating-point number representing the integer exponent, store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". This intrinsic essentially calculates "floor(log2(x))" for the lower element. - [round_note] - + Convert + + + Convert packed single-precision (32-bit) floating-point elements in "a" to packed unsigned 32-bit integers with truncation, and store the results in "dst". -IF k[0] - dst[31:0] := ConvertExpFP32(b[31:0]) -ELSE - dst[31:0] := 0 -FI -dst[127:32] := a[127:32] +FOR j := 0 to 3 + i := 32*j + dst[i+31:i] := Convert_FP32_To_UInt32_Truncate(a[i+31:i]) +ENDFOR dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point + Integer + AVX512VL AVX512F - Miscellaneous - - - - Convert the exponent of the lower single-precision (32-bit) floating-point element in "b" to a single-precision (32-bit) floating-point number representing the integer exponent, store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". This intrinsic essentially calculates "floor(log2(x))" for the lower element. + Convert + + + + + Convert packed double-precision (32-bit) floating-point elements in "a" to packed unsigned 32-bit integers with truncation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -IF k[0] - dst[31:0] := ConvertExpFP32(b[31:0]) -ELSE - dst[31:0] := 0 -FI -dst[127:32] := a[127:32] +FOR j := 0 to 3 + i := 32*j + IF k[j] + dst[i+31:i] := Convert_FP64_To_UInt32_Truncate(a[i+31:i]) + ELSE + dst[i+31:i] := src[i+31:i] + FI +ENDFOR dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point - AVX512F/KNCNI - Miscellaneous - - - - Normalize the mantissas of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst". This intrinsic essentially calculates "±(2^k)*|x.significand|", where "k" depends on the interval range defined by "interv" and the sign depends on "sc" and the source sign. - [getmant_note] + Integer + AVX512VL + AVX512F + Convert + + + + Convert packed double-precision (32-bit) floating-point elements in "a" to packed unsigned 32-bit integers with truncation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := j*64 - dst[i+63:i] := GetNormalizedMantissa(a[i+63:i], sc, interv) +FOR j := 0 to 3 + i := 32*j + IF k[j] + dst[i+31:i] := Convert_FP64_To_UInt32_Truncate(a[i+31:i]) + ELSE + dst[i+31:i] := 0 + FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point - AVX512F/KNCNI - Miscellaneous - - - - - Normalize the mantissas of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst". This intrinsic essentially calculates "±(2^k)*|x.significand|", where "k" depends on the interval range defined by "interv" and the sign depends on "sc" and the source sign. - [getmant_note][round_note] + Integer + AVX512VL + AVX512F + Convert + + + Convert packed unsigned 32-bit integers in "a" to packed double-precision (64-bit) floating-point elements, and store the results in "dst". -FOR j := 0 to 7 +FOR j := 0 to 3 i := j*64 - dst[i+63:i] := GetNormalizedMantissa(a[i+63:i], sc, interv) + l := j*32 + dst[i+63:i] := Convert_Int32_To_FP64(a[l+31:l]) ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Floating Point - AVX512F/KNCNI - Miscellaneous - - - - - - Normalize the mantissas of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). This intrinsic essentially calculates "±(2^k)*|x.significand|", where "k" depends on the interval range defined by "interv" and the sign depends on "sc" and the source sign. - [getmant_note] + Integer + AVX512VL + AVX512F + Convert + + + + + Convert packed unsigned 32-bit integers in "a" to packed double-precision (64-bit) floating-point elements, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 7 +FOR j := 0 to 3 i := j*64 + l := j*32 IF k[j] - dst[i+63:i] := GetNormalizedMantissa(a[i+63:i], sc, interv) + dst[i+63:i] := Convert_Int32_To_FP64(a[l+31:l]) ELSE dst[i+63:i] := src[i+63:i] - FI + FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Floating Point - AVX512F/KNCNI - Miscellaneous - - - - - - - Normalize the mantissas of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). This intrinsic essentially calculates "±(2^k)*|x.significand|", where "k" depends on the interval range defined by "interv" and the sign depends on "sc" and the source sign. - [getmant_note][round_note] + Integer + AVX512VL + AVX512F + Convert + + + + Convert packed unsigned 32-bit integers in "a" to packed double-precision (64-bit) floating-point elements, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 +FOR j := 0 to 3 i := j*64 + l := j*32 IF k[j] - dst[i+63:i] := GetNormalizedMantissa(a[i+63:i], sc, interv) + dst[i+63:i] := Convert_Int64_To_FP64(a[l+31:l]) ELSE - dst[i+63:i] := src[i+63:i] - FI + dst[i+63:i] := 0 + FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Floating Point + Integer + AVX512VL AVX512F - Miscellaneous - - - - - Normalize the mantissas of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). This intrinsic essentially calculates "±(2^k)*|x.significand|", where "k" depends on the interval range defined by "interv" and the sign depends on "sc" and the source sign. - [getmant_note] + Convert + + + Convert packed unsigned 32-bit integers in "a" to packed double-precision (64-bit) floating-point elements, and store the results in "dst". -FOR j := 0 to 7 +FOR j := 0 to 1 + i := j*64 + l := j*32 + dst[i+63:i] := Convert_Int64_To_FP64(a[l+31:l]) +ENDFOR +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Floating Point + Integer + AVX512VL + AVX512F + Convert + + + + + Convert packed unsigned 32-bit integers in "a" to packed double-precision (64-bit) floating-point elements, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +FOR j := 0 to 1 i := j*64 + l := j*32 IF k[j] - dst[i+63:i] := GetNormalizedMantissa(a[i+63:i], sc, interv) + dst[i+63:i] := Convert_Int64_To_FP64(a[l+31:l]) ELSE - dst[i+63:i] := 0 - FI + dst[i+63:i] := src[i+63:i] + FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point + Integer + AVX512VL AVX512F - Miscellaneous - - - - - - Normalize the mantissas of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). This intrinsic essentially calculates "±(2^k)*|x.significand|", where "k" depends on the interval range defined by "interv" and the sign depends on "sc" and the source sign. - [getmant_note][round_note] + Convert + + + + Convert packed unsigned 32-bit integers in "a" to packed double-precision (64-bit) floating-point elements, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 +FOR j := 0 to 1 i := j*64 + l := j*32 IF k[j] - dst[i+63:i] := GetNormalizedMantissa(a[i+63:i], sc, interv) + dst[i+63:i] := Convert_Int64_To_FP64(a[l+31:l]) ELSE dst[i+63:i] := 0 - FI + FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point - AVX512F/KNCNI - Miscellaneous - - - - Normalize the mantissas of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst". This intrinsic essentially calculates "±(2^k)*|x.significand|", where "k" depends on the interval range defined by "interv" and the sign depends on "sc" and the source sign. - [getmant_note] + AVX512VL + AVX512F + Arithmetic + + + + + + Divide packed double-precision (64-bit) floating-point elements in "a" by packed elements in "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*32 - dst[i+31:i] := GetNormalizedMantissa(a[i+31:i], sc, interv) +FOR j := 0 to 3 + i := 64*j + IF k[j] + dst[i+63:i] := a[i+63:i] / b[i+63:i] + ELSE + dst[i+63:i] := src[i+63:i] + FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Floating Point - AVX512F/KNCNI - Miscellaneous - - - - - Normalize the mantissas of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst". This intrinsic essentially calculates "±(2^k)*|x.significand|", where "k" depends on the interval range defined by "interv" and the sign depends on "sc" and the source sign. - [getmant_note][round_note] + AVX512VL + AVX512F + Arithmetic + + + + + Divide packed double-precision (64-bit) floating-point elements in "a" by packed elements in "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*32 - dst[i+31:i] := GetNormalizedMantissa(a[i+31:i], sc, interv) +FOR j := 0 to 3 + i := 64*j + IF k[j] + dst[i+63:i] := a[i+63:i] / b[i+63:i] + ELSE + dst[i+63:i] := 0 + FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Floating Point - AVX512F/KNCNI - Miscellaneous - - - - - - Normalize the mantissas of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). This intrinsic essentially calculates "±(2^k)*|x.significand|", where "k" depends on the interval range defined by "interv" and the sign depends on "sc" and the source sign. - [getmant_note] + AVX512VL + AVX512F + Arithmetic + + + + + + Divide packed double-precision (64-bit) floating-point elements in "a" by packed elements in "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*32 +FOR j := 0 to 1 + i := 64*j IF k[j] - dst[i+31:i] := GetNormalizedMantissa(a[i+31:i], sc, interv) + dst[i+63:i] := a[i+63:i] / b[i+63:i] ELSE - dst[i+31:i] := src[i+31:i] + dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point - AVX512F/KNCNI - Miscellaneous - - - - - - - Normalize the mantissas of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). This intrinsic essentially calculates "±(2^k)*|x.significand|", where "k" depends on the interval range defined by "interv" and the sign depends on "sc" and the source sign. - [getmant_note][round_note] + AVX512VL + AVX512F + Arithmetic + + + + + Divide packed double-precision (64-bit) floating-point elements in "a" by packed elements in "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*32 +FOR j := 0 to 1 + i := 64*j IF k[j] - dst[i+31:i] := GetNormalizedMantissa(a[i+31:i], sc, interv) + dst[i+63:i] := a[i+63:i] / b[i+63:i] ELSE - dst[i+31:i] := src[i+31:i] + dst[i+63:i] := 0 FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point + AVX512VL AVX512F - Miscellaneous - - - - - Normalize the mantissas of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). This intrinsic essentially calculates "±(2^k)*|x.significand|", where "k" depends on the interval range defined by "interv" and the sign depends on "sc" and the source sign. - [getmant_note] + Arithmetic + + + + + + Divide packed single-precision (32-bit) floating-point elements in "a" by packed elements in "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*32 +FOR j := 0 to 7 + i := 32*j IF k[j] - dst[i+31:i] := GetNormalizedMantissa(a[i+31:i], sc, interv) + dst[i+31:i] := a[i+31:i] / b[i+31:i] ELSE - dst[i+31:i] := 0 + dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Floating Point + AVX512VL AVX512F - Miscellaneous - - - - - - Normalize the mantissas of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). This intrinsic essentially calculates "±(2^k)*|x.significand|", where "k" depends on the interval range defined by "interv" and the sign depends on "sc" and the source sign. - [getmant_note][round_note] + Arithmetic + + + + + Divide packed single-precision (32-bit) floating-point elements in "a" by packed elements in "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*32 +FOR j := 0 to 7 + i := 32*j IF k[j] - dst[i+31:i] := GetNormalizedMantissa(a[i+31:i], sc, interv) + dst[i+31:i] := a[i+31:i] / b[i+31:i] ELSE dst[i+31:i] := 0 FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Floating Point + AVX512VL AVX512F - Miscellaneous - - - - - - Normalize the mantissas of the lower double-precision (64-bit) floating-point element in "a", store the result in the lower element of "dst", and copy the upper element from "b" to the upper element of "dst". This intrinsic essentially calculates "±(2^k)*|x.significand|", where "k" depends on the interval range defined by "interv" and the sign depends on "sc" and the source sign. - [getmant_note][round_note] + Arithmetic + + + + + + Divide packed single-precision (32-bit) floating-point elements in "a" by packed elements in "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -dst[63:0] := GetNormalizedMantissa(a[63:0], sc, interv) -dst[127:64] := b[127:64] +FOR j := 0 to 3 + i := 32*j + IF k[j] + dst[i+31:i] := a[i+31:i] / b[i+31:i] + ELSE + dst[i+31:i] := src[i+31:i] + FI +ENDFOR dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point + AVX512VL AVX512F - Miscellaneous - - - - - Normalize the mantissas of the lower double-precision (64-bit) floating-point element in "a", store the result in the lower element of "dst", and copy the upper element from "b" to the upper element of "dst". This intrinsic essentially calculates "±(2^k)*|x.significand|", where "k" depends on the interval range defined by "interv" and the sign depends on "sc" and the source sign. - [getmant_note] + Arithmetic + + + + + Divide packed single-precision (32-bit) floating-point elements in "a" by packed elements in "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -dst[63:0] := GetNormalizedMantissa(a[63:0], sc, interv) -dst[127:64] := b[127:64] +FOR j := 0 to 3 + i := 32*j + IF k[j] + dst[i+31:i] := a[i+31:i] / b[i+31:i] + ELSE + dst[i+31:i] := 0 + FI +ENDFOR dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point + AVX512VL AVX512F Miscellaneous - - - - - - - - Normalize the mantissas of the lower double-precision (64-bit) floating-point element in "a", store the result in the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and copy the upper element from "b" to the upper element of "dst". This intrinsic essentially calculates "±(2^k)*|x.significand|", where "k" depends on the interval range defined by "interv" and the sign depends on "sc" and the source sign. - [getmant_note][round_note] + + + + + Load contiguous active double-precision (64-bit) floating-point elements from "a" (those with their respective bit set in mask "k"), and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -IF k[0] - dst[63:0] := GetNormalizedMantissa(a[63:0], sc, interv) -ELSE - dst[63:0] := src[63:0] -FI -dst[127:64] := b[127:64] -dst[MAX:128] := 0 +m := 0 +FOR j := 0 to 3 + i := j*64 + IF k[j] + dst[i+63:i] := a[m+63:m] + m := m + 64 + ELSE + dst[i+63:i] := src[i+63:i] + FI +ENDFOR +dst[MAX:256] := 0 - +
immintrin.h
 - + Floating Point + AVX512VL AVX512F - Miscellaneous - - - - - - - Normalize the mantissas of the lower double-precision (64-bit) floating-point element in "a", store the result in the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and copy the upper element from "b" to the upper element of "dst". This intrinsic essentially calculates "±(2^k)*|x.significand|", where "k" depends on the interval range defined by "interv" and the sign depends on "sc" and the source sign. - [getmant_note] + Load + + + + + Load contiguous active double-precision (64-bit) floating-point elements from unaligned memory at "mem_addr" (those with their respective bit set in mask "k"), and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -IF k[0] - dst[63:0] := GetNormalizedMantissa(a[63:0], sc, interv) -ELSE - dst[63:0] := src[63:0] -FI -dst[127:64] := b[127:64] -dst[MAX:128] := 0 +m := 0 +FOR j := 0 to 3 + i := j*64 + IF k[j] + dst[i+63:i] := MEM[mem_addr+m+63:mem_addr+m] + m := m + 64 + ELSE + dst[i+63:i] := src[i+63:i] + FI +ENDFOR +dst[MAX:256] := 0 - +
immintrin.h
 - + Floating Point + AVX512VL AVX512F Miscellaneous - - - - - - - Normalize the mantissas of the lower double-precision (64-bit) floating-point element in "a", store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper element from "b" to the upper element of "dst". This intrinsic essentially calculates "±(2^k)*|x.significand|", where "k" depends on the interval range defined by "interv" and the sign depends on "sc" and the source sign. - [getmant_note][round_note] + + + + Load contiguous active double-precision (64-bit) floating-point elements from "a" (those with their respective bit set in mask "k"), and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -IF k[0] - dst[63:0] := GetNormalizedMantissa(a[63:0], sc, interv) -ELSE - dst[63:0] := 0 -FI -dst[127:64] := b[127:64] -dst[MAX:128] := 0 +m := 0 +FOR j := 0 to 3 + i := j*64 + IF k[j] + dst[i+63:i] := a[m+63:m] + m := m + 64 + ELSE + dst[i+63:i] := 0 + FI +ENDFOR +dst[MAX:256] := 0 - +
immintrin.h
 - + Floating Point + AVX512VL AVX512F - Miscellaneous - - - - - - Normalize the mantissas of the lower double-precision (64-bit) floating-point element in "a", store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper element from "b" to the upper element of "dst". This intrinsic essentially calculates "±(2^k)*|x.significand|", where "k" depends on the interval range defined by "interv" and the sign depends on "sc" and the source sign. - [getmant_note] + Load + + + + Load contiguous active double-precision (64-bit) floating-point elements from unaligned memory at "mem_addr" (those with their respective bit set in mask "k"), and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -IF k[0] - dst[63:0] := GetNormalizedMantissa(a[63:0], sc, interv) -ELSE - dst[63:0] := 0 -FI -dst[127:64] := b[127:64] -dst[MAX:128] := 0 +m := 0 +FOR j := 0 to 3 + i := j*64 + IF k[j] + dst[i+63:i] := MEM[mem_addr+m+63:mem_addr+m] + m := m + 64 + ELSE + dst[i+63:i] := 0 + FI +ENDFOR +dst[MAX:256] := 0 - +
immintrin.h
 - + Floating Point + AVX512VL AVX512F Miscellaneous - - - - - - Normalize the mantissas of the lower single-precision (32-bit) floating-point element in "a", store the result in the lower element of "dst", and copy the upper 3 packed elements from "b" to the upper elements of "dst". This intrinsic essentially calculates "±(2^k)*|x.significand|", where "k" depends on the interval range defined by "interv" and the sign depends on "sc" and the source sign. - [getmant_note][round_note] + + + + + Load contiguous active double-precision (64-bit) floating-point elements from "a" (those with their respective bit set in mask "k"), and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -dst[31:0] := GetNormalizedMantissa(a[31:0], sc, interv) -dst[127:32] := b[127:32] +m := 0 +FOR j := 0 to 1 + i := j*64 + IF k[j] + dst[i+63:i] := a[m+63:m] + m := m + 64 + ELSE + dst[i+63:i] := src[i+63:i] + FI +ENDFOR dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point + AVX512VL AVX512F - Miscellaneous - - - - - Normalize the mantissas of the lower single-precision (32-bit) floating-point element in "a", store the result in the lower element of "dst", and copy the upper 3 packed elements from "b" to the upper elements of "dst". This intrinsic essentially calculates "±(2^k)*|x.significand|", where "k" depends on the interval range defined by "interv" and the sign depends on "sc" and the source sign. - [getmant_note] + Load + + + + + Load contiguous active double-precision (64-bit) floating-point elements from unaligned memory at "mem_addr" (those with their respective bit set in mask "k"), and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -dst[31:0] := GetNormalizedMantissa(a[31:0], sc, interv) -dst[127:32] := b[127:32] +m := 0 +FOR j := 0 to 1 + i := j*64 + IF k[j] + dst[i+63:i] := MEM[mem_addr+m+63:mem_addr+m] + m := m + 64 + ELSE + dst[i+63:i] := src[i+63:i] + FI +ENDFOR dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point + AVX512VL AVX512F Miscellaneous - - - - - - - - Normalize the mantissas of the lower single-precision (32-bit) floating-point element in "a", store the result in the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and copy the upper 3 packed elements from "b" to the upper elements of "dst". This intrinsic essentially calculates "±(2^k)*|x.significand|", where "k" depends on the interval range defined by "interv" and the sign depends on "sc" and the source sign. - [getmant_note][round_note] + + + + Load contiguous active double-precision (64-bit) floating-point elements from "a" (those with their respective bit set in mask "k"), and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -IF k[0] - dst[31:0] := GetNormalizedMantissa(a[31:0], sc, interv) -ELSE - dst[31:0] := src[31:0] -FI -dst[127:32] := b[127:32] +m := 0 +FOR j := 0 to 1 + i := j*64 + IF k[j] + dst[i+63:i] := a[m+63:m] + m := m + 64 + ELSE + dst[i+63:i] := 0 + FI +ENDFOR dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point + AVX512VL AVX512F - Miscellaneous - - - - - - - Normalize the mantissas of the lower single-precision (32-bit) floating-point element in "a", store the result in the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and copy the upper 3 packed elements from "b" to the upper elements of "dst". This intrinsic essentially calculates "±(2^k)*|x.significand|", where "k" depends on the interval range defined by "interv" and the sign depends on "sc" and the source sign. - [getmant_note] + Load + + + + Load contiguous active double-precision (64-bit) floating-point elements from unaligned memory at "mem_addr" (those with their respective bit set in mask "k"), and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -IF k[0] - dst[31:0] := GetNormalizedMantissa(a[31:0], sc, interv) -ELSE - dst[31:0] := src[31:0] -FI -dst[127:32] := b[127:32] +m := 0 +FOR j := 0 to 1 + i := j*64 + IF k[j] + dst[i+63:i] := MEM[mem_addr+m+63:mem_addr+m] + m := m + 64 + ELSE + dst[i+63:i] := 0 + FI +ENDFOR dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point + AVX512VL AVX512F Miscellaneous - - - - - - - Normalize the mantissas of the lower single-precision (32-bit) floating-point element in "a", store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper 3 packed elements from "b" to the upper elements of "dst". This intrinsic essentially calculates "±(2^k)*|x.significand|", where "k" depends on the interval range defined by "interv" and the sign depends on "sc" and the source sign. - [getmant_note][round_note] + + + + + Load contiguous active single-precision (32-bit) floating-point elements from "a" (those with their respective bit set in mask "k"), and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -IF k[0] - dst[31:0] := GetNormalizedMantissa(a[31:0], sc, interv) -ELSE - dst[31:0] := 0 -FI -dst[127:32] := b[127:32] -dst[MAX:128] := 0 +m := 0 +FOR j := 0 to 7 + i := j*32 + IF k[j] + dst[i+31:i] := a[m+31:m] + m := m + 32 + ELSE + dst[i+31:i] := src[i+31:i] + FI +ENDFOR +dst[MAX:256] := 0 - +
immintrin.h
 - + Floating Point + AVX512VL AVX512F - Miscellaneous - - - - - - Normalize the mantissas of the lower single-precision (32-bit) floating-point element in "a", store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper 3 packed elements from "b" to the upper elements of "dst". This intrinsic essentially calculates "±(2^k)*|x.significand|", where "k" depends on the interval range defined by "interv" and the sign depends on "sc" and the source sign. - [getmant_note] + Load + + + + + Load contiguous active single-precision (32-bit) floating-point elements from unaligned memory at "mem_addr" (those with their respective bit set in mask "k"), and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -IF k[0] - dst[31:0] := GetNormalizedMantissa(a[31:0], sc, interv) -ELSE - dst[31:0] := 0 -FI -dst[127:32] := b[127:32] -dst[MAX:128] := 0 +m := 0 +FOR j := 0 to 7 + i := j*32 + IF k[j] + dst[i+31:i] := MEM[mem_addr+m+31:mem_addr+m] + m := m + 32 + ELSE + dst[i+31:i] := src[i+31:i] + FI +ENDFOR +dst[MAX:256] := 0 - +
immintrin.h
 - + Floating Point + AVX512VL AVX512F - Swizzle - - - - Copy "a" to "dst", then insert 128 bits (composed of 4 packed single-precision (32-bit) floating-point elements) from "b" into "dst" at the location specified by "imm8". + Miscellaneous + + + + Load contiguous active single-precision (32-bit) floating-point elements from "a" (those with their respective bit set in mask "k"), and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -dst[511:0] := a[511:0] -CASE (imm8[1:0]) of -0: dst[127:0] := b[127:0] -1: dst[255:128] := b[127:0] -2: dst[383:256] := b[127:0] -3: dst[511:384] := b[127:0] -ESAC -dst[MAX:512] := 0 +m := 0 +FOR j := 0 to 7 + i := j*32 + IF k[j] + dst[i+31:i] := a[m+31:m] + m := m + 32 + ELSE + dst[i+31:i] := 0 + FI +ENDFOR +dst[MAX:256] := 0 - +
immintrin.h
 - + Floating Point + AVX512VL AVX512F - Swizzle - - - - - - Copy "a" to "tmp", then insert 128 bits (composed of 4 packed single-precision (32-bit) floating-point elements) from "b" into "tmp" at the location specified by "imm8". Store "tmp" to "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - + Load + + + + Load contiguous active single-precision (32-bit) floating-point elements from unaligned memory at "mem_addr" (those with their respective bit set in mask "k"), and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -tmp[511:0] := a[511:0] -CASE (imm8[1:0]) of -0: tmp[127:0] := b[127:0] -1: tmp[255:128] := b[127:0] -2: tmp[383:256] := b[127:0] -3: tmp[511:384] := b[127:0] -ESAC -FOR j := 0 to 15 +m := 0 +FOR j := 0 to 7 i := j*32 IF k[j] - dst[i+31:i] := tmp[i+31:i] + dst[i+31:i] := MEM[mem_addr+m+31:mem_addr+m] + m := m + 32 ELSE - dst[i+31:i] := src[i+31:i] + dst[i+31:i] := 0 FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Floating Point + AVX512VL AVX512F - Swizzle - - - - - Copy "a" to "tmp", then insert 128 bits (composed of 4 packed single-precision (32-bit) floating-point elements) from "b" into "tmp" at the location specified by "imm8". Store "tmp" to "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - + Miscellaneous + + + + + Load contiguous active single-precision (32-bit) floating-point elements from "a" (those with their respective bit set in mask "k"), and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -tmp[511:0] := a[511:0] -CASE (imm8[1:0]) of -0: tmp[127:0] := b[127:0] -1: tmp[255:128] := b[127:0] -2: tmp[383:256] := b[127:0] -3: tmp[511:384] := b[127:0] -ESAC -FOR j := 0 to 15 +m := 0 +FOR j := 0 to 3 i := j*32 IF k[j] - dst[i+31:i] := tmp[i+31:i] + dst[i+31:i] := a[m+31:m] + m := m + 32 ELSE - dst[i+31:i] := 0 + dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point + AVX512VL AVX512F - Swizzle - - - - Copy "a" to "dst", then insert 256 bits (composed of 4 packed double-precision (64-bit) floating-point elements) from "b" into "dst" at the location specified by "imm8". + Load + + + + + Load contiguous active single-precision (32-bit) floating-point elements from unaligned memory at "mem_addr" (those with their respective bit set in mask "k"), and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -dst[511:0] := a[511:0] -CASE (imm8[0]) of -0: dst[255:0] := b[255:0] -1: dst[511:256] := b[255:0] -ESAC -dst[MAX:512] := 0 +m := 0 +FOR j := 0 to 3 + i := j*32 + IF k[j] + dst[i+31:i] := MEM[mem_addr+m+31:mem_addr+m] + m := m + 32 + ELSE + dst[i+31:i] := src[i+31:i] + FI +ENDFOR +dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point + AVX512VL AVX512F - Swizzle - - - - - - Copy "a" to "tmp", then insert 256 bits (composed of 4 packed double-precision (64-bit) floating-point elements) from "b" into "tmp" at the location specified by "imm8". Store "tmp" to "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - + Miscellaneous + + + + Load contiguous active single-precision (32-bit) floating-point elements from "a" (those with their respective bit set in mask "k"), and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -tmp[511:0] := a[511:0] -CASE (imm8[0]) of -0: tmp[255:0] := b[255:0] -1: tmp[511:256] := b[255:0] -ESAC -FOR j := 0 to 7 - i := j*64 +m := 0 +FOR j := 0 to 3 + i := j*32 IF k[j] - dst[i+63:i] := tmp[i+63:i] + dst[i+31:i] := a[m+31:m] + m := m + 32 ELSE - dst[i+63:i] := src[i+63:i] + dst[i+31:i] := 0 FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point + AVX512VL AVX512F - Swizzle - - - - - Copy "a" to "tmp", then insert 256 bits (composed of 4 packed double-precision (64-bit) floating-point elements) from "b" into "tmp" at the location specified by "imm8". Store "tmp" to "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - + Load + + + + Load contiguous active single-precision (32-bit) floating-point elements from unaligned memory at "mem_addr" (those with their respective bit set in mask "k"), and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -tmp[511:0] := a[511:0] -CASE (imm8[0]) of -0: tmp[255:0] := b[255:0] -1: tmp[511:256] := b[255:0] -ESAC -FOR j := 0 to 7 - i := j*64 +m := 0 +FOR j := 0 to 3 + i := j*32 IF k[j] - dst[i+63:i] := tmp[i+63:i] + dst[i+31:i] := MEM[mem_addr+m+31:mem_addr+m] + m := m + 32 ELSE - dst[i+63:i] := 0 + dst[i+31:i] := 0 FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - - Integer + + Floating Point + AVX512VL AVX512F - Swizzle - - - - Copy "a" to "dst", then insert 128 bits (composed of 4 packed 32-bit integers) from "b" into "dst" at the location specified by "imm8". + Miscellaneous + + + + Extract 128 bits (composed of 4 packed single-precision (32-bit) floating-point elements) from "a", selected with "imm8", and store the result in "dst". -dst[511:0] := a[511:0] -CASE (imm8[1:0]) of -0: dst[127:0] := b[127:0] -1: dst[255:128] := b[127:0] -2: dst[383:256] := b[127:0] -3: dst[511:384] := b[127:0] +CASE imm8[0] OF +0: dst[127:0] := a[127:0] +1: dst[127:0] := a[255:128] ESAC -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - - Integer + + Floating Point + AVX512VL AVX512F - Swizzle - - - - - - Copy "a" to "tmp", then insert 128 bits (composed of 4 packed 32-bit integers) from "b" into "tmp" at the location specified by "imm8". Store "tmp" to "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - + Miscellaneous + + + + + + Extract 128 bits (composed of 4 packed single-precision (32-bit) floating-point elements) from "a", selected with "imm8", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -tmp[511:0] := a[511:0] -CASE (imm8[1:0]) of -0: tmp[127:0] := b[127:0] -1: tmp[255:128] := b[127:0] -2: tmp[383:256] := b[127:0] -3: tmp[511:384] := b[127:0] +CASE imm8[0] OF +0: tmp[127:0] := a[127:0] +1: tmp[127:0] := a[255:128] ESAC -FOR j := 0 to 15 +FOR j := 0 to 3 i := j*32 IF k[j] dst[i+31:i] := tmp[i+31:i] @@ -40087,30 +45625,27 @@ FOR j := 0 to 15 dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - - Integer + + Floating Point + AVX512VL AVX512F - Swizzle - - - - - Copy "a" to "tmp", then insert 128 bits (composed of 4 packed 32-bit integers) from "b" into "tmp" at the location specified by "imm8". Store "tmp" to "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - + Miscellaneous + + + + + Extract 128 bits (composed of 4 packed single-precision (32-bit) floating-point elements) from "a", selected with "imm8", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -tmp[511:0] := a[511:0] -CASE (imm8[1:0]) of -0: tmp[127:0] := b[127:0] -1: tmp[255:128] := b[127:0] -2: tmp[383:256] := b[127:0] -3: tmp[511:384] := b[127:0] +CASE imm8[0] OF +0: tmp[127:0] := a[127:0] +1: tmp[127:0] := a[255:128] ESAC -FOR j := 0 to 15 +FOR j := 0 to 3 i := j*32 IF k[j] dst[i+31:i] := tmp[i+31:i] @@ -40118,47239 +45653,57970 @@ FOR j := 0 to 15 dst[i+31:i] := 0 FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer + AVX512VL AVX512F - Swizzle - - - - Copy "a" to "dst", then insert 256 bits (composed of 4 packed 64-bit integers) from "b" into "dst" at the location specified by "imm8". + Miscellaneous + + + + Extract 128 bits (composed of 4 packed 32-bit integers) from "a", selected with "imm8", and store the result in "dst". -dst[511:0] := a[511:0] -CASE (imm8[7:0]) OF -0: dst[255:0] := b[255:0] -1: dst[511:256] := b[255:0] +CASE imm8[0] OF +0: dst[127:0] := a[127:0] +1: dst[127:0] := a[255:128] ESAC -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer + AVX512VL AVX512F - Swizzle - - - - - - Copy "a" to "tmp", then insert 256 bits (composed of 4 packed 64-bit integers) from "b" into "tmp" at the location specified by "imm8". Store "tmp" to "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - + Miscellaneous + + + + + + Extract 128 bits (composed of 4 packed 32-bit integers) from "a", selected with "imm8", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -tmp[511:0] := a[511:0] -CASE (imm8[0]) of -0: tmp[255:0] := b[255:0] -1: tmp[511:256] := b[255:0] +CASE imm8[0] OF +0: tmp[127:0] := a[127:0] +1: tmp[127:0] := a[255:128] ESAC -FOR j := 0 to 7 - i := j*64 +FOR j := 0 to 3 + i := j*32 IF k[j] - dst[i+63:i] := tmp[i+63:i] + dst[i+31:i] := tmp[i+31:i] ELSE - dst[i+63:i] := src[i+63:i] + dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer + AVX512VL AVX512F - Swizzle - - - - - Copy "a" to "tmp", then insert 256 bits (composed of 4 packed 64-bit integers) from "b" into "tmp" at the location specified by "imm8". Store "tmp" to "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - + Miscellaneous + + + + + Extract 128 bits (composed of 4 packed 32-bit integers) from "a", selected with "imm8", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -tmp[511:0] := a[511:0] -CASE (imm8[0]) of -0: tmp[255:0] := b[255:0] -1: tmp[511:256] := b[255:0] +CASE imm8[0] OF +0: tmp[127:0] := a[127:0] +1: tmp[127:0] := a[255:128] ESAC -FOR j := 0 to 7 - i := j*64 +FOR j := 0 to 3 + i := j*32 IF k[j] - dst[i+63:i] := tmp[i+63:i] + dst[i+31:i] := tmp[i+31:i] ELSE - dst[i+63:i] := 0 + dst[i+31:i] := 0 FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point + AVX512VL AVX512F - Special Math Functions - - - - - Compare packed double-precision (64-bit) floating-point elements in "a" and "b", and store packed maximum values in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -FOR j := 0 to 7 + Miscellaneous + + + + + + Fix up packed double-precision (64-bit) floating-point elements in "a" and "b" using packed 64-bit integers in "c", and store the results in "dst". "imm8" is used to set the required flags reporting. + enum TOKEN_TYPE { + QNAN_TOKEN := 0, \ + SNAN_TOKEN := 1, \ + ZERO_VALUE_TOKEN := 2, \ + ONE_VALUE_TOKEN := 3, \ + NEG_INF_TOKEN := 4, \ + POS_INF_TOKEN := 5, \ + NEG_VALUE_TOKEN := 6, \ + POS_VALUE_TOKEN := 7 +} +DEFINE FIXUPIMMPD(src1[63:0], src2[63:0], src3[63:0], imm8[7:0]) { + tsrc[63:0] := ((src2[62:52] == 0) AND (MXCSR.DAZ == 1)) ? 0.0 : src2[63:0] + CASE(tsrc[63:0]) OF + QNAN_TOKEN: j := 0 + SNAN_TOKEN: j := 1 + ZERO_VALUE_TOKEN: j := 2 + ONE_VALUE_TOKEN: j := 3 + NEG_INF_TOKEN: j := 4 + POS_INF_TOKEN: j := 5 + NEG_VALUE_TOKEN: j := 6 + POS_VALUE_TOKEN: j := 7 + ESAC + + token_response[3:0] := src3[3+4*j:4*j] + + CASE(token_response[3:0]) OF + 0 : dest[63:0] := src1[63:0] + 1 : dest[63:0] := tsrc[63:0] + 2 : dest[63:0] := QNaN(tsrc[63:0]) + 3 : dest[63:0] := QNAN_Indefinite + 4 : dest[63:0] := -INF + 5 : dest[63:0] := +INF + 6 : dest[63:0] := tsrc.sign? -INF : +INF + 7 : dest[63:0] := -0 + 8 : dest[63:0] := +0 + 9 : dest[63:0] := -1 + 10: dest[63:0] := +1 + 11: dest[63:0] := 1/2 + 12: dest[63:0] := 90.0 + 13: dest[63:0] := PI/2 + 14: dest[63:0] := MAX_FLOAT + 15: dest[63:0] := -MAX_FLOAT + ESAC + + CASE(tsrc[31:0]) OF + ZERO_VALUE_TOKEN: + IF (imm8[0]) #ZE; FI + ZERO_VALUE_TOKEN: + IF (imm8[1]) #IE; FI + ONE_VALUE_TOKEN: + IF (imm8[2]) #ZE; FI + ONE_VALUE_TOKEN: + IF (imm8[3]) #IE; FI + SNAN_TOKEN: + IF (imm8[4]) #IE; FI + NEG_INF_TOKEN: + IF (imm8[5]) #IE; FI + NEG_VALUE_TOKEN: + IF (imm8[6]) #IE; FI + POS_INF_TOKEN: + IF (imm8[7]) #IE; FI + ESAC + RETURN dest[63:0] +} +FOR j := 0 to 3 i := j*64 - IF k[j] - dst[i+63:i] := MAX(a[i+63:i], b[i+63:i]) - ELSE - dst[i+63:i] := src[i+63:i] - FI + dst[i+63:i] := FIXUPIMMPD(a[i+63:i], b[i+63:i], c[i+63:i], imm8[7:0]) ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Floating Point + AVX512VL AVX512F - Special Math Functions - - - - - - Compare packed double-precision (64-bit) floating-point elements in "a" and "b", and store packed maximum values in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - Pass __MM_FROUND_NO_EXC to "sae" to suppress all exceptions. - -FOR j := 0 to 7 + Miscellaneous + + + + + + + Fix up packed double-precision (64-bit) floating-point elements in "a" and "b" using packed 64-bit integers in "c", and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). "imm8" is used to set the required flags reporting. + enum TOKEN_TYPE { + QNAN_TOKEN := 0, \ + SNAN_TOKEN := 1, \ + ZERO_VALUE_TOKEN := 2, \ + ONE_VALUE_TOKEN := 3, \ + NEG_INF_TOKEN := 4, \ + POS_INF_TOKEN := 5, \ + NEG_VALUE_TOKEN := 6, \ + POS_VALUE_TOKEN := 7 +} +DEFINE FIXUPIMMPD(src1[63:0], src2[63:0], src3[63:0], imm8[7:0]) { + tsrc[63:0] := ((src2[62:52] == 0) AND (MXCSR.DAZ == 1)) ? 0.0 : src2[63:0] + CASE(tsrc[63:0]) OF + QNAN_TOKEN:j := 0 + SNAN_TOKEN:j := 1 + ZERO_VALUE_TOKEN: j := 2 + ONE_VALUE_TOKEN: j := 3 + NEG_INF_TOKEN: j := 4 + POS_INF_TOKEN: j := 5 + NEG_VALUE_TOKEN: j := 6 + POS_VALUE_TOKEN: j := 7 + ESAC + + token_response[3:0] := src3[3+4*j:4*j] + + CASE(token_response[3:0]) OF + 0 : dest[63:0] := src1[63:0] + 1 : dest[63:0] := tsrc[63:0] + 2 : dest[63:0] := QNaN(tsrc[63:0]) + 3 : dest[63:0] := QNAN_Indefinite + 4 : dest[63:0] := -INF + 5 : dest[63:0] := +INF + 6 : dest[63:0] := tsrc.sign? -INF : +INF + 7 : dest[63:0] := -0 + 8 : dest[63:0] := +0 + 9 : dest[63:0] := -1 + 10: dest[63:0] := +1 + 11: dest[63:0] := 1/2 + 12: dest[63:0] := 90.0 + 13: dest[63:0] := PI/2 + 14: dest[63:0] := MAX_FLOAT + 15: dest[63:0] := -MAX_FLOAT + ESAC + + CASE(tsrc[31:0]) OF + ZERO_VALUE_TOKEN: + IF (imm8[0]) #ZE; FI + ZERO_VALUE_TOKEN: + IF (imm8[1]) #IE; FI + ONE_VALUE_TOKEN: + IF (imm8[2]) #ZE; FI + ONE_VALUE_TOKEN: + IF (imm8[3]) #IE; FI + SNAN_TOKEN: + IF (imm8[4]) #IE; FI + NEG_INF_TOKEN: + IF (imm8[5]) #IE; FI + NEG_VALUE_TOKEN: + IF (imm8[6]) #IE; FI + POS_INF_TOKEN: + IF (imm8[7]) #IE; FI + ESAC + RETURN dest[63:0] +} +FOR j := 0 to 3 i := j*64 IF k[j] - dst[i+63:i] := MAX(a[i+63:i], b[i+63:i]) + dst[i+63:i] := FIXUPIMMPD(a[i+63:i], b[i+63:i], c[i+63:i], imm8[7:0]) ELSE - dst[i+63:i] := src[i+63:i] + dst[i+63:i] := a[i+63:i] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Floating Point + AVX512VL AVX512F - Special Math Functions - - - - Compare packed double-precision (64-bit) floating-point elements in "a" and "b", and store packed maximum values in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - -FOR j := 0 to 7 - i := j*64 - IF k[j] - dst[i+63:i] := MAX(a[i+63:i], b[i+63:i]) + Miscellaneous + + + + + + + Fix up packed double-precision (64-bit) floating-point elements in "a" and "b" using packed 64-bit integers in "c", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). "imm8" is used to set the required flags reporting. + enum TOKEN_TYPE { + QNAN_TOKEN := 0, \ + SNAN_TOKEN := 1, \ + ZERO_VALUE_TOKEN := 2, \ + ONE_VALUE_TOKEN := 3, \ + NEG_INF_TOKEN := 4, \ + POS_INF_TOKEN := 5, \ + NEG_VALUE_TOKEN := 6, \ + POS_VALUE_TOKEN := 7 +} +DEFINE FIXUPIMMPD(src1[63:0], src2[63:0], src3[63:0], imm8[7:0]) { + tsrc[63:0] := ((src2[62:52] == 0) AND (MXCSR.DAZ == 1)) ? 0.0 : src2[63:0] + CASE(tsrc[63:0]) OF + QNAN_TOKEN:j := 0 + SNAN_TOKEN:j := 1 + ZERO_VALUE_TOKEN: j := 2 + ONE_VALUE_TOKEN: j := 3 + NEG_INF_TOKEN: j := 4 + POS_INF_TOKEN: j := 5 + NEG_VALUE_TOKEN: j := 6 + POS_VALUE_TOKEN: j := 7 + ESAC + + token_response[3:0] := src3[3+4*j:4*j] + + CASE(token_response[3:0]) OF + 0 : dest[63:0] := src1[63:0] + 1 : dest[63:0] := tsrc[63:0] + 2 : dest[63:0] := QNaN(tsrc[63:0]) + 3 : dest[63:0] := QNAN_Indefinite + 4 : dest[63:0] := -INF + 5 : dest[63:0] := +INF + 6 : dest[63:0] := tsrc.sign? -INF : +INF + 7 : dest[63:0] := -0 + 8 : dest[63:0] := +0 + 9 : dest[63:0] := -1 + 10: dest[63:0] := +1 + 11: dest[63:0] := 1/2 + 12: dest[63:0] := 90.0 + 13: dest[63:0] := PI/2 + 14: dest[63:0] := MAX_FLOAT + 15: dest[63:0] := -MAX_FLOAT + ESAC + + CASE(tsrc[31:0]) OF + ZERO_VALUE_TOKEN: + IF (imm8[0]) #ZE; FI + ZERO_VALUE_TOKEN: + IF (imm8[1]) #IE; FI + ONE_VALUE_TOKEN: + IF (imm8[2]) #ZE; FI + ONE_VALUE_TOKEN: + IF (imm8[3]) #IE; FI + SNAN_TOKEN: + IF (imm8[4]) #IE; FI + NEG_INF_TOKEN: + IF (imm8[5]) #IE; FI + NEG_VALUE_TOKEN: + IF (imm8[6]) #IE; FI + POS_INF_TOKEN: + IF (imm8[7]) #IE; FI + ESAC + RETURN dest[63:0] +} +FOR j := 0 to 3 + i := j*64 + IF k[j] + dst[i+63:i] := FIXUPIMMPD(a[i+63:i], b[i+63:i], c[i+63:i], imm8[7:0]) ELSE dst[i+63:i] := 0 FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Floating Point + AVX512VL AVX512F - Special Math Functions - - - - - Compare packed double-precision (64-bit) floating-point elements in "a" and "b", and store packed maximum values in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - Pass __MM_FROUND_NO_EXC to "sae" to suppress all exceptions. - -FOR j := 0 to 7 + Miscellaneous + + + + + + Fix up packed double-precision (64-bit) floating-point elements in "a" and "b" using packed 64-bit integers in "c", and store the results in "dst". "imm8" is used to set the required flags reporting. + enum TOKEN_TYPE { + QNAN_TOKEN := 0, \ + SNAN_TOKEN := 1, \ + ZERO_VALUE_TOKEN := 2, \ + ONE_VALUE_TOKEN := 3, \ + NEG_INF_TOKEN := 4, \ + POS_INF_TOKEN := 5, \ + NEG_VALUE_TOKEN := 6, \ + POS_VALUE_TOKEN := 7 +} +DEFINE FIXUPIMMPD(src1[63:0], src2[63:0], src3[63:0], imm8[7:0]) { + tsrc[63:0] := ((src2[62:52] == 0) AND (MXCSR.DAZ == 1)) ? 0.0 : src2[63:0] + CASE(tsrc[63:0]) OF + QNAN_TOKEN:j := 0 + SNAN_TOKEN:j := 1 + ZERO_VALUE_TOKEN: j := 2 + ONE_VALUE_TOKEN: j := 3 + NEG_INF_TOKEN: j := 4 + POS_INF_TOKEN: j := 5 + NEG_VALUE_TOKEN: j := 6 + POS_VALUE_TOKEN: j := 7 + ESAC + + token_response[3:0] := src3[3+4*j:4*j] + + CASE(token_response[3:0]) OF + 0 : dest[63:0] := src1[63:0] + 1 : dest[63:0] := tsrc[63:0] + 2 : dest[63:0] := QNaN(tsrc[63:0]) + 3 : dest[63:0] := QNAN_Indefinite + 4 : dest[63:0] := -INF + 5 : dest[63:0] := +INF + 6 : dest[63:0] := tsrc.sign? -INF : +INF + 7 : dest[63:0] := -0 + 8 : dest[63:0] := +0 + 9 : dest[63:0] := -1 + 10: dest[63:0] := +1 + 11: dest[63:0] := 1/2 + 12: dest[63:0] := 90.0 + 13: dest[63:0] := PI/2 + 14: dest[63:0] := MAX_FLOAT + 15: dest[63:0] := -MAX_FLOAT + ESAC + + CASE(tsrc[31:0]) OF + ZERO_VALUE_TOKEN: + IF (imm8[0]) #ZE; FI + ZERO_VALUE_TOKEN: + IF (imm8[1]) #IE; FI + ONE_VALUE_TOKEN: + IF (imm8[2]) #ZE; FI + ONE_VALUE_TOKEN: + IF (imm8[3]) #IE; FI + SNAN_TOKEN: + IF (imm8[4]) #IE; FI + NEG_INF_TOKEN: + IF (imm8[5]) #IE; FI + NEG_VALUE_TOKEN: + IF (imm8[6]) #IE; FI + POS_INF_TOKEN: + IF (imm8[7]) #IE; FI + ESAC + RETURN dest[63:0] +} +FOR j := 0 to 1 + i := j*64 + dst[i+63:i] := FIXUPIMMPD(a[i+63:i], b[i+63:i], c[i+63:i], imm8[7:0]) +ENDFOR +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512VL + AVX512F + Miscellaneous + + + + + + + Fix up packed double-precision (64-bit) floating-point elements in "a" and "b" using packed 64-bit integers in "c", and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). "imm8" is used to set the required flags reporting. + enum TOKEN_TYPE { + QNAN_TOKEN := 0, \ + SNAN_TOKEN := 1, \ + ZERO_VALUE_TOKEN := 2, \ + ONE_VALUE_TOKEN := 3, \ + NEG_INF_TOKEN := 4, \ + POS_INF_TOKEN := 5, \ + NEG_VALUE_TOKEN := 6, \ + POS_VALUE_TOKEN := 7 +} +DEFINE FIXUPIMMPD(src1[63:0], src2[63:0], src3[63:0], imm8[7:0]) { + tsrc[63:0] := ((src2[62:52] == 0) AND (MXCSR.DAZ == 1)) ? 0.0 : src2[63:0] + CASE(tsrc[63:0]) OF + QNAN_TOKEN:j := 0 + SNAN_TOKEN:j := 1 + ZERO_VALUE_TOKEN: j := 2 + ONE_VALUE_TOKEN: j := 3 + NEG_INF_TOKEN: j := 4 + POS_INF_TOKEN: j := 5 + NEG_VALUE_TOKEN: j := 6 + POS_VALUE_TOKEN: j := 7 + ESAC + + token_response[3:0] := src3[3+4*j:4*j] + + CASE(token_response[3:0]) OF + 0 : dest[63:0] := src1[63:0] + 1 : dest[63:0] := tsrc[63:0] + 2 : dest[63:0] := QNaN(tsrc[63:0]) + 3 : dest[63:0] := QNAN_Indefinite + 4 : dest[63:0] := -INF + 5 : dest[63:0] := +INF + 6 : dest[63:0] := tsrc.sign? -INF : +INF + 7 : dest[63:0] := -0 + 8 : dest[63:0] := +0 + 9 : dest[63:0] := -1 + 10: dest[63:0] := +1 + 11: dest[63:0] := 1/2 + 12: dest[63:0] := 90.0 + 13: dest[63:0] := PI/2 + 14: dest[63:0] := MAX_FLOAT + 15: dest[63:0] := -MAX_FLOAT + ESAC + + CASE(tsrc[31:0]) OF + ZERO_VALUE_TOKEN: + IF (imm8[0]) #ZE; FI + ZERO_VALUE_TOKEN: + IF (imm8[1]) #IE; FI + ONE_VALUE_TOKEN: + IF (imm8[2]) #ZE; FI + ONE_VALUE_TOKEN: + IF (imm8[3]) #IE; FI + SNAN_TOKEN: + IF (imm8[4]) #IE; FI + NEG_INF_TOKEN: + IF (imm8[5]) #IE; FI + NEG_VALUE_TOKEN: + IF (imm8[6]) #IE; FI + POS_INF_TOKEN: + IF (imm8[7]) #IE; FI + ESAC + RETURN dest[63:0] +} +FOR j := 0 to 1 i := j*64 IF k[j] - dst[i+63:i] := MAX(a[i+63:i], b[i+63:i]) + dst[i+63:i] := FIXUPIMMPD(a[i+63:i], b[i+63:i], c[i+63:i], imm8[7:0]) ELSE - dst[i+63:i] := 0 + dst[i+63:i] := a[i+63:i] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point + AVX512VL AVX512F - Special Math Functions - - - Compare packed double-precision (64-bit) floating-point elements in "a" and "b", and store packed maximum values in "dst". - -FOR j := 0 to 7 + Miscellaneous + + + + + + + Fix up packed double-precision (64-bit) floating-point elements in "a" and "b" using packed 64-bit integers in "c", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). "imm8" is used to set the required flags reporting. + enum TOKEN_TYPE { + QNAN_TOKEN := 0, \ + SNAN_TOKEN := 1, \ + ZERO_VALUE_TOKEN := 2, \ + ONE_VALUE_TOKEN := 3, \ + NEG_INF_TOKEN := 4, \ + POS_INF_TOKEN := 5, \ + NEG_VALUE_TOKEN := 6, \ + POS_VALUE_TOKEN := 7 +} +DEFINE FIXUPIMMPD(src1[63:0], src2[63:0], src3[63:0], imm8[7:0]) { + tsrc[63:0] := ((src2[62:52] == 0) AND (MXCSR.DAZ == 1)) ? 0.0 : src2[63:0] + CASE(tsrc[63:0]) OF + QNAN_TOKEN:j := 0 + SNAN_TOKEN:j := 1 + ZERO_VALUE_TOKEN: j := 2 + ONE_VALUE_TOKEN: j := 3 + NEG_INF_TOKEN: j := 4 + POS_INF_TOKEN: j := 5 + NEG_VALUE_TOKEN: j := 6 + POS_VALUE_TOKEN: j := 7 + ESAC + + token_response[3:0] := src3[3+4*j:4*j] + + CASE(token_response[3:0]) OF + 0 : dest[63:0] := src1[63:0] + 1 : dest[63:0] := tsrc[63:0] + 2 : dest[63:0] := QNaN(tsrc[63:0]) + 3 : dest[63:0] := QNAN_Indefinite + 4 : dest[63:0] := -INF + 5 : dest[63:0] := +INF + 6 : dest[63:0] := tsrc.sign? -INF : +INF + 7 : dest[63:0] := -0 + 8 : dest[63:0] := +0 + 9 : dest[63:0] := -1 + 10: dest[63:0] := +1 + 11: dest[63:0] := 1/2 + 12: dest[63:0] := 90.0 + 13: dest[63:0] := PI/2 + 14: dest[63:0] := MAX_FLOAT + 15: dest[63:0] := -MAX_FLOAT + ESAC + + CASE(tsrc[31:0]) OF + ZERO_VALUE_TOKEN: + IF (imm8[0]) #ZE; FI + ZERO_VALUE_TOKEN: + IF (imm8[1]) #IE; FI + ONE_VALUE_TOKEN: + IF (imm8[2]) #ZE; FI + ONE_VALUE_TOKEN: + IF (imm8[3]) #IE; FI + SNAN_TOKEN: + IF (imm8[4]) #IE; FI + NEG_INF_TOKEN: + IF (imm8[5]) #IE; FI + NEG_VALUE_TOKEN: + IF (imm8[6]) #IE; FI + POS_INF_TOKEN: + IF (imm8[7]) #IE; FI + ESAC + RETURN dest[63:0] +} +FOR j := 0 to 1 i := j*64 - dst[i+63:i] := MAX(a[i+63:i], b[i+63:i]) + IF k[j] + dst[i+63:i] := FIXUPIMMPD(a[i+63:i], b[i+63:i], c[i+63:i], imm8[7:0]) + ELSE + dst[i+63:i] := 0 + FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point + AVX512VL AVX512F - Special Math Functions - - - - Compare packed double-precision (64-bit) floating-point elements in "a" and "b", and store packed maximum values in "dst". - Pass __MM_FROUND_NO_EXC to "sae" to suppress all exceptions. - + Miscellaneous + + + + + + Fix up packed single-precision (32-bit) floating-point elements in "a" and "b" using packed 32-bit integers in "c", and store the results in "dst". "imm8" is used to set the required flags reporting. + enum TOKEN_TYPE { + QNAN_TOKEN := 0, \ + SNAN_TOKEN := 1, \ + ZERO_VALUE_TOKEN := 2, \ + ONE_VALUE_TOKEN := 3, \ + NEG_INF_TOKEN := 4, \ + POS_INF_TOKEN := 5, \ + NEG_VALUE_TOKEN := 6, \ + POS_VALUE_TOKEN := 7 +} +DEFINE FIXUPIMMPD(src1[31:0], src2[31:0], src3[31:0], imm8[7:0]) { + tsrc[31:0] := ((src2[30:23] == 0) AND (MXCSR.DAZ == 1)) ? 0.0 : src2[31:0] + CASE(tsrc[31:0]) OF + QNAN_TOKEN:j := 0 + SNAN_TOKEN:j := 1 + ZERO_VALUE_TOKEN: j := 2 + ONE_VALUE_TOKEN: j := 3 + NEG_INF_TOKEN: j := 4 + POS_INF_TOKEN: j := 5 + NEG_VALUE_TOKEN: j := 6 + POS_VALUE_TOKEN: j := 7 + ESAC + + token_response[3:0] := src3[3+4*j:4*j] + + CASE(token_response[3:0]) OF + 0 : dest[31:0] := src1[31:0] + 1 : dest[31:0] := tsrc[31:0] + 2 : dest[31:0] := QNaN(tsrc[31:0]) + 3 : dest[31:0] := QNAN_Indefinite + 4 : dest[31:0] := -INF + 5 : dest[31:0] := +INF + 6 : dest[31:0] := tsrc.sign? -INF : +INF + 7 : dest[31:0] := -0 + 8 : dest[31:0] := +0 + 9 : dest[31:0] := -1 + 10: dest[31:0] := +1 + 11: dest[31:0] := 1/2 + 12: dest[31:0] := 90.0 + 13: dest[31:0] := PI/2 + 14: dest[31:0] := MAX_FLOAT + 15: dest[31:0] := -MAX_FLOAT + ESAC + + CASE(tsrc[31:0]) OF + ZERO_VALUE_TOKEN: + IF (imm8[0]) #ZE; FI + ZERO_VALUE_TOKEN: + IF (imm8[1]) #IE; FI + ONE_VALUE_TOKEN: + IF (imm8[2]) #ZE; FI + ONE_VALUE_TOKEN: + IF (imm8[3]) #IE; FI + SNAN_TOKEN: + IF (imm8[4]) #IE; FI + NEG_INF_TOKEN: + IF (imm8[5]) #IE; FI + NEG_VALUE_TOKEN: + IF (imm8[6]) #IE; FI + POS_INF_TOKEN: + IF (imm8[7]) #IE; FI + ESAC + RETURN dest[31:0] +} FOR j := 0 to 7 - i := j*64 - dst[i+63:i] := MAX(a[i+63:i], b[i+63:i]) + i := j*32 + dst[i+31:i] := FIXUPIMMPD(a[i+31:i], b[i+31:i], c[i+31:i], imm8[7:0]) ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Floating Point + AVX512VL AVX512F - Special Math Functions - - - - - Compare packed single-precision (32-bit) floating-point elements in "a" and "b", and store packed maximum values in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -FOR j := 0 to 15 + Miscellaneous + + + + + + + Fix up packed single-precision (32-bit) floating-point elements in "a" and "b" using packed 32-bit integers in "c", and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). "imm8" is used to set the required flags reporting. + enum TOKEN_TYPE { + QNAN_TOKEN := 0, \ + SNAN_TOKEN := 1, \ + ZERO_VALUE_TOKEN := 2, \ + ONE_VALUE_TOKEN := 3, \ + NEG_INF_TOKEN := 4, \ + POS_INF_TOKEN := 5, \ + NEG_VALUE_TOKEN := 6, \ + POS_VALUE_TOKEN := 7 +} +DEFINE FIXUPIMMPD(src1[31:0], src2[31:0], src3[31:0], imm8[7:0]) { + tsrc[31:0] := ((src2[30:23] == 0) AND (MXCSR.DAZ == 1)) ? 0.0 : src2[31:0] + CASE(tsrc[31:0]) OF + QNAN_TOKEN:j := 0 + SNAN_TOKEN:j := 1 + ZERO_VALUE_TOKEN: j := 2 + ONE_VALUE_TOKEN: j := 3 + NEG_INF_TOKEN: j := 4 + POS_INF_TOKEN: j := 5 + NEG_VALUE_TOKEN: j := 6 + POS_VALUE_TOKEN: j := 7 + ESAC + + token_response[3:0] := src3[3+4*j:4*j] + + CASE(token_response[3:0]) OF + 0 : dest[31:0] := src1[31:0] + 1 : dest[31:0] := tsrc[31:0] + 2 : dest[31:0] := QNaN(tsrc[31:0]) + 3 : dest[31:0] := QNAN_Indefinite + 4 : dest[31:0] := -INF + 5 : dest[31:0] := +INF + 6 : dest[31:0] := tsrc.sign? -INF : +INF + 7 : dest[31:0] := -0 + 8 : dest[31:0] := +0 + 9 : dest[31:0] := -1 + 10: dest[31:0] := +1 + 11: dest[31:0] := 1/2 + 12: dest[31:0] := 90.0 + 13: dest[31:0] := PI/2 + 14: dest[31:0] := MAX_FLOAT + 15: dest[31:0] := -MAX_FLOAT + ESAC + + CASE(tsrc[31:0]) OF + ZERO_VALUE_TOKEN: + IF (imm8[0]) #ZE; FI + ZERO_VALUE_TOKEN: + IF (imm8[1]) #IE; FI + ONE_VALUE_TOKEN: + IF (imm8[2]) #ZE; FI + ONE_VALUE_TOKEN: + IF (imm8[3]) #IE; FI + SNAN_TOKEN: + IF (imm8[4]) #IE; FI + NEG_INF_TOKEN: + IF (imm8[5]) #IE; FI + NEG_VALUE_TOKEN: + IF (imm8[6]) #IE; FI + POS_INF_TOKEN: + IF (imm8[7]) #IE; FI + ESAC + RETURN dest[31:0] +} +FOR j := 0 to 7 i := j*32 IF k[j] - dst[i+31:i] := MAX(a[i+31:i], b[i+31:i]) + dst[i+31:i] := FIXUPIMMPD(a[i+31:i], b[i+31:i], c[i+31:i], imm8[7:0]) ELSE - dst[i+31:i] := src[i+31:i] + dst[i+31:i] := a[i+31:i] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Floating Point + AVX512VL AVX512F - Special Math Functions - - - - - - Compare packed single-precision (32-bit) floating-point elements in "a" and "b", and store packed maximum values in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - Pass __MM_FROUND_NO_EXC to "sae" to suppress all exceptions. - -FOR j := 0 to 15 + Miscellaneous + + + + + + + Fix up packed single-precision (32-bit) floating-point elements in "a" and "b" using packed 32-bit integers in "c", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). "imm8" is used to set the required flags reporting. + enum TOKEN_TYPE { + QNAN_TOKEN := 0, \ + SNAN_TOKEN := 1, \ + ZERO_VALUE_TOKEN := 2, \ + ONE_VALUE_TOKEN := 3, \ + NEG_INF_TOKEN := 4, \ + POS_INF_TOKEN := 5, \ + NEG_VALUE_TOKEN := 6, \ + POS_VALUE_TOKEN := 7 +} +DEFINE FIXUPIMMPD(src1[31:0], src2[31:0], src3[31:0], imm8[7:0]) { + tsrc[31:0] := ((src2[30:23] == 0) AND (MXCSR.DAZ == 1)) ? 0.0 : src2[31:0] + CASE(tsrc[31:0]) OF + QNAN_TOKEN:j := 0 + SNAN_TOKEN:j := 1 + ZERO_VALUE_TOKEN: j := 2 + ONE_VALUE_TOKEN: j := 3 + NEG_INF_TOKEN: j := 4 + POS_INF_TOKEN: j := 5 + NEG_VALUE_TOKEN: j := 6 + POS_VALUE_TOKEN: j := 7 + ESAC + + token_response[3:0] := src3[3+4*j:4*j] + + CASE(token_response[3:0]) OF + 0 : dest[31:0] := src1[31:0] + 1 : dest[31:0] := tsrc[31:0] + 2 : dest[31:0] := QNaN(tsrc[31:0]) + 3 : dest[31:0] := QNAN_Indefinite + 4 : dest[31:0] := -INF + 5 : dest[31:0] := +INF + 6 : dest[31:0] := tsrc.sign? -INF : +INF + 7 : dest[31:0] := -0 + 8 : dest[31:0] := +0 + 9 : dest[31:0] := -1 + 10: dest[31:0] := +1 + 11: dest[31:0] := 1/2 + 12: dest[31:0] := 90.0 + 13: dest[31:0] := PI/2 + 14: dest[31:0] := MAX_FLOAT + 15: dest[31:0] := -MAX_FLOAT + ESAC + + CASE(tsrc[31:0]) OF + ZERO_VALUE_TOKEN: + IF (imm8[0]) #ZE; FI + ZERO_VALUE_TOKEN: + IF (imm8[1]) #IE; FI + ONE_VALUE_TOKEN: + IF (imm8[2]) #ZE; FI + ONE_VALUE_TOKEN: + IF (imm8[3]) #IE; FI + SNAN_TOKEN: + IF (imm8[4]) #IE; FI + NEG_INF_TOKEN: + IF (imm8[5]) #IE; FI + NEG_VALUE_TOKEN: + IF (imm8[6]) #IE; FI + POS_INF_TOKEN: + IF (imm8[7]) #IE; FI + ESAC + RETURN dest[31:0] +} +FOR j := 0 to 7 i := j*32 IF k[j] - dst[i+31:i] := MAX(a[i+31:i], b[i+31:i]) + dst[i+31:i] := FIXUPIMMPD(a[i+31:i], b[i+31:i], c[i+31:i], imm8[7:0]) ELSE - dst[i+31:i] := src[i+31:i] + dst[i+31:i] := 0 FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Floating Point + AVX512VL AVX512F - Special Math Functions - - - - Compare packed single-precision (32-bit) floating-point elements in "a" and "b", and store packed maximum values in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - -FOR j := 0 to 15 + Miscellaneous + + + + + + Fix up packed single-precision (32-bit) floating-point elements in "a" and "b" using packed 32-bit integers in "c", and store the results in "dst". "imm8" is used to set the required flags reporting. + enum TOKEN_TYPE { + QNAN_TOKEN := 0, \ + SNAN_TOKEN := 1, \ + ZERO_VALUE_TOKEN := 2, \ + ONE_VALUE_TOKEN := 3, \ + NEG_INF_TOKEN := 4, \ + POS_INF_TOKEN := 5, \ + NEG_VALUE_TOKEN := 6, \ + POS_VALUE_TOKEN := 7 +} +DEFINE FIXUPIMMPD(src1[31:0], src2[31:0], src3[31:0], imm8[7:0]) { + tsrc[31:0] := ((src2[30:23] == 0) AND (MXCSR.DAZ == 1)) ? 0.0 : src2[31:0] + CASE(tsrc[31:0]) OF + QNAN_TOKEN:j := 0 + SNAN_TOKEN:j := 1 + ZERO_VALUE_TOKEN: j := 2 + ONE_VALUE_TOKEN: j := 3 + NEG_INF_TOKEN: j := 4 + POS_INF_TOKEN: j := 5 + NEG_VALUE_TOKEN: j := 6 + POS_VALUE_TOKEN: j := 7 + ESAC + + token_response[3:0] := src3[3+4*j:4*j] + + CASE(token_response[3:0]) OF + 0 : dest[31:0] := src1[31:0] + 1 : dest[31:0] := tsrc[31:0] + 2 : dest[31:0] := QNaN(tsrc[31:0]) + 3 : dest[31:0] := QNAN_Indefinite + 4 : dest[31:0] := -INF + 5 : dest[31:0] := +INF + 6 : dest[31:0] := tsrc.sign? -INF : +INF + 7 : dest[31:0] := -0 + 8 : dest[31:0] := +0 + 9 : dest[31:0] := -1 + 10: dest[31:0] := +1 + 11: dest[31:0] := 1/2 + 12: dest[31:0] := 90.0 + 13: dest[31:0] := PI/2 + 14: dest[31:0] := MAX_FLOAT + 15: dest[31:0] := -MAX_FLOAT + ESAC + + CASE(tsrc[31:0]) OF + ZERO_VALUE_TOKEN: + IF (imm8[0]) #ZE; FI + ZERO_VALUE_TOKEN: + IF (imm8[1]) #IE; FI + ONE_VALUE_TOKEN: + IF (imm8[2]) #ZE; FI + ONE_VALUE_TOKEN: + IF (imm8[3]) #IE; FI + SNAN_TOKEN: + IF (imm8[4]) #IE; FI + NEG_INF_TOKEN: + IF (imm8[5]) #IE; FI + NEG_VALUE_TOKEN: + IF (imm8[6]) #IE; FI + POS_INF_TOKEN: + IF (imm8[7]) #IE; FI + ESAC + RETURN dest[31:0] +} +FOR j := 0 to 3 i := j*32 - IF k[j] - dst[i+31:i] := MAX(a[i+31:i], b[i+31:i]) - ELSE - dst[i+31:i] := 0 - FI + dst[i+31:i] := FIXUPIMMPD(a[i+31:i], b[i+31:i], c[i+31:i], imm8[7:0]) ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point + AVX512VL AVX512F - Special Math Functions - - - - - Compare packed single-precision (32-bit) floating-point elements in "a" and "b", and store packed maximum values in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - Pass __MM_FROUND_NO_EXC to "sae" to suppress all exceptions. - -FOR j := 0 to 15 + Miscellaneous + + + + + + + Fix up packed single-precision (32-bit) floating-point elements in "a" and "b" using packed 32-bit integers in "c", and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). "imm8" is used to set the required flags reporting. + enum TOKEN_TYPE { + QNAN_TOKEN := 0, \ + SNAN_TOKEN := 1, \ + ZERO_VALUE_TOKEN := 2, \ + ONE_VALUE_TOKEN := 3, \ + NEG_INF_TOKEN := 4, \ + POS_INF_TOKEN := 5, \ + NEG_VALUE_TOKEN := 6, \ + POS_VALUE_TOKEN := 7 +} +DEFINE FIXUPIMMPD(src1[31:0], src2[31:0], src3[31:0], imm8[7:0]) { + tsrc[31:0] := ((src2[30:23] == 0) AND (MXCSR.DAZ == 1)) ? 0.0 : src2[31:0] + CASE(tsrc[31:0]) OF + QNAN_TOKEN:j := 0 + SNAN_TOKEN:j := 1 + ZERO_VALUE_TOKEN: j := 2 + ONE_VALUE_TOKEN: j := 3 + NEG_INF_TOKEN: j := 4 + POS_INF_TOKEN: j := 5 + NEG_VALUE_TOKEN: j := 6 + POS_VALUE_TOKEN: j := 7 + ESAC + + token_response[3:0] := src3[3+4*j:4*j] + + CASE(token_response[3:0]) OF + 0 : dest[31:0] := src1[31:0] + 1 : dest[31:0] := tsrc[31:0] + 2 : dest[31:0] := QNaN(tsrc[31:0]) + 3 : dest[31:0] := QNAN_Indefinite + 4 : dest[31:0] := -INF + 5 : dest[31:0] := +INF + 6 : dest[31:0] := tsrc.sign? -INF : +INF + 7 : dest[31:0] := -0 + 8 : dest[31:0] := +0 + 9 : dest[31:0] := -1 + 10: dest[31:0] := +1 + 11: dest[31:0] := 1/2 + 12: dest[31:0] := 90.0 + 13: dest[31:0] := PI/2 + 14: dest[31:0] := MAX_FLOAT + 15: dest[31:0] := -MAX_FLOAT + ESAC + + CASE(tsrc[31:0]) OF + ZERO_VALUE_TOKEN: + IF (imm8[0]) #ZE; FI + ZERO_VALUE_TOKEN: + IF (imm8[1]) #IE; FI + ONE_VALUE_TOKEN: + IF (imm8[2]) #ZE; FI + ONE_VALUE_TOKEN: + IF (imm8[3]) #IE; FI + SNAN_TOKEN: + IF (imm8[4]) #IE; FI + NEG_INF_TOKEN: + IF (imm8[5]) #IE; FI + NEG_VALUE_TOKEN: + IF (imm8[6]) #IE; FI + POS_INF_TOKEN: + IF (imm8[7]) #IE; FI + ESAC + RETURN dest[31:0] +} +FOR j := 0 to 3 i := j*32 IF k[j] - dst[i+31:i] := MAX(a[i+31:i], b[i+31:i]) + dst[i+31:i] := FIXUPIMMPD(a[i+31:i], b[i+31:i], c[i+31:i], imm8[7:0]) ELSE - dst[i+31:i] := 0 + dst[i+31:i] := a[i+31:i] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point + AVX512VL AVX512F - Special Math Functions - - - Compare packed single-precision (32-bit) floating-point elements in "a" and "b", and store packed maximum values in "dst". - -FOR j := 0 to 15 + Miscellaneous + + + + + + + Fix up packed single-precision (32-bit) floating-point elements in "a" and "b" using packed 32-bit integers in "c", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). "imm8" is used to set the required flags reporting. + enum TOKEN_TYPE { + QNAN_TOKEN := 0, \ + SNAN_TOKEN := 1, \ + ZERO_VALUE_TOKEN := 2, \ + ONE_VALUE_TOKEN := 3, \ + NEG_INF_TOKEN := 4, \ + POS_INF_TOKEN := 5, \ + NEG_VALUE_TOKEN := 6, \ + POS_VALUE_TOKEN := 7 +} +DEFINE FIXUPIMMPD(src1[31:0], src2[31:0], src3[31:0], imm8[7:0]) { + tsrc[31:0] := ((src2[30:23] == 0) AND (MXCSR.DAZ == 1)) ? 0.0 : src2[31:0] + CASE(tsrc[31:0]) OF + QNAN_TOKEN:j := 0 + SNAN_TOKEN:j := 1 + ZERO_VALUE_TOKEN: j := 2 + ONE_VALUE_TOKEN: j := 3 + NEG_INF_TOKEN: j := 4 + POS_INF_TOKEN: j := 5 + NEG_VALUE_TOKEN: j := 6 + POS_VALUE_TOKEN: j := 7 + ESAC + + token_response[3:0] := src3[3+4*j:4*j] + + CASE(token_response[3:0]) OF + 0 : dest[31:0] := src1[31:0] + 1 : dest[31:0] := tsrc[31:0] + 2 : dest[31:0] := QNaN(tsrc[31:0]) + 3 : dest[31:0] := QNAN_Indefinite + 4 : dest[31:0] := -INF + 5 : dest[31:0] := +INF + 6 : dest[31:0] := tsrc.sign? -INF : +INF + 7 : dest[31:0] := -0 + 8 : dest[31:0] := +0 + 9 : dest[31:0] := -1 + 10: dest[31:0] := +1 + 11: dest[31:0] := 1/2 + 12: dest[31:0] := 90.0 + 13: dest[31:0] := PI/2 + 14: dest[31:0] := MAX_FLOAT + 15: dest[31:0] := -MAX_FLOAT + ESAC + + CASE(tsrc[31:0]) OF + ZERO_VALUE_TOKEN: + IF (imm8[0]) #ZE; FI + ZERO_VALUE_TOKEN: + IF (imm8[1]) #IE; FI + ONE_VALUE_TOKEN: + IF (imm8[2]) #ZE; FI + ONE_VALUE_TOKEN: + IF (imm8[3]) #IE; FI + SNAN_TOKEN: + IF (imm8[4]) #IE; FI + NEG_INF_TOKEN: + IF (imm8[5]) #IE; FI + NEG_VALUE_TOKEN: + IF (imm8[6]) #IE; FI + POS_INF_TOKEN: + IF (imm8[7]) #IE; FI + ESAC + RETURN dest[31:0] +} +FOR j := 0 to 3 i := j*32 - dst[i+31:i] := MAX(a[i+31:i], b[i+31:i]) + IF k[j] + dst[i+31:i] := FIXUPIMMPD(a[i+31:i], b[i+31:i], c[i+31:i], imm8[7:0]) + ELSE + dst[i+31:i] := 0 + FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point + AVX512VL AVX512F - Special Math Functions - - - - Compare packed single-precision (32-bit) floating-point elements in "a" and "b", and store packed maximum values in "dst". - Pass __MM_FROUND_NO_EXC to "sae" to suppress all exceptions. + Arithmetic + + + + + + Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", add the intermediate result to packed elements in "c", and store the results in "dst" using writemask "k" (elements are copied from "c" when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*32 - dst[i+31:i] := MAX(a[i+31:i], b[i+31:i]) +FOR j := 0 to 3 + i := j*64 + IF k[j] + dst[i+63:i] := (a[i+63:i] * b[i+63:i]) + c[i+63:i] + ELSE + dst[i+63:i] := c[i+63:i] + FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - + + +
immintrin.h
 - + Floating Point + AVX512VL AVX512F - Special Math Functions - - - - - - Compare the lower double-precision (64-bit) floating-point elements in "a" and "b", store the maximum value in the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". - Pass __MM_FROUND_NO_EXC to "sae" to suppress all exceptions. + Arithmetic + + + + + + Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", add the intermediate result to packed elements in "c", and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). -IF k[0] - dst[63:0] := MAX(a[63:0], b[63:0]) -ELSE - dst[63:0] := src[63:0] -FI -dst[127:64] := a[127:64] -dst[MAX:128] := 0 +FOR j := 0 to 3 + i := j*64 + IF k[j] + dst[i+63:i] := (a[i+63:i] * b[i+63:i]) + c[i+63:i] + ELSE + dst[i+63:i] := a[i+63:i] + FI +ENDFOR +dst[MAX:256] := 0 - + + +
immintrin.h
 - + Floating Point + AVX512VL AVX512F - Special Math Functions - - - - - Compare the lower double-precision (64-bit) floating-point elements in "a" and "b", store the maximum value in the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". + Arithmetic + + + + + + Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", add the intermediate result to packed elements in "c", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -IF k[0] - dst[63:0] := MAX(a[63:0], b[63:0]) -ELSE - dst[63:0] := src[63:0] -FI -dst[127:64] := a[127:64] -dst[MAX:128] := 0 +FOR j := 0 to 3 + i := j*64 + IF k[j] + dst[i+63:i] := (a[i+63:i] * b[i+63:i]) + c[i+63:i] + ELSE + dst[i+63:i] := 0 + FI +ENDFOR +dst[MAX:256] := 0 - + + +
immintrin.h
 - + Floating Point + AVX512VL AVX512F - Special Math Functions - - - - - Compare the lower double-precision (64-bit) floating-point elements in "a" and "b", store the maximum value in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". - Pass __MM_FROUND_NO_EXC to "sae" to suppress all exceptions. + Arithmetic + + + + + + Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", add the intermediate result to packed elements in "c", and store the results in "dst" using writemask "k" (elements are copied from "c" when the corresponding mask bit is not set). -IF k[0] - dst[63:0] := MAX(a[63:0], b[63:0]) -ELSE - dst[63:0] := 0 -FI -dst[127:64] := a[127:64] +FOR j := 0 to 1 + i := j*64 + IF k[j] + dst[i+63:i] := (a[i+63:i] * b[i+63:i]) + c[i+63:i] + ELSE + dst[i+63:i] := c[i+63:i] + FI +ENDFOR dst[MAX:128] := 0 - + + +
immintrin.h
 - + Floating Point + AVX512VL AVX512F - Special Math Functions - - - - Compare the lower double-precision (64-bit) floating-point elements in "a" and "b", store the maximum value in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". + Arithmetic + + + + + + Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", add the intermediate result to packed elements in "c", and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). -IF k[0] - dst[63:0] := MAX(a[63:0], b[63:0]) -ELSE - dst[63:0] := 0 -FI -dst[127:64] := a[127:64] +FOR j := 0 to 1 + i := j*64 + IF k[j] + dst[i+63:i] := (a[i+63:i] * b[i+63:i]) + c[i+63:i] + ELSE + dst[i+63:i] := a[i+63:i] + FI +ENDFOR dst[MAX:128] := 0 - + + +
immintrin.h
 - + Floating Point + AVX512VL AVX512F - Special Math Functions - - - - Compare the lower double-precision (64-bit) floating-point elements in "a" and "b", store the maximum value in the lower element of "dst", and copy the upper element from "a" to the upper element of "dst". - Pass __MM_FROUND_NO_EXC to "sae" to suppress all exceptions. + Arithmetic + + + + + + Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", add the intermediate result to packed elements in "c", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -dst[63:0] := MAX(a[63:0], b[63:0]) -dst[127:64] := a[127:64] +FOR j := 0 to 1 + i := j*64 + IF k[j] + dst[i+63:i] := (a[i+63:i] * b[i+63:i]) + c[i+63:i] + ELSE + dst[i+63:i] := 0 + FI +ENDFOR dst[MAX:128] := 0 - + + +
immintrin.h
 - + Floating Point + AVX512VL AVX512F - Special Math Functions - - - - - - Compare the lower single-precision (32-bit) floating-point elements in "a" and "b", store the maximum value in the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". - Pass __MM_FROUND_NO_EXC to "sae" to suppress all exceptions. + Arithmetic + + + + + + Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", add the intermediate result to packed elements in "c", and store the results in "dst" using writemask "k" (elements are copied from "c" when the corresponding mask bit is not set). -IF k[0] - dst[31:0] := MAX(a[31:0], b[31:0]) -ELSE - dst[31:0] := src[31:0] -FI -dst[127:32] := a[127:32] -dst[MAX:128] := 0 +FOR j := 0 to 7 + i := j*32 + IF k[j] + dst[i+31:i] := (a[i+31:i] * b[i+31:i]) + c[i+31:i] + ELSE + dst[i+31:i] := c[i+31:i] + FI +ENDFOR +dst[MAX:256] := 0 - + + +
immintrin.h
 - + Floating Point + AVX512VL AVX512F - Special Math Functions - - - - - Compare the lower single-precision (32-bit) floating-point elements in "a" and "b", store the maximum value in the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". + Arithmetic + + + + + + Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", add the intermediate result to packed elements in "c", and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). -IF k[0] - dst[31:0] := MAX(a[31:0], b[31:0]) -ELSE - dst[31:0] := src[31:0] -FI -dst[127:32] := a[127:32] -dst[MAX:128] := 0 +FOR j := 0 to 7 + i := j*32 + IF k[j] + dst[i+31:i] := (a[i+31:i] * b[i+31:i]) + c[i+31:i] + ELSE + dst[i+31:i] := a[i+31:i] + FI +ENDFOR +dst[MAX:256] := 0 - + + +
immintrin.h
 - + Floating Point + AVX512VL AVX512F - Special Math Functions - - - - - Compare the lower single-precision (32-bit) floating-point elements in "a" and "b", store the maximum value in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". - Pass __MM_FROUND_NO_EXC to "sae" to suppress all exceptions. + Arithmetic + + + + + + Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", add the intermediate result to packed elements in "c", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -IF k[0] - dst[31:0] := MAX(a[31:0], b[31:0]) -ELSE - dst[31:0] := 0 -FI -dst[127:32] := a[127:32] -dst[MAX:128] := 0 +FOR j := 0 to 7 + i := j*32 + IF k[j] + dst[i+31:i] := (a[i+31:i] * b[i+31:i]) + c[i+31:i] + ELSE + dst[i+31:i] := 0 + FI +ENDFOR +dst[MAX:256] := 0 - + + +
immintrin.h
 - + Floating Point + AVX512VL AVX512F - Special Math Functions - - - - Compare the lower single-precision (32-bit) floating-point elements in "a" and "b", store the maximum value in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". + Arithmetic + + + + + + Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", add the intermediate result to packed elements in "c", and store the results in "dst" using writemask "k" (elements are copied from "c" when the corresponding mask bit is not set). -IF k[0] - dst[31:0] := MAX(a[31:0], b[31:0]) -ELSE - dst[31:0] := 0 -FI -dst[127:32] := a[127:32] +FOR j := 0 to 3 + i := j*32 + IF k[j] + dst[i+31:i] := (a[i+31:i] * b[i+31:i]) + c[i+31:i] + ELSE + dst[i+31:i] := c[i+31:i] + FI +ENDFOR dst[MAX:128] := 0 - + + +
immintrin.h
 - + Floating Point + AVX512VL AVX512F - Special Math Functions - - - - Compare the lower single-precision (32-bit) floating-point elements in "a" and "b", store the maximum value in the lower element of "dst", and copy the upper element from "a" to the upper element of "dst". - Pass __MM_FROUND_NO_EXC to "sae" to suppress all exceptions. + Arithmetic + + + + + + Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", add the intermediate result to packed elements in "c", and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). -dst[31:0] := MAX(a[31:0], b[31:0]) -dst[127:32] := a[127:32] +FOR j := 0 to 3 + i := j*32 + IF k[j] + dst[i+31:i] := (a[i+31:i] * b[i+31:i]) + c[i+31:i] + ELSE + dst[i+31:i] := a[i+31:i] + FI +ENDFOR dst[MAX:128] := 0 - + + +
immintrin.h
 - + Floating Point + AVX512VL AVX512F - Special Math Functions - - - - - Compare packed double-precision (64-bit) floating-point elements in "a" and "b", and store packed minimum values in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Arithmetic + + + + + + Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", add the intermediate result to packed elements in "c", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := j*64 +FOR j := 0 to 3 + i := j*32 IF k[j] - dst[i+63:i] := MIN(a[i+63:i], b[i+63:i]) + dst[i+31:i] := (a[i+31:i] * b[i+31:i]) + c[i+31:i] ELSE - dst[i+63:i] := src[i+63:i] + dst[i+31:i] := 0 FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - + + +
immintrin.h
 - + Floating Point + AVX512VL AVX512F - Special Math Functions - - - - - - Compare packed double-precision (64-bit) floating-point elements in "a" and "b", and store packed minimum values in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - Pass __MM_FROUND_NO_EXC to "sae" to suppress all exceptions. + Arithmetic + + + + + + Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", alternatively add and subtract packed elements in "c" to/from the intermediate result, and store the results in "dst" using writemask "k" (elements are copied from "c" when the corresponding mask bit is not set). -FOR j := 0 to 7 +FOR j := 0 to 3 i := j*64 IF k[j] - dst[i+63:i] := MIN(a[i+63:i], b[i+63:i]) + IF ((j & 1) == 0) + dst[i+63:i] := (a[i+63:i] * b[i+63:i]) - c[i+63:i] + ELSE + dst[i+63:i] := (a[i+63:i] * b[i+63:i]) + c[i+63:i] + FI ELSE - dst[i+63:i] := src[i+63:i] + dst[i+63:i] := c[i+63:i] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - + + +
immintrin.h
 - + Floating Point + AVX512VL AVX512F - Special Math Functions - - - - Compare packed double-precision (64-bit) floating-point elements in "a" and "b", and store packed minimum values in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Arithmetic + + + + + + Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", alternatively add and subtract packed elements in "c" to/from the intermediate result, and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). -FOR j := 0 to 7 +FOR j := 0 to 3 i := j*64 IF k[j] - dst[i+63:i] := MIN(a[i+63:i], b[i+63:i]) + IF ((j & 1) == 0) + dst[i+63:i] := (a[i+63:i] * b[i+63:i]) - c[i+63:i] + ELSE + dst[i+63:i] := (a[i+63:i] * b[i+63:i]) + c[i+63:i] + FI ELSE - dst[i+63:i] := 0 + dst[i+63:i] := a[i+63:i] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - + + +
immintrin.h
 - + Floating Point + AVX512VL AVX512F - Special Math Functions - - - - - Compare packed double-precision (64-bit) floating-point elements in "a" and "b", and store packed minimum values in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - Pass __MM_FROUND_NO_EXC to "sae" to suppress all exceptions. + Arithmetic + + + + + + Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", alternatively add and subtract packed elements in "c" to/from the intermediate result, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 +FOR j := 0 to 3 i := j*64 IF k[j] - dst[i+63:i] := MIN(a[i+63:i], b[i+63:i]) + IF ((j & 1) == 0) + dst[i+63:i] := (a[i+63:i] * b[i+63:i]) - c[i+63:i] + ELSE + dst[i+63:i] := (a[i+63:i] * b[i+63:i]) + c[i+63:i] + FI ELSE dst[i+63:i] := 0 FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - + + +
immintrin.h
 - + Floating Point + AVX512VL AVX512F - Special Math Functions - - - Compare packed double-precision (64-bit) floating-point elements in "a" and "b", and store packed minimum values in "dst". - + Arithmetic + + + + + + Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", alternatively add and subtract packed elements in "c" to/from the intermediate result, and store the results in "dst" using writemask "k" (elements are copied from "c" when the corresponding mask bit is not set). -FOR j := 0 to 7 +FOR j := 0 to 1 i := j*64 - dst[i+63:i] := MIN(a[i+63:i], b[i+63:i]) + IF k[j] + IF ((j & 1) == 0) + dst[i+63:i] := (a[i+63:i] * b[i+63:i]) - c[i+63:i] + ELSE + dst[i+63:i] := (a[i+63:i] * b[i+63:i]) + c[i+63:i] + FI + ELSE + dst[i+63:i] := c[i+63:i] + FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - + + +
immintrin.h
 - + Floating Point + AVX512VL AVX512F - Special Math Functions - - - - Compare packed double-precision (64-bit) floating-point elements in "a" and "b", and store packed minimum values in "dst". - Pass __MM_FROUND_NO_EXC to "sae" to suppress all exceptions. + Arithmetic + + + + + + Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", alternatively add and subtract packed elements in "c" to/from the intermediate result, and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). -FOR j := 0 to 7 +FOR j := 0 to 1 i := j*64 - dst[i+63:i] := MIN(a[i+63:i], b[i+63:i]) + IF k[j] + IF ((j & 1) == 0) + dst[i+63:i] := (a[i+63:i] * b[i+63:i]) - c[i+63:i] + ELSE + dst[i+63:i] := (a[i+63:i] * b[i+63:i]) + c[i+63:i] + FI + ELSE + dst[i+63:i] := a[i+63:i] + FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - + + +
immintrin.h
 - + Floating Point + AVX512VL AVX512F - Special Math Functions - - - - - Compare packed single-precision (32-bit) floating-point elements in "a" and "b", and store packed minimum values in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Arithmetic + + + + + + Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", alternatively add and subtract packed elements in "c" to/from the intermediate result, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*32 +FOR j := 0 to 1 + i := j*64 IF k[j] - dst[i+31:i] := MIN(a[i+31:i], b[i+31:i]) + IF ((j & 1) == 0) + dst[i+63:i] := (a[i+63:i] * b[i+63:i]) - c[i+63:i] + ELSE + dst[i+63:i] := (a[i+63:i] * b[i+63:i]) + c[i+63:i] + FI ELSE - dst[i+31:i] := src[i+31:i] + dst[i+63:i] := 0 FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - + + +
immintrin.h
 - + Floating Point + AVX512VL AVX512F - Special Math Functions - - - - - - Compare packed single-precision (32-bit) floating-point elements in "a" and "b", and store packed minimum values in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - Pass __MM_FROUND_NO_EXC to "sae" to suppress all exceptions. + Arithmetic + + + + + + Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", alternatively add and subtract packed elements in "c" to/from the intermediate result, and store the results in "dst" using writemask "k" (elements are copied from "c" when the corresponding mask bit is not set). -FOR j := 0 to 15 +FOR j := 0 to 7 i := j*32 IF k[j] - dst[i+31:i] := MIN(a[i+31:i], b[i+31:i]) + IF ((j & 1) == 0) + dst[i+31:i] := (a[i+31:i] * b[i+31:i]) - c[i+31:i] + ELSE + dst[i+31:i] := (a[i+31:i] * b[i+31:i]) + c[i+31:i] + FI ELSE - dst[i+31:i] := src[i+31:i] + dst[i+31:i] := c[i+31:i] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - + + +
immintrin.h
 - + Floating Point + AVX512VL AVX512F - Special Math Functions - - - - Compare packed single-precision (32-bit) floating-point elements in "a" and "b", and store packed minimum values in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Arithmetic + + + + + + Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", alternatively add and subtract packed elements in "c" to/from the intermediate result, and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). -FOR j := 0 to 15 +FOR j := 0 to 7 i := j*32 IF k[j] - dst[i+31:i] := MIN(a[i+31:i], b[i+31:i]) + IF ((j & 1) == 0) + dst[i+31:i] := (a[i+31:i] * b[i+31:i]) - c[i+31:i] + ELSE + dst[i+31:i] := (a[i+31:i] * b[i+31:i]) + c[i+31:i] + FI ELSE - dst[i+31:i] := 0 + dst[i+31:i] := a[i+31:i] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - + + +
immintrin.h
 - + Floating Point + AVX512VL AVX512F - Special Math Functions - - - - - Compare packed single-precision (32-bit) floating-point elements in "a" and "b", and store packed minimum values in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - Pass __MM_FROUND_NO_EXC to "sae" to suppress all exceptions. + Arithmetic + + + + + + Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", alternatively add and subtract packed elements in "c" to/from the intermediate result, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 15 +FOR j := 0 to 7 i := j*32 IF k[j] - dst[i+31:i] := MIN(a[i+31:i], b[i+31:i]) + IF ((j & 1) == 0) + dst[i+31:i] := (a[i+31:i] * b[i+31:i]) - c[i+31:i] + ELSE + dst[i+31:i] := (a[i+31:i] * b[i+31:i]) + c[i+31:i] + FI ELSE dst[i+31:i] := 0 FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - + + +
immintrin.h
 - + Floating Point + AVX512VL AVX512F - Special Math Functions - - - Compare packed single-precision (32-bit) floating-point elements in "a" and "b", and store packed minimum values in "dst". - + Arithmetic + + + + + + Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", alternatively add and subtract packed elements in "c" to/from the intermediate result, and store the results in "dst" using writemask "k" (elements are copied from "c" when the corresponding mask bit is not set). -FOR j := 0 to 15 +FOR j := 0 to 3 i := j*32 - dst[i+31:i] := MIN(a[i+31:i], b[i+31:i]) + IF k[j] + IF ((j & 1) == 0) + dst[i+31:i] := (a[i+31:i] * b[i+31:i]) - c[i+31:i] + ELSE + dst[i+31:i] := (a[i+31:i] * b[i+31:i]) + c[i+31:i] + FI + ELSE + dst[i+31:i] := c[i+31:i] + FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - + + +
immintrin.h
 - + Floating Point + AVX512VL AVX512F - Special Math Functions - - - - Compare packed single-precision (32-bit) floating-point elements in "a" and "b", and store packed minimum values in "dst". - Pass __MM_FROUND_NO_EXC to "sae" to suppress all exceptions. + Arithmetic + + + + + + Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", alternatively add and subtract packed elements in "c" to/from the intermediate result, and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). -FOR j := 0 to 15 +FOR j := 0 to 3 i := j*32 - dst[i+31:i] := MIN(a[i+31:i], b[i+31:i]) + IF k[j] + IF ((j & 1) == 0) + dst[i+31:i] := (a[i+31:i] * b[i+31:i]) - c[i+31:i] + ELSE + dst[i+31:i] := (a[i+31:i] * b[i+31:i]) + c[i+31:i] + FI + ELSE + dst[i+31:i] := a[i+31:i] + FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - + + +
immintrin.h
 - + Floating Point + AVX512VL AVX512F - Special Math Functions - - - - - - Compare the lower double-precision (64-bit) floating-point elements in "a" and "b", store the minimum value in the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". - Pass __MM_FROUND_NO_EXC to "sae" to suppress all exceptions. + Arithmetic + + + + + + Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", alternatively add and subtract packed elements in "c" to/from the intermediate result, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -IF k[0] - dst[63:0] := MIN(a[63:0], b[63:0]) -ELSE - dst[63:0] := src[63:0] -FI -dst[127:64] := a[127:64] +FOR j := 0 to 3 + i := j*32 + IF k[j] + IF ((j & 1) == 0) + dst[i+31:i] := (a[i+31:i] * b[i+31:i]) - c[i+31:i] + ELSE + dst[i+31:i] := (a[i+31:i] * b[i+31:i]) + c[i+31:i] + FI + ELSE + dst[i+31:i] := 0 + FI +ENDFOR dst[MAX:128] := 0 - + + +
immintrin.h
 - + Floating Point + AVX512VL AVX512F - Special Math Functions - - - - - Compare the lower double-precision (64-bit) floating-point elements in "a" and "b", store the minimum value in the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". + Arithmetic + + + + + + Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", subtract packed elements in "c" from the intermediate result, and store the results in "dst" using writemask "k" (elements are copied from "c" when the corresponding mask bit is not set). -IF k[0] - dst[63:0] := MIN(a[63:0], b[63:0]) -ELSE - dst[63:0] := src[63:0] -FI -dst[127:64] := a[127:64] -dst[MAX:128] := 0 +FOR j := 0 to 3 + i := j*64 + IF k[j] + dst[i+63:i] := (a[i+63:i] * b[i+63:i]) - c[i+63:i] + ELSE + dst[i+63:i] := c[i+63:i] + FI +ENDFOR +dst[MAX:256] := 0 - + + +
immintrin.h
 - + Floating Point + AVX512VL AVX512F - Special Math Functions - - - - - Compare the lower double-precision (64-bit) floating-point elements in "a" and "b", store the minimum value in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". - Pass __MM_FROUND_NO_EXC to "sae" to suppress all exceptions. + Arithmetic + + + + + + Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", subtract packed elements in "c" from the intermediate result, and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). -IF k[0] - dst[63:0] := MIN(a[63:0], b[63:0]) -ELSE - dst[63:0] := 0 -FI -dst[127:64] := a[127:64] -dst[MAX:128] := 0 +FOR j := 0 to 3 + i := j*64 + IF k[j] + dst[i+63:i] := (a[i+63:i] * b[i+63:i]) - c[i+63:i] + ELSE + dst[i+63:i] := a[i+63:i] + FI +ENDFOR +dst[MAX:256] := 0 - + + +
immintrin.h
 - + Floating Point + AVX512VL AVX512F - Special Math Functions - - - - Compare the lower double-precision (64-bit) floating-point elements in "a" and "b", store the minimum value in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". + Arithmetic + + + + + + Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", subtract packed elements in "c" from the intermediate result, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -IF k[0] - dst[63:0] := MIN(a[63:0], b[63:0]) -ELSE - dst[63:0] := 0 -FI -dst[127:64] := a[127:64] -dst[MAX:128] := 0 +FOR j := 0 to 3 + i := j*64 + IF k[j] + dst[i+63:i] := (a[i+63:i] * b[i+63:i]) - c[i+63:i] + ELSE + dst[i+63:i] := 0 + FI +ENDFOR +dst[MAX:256] := 0 - + + +
immintrin.h
 - + Floating Point + AVX512VL AVX512F - Special Math Functions - - - - Compare the lower double-precision (64-bit) floating-point elements in "a" and "b", store the minimum value in the lower element of "dst" , and copy the upper element from "a" to the upper element of "dst". - Pass __MM_FROUND_NO_EXC to "sae" to suppress all exceptions. + Arithmetic + + + + + + Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", subtract packed elements in "c" from the intermediate result, and store the results in "dst" using writemask "k" (elements are copied from "c" when the corresponding mask bit is not set). -dst[63:0] := MIN(a[63:0], b[63:0]) -dst[127:64] := a[127:64] +FOR j := 0 to 1 + i := j*64 + IF k[j] + dst[i+63:i] := (a[i+63:i] * b[i+63:i]) - c[i+63:i] + ELSE + dst[i+63:i] := c[i+63:i] + FI +ENDFOR dst[MAX:128] := 0 - + + +
immintrin.h
 - + Floating Point + AVX512VL AVX512F - Special Math Functions - - - - - - Compare the lower single-precision (32-bit) floating-point elements in "a" and "b", store the minimum value in the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". - Pass __MM_FROUND_NO_EXC to "sae" to suppress all exceptions. + Arithmetic + + + + + + Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", subtract packed elements in "c" from the intermediate result, and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). -IF k[0] - dst[31:0] := MIN(a[31:0], b[31:0]) -ELSE - dst[31:0] := src[31:0] -FI -dst[127:32] := a[127:32] +FOR j := 0 to 1 + i := j*64 + IF k[j] + dst[i+63:i] := (a[i+63:i] * b[i+63:i]) - c[i+63:i] + ELSE + dst[i+63:i] := a[i+63:i] + FI +ENDFOR dst[MAX:128] := 0 - + + +
immintrin.h
 - + Floating Point + AVX512VL AVX512F - Special Math Functions - - - - - Compare the lower single-precision (32-bit) floating-point elements in "a" and "b", store the minimum value in the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". + Arithmetic + + + + + + Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", subtract packed elements in "c" from the intermediate result, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -IF k[0] - dst[31:0] := MIN(a[31:0], b[31:0]) -ELSE - dst[31:0] := src[31:0] -FI -dst[127:32] := a[127:32] +FOR j := 0 to 1 + i := j*64 + IF k[j] + dst[i+63:i] := (a[i+63:i] * b[i+63:i]) - c[i+63:i] + ELSE + dst[i+63:i] := 0 + FI +ENDFOR dst[MAX:128] := 0 - + + +
immintrin.h
 - + Floating Point + AVX512VL AVX512F - Special Math Functions - - - - - Compare the lower single-precision (32-bit) floating-point elements in "a" and "b", store the minimum value in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". - Pass __MM_FROUND_NO_EXC to "sae" to suppress all exceptions. + Arithmetic + + + + + + Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", subtract packed elements in "c" from the intermediate result, and store the results in "dst" using writemask "k" (elements are copied from "c" when the corresponding mask bit is not set). -IF k[0] - dst[31:0] := MIN(a[31:0], b[31:0]) -ELSE - dst[31:0] := 0 -FI -dst[127:32] := a[127:32] -dst[MAX:128] := 0 +FOR j := 0 to 7 + i := j*32 + IF k[j] + dst[i+31:i] := (a[i+31:i] * b[i+31:i]) - c[i+31:i] + ELSE + dst[i+31:i] := c[i+31:i] + FI +ENDFOR +dst[MAX:256] := 0 - + + +
immintrin.h
 - + Floating Point + AVX512VL AVX512F - Special Math Functions - - - - Compare the lower single-precision (32-bit) floating-point elements in "a" and "b", store the minimum value in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". + Arithmetic + + + + + + Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", subtract packed elements in "c" from the intermediate result, and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). -IF k[0] - dst[31:0] := MIN(a[31:0], b[31:0]) -ELSE - dst[31:0] := 0 -FI -dst[127:32] := a[127:32] -dst[MAX:128] := 0 +FOR j := 0 to 7 + i := j*32 + IF k[j] + dst[i+31:i] := (a[i+31:i] * b[i+31:i]) - c[i+31:i] + ELSE + dst[i+31:i] := a[i+31:i] + FI +ENDFOR +dst[MAX:256] := 0 - + + +
immintrin.h
 - + Floating Point + AVX512VL AVX512F - Special Math Functions - - - - Compare the lower single-precision (32-bit) floating-point elements in "a" and "b", store the minimum value in the lower element of "dst", and copy the upper element from "a" to the upper element of "dst". - Pass __MM_FROUND_NO_EXC to "sae" to suppress all exceptions. + Arithmetic + + + + + + Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", subtract packed elements in "c" from the intermediate result, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -dst[31:0] := MIN(a[31:0], b[31:0]) -dst[127:32] := a[127:32] -dst[MAX:128] := 0 +FOR j := 0 to 7 + i := j*32 + IF k[j] + dst[i+31:i] := (a[i+31:i] * b[i+31:i]) - c[i+31:i] + ELSE + dst[i+31:i] := 0 + FI +ENDFOR +dst[MAX:256] := 0 - + + +
immintrin.h
 - + Floating Point - AVX512F/KNCNI - Load - - Load 512-bits (composed of 8 packed double-precision (64-bit) floating-point elements) from memory into "dst". - "mem_addr" must be aligned on a 64-byte boundary or a general-protection exception may be generated. + AVX512VL + AVX512F + Arithmetic + + + + + + Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", subtract packed elements in "c" from the intermediate result, and store the results in "dst" using writemask "k" (elements are copied from "c" when the corresponding mask bit is not set). -dst[511:0] := MEM[mem_addr+511:mem_addr] -dst[MAX:512] := 0 +FOR j := 0 to 3 + i := j*32 + IF k[j] + dst[i+31:i] := (a[i+31:i] * b[i+31:i]) - c[i+31:i] + ELSE + dst[i+31:i] := c[i+31:i] + FI +ENDFOR +dst[MAX:128] := 0 - + + +
immintrin.h
 - + Floating Point - AVX512F/KNCNI - Load - - - - Load packed double-precision (64-bit) floating-point elements from memory into "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). "mem_addr" must be aligned on a 64-byte boundary or a general-protection exception may be generated. + AVX512VL + AVX512F + Arithmetic + + + + + + Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", subtract packed elements in "c" from the intermediate result, and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := j*64 +FOR j := 0 to 3 + i := j*32 IF k[j] - dst[i+63:i] := MEM[mem_addr+i+63:mem_addr+i] + dst[i+31:i] := (a[i+31:i] * b[i+31:i]) - c[i+31:i] ELSE - dst[i+63:i] := src[i+63:i] + dst[i+31:i] := a[i+31:i] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - + + +
immintrin.h
 - + Floating Point - AVX512F/KNCNI - Move - - - - Move packed double-precision (64-bit) floating-point elements from "a" to "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + AVX512VL + AVX512F + Arithmetic + + + + + + Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", subtract packed elements in "c" from the intermediate result, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := j*64 +FOR j := 0 to 3 + i := j*32 IF k[j] - dst[i+63:i] := a[i+63:i] + dst[i+31:i] := (a[i+31:i] * b[i+31:i]) - c[i+31:i] ELSE - dst[i+63:i] := src[i+63:i] + dst[i+31:i] := 0 FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - + + +
immintrin.h
 - + Floating Point - AVX512F/KNCNI - Store - - - - Store packed double-precision (64-bit) floating-point elements from "a" into memory using writemask "k". - "mem_addr" must be aligned on a 64-byte boundary or a general-protection exception may be generated. + AVX512VL + AVX512F + Arithmetic + + + + + + Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", alternatively subtract and add packed elements in "c" from/to the intermediate result, and store the results in "dst" using writemask "k" (elements are copied from "c" when the corresponding mask bit is not set). -FOR j := 0 to 7 +FOR j := 0 to 3 i := j*64 IF k[j] - MEM[mem_addr+i+63:mem_addr+i] := a[i+63:i] + IF ((j & 1) == 0) + dst[i+63:i] := (a[i+63:i] * b[i+63:i]) + c[i+63:i] + ELSE + dst[i+63:i] := (a[i+63:i] * b[i+63:i]) - c[i+63:i] + FI + ELSE + dst[i+63:i] := c[i+63:i] FI ENDFOR +dst[MAX:256] := 0 - + + +
immintrin.h
 - + Floating Point + AVX512VL AVX512F - Load - - - Load packed double-precision (64-bit) floating-point elements from memory into "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). "mem_addr" must be aligned on a 64-byte boundary or a general-protection exception may be generated. + Arithmetic + + + + + + Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", alternatively subtract and add packed elements in "c" from/to the intermediate result, and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). -FOR j := 0 to 7 +FOR j := 0 to 3 i := j*64 IF k[j] - dst[i+63:i] := MEM[mem_addr+i+63:mem_addr+i] + IF ((j & 1) == 0) + dst[i+63:i] := (a[i+63:i] * b[i+63:i]) + c[i+63:i] + ELSE + dst[i+63:i] := (a[i+63:i] * b[i+63:i]) - c[i+63:i] + FI ELSE - dst[i+63:i] := 0 + dst[i+63:i] := a[i+63:i] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - + + +
immintrin.h
 - + Floating Point + AVX512VL AVX512F - Move - - - Move packed double-precision (64-bit) floating-point elements from "a" into "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Arithmetic + + + + + + Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", alternatively subtract and add packed elements in "c" from/to the intermediate result, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 +FOR j := 0 to 3 i := j*64 IF k[j] - dst[i+63:i] := a[i+63:i] + IF ((j & 1) == 0) + dst[i+63:i] := (a[i+63:i] * b[i+63:i]) + c[i+63:i] + ELSE + dst[i+63:i] := (a[i+63:i] * b[i+63:i]) - c[i+63:i] + FI ELSE dst[i+63:i] := 0 FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - + + +
immintrin.h
 - + Floating Point - AVX512F/KNCNI - Store - - - Store 512-bits (composed of 8 packed double-precision (64-bit) floating-point elements) from "a" into memory. - "mem_addr" must be aligned on a 64-byte boundary or a general-protection exception may be generated. + AVX512VL + AVX512F + Arithmetic + + + + + + Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", alternatively subtract and add packed elements in "c" from/to the intermediate result, and store the results in "dst" using writemask "k" (elements are copied from "c" when the corresponding mask bit is not set). -MEM[mem_addr+511:mem_addr] := a[511:0] +FOR j := 0 to 1 + i := j*64 + IF k[j] + IF ((j & 1) == 0) + dst[i+63:i] := (a[i+63:i] * b[i+63:i]) + c[i+63:i] + ELSE + dst[i+63:i] := (a[i+63:i] * b[i+63:i]) - c[i+63:i] + FI + ELSE + dst[i+63:i] := c[i+63:i] + FI +ENDFOR +dst[MAX:128] := 0 - + + +
immintrin.h
 - + Floating Point - AVX512F/KNCNI - Load - - Load 512-bits (composed of 16 packed single-precision (32-bit) floating-point elements) from memory into "dst". - "mem_addr" must be aligned on a 64-byte boundary or a general-protection exception may be generated. + AVX512VL + AVX512F + Arithmetic + + + + + + Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", alternatively subtract and add packed elements in "c" from/to the intermediate result, and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). -dst[511:0] := MEM[mem_addr+511:mem_addr] -dst[MAX:512] := 0 +FOR j := 0 to 1 + i := j*64 + IF k[j] + IF ((j & 1) == 0) + dst[i+63:i] := (a[i+63:i] * b[i+63:i]) + c[i+63:i] + ELSE + dst[i+63:i] := (a[i+63:i] * b[i+63:i]) - c[i+63:i] + FI + ELSE + dst[i+63:i] := a[i+63:i] + FI +ENDFOR +dst[MAX:128] := 0 - + + +
immintrin.h
 - + Floating Point - AVX512F/KNCNI - Load - - - - Load packed single-precision (32-bit) floating-point elements from memory into "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). "mem_addr" must be aligned on a 64-byte boundary or a general-protection exception may be generated. + AVX512VL + AVX512F + Arithmetic + + + + + + Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", alternatively subtract and add packed elements in "c" from/to the intermediate result, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*32 +FOR j := 0 to 1 + i := j*64 IF k[j] - dst[i+31:i] := MEM[mem_addr+i+31:mem_addr+i] + IF ((j & 1) == 0) + dst[i+63:i] := (a[i+63:i] * b[i+63:i]) + c[i+63:i] + ELSE + dst[i+63:i] := (a[i+63:i] * b[i+63:i]) - c[i+63:i] + FI ELSE - dst[i+31:i] := src[i+31:i] + dst[i+63:i] := 0 FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - + + +
immintrin.h
 - + Floating Point - AVX512F/KNCNI - Move - - - - Move packed single-precision (32-bit) floating-point elements from "a" to "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + AVX512VL + AVX512F + Arithmetic + + + + + + Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", alternatively subtract and add packed elements in "c" from/to the intermediate result, and store the results in "dst" using writemask "k" (elements are copied from "c" when the corresponding mask bit is not set). -FOR j := 0 to 15 +FOR j := 0 to 7 i := j*32 IF k[j] - dst[i+31:i] := a[i+31:i] + IF ((j & 1) == 0) + dst[i+31:i] := (a[i+31:i] * b[i+31:i]) + c[i+31:i] + ELSE + dst[i+31:i] := (a[i+31:i] * b[i+31:i]) - c[i+31:i] + FI ELSE - dst[i+31:i] := src[i+31:i] + dst[i+31:i] := c[i+31:i] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - + + +
immintrin.h
 - + Floating Point - AVX512F/KNCNI - Store - - - - Store packed single-precision (32-bit) floating-point elements from "a" into memory using writemask "k". - "mem_addr" must be aligned on a 64-byte boundary or a general-protection exception may be generated. + AVX512VL + AVX512F + Arithmetic + + + + + + Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", alternatively subtract and add packed elements in "c" from/to the intermediate result, and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). -FOR j := 0 to 15 +FOR j := 0 to 7 i := j*32 IF k[j] - MEM[mem_addr+i+31:mem_addr+i] := a[i+31:i] + IF ((j & 1) == 0) + dst[i+31:i] := (a[i+31:i] * b[i+31:i]) + c[i+31:i] + ELSE + dst[i+31:i] := (a[i+31:i] * b[i+31:i]) - c[i+31:i] + FI + ELSE + dst[i+31:i] := a[i+31:i] FI ENDFOR +dst[MAX:256] := 0 - + + +
immintrin.h
 - + Floating Point + AVX512VL AVX512F - Load - - - Load packed single-precision (32-bit) floating-point elements from memory into "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). "mem_addr" must be aligned on a 64-byte boundary or a general-protection exception may be generated. + Arithmetic + + + + + + Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", alternatively subtract and add packed elements in "c" from/to the intermediate result, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 15 +FOR j := 0 to 7 i := j*32 IF k[j] - dst[i+31:i] := MEM[mem_addr+i+31:mem_addr+i] + IF ((j & 1) == 0) + dst[i+31:i] := (a[i+31:i] * b[i+31:i]) + c[i+31:i] + ELSE + dst[i+31:i] := (a[i+31:i] * b[i+31:i]) - c[i+31:i] + FI ELSE dst[i+31:i] := 0 FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - + + +
immintrin.h
 - + Floating Point + AVX512VL AVX512F - Move - - - Move packed single-precision (32-bit) floating-point elements from "a" into "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Arithmetic + + + + + + Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", alternatively subtract and add packed elements in "c" from/to the intermediate result, and store the results in "dst" using writemask "k" (elements are copied from "c" when the corresponding mask bit is not set). -FOR j := 0 to 15 +FOR j := 0 to 3 i := j*32 IF k[j] - dst[i+31:i] := a[i+31:i] + IF ((j & 1) == 0) + dst[i+31:i] := (a[i+31:i] * b[i+31:i]) + c[i+31:i] + ELSE + dst[i+31:i] := (a[i+31:i] * b[i+31:i]) - c[i+31:i] + FI ELSE - dst[i+31:i] := 0 + dst[i+31:i] := c[i+31:i] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - + + +
immintrin.h
 - + Floating Point - AVX512F/KNCNI - Store - - - Store 512-bits (composed of 16 packed single-precision (32-bit) floating-point elements) from "a" into memory. - "mem_addr" must be aligned on a 64-byte boundary or a general-protection exception may be generated. + AVX512VL + AVX512F + Arithmetic + + + + + + Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", alternatively subtract and add packed elements in "c" from/to the intermediate result, and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). -MEM[mem_addr+511:mem_addr] := a[511:0] +FOR j := 0 to 3 + i := j*32 + IF k[j] + IF ((j & 1) == 0) + dst[i+31:i] := (a[i+31:i] * b[i+31:i]) + c[i+31:i] + ELSE + dst[i+31:i] := (a[i+31:i] * b[i+31:i]) - c[i+31:i] + FI + ELSE + dst[i+31:i] := a[i+31:i] + FI +ENDFOR +dst[MAX:128] := 0 - + + +
immintrin.h
 - + Floating Point + AVX512VL AVX512F - Move - - - - Duplicate even-indexed double-precision (64-bit) floating-point elements from "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Arithmetic + + + + + + Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", alternatively subtract and add packed elements in "c" from/to the intermediate result, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -tmp[63:0] := a[63:0] -tmp[127:64] := a[63:0] -tmp[191:128] := a[191:128] -tmp[255:192] := a[191:128] -tmp[319:256] := a[319:256] -tmp[383:320] := a[319:256] -tmp[447:384] := a[447:384] -tmp[511:448] := a[447:384] -FOR j := 0 to 7 - i := j*64 +FOR j := 0 to 3 + i := j*32 IF k[j] - dst[i+63:i] := tmp[i+63:i] + IF ((j & 1) == 0) + dst[i+31:i] := (a[i+31:i] * b[i+31:i]) + c[i+31:i] + ELSE + dst[i+31:i] := (a[i+31:i] * b[i+31:i]) - c[i+31:i] + FI ELSE - dst[i+63:i] := src[i+63:i] + dst[i+31:i] := 0 FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - + + +
immintrin.h
 - + Floating Point + AVX512VL AVX512F - Move - - - Duplicate even-indexed double-precision (64-bit) floating-point elements from "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Arithmetic + + + + + + Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", add the negated intermediate result to packed elements in "c", and store the results in "dst" using writemask "k" (elements are copied from "c" when the corresponding mask bit is not set). -tmp[63:0] := a[63:0] -tmp[127:64] := a[63:0] -tmp[191:128] := a[191:128] -tmp[255:192] := a[191:128] -tmp[319:256] := a[319:256] -tmp[383:320] := a[319:256] -tmp[447:384] := a[447:384] -tmp[511:448] := a[447:384] -FOR j := 0 to 7 +FOR j := 0 to 3 i := j*64 IF k[j] - dst[i+63:i] := tmp[i+63:i] + dst[i+63:i] := -(a[i+63:i] * b[i+63:i]) + c[i+63:i] ELSE - dst[i+63:i] := 0 + dst[i+63:i] := c[i+63:i] FI -ENDFOR -dst[MAX:512] := 0 +ENDFOR +dst[MAX:256] := 0 - + + +
immintrin.h
 - + Floating Point + AVX512VL AVX512F - Move - - Duplicate even-indexed double-precision (64-bit) floating-point elements from "a", and store the results in "dst". - + Arithmetic + + + + + + Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", add the negated intermediate result to packed elements in "c", and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). -tmp[63:0] := a[63:0] -tmp[127:64] := a[63:0] -tmp[191:128] := a[191:128] -tmp[255:192] := a[191:128] -tmp[319:256] := a[319:256] -tmp[383:320] := a[319:256] -tmp[447:384] := a[447:384] -tmp[511:448] := a[447:384] -dst[MAX:512] := 0 +FOR j := 0 to 3 + i := j*64 + IF k[j] + dst[i+63:i] := -(a[i+63:i] * b[i+63:i]) + c[i+63:i] + ELSE + dst[i+63:i] := a[i+63:i] + FI +ENDFOR +dst[MAX:256] := 0 - + + +
immintrin.h
 - - Integer - AVX512F/KNCNI - Load - - Load 512-bits (composed of 16 packed 32-bit integers) from memory into "dst". - "mem_addr" must be aligned on a 64-byte boundary or a general-protection exception may be generated. + + Floating Point + AVX512VL + AVX512F + Arithmetic + + + + + + Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", add the negated intermediate result to packed elements in "c", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -dst[511:0] := MEM[mem_addr+511:mem_addr] -dst[MAX:512] := 0 +FOR j := 0 to 3 + i := j*64 + IF k[j] + dst[i+63:i] := -(a[i+63:i] * b[i+63:i]) + c[i+63:i] + ELSE + dst[i+63:i] := 0 + FI +ENDFOR +dst[MAX:256] := 0 - + + +
immintrin.h
 - - Integer - AVX512F/KNCNI - Load - - Load 512-bits of integer data from memory into "dst". - "mem_addr" must be aligned on a 64-byte boundary or a general-protection exception may be generated. + + Floating Point + AVX512VL + AVX512F + Arithmetic + + + + + + Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", add the negated intermediate result to packed elements in "c", and store the results in "dst" using writemask "k" (elements are copied from "c" when the corresponding mask bit is not set). -dst[511:0] := MEM[mem_addr+511:mem_addr] -dst[MAX:512] := 0 +FOR j := 0 to 1 + i := j*64 + IF k[j] + dst[i+63:i] := -(a[i+63:i] * b[i+63:i]) + c[i+63:i] + ELSE + dst[i+63:i] := c[i+63:i] + FI +ENDFOR +dst[MAX:128] := 0 - + + +
immintrin.h
 - - Integer - AVX512F/KNCNI - Load - - - - Load packed 32-bit integers from memory into "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - "mem_addr" must be aligned on a 64-byte boundary or a general-protection exception may be generated. + + Floating Point + AVX512VL + AVX512F + Arithmetic + + + + + + Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", add the negated intermediate result to packed elements in "c", and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*32 +FOR j := 0 to 1 + i := j*64 IF k[j] - dst[i+31:i] := MEM[mem_addr+i+31:mem_addr+i] + dst[i+63:i] := -(a[i+63:i] * b[i+63:i]) + c[i+63:i] ELSE - dst[i+31:i] := src[i+31:i] + dst[i+63:i] := a[i+63:i] FI -ENDFOR -dst[MAX:512] := 0 +ENDFOR +dst[MAX:128] := 0 - + + +
immintrin.h
 - - Integer - AVX512F/KNCNI - Move - - - - Move packed 32-bit integers from "a" to "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Floating Point + AVX512VL + AVX512F + Arithmetic + + + + + + Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", add the negated intermediate result to packed elements in "c", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*32 +FOR j := 0 to 1 + i := j*64 IF k[j] - dst[i+31:i] := a[i+31:i] + dst[i+63:i] := -(a[i+63:i] * b[i+63:i]) + c[i+63:i] ELSE - dst[i+31:i] := src[i+31:i] + dst[i+63:i] := 0 FI -ENDFOR -dst[MAX:512] := 0 +ENDFOR +dst[MAX:128] := 0 - + + +
immintrin.h
 - - Integer - AVX512F/KNCNI - Store - - - - Store packed 32-bit integers from "a" into memory using writemask "k". - "mem_addr" must be aligned on a 64-byte boundary or a general-protection exception may be generated. - + + Floating Point + AVX512VL + AVX512F + Arithmetic + + + + + + Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", add the negated intermediate result to packed elements in "c", and store the results in "dst" using writemask "k" (elements are copied from "c" when the corresponding mask bit is not set). -FOR j := 0 to 15 +FOR j := 0 to 7 i := j*32 IF k[j] - MEM[mem_addr+i+31:mem_addr+i] := a[i+31:i] + dst[i+31:i] := -(a[i+31:i] * b[i+31:i]) + c[i+31:i] + ELSE + dst[i+31:i] := c[i+31:i] FI -ENDFOR +ENDFOR +dst[MAX:256] := 0 - + + +
immintrin.h
 - - Integer + + Floating Point + AVX512VL AVX512F - Load - - - Load packed 32-bit integers from memory into "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - "mem_addr" must be aligned on a 64-byte boundary or a general-protection exception may be generated. + Arithmetic + + + + + + Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", add the negated intermediate result to packed elements in "c", and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). -FOR j := 0 to 15 +FOR j := 0 to 7 i := j*32 IF k[j] - dst[i+31:i] := MEM[mem_addr+i+31:mem_addr+i] + dst[i+31:i] := -(a[i+31:i] * b[i+31:i]) + c[i+31:i] ELSE - dst[i+31:i] := 0 + dst[i+31:i] := a[i+31:i] FI -ENDFOR -dst[MAX:512] := 0 +ENDFOR +dst[MAX:256] := 0 - + + +
immintrin.h
 - - Integer + + Floating Point + AVX512VL AVX512F - Move - - - Move packed 32-bit integers from "a" into "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Arithmetic + + + + + + Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", add the negated intermediate result to packed elements in "c", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 15 +FOR j := 0 to 7 i := j*32 IF k[j] - dst[i+31:i] := a[i+31:i] + dst[i+31:i] := -(a[i+31:i] * b[i+31:i]) + c[i+31:i] ELSE dst[i+31:i] := 0 FI -ENDFOR -dst[MAX:512] := 0 +ENDFOR +dst[MAX:256] := 0 - + + +
immintrin.h
 - - Integer - AVX512F/KNCNI - Store - - - Store 512-bits (composed of 16 packed 32-bit integers) from "a" into memory. - "mem_addr" must be aligned on a 64-byte boundary or a general-protection exception may be generated. + + Floating Point + AVX512VL + AVX512F + Arithmetic + + + + + + Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", add the negated intermediate result to packed elements in "c", and store the results in "dst" using writemask "k" (elements are copied from "c" when the corresponding mask bit is not set). -MEM[mem_addr+511:mem_addr] := a[511:0] +FOR j := 0 to 3 + i := j*32 + IF k[j] + dst[i+31:i] := -(a[i+31:i] * b[i+31:i]) + c[i+31:i] + ELSE + dst[i+31:i] := c[i+31:i] + FI +ENDFOR +dst[MAX:128] := 0 - + + +
immintrin.h
 - - Integer - AVX512F/KNCNI - Store - - - Store 512-bits of integer data from "a" into memory. - "mem_addr" must be aligned on a 64-byte boundary or a general-protection exception may be generated. + + Floating Point + AVX512VL + AVX512F + Arithmetic + + + + + + Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", add the negated intermediate result to packed elements in "c", and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). -MEM[mem_addr+511:mem_addr] := a[511:0] +FOR j := 0 to 3 + i := j*32 + IF k[j] + dst[i+31:i] := -(a[i+31:i] * b[i+31:i]) + c[i+31:i] + ELSE + dst[i+31:i] := a[i+31:i] + FI +ENDFOR +dst[MAX:128] := 0 - + + +
immintrin.h
 - - Integer - AVX512F/KNCNI - Load - - Load 512-bits (composed of 8 packed 64-bit integers) from memory into "dst". - "mem_addr" must be aligned on a 64-byte boundary or a general-protection exception may be generated. + + Floating Point + AVX512VL + AVX512F + Arithmetic + + + + + + Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", add the negated intermediate result to packed elements in "c", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -dst[511:0] := MEM[mem_addr+511:mem_addr] -dst[MAX:512] := 0 +FOR j := 0 to 3 + i := j*32 + IF k[j] + dst[i+31:i] := -(a[i+31:i] * b[i+31:i]) + c[i+31:i] + ELSE + dst[i+31:i] := 0 + FI +ENDFOR +dst[MAX:128] := 0 - + + +
immintrin.h
 - - Integer - AVX512F/KNCNI - Load - - - - Load packed 64-bit integers from memory into "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - "mem_addr" must be aligned on a 64-byte boundary or a general-protection exception may be generated. + + Floating Point + AVX512VL + AVX512F + Arithmetic + + + + + + Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", subtract packed elements in "c" from the negated intermediate result, and store the results in "dst" using writemask "k" (elements are copied from "c" when the corresponding mask bit is not set). -FOR j := 0 to 7 +FOR j := 0 to 3 i := j*64 IF k[j] - dst[i+63:i] := MEM[mem_addr+i+63:mem_addr+i] + dst[i+63:i] := -(a[i+63:i] * b[i+63:i]) - c[i+63:i] ELSE - dst[i+63:i] := src[i+63:i] + dst[i+63:i] := c[i+63:i] FI -ENDFOR -dst[MAX:512] := 0 +ENDFOR +dst[MAX:256] := 0 - + + +
immintrin.h
 - - Integer - AVX512F/KNCNI - Move - - - - Move packed 64-bit integers from "a" to "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Floating Point + AVX512VL + AVX512F + Arithmetic + + + + + + Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", subtract packed elements in "c" from the negated intermediate result, and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). -FOR j := 0 to 7 +FOR j := 0 to 3 i := j*64 IF k[j] - dst[i+63:i] := a[i+63:i] + dst[i+63:i] := -(a[i+63:i] * b[i+63:i]) - c[i+63:i] ELSE - dst[i+63:i] := src[i+63:i] + dst[i+63:i] := a[i+63:i] FI -ENDFOR -dst[MAX:512] := 0 +ENDFOR +dst[MAX:256] := 0 - + + +
immintrin.h
 - - Integer - AVX512F/KNCNI - Store - - - - Store packed 64-bit integers from "a" into memory using writemask "k". - "mem_addr" must be aligned on a 64-byte boundary or a general-protection exception may be generated. - + + Floating Point + AVX512VL + AVX512F + Arithmetic + + + + + + Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", subtract packed elements in "c" from the negated intermediate result, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 +FOR j := 0 to 3 i := j*64 IF k[j] - MEM[mem_addr+i+63:mem_addr+i] := a[i+63:i] + dst[i+63:i] := -(a[i+63:i] * b[i+63:i]) - c[i+63:i] + ELSE + dst[i+63:i] := 0 FI -ENDFOR +ENDFOR +dst[MAX:256] := 0 - + + +
immintrin.h
 - - Integer + + Floating Point + AVX512VL AVX512F - Load - - - Load packed 64-bit integers from memory into "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - "mem_addr" must be aligned on a 64-byte boundary or a general-protection exception may be generated. - + Arithmetic + + + + + + Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", subtract packed elements in "c" from the negated intermediate result, and store the results in "dst" using writemask "k" (elements are copied from "c" when the corresponding mask bit is not set). -FOR j := 0 to 7 +FOR j := 0 to 1 i := j*64 IF k[j] - dst[i+63:i] := MEM[mem_addr+i+63:mem_addr+i] + dst[i+63:i] := -(a[i+63:i] * b[i+63:i]) - c[i+63:i] ELSE - dst[i+63:i] := 0 + dst[i+63:i] := c[i+63:i] FI -ENDFOR -dst[MAX:512] := 0 +ENDFOR +dst[MAX:128] := 0 - + + +
immintrin.h
 - - Integer + + Floating Point + AVX512VL AVX512F - Move - - - Move packed 64-bit integers from "a" into "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Arithmetic + + + + + + Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", subtract packed elements in "c" from the negated intermediate result, and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). -FOR j := 0 to 7 +FOR j := 0 to 1 i := j*64 IF k[j] + dst[i+63:i] := -(a[i+63:i] * b[i+63:i]) - c[i+63:i] + ELSE dst[i+63:i] := a[i+63:i] + FI +ENDFOR +dst[MAX:128] := 0 + + + + +
immintrin.h
+ + + Floating Point + AVX512VL + AVX512F + Arithmetic + + + + + + Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", subtract packed elements in "c" from the negated intermediate result, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + +FOR j := 0 to 1 + i := j*64 + IF k[j] + dst[i+63:i] := -(a[i+63:i] * b[i+63:i]) - c[i+63:i] ELSE dst[i+63:i] := 0 FI -ENDFOR -dst[MAX:512] := 0 +ENDFOR +dst[MAX:128] := 0 - + + +
immintrin.h
 - - Integer - AVX512F/KNCNI - Store - - - Store 512-bits (composed of 8 packed 64-bit integers) from "a" into memory. - "mem_addr" must be aligned on a 64-byte boundary or a general-protection exception may be generated. + + Floating Point + AVX512VL + AVX512F + Arithmetic + + + + + + Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", subtract packed elements in "c" from the negated intermediate result, and store the results in "dst" using writemask "k" (elements are copied from "c" when the corresponding mask bit is not set). -MEM[mem_addr+511:mem_addr] := a[511:0] +FOR j := 0 to 7 + i := j*32 + IF k[j] + dst[i+31:i] := -(a[i+31:i] * b[i+31:i]) - c[i+31:i] + ELSE + dst[i+31:i] := c[i+31:i] + FI +ENDFOR +dst[MAX:256] := 0 - + + +
immintrin.h
 - - Integer + + Floating Point + AVX512VL AVX512F - Load - - Load 512-bits of integer data from memory into "dst". - "mem_addr" does not need to be aligned on any particular boundary. - + Arithmetic + + + + + + Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", subtract packed elements in "c" from the negated intermediate result, and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). -dst[511:0] := MEM[mem_addr+511:mem_addr] -dst[MAX:512] := 0 +FOR j := 0 to 7 + i := j*32 + IF k[j] + dst[i+31:i] := -(a[i+31:i] * b[i+31:i]) - c[i+31:i] + ELSE + dst[i+31:i] := a[i+31:i] + FI +ENDFOR +dst[MAX:256] := 0 - + + +
immintrin.h
 - - Integer + + Floating Point + AVX512VL AVX512F - Load - - - - Load packed 32-bit integers from memory into "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - "mem_addr" does not need to be aligned on any particular boundary. - + Arithmetic + + + + + + Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", subtract packed elements in "c" from the negated intermediate result, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 15 +FOR j := 0 to 7 i := j*32 IF k[j] - dst[i+31:i] := MEM[mem_addr+i+31:mem_addr+i] + dst[i+31:i] := -(a[i+31:i] * b[i+31:i]) - c[i+31:i] ELSE - dst[i+31:i] := src[i+31:i] + dst[i+31:i] := 0 FI -ENDFOR -dst[MAX:512] := 0 +ENDFOR +dst[MAX:256] := 0 - + + +
immintrin.h
 - - Integer + + Floating Point + AVX512VL AVX512F - Store - - - - Store packed 32-bit integers from "a" into memory using writemask "k". - "mem_addr" does not need to be aligned on any particular boundary. - + Arithmetic + + + + + + Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", subtract packed elements in "c" from the negated intermediate result, and store the results in "dst" using writemask "k" (elements are copied from "c" when the corresponding mask bit is not set). -FOR j := 0 to 15 +FOR j := 0 to 3 i := j*32 IF k[j] - MEM[mem_addr+i+31:mem_addr+i] := a[i+31:i] + dst[i+31:i] := -(a[i+31:i] * b[i+31:i]) - c[i+31:i] + ELSE + dst[i+31:i] := c[i+31:i] FI -ENDFOR +ENDFOR +dst[MAX:128] := 0 - + + +
immintrin.h
 - - Integer + + Floating Point + AVX512VL AVX512F - Load - - - Load packed 32-bit integers from memory into "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - "mem_addr" does not need to be aligned on any particular boundary. - + Arithmetic + + + + + + Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", subtract packed elements in "c" from the negated intermediate result, and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). -FOR j := 0 to 15 +FOR j := 0 to 3 i := j*32 IF k[j] - dst[i+31:i] := MEM[mem_addr+i+31:mem_addr+i] + dst[i+31:i] := -(a[i+31:i] * b[i+31:i]) - c[i+31:i] ELSE - dst[i+31:i] := 0 + dst[i+31:i] := a[i+31:i] FI -ENDFOR -dst[MAX:512] := 0 +ENDFOR +dst[MAX:128] := 0 - + + +
immintrin.h
 - - Integer + + Floating Point + AVX512VL AVX512F - Store - - - Store 512-bits of integer data from "a" into memory. - "mem_addr" does not need to be aligned on any particular boundary. - + Arithmetic + + + + + + Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", subtract packed elements in "c" from the negated intermediate result, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -MEM[mem_addr+511:mem_addr] := a[511:0] +FOR j := 0 to 3 + i := j*32 + IF k[j] + dst[i+31:i] := -(a[i+31:i] * b[i+31:i]) - c[i+31:i] + ELSE + dst[i+31:i] := 0 + FI +ENDFOR +dst[MAX:128] := 0 - + + +
immintrin.h
 - - Integer + + Floating Point + AVX512VL AVX512F Load - - - - Load packed 64-bit integers from memory into "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - "mem_addr" does not need to be aligned on any particular boundary. - + + + + + + + Gather double-precision (64-bit) floating-point elements from memory using 32-bit indices. 64-bit elements are loaded from addresses starting at "base_addr" and offset by each 32-bit element in "vindex" (each index is scaled by the factor in "scale"). Gathered elements are merged into "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). "scale" should be 1, 2, 4 or 8. -FOR j := 0 to 7 +FOR j := 0 to 3 i := j*64 + m := j*32 IF k[j] - dst[i+63:i] := MEM[mem_addr+i+63:mem_addr+i] + addr := base_addr + SignExtend64(vindex[m+31:m]) * ZeroExtend64(scale) * 8 + dst[i+63:i] := MEM[addr+63:addr] ELSE dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - - Integer + + Floating Point + AVX512VL AVX512F - Store - - - - Store packed 64-bit integers from "a" into memory using writemask "k". - "mem_addr" does not need to be aligned on any particular boundary. - + Load + + + + + + + Gather double-precision (64-bit) floating-point elements from memory using 32-bit indices. 64-bit elements are loaded from addresses starting at "base_addr" and offset by each 32-bit element in "vindex" (each index is scaled by the factor in "scale"). Gathered elements are merged into "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). "scale" should be 1, 2, 4 or 8. -FOR j := 0 to 7 +FOR j := 0 to 1 i := j*64 + m := j*32 IF k[j] - MEM[mem_addr+i+63:mem_addr+i] := a[i+63:i] + addr := base_addr + SignExtend64(vindex[m+31:m]) * ZeroExtend64(scale) * 8 + dst[i+63:i] := MEM[addr+63:addr] + ELSE + dst[i+63:i] := src[i+63:i] FI ENDFOR +dst[MAX:128] := 0 - +
immintrin.h
 - - Integer + + Floating Point + AVX512VL AVX512F Load - - - Load packed 64-bit integers from memory into "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - "mem_addr" does not need to be aligned on any particular boundary. - + + + + + + + Gather single-precision (32-bit) floating-point elements from memory using 32-bit indices. 32-bit elements are loaded from addresses starting at "base_addr" and offset by each 32-bit element in "vindex" (each index is scaled by the factor in "scale"). Gathered elements are merged into "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). "scale" should be 1, 2, 4 or 8. FOR j := 0 to 7 - i := j*64 + i := j*32 + m := j*32 IF k[j] - dst[i+63:i] := MEM[mem_addr+i+63:mem_addr+i] + addr := base_addr + SignExtend64(vindex[m+31:m]) * ZeroExtend64(scale) * 8 + dst[i+31:i] := MEM[addr+31:addr] ELSE - dst[i+63:i] := 0 + dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - - Integer + + Floating Point + AVX512VL AVX512F Load - - Load 512-bits of integer data from memory into "dst" using a non-temporal memory hint. - "mem_addr" must be aligned on a 64-byte boundary or a general-protection exception may be generated. + + + + + + + Gather single-precision (32-bit) floating-point elements from memory using 32-bit indices. 32-bit elements are loaded from addresses starting at "base_addr" and offset by each 32-bit element in "vindex" (each index is scaled by the factor in "scale"). Gathered elements are merged into "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). "scale" should be 1, 2, 4 or 8. -dst[511:0] := MEM[mem_addr+511:mem_addr] -dst[MAX:512] := 0 +FOR j := 0 to 3 + i := j*32 + m := j*32 + IF k[j] + addr := base_addr + SignExtend64(vindex[m+31:m]) * ZeroExtend64(scale) * 8 + dst[i+31:i] := MEM[addr+31:addr] + ELSE + dst[i+31:i] := src[i+31:i] + FI +ENDFOR +dst[MAX:128] := 0 - +
immintrin.h
 - - Integer + + Floating Point + AVX512VL AVX512F - Store - - - Store 512-bits of integer data from "a" into memory using a non-temporal memory hint. - "mem_addr" must be aligned on a 64-byte boundary or a general-protection exception may be generated. + Load + + + + + + + Gather double-precision (64-bit) floating-point elements from memory using 64-bit indices. 64-bit elements are loaded from addresses starting at "base_addr" and offset by each 64-bit element in "vindex" (each index is scaled by the factor in "scale"). Gathered elements are merged into "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). "scale" should be 1, 2, 4 or 8. -MEM[mem_addr+511:mem_addr] := a[511:0] +FOR j := 0 to 3 + i := j*64 + m := j*64 + IF k[j] + addr := base_addr + vindex[m+63:m] * ZeroExtend64(scale) * 8 + dst[i+63:i] := MEM[addr+63:addr] + ELSE + dst[i+63:i] := src[i+63:i] + FI +ENDFOR +dst[MAX:256] := 0 - +
immintrin.h
 - + Floating Point + AVX512VL AVX512F - Store - - - Store 512-bits (composed of 8 packed double-precision (64-bit) floating-point elements) from "a" into memory using a non-temporal memory hint. - "mem_addr" must be aligned on a 64-byte boundary or a general-protection exception may be generated. + Load + + + + + + + Gather double-precision (64-bit) floating-point elements from memory using 64-bit indices. 64-bit elements are loaded from addresses starting at "base_addr" and offset by each 64-bit element in "vindex" (each index is scaled by the factor in "scale"). Gathered elements are merged into "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). "scale" should be 1, 2, 4 or 8. -MEM[mem_addr+511:mem_addr] := a[511:0] +FOR j := 0 to 1 + i := j*64 + m := j*64 + IF k[j] + addr := base_addr + vindex[m+63:m] * ZeroExtend64(scale) * 8 + dst[i+63:i] := MEM[addr+63:addr] + ELSE + dst[i+63:i] := src[i+63:i] + FI +ENDFOR +dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point + AVX512VL AVX512F - Store - - - Store 512-bits (composed of 16 packed single-precision (32-bit) floating-point elements) from "a" into memory using a non-temporal memory hint. - "mem_addr" must be aligned on a 64-byte boundary or a general-protection exception may be generated. + Load + + + + + + + Gather single-precision (32-bit) floating-point elements from memory using 64-bit indices. 32-bit elements are loaded from addresses starting at "base_addr" and offset by each 64-bit element in "vindex" (each index is scaled by the factor in "scale"). Gathered elements are merged into "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). "scale" should be 1, 2, 4 or 8. -MEM[mem_addr+511:mem_addr] := a[511:0] +FOR j := 0 to 3 + i := j*32 + m := j*64 + IF k[j] + addr := base_addr + vindex[m+63:m] * ZeroExtend64(scale) * 8 + dst[i+31:i] := MEM[addr+31:addr] + ELSE + dst[i+31:i] := src[i+31:i] + FI +ENDFOR +dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point + AVX512VL AVX512F Load - - - - Load a double-precision (64-bit) floating-point element from memory into the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and set the upper element of "dst" to zero. "mem_addr" must be aligned on a 16-byte boundary or a general-protection exception may be generated. + + + + + + + Gather single-precision (32-bit) floating-point elements from memory using 64-bit indices. 32-bit elements are loaded from addresses starting at "base_addr" and offset by each 64-bit element in "vindex" (each index is scaled by the factor in "scale"). Gathered elements are merged into "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). "scale" should be 1, 2, 4 or 8. -IF k[0] - dst[63:0] := MEM[mem_addr+63:mem_addr] -ELSE - dst[63:0] := src[63:0] -FI +FOR j := 0 to 1 + i := j*32 + m := j*64 + IF k[j] + addr := base_addr + vindex[m+63:m] * ZeroExtend64(scale) * 8 + dst[i+31:i] := MEM[addr+31:addr] + ELSE + dst[i+31:i] := src[i+31:i] + FI +ENDFOR dst[MAX:64] := 0 - +
immintrin.h
 - + Floating Point + AVX512VL AVX512F - Move - - - - - Move the lower double-precision (64-bit) floating-point element from "b" to the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". - -IF k[0] - dst[63:0] := b[63:0] -ELSE - dst[63:0] := src[63:0] -FI -dst[127:64] := a[127:64] -dst[MAX:128] := 0 + Miscellaneous + + + Convert the exponent of each packed double-precision (64-bit) floating-point element in "a" to a double-precision (64-bit) floating-point number representing the integer exponent, and store the results in "dst". This intrinsic essentially calculates "floor(log2(x))" for each element. + FOR j := 0 to 3 + i := j*64 + dst[i+63:i] := ConvertExpFP64(a[i+63:i]) +ENDFOR +dst[MAX:256] := 0 - +
immintrin.h
 - + Floating Point + AVX512VL AVX512F - Store - - - - Store the lower double-precision (64-bit) floating-point element from "a" into memory using writemask "k". - "mem_addr" must be aligned on a 16-byte boundary or a general-protection exception may be generated. - -IF k[0] - MEM[mem_addr+63:mem_addr] := a[63:0] -FI + Miscellaneous + + + + + Convert the exponent of each packed double-precision (64-bit) floating-point element in "a" to a double-precision (64-bit) floating-point number representing the integer exponent, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). This intrinsic essentially calculates "floor(log2(x))" for each element. + FOR j := 0 to 3 + i := j*64 + IF k[j] + dst[i+63:i] := ConvertExpFP64(a[i+63:i]) + ELSE + dst[i+63:i] := src[i+63:i] + FI +ENDFOR +dst[MAX:256] := 0 - +
immintrin.h
 - + Floating Point + AVX512VL AVX512F - Load - - - Load a double-precision (64-bit) floating-point element from memory into the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and set the upper element of "dst" to zero. "mem_addr" must be aligned on a 16-byte boundary or a general-protection exception may be generated. - -IF k[0] - dst[63:0] := MEM[mem_addr+63:mem_addr] -ELSE - dst[63:0] := 0 -FI -dst[MAX:64] := 0 + Miscellaneous + + + + Convert the exponent of each packed double-precision (64-bit) floating-point element in "a" to a double-precision (64-bit) floating-point number representing the integer exponent, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). This intrinsic essentially calculates "floor(log2(x))" for each element. + FOR j := 0 to 3 + i := j*64 + IF k[j] + dst[i+63:i] := ConvertExpFP64(a[i+63:i]) + ELSE + dst[i+63:i] := 0 + FI +ENDFOR +dst[MAX:256] := 0 - +
immintrin.h
 - + Floating Point + AVX512VL AVX512F - Move - - - - Move the lower double-precision (64-bit) floating-point element from "b" to the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". - -IF k[0] - dst[63:0] := b[63:0] -ELSE - dst[63:0] := 0 -FI -dst[127:64] := a[127:64] + Miscellaneous + + + Convert the exponent of each packed double-precision (64-bit) floating-point element in "a" to a double-precision (64-bit) floating-point number representing the integer exponent, and store the results in "dst". This intrinsic essentially calculates "floor(log2(x))" for each element. + FOR j := 0 to 1 + i := j*64 + dst[i+63:i] := ConvertExpFP64(a[i+63:i]) +ENDFOR dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point + AVX512VL AVX512F - Move - - - - Duplicate odd-indexed single-precision (32-bit) floating-point elements from "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -tmp[31:0] := a[63:32] -tmp[63:32] := a[63:32] -tmp[95:64] := a[127:96] -tmp[127:96] := a[127:96] -tmp[159:128] := a[191:160] -tmp[191:160] := a[191:160] -tmp[223:192] := a[255:224] -tmp[255:224] := a[255:224] -tmp[287:256] := a[319:288] -tmp[319:288] := a[319:288] -tmp[351:320] := a[383:352] -tmp[383:352] := a[383:352] -tmp[415:384] := a[447:416] -tmp[447:416] := a[447:416] -tmp[479:448] := a[511:480] -tmp[511:480] := a[511:480] -FOR j := 0 to 15 - i := j*32 + Miscellaneous + + + + + Convert the exponent of each packed double-precision (64-bit) floating-point element in "a" to a double-precision (64-bit) floating-point number representing the integer exponent, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). This intrinsic essentially calculates "floor(log2(x))" for each element. + FOR j := 0 to 1 + i := j*64 IF k[j] - dst[i+31:i] := tmp[i+31:i] + dst[i+63:i] := ConvertExpFP64(a[i+63:i]) ELSE - dst[i+31:i] := src[i+31:i] + dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point + AVX512VL AVX512F - Move - - - Duplicate odd-indexed single-precision (32-bit) floating-point elements from "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - -tmp[31:0] := a[63:32] -tmp[63:32] := a[63:32] -tmp[95:64] := a[127:96] -tmp[127:96] := a[127:96] -tmp[159:128] := a[191:160] -tmp[191:160] := a[191:160] -tmp[223:192] := a[255:224] -tmp[255:224] := a[255:224] -tmp[287:256] := a[319:288] -tmp[319:288] := a[319:288] -tmp[351:320] := a[383:352] -tmp[383:352] := a[383:352] -tmp[415:384] := a[447:416] -tmp[447:416] := a[447:416] -tmp[479:448] := a[511:480] -tmp[511:480] := a[511:480] -FOR j := 0 to 15 - i := j*32 + Miscellaneous + + + + Convert the exponent of each packed double-precision (64-bit) floating-point element in "a" to a double-precision (64-bit) floating-point number representing the integer exponent, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). This intrinsic essentially calculates "floor(log2(x))" for each element. + FOR j := 0 to 1 + i := j*64 IF k[j] - dst[i+31:i] := tmp[i+31:i] + dst[i+63:i] := ConvertExpFP64(a[i+63:i]) ELSE - dst[i+31:i] := 0 + dst[i+63:i] := 0 FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point + AVX512VL AVX512F - Move - - Duplicate odd-indexed single-precision (32-bit) floating-point elements from "a", and store the results in "dst". - - -dst[31:0] := a[63:32] -dst[63:32] := a[63:32] -dst[95:64] := a[127:96] -dst[127:96] := a[127:96] -dst[159:128] := a[191:160] -dst[191:160] := a[191:160] -dst[223:192] := a[255:224] -dst[255:224] := a[255:224] -dst[287:256] := a[319:288] -dst[319:288] := a[319:288] -dst[351:320] := a[383:352] -dst[383:352] := a[383:352] -dst[415:384] := a[447:416] -dst[447:416] := a[447:416] -dst[479:448] := a[511:480] -dst[511:480] := a[511:480] -dst[MAX:512] := 0 + Miscellaneous + + + Convert the exponent of each packed single-precision (32-bit) floating-point element in "a" to a single-precision (32-bit) floating-point number representing the integer exponent, and store the results in "dst". This intrinsic essentially calculates "floor(log2(x))" for each element. + FOR j := 0 to 7 + i := j*32 + dst[i+31:i] := ConvertExpFP32(a[i+31:i]) +ENDFOR +dst[MAX:256] := 0 - +
immintrin.h
 - + Floating Point + AVX512VL AVX512F - Move - - - - Duplicate even-indexed single-precision (32-bit) floating-point elements from "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -tmp[31:0] := a[31:0] -tmp[63:32] := a[31:0] -tmp[95:64] := a[95:64] -tmp[127:96] := a[95:64] -tmp[159:128] := a[159:128] -tmp[191:160] := a[159:128] -tmp[223:192] := a[223:192] -tmp[255:224] := a[223:192] -tmp[287:256] := a[287:256] -tmp[319:288] := a[287:256] -tmp[351:320] := a[351:320] -tmp[383:352] := a[351:320] -tmp[415:384] := a[415:384] -tmp[447:416] := a[415:384] -tmp[479:448] := a[479:448] -tmp[511:480] := a[479:448] -FOR j := 0 to 15 + Miscellaneous + + + + + Convert the exponent of each packed single-precision (32-bit) floating-point element in "a" to a single-precision (32-bit) floating-point number representing the integer exponent, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). This intrinsic essentially calculates "floor(log2(x))" for each element. + FOR j := 0 to 7 i := j*32 IF k[j] - dst[i+31:i] := tmp[i+31:i] + dst[i+31:i] := ConvertExpFP32(a[i+31:i]) ELSE dst[i+31:i] := src[i+31:i] FI -ENDFOR -dst[MAX:512] := 0 +ENDFOR +dst[MAX:256] := 0 - +
immintrin.h
 - + Floating Point + AVX512VL AVX512F - Move - - - Duplicate even-indexed single-precision (32-bit) floating-point elements from "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - -tmp[31:0] := a[31:0] -tmp[63:32] := a[31:0] -tmp[95:64] := a[95:64] -tmp[127:96] := a[95:64] -tmp[159:128] := a[159:128] -tmp[191:160] := a[159:128] -tmp[223:192] := a[223:192] -tmp[255:224] := a[223:192] -tmp[287:256] := a[287:256] -tmp[319:288] := a[287:256] -tmp[351:320] := a[351:320] -tmp[383:352] := a[351:320] -tmp[415:384] := a[415:384] -tmp[447:416] := a[415:384] -tmp[479:448] := a[479:448] -tmp[511:480] := a[479:448] -FOR j := 0 to 15 + Miscellaneous + + + + Convert the exponent of each packed single-precision (32-bit) floating-point element in "a" to a single-precision (32-bit) floating-point number representing the integer exponent, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). This intrinsic essentially calculates "floor(log2(x))" for each element. + FOR j := 0 to 7 i := j*32 IF k[j] - dst[i+31:i] := tmp[i+31:i] + dst[i+31:i] := ConvertExpFP32(a[i+31:i]) ELSE dst[i+31:i] := 0 FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Floating Point + AVX512VL AVX512F - Move - - Duplicate even-indexed single-precision (32-bit) floating-point elements from "a", and store the results in "dst". - - -dst[31:0] := a[31:0] -dst[63:32] := a[31:0] -dst[95:64] := a[95:64] -dst[127:96] := a[95:64] -dst[159:128] := a[159:128] -dst[191:160] := a[159:128] -dst[223:192] := a[223:192] -dst[255:224] := a[223:192] -dst[287:256] := a[287:256] -dst[319:288] := a[287:256] -dst[351:320] := a[351:320] -dst[383:352] := a[351:320] -dst[415:384] := a[415:384] -dst[447:416] := a[415:384] -dst[479:448] := a[479:448] -dst[511:480] := a[479:448] -dst[MAX:512] := 0 - - -
immintrin.h
+ Miscellaneous + + + Convert the exponent of each packed single-precision (32-bit) floating-point element in "a" to a single-precision (32-bit) floating-point number representing the integer exponent, and store the results in "dst". This intrinsic essentially calculates "floor(log2(x))" for each element. + FOR j := 0 to 3 + i := j*32 + dst[i+31:i] := ConvertExpFP32(a[i+31:i]) +ENDFOR +dst[MAX:128] := 0 + + +
immintrin.h
 - + Floating Point + AVX512VL AVX512F - Load - - - - Load a single-precision (32-bit) floating-point element from memory into the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and set the upper elements of "dst" to zero. "mem_addr" must be aligned on a 16-byte boundary or a general-protection exception may be generated. - -IF k[0] - dst[31:0] := MEM[mem_addr+31:mem_addr] -ELSE - dst[31:0] := src[31:0] -FI -dst[MAX:32] := 0 + Miscellaneous + + + + + Convert the exponent of each packed single-precision (32-bit) floating-point element in "a" to a single-precision (32-bit) floating-point number representing the integer exponent, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). This intrinsic essentially calculates "floor(log2(x))" for each element. + FOR j := 0 to 3 + i := j*32 + IF k[j] + dst[i+31:i] := ConvertExpFP32(a[i+31:i]) + ELSE + dst[i+31:i] := src[i+31:i] + FI +ENDFOR +dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point + AVX512VL AVX512F - Move - - - - - Move the lower single-precision (32-bit) floating-point element from "b" to the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". - -IF k[0] - dst[31:0] := b[31:0] -ELSE - dst[31:0] := src[31:0] -FI -dst[127:32] := a[127:32] + Miscellaneous + + + + Convert the exponent of each packed single-precision (32-bit) floating-point element in "a" to a single-precision (32-bit) floating-point number representing the integer exponent, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). This intrinsic essentially calculates "floor(log2(x))" for each element. + FOR j := 0 to 3 + i := j*32 + IF k[j] + dst[i+31:i] := ConvertExpFP32(a[i+31:i]) + ELSE + dst[i+31:i] := 0 + FI +ENDFOR dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point + AVX512VL AVX512F - Store - - - - Store the lower single-precision (32-bit) floating-point element from "a" into memory using writemask "k". - "mem_addr" must be aligned on a 16-byte boundary or a general-protection exception may be generated. - -IF k[0] - MEM[mem_addr+31:mem_addr] := a[31:0] -FI + Miscellaneous + + + + + Normalize the mantissas of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst". This intrinsic essentially calculates "±(2^k)*|x.significand|", where "k" depends on the interval range defined by "interv" and the sign depends on "sc" and the source sign. + [getmant_note] + FOR j := 0 to 3 + i := j*64 + dst[i+63:i] := GetNormalizedMantissa(a[i+63:i], sc, interv) +ENDFOR +dst[MAX:256] := 0 - +
immintrin.h
 - + Floating Point + AVX512VL AVX512F - Load - - - Load a single-precision (32-bit) floating-point element from memory into the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and set the upper elements of "dst" to zero. "mem_addr" must be aligned on a 16-byte boundary or a general-protection exception may be generated. - -IF k[0] - dst[31:0] := MEM[mem_addr+31:mem_addr] -ELSE - dst[31:0] := 0 -FI -dst[MAX:32] := 0 + Miscellaneous + + + + + + + Normalize the mantissas of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). This intrinsic essentially calculates "±(2^k)*|x.significand|", where "k" depends on the interval range defined by "interv" and the sign depends on "sc" and the source sign. + [getmant_note] + FOR j := 0 to 3 + i := j*64 + IF k[j] + dst[i+63:i] := GetNormalizedMantissa(a[i+63:i], sc, interv) + ELSE + dst[i+63:i] := src[i+63:i] + FI +ENDFOR +dst[MAX:256] := 0 - +
immintrin.h
 - + Floating Point + AVX512VL AVX512F - Move - - - - Move the lower single-precision (32-bit) floating-point element from "b" to the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". - -IF k[0] - dst[31:0] := b[31:0] -ELSE - dst[31:0] := 0 -FI -dst[127:32] := a[127:32] -dst[MAX:128] := 0 + Miscellaneous + + + + + + Normalize the mantissas of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). This intrinsic essentially calculates "±(2^k)*|x.significand|", where "k" depends on the interval range defined by "interv" and the sign depends on "sc" and the source sign. + [getmant_note] + FOR j := 0 to 3 + i := j*64 + IF k[j] + dst[i+63:i] := GetNormalizedMantissa(a[i+63:i], sc, interv) + ELSE + dst[i+63:i] := 0 + FI +ENDFOR +dst[MAX:256] := 0 - +
immintrin.h
 - + Floating Point + AVX512VL AVX512F - Load - - Load 512-bits (composed of 8 packed double-precision (64-bit) floating-point elements) from memory into "dst". - "mem_addr" does not need to be aligned on any particular boundary. - -dst[511:0] := MEM[mem_addr+511:mem_addr] -dst[MAX:512] := 0 + Miscellaneous + + + + + Normalize the mantissas of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst". This intrinsic essentially calculates "±(2^k)*|x.significand|", where "k" depends on the interval range defined by "interv" and the sign depends on "sc" and the source sign. + [getmant_note] + FOR j := 0 to 1 + i := j*64 + dst[i+63:i] := GetNormalizedMantissa(a[i+63:i], sc, interv) +ENDFOR +dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point + AVX512VL AVX512F - Load - - - - Load packed double-precision (64-bit) floating-point elements from memoy into "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - "mem_addr" does not need to be aligned on any particular boundary. - -FOR j := 0 to 7 + Miscellaneous + + + + + + + Normalize the mantissas of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). This intrinsic essentially calculates "±(2^k)*|x.significand|", where "k" depends on the interval range defined by "interv" and the sign depends on "sc" and the source sign. + [getmant_note] + FOR j := 0 to 1 i := j*64 IF k[j] - dst[i+63:i] := MEM[mem_addr+i+63:mem_addr+i] + dst[i+63:i] := GetNormalizedMantissa(a[i+63:i], sc, interv) ELSE dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point + AVX512VL AVX512F - Store - - - - Store packed double-precision (64-bit) floating-point elements from "a" into memory using writemask "k". - "mem_addr" does not need to be aligned on any particular boundary. - -FOR j := 0 to 7 + Miscellaneous + + + + + + Normalize the mantissas of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). This intrinsic essentially calculates "±(2^k)*|x.significand|", where "k" depends on the interval range defined by "interv" and the sign depends on "sc" and the source sign. + [getmant_note] + FOR j := 0 to 1 i := j*64 IF k[j] - MEM[mem_addr+i+63:mem_addr+i] := a[i+63:i] + dst[i+63:i] := GetNormalizedMantissa(a[i+63:i], sc, interv) + ELSE + dst[i+63:i] := 0 FI ENDFOR +dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point + AVX512VL AVX512F - Load - - - Load packed double-precision (64-bit) floating-point elements from memoy into "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - "mem_addr" does not need to be aligned on any particular boundary. - - -FOR j := 0 to 7 - i := j*64 + Miscellaneous + + + + + Normalize the mantissas of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst". This intrinsic essentially calculates "±(2^k)*|x.significand|", where "k" depends on the interval range defined by "interv" and the sign depends on "sc" and the source sign. + [getmant_note] + FOR j := 0 to 7 + i := j*32 + dst[i+31:i] := GetNormalizedMantissa(a[i+31:i], sc, interv) +ENDFOR +dst[MAX:256] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512VL + AVX512F + Miscellaneous + + + + + + + Normalize the mantissas of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). This intrinsic essentially calculates "±(2^k)*|x.significand|", where "k" depends on the interval range defined by "interv" and the sign depends on "sc" and the source sign. + [getmant_note] + FOR j := 0 to 7 + i := j*32 IF k[j] - dst[i+63:i] := MEM[mem_addr+i+63:mem_addr+i] + dst[i+31:i] := GetNormalizedMantissa(a[i+31:i], sc, interv) ELSE - dst[i+63:i] := 0 + dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Floating Point + AVX512VL AVX512F - Store - - - Store 512-bits (composed of 8 packed double-precision (64-bit) floating-point elements) from "a" into memory. - "mem_addr" does not need to be aligned on any particular boundary. - -MEM[mem_addr+511:mem_addr] := a[511:0] + Miscellaneous + + + + + + Normalize the mantissas of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). This intrinsic essentially calculates "±(2^k)*|x.significand|", where "k" depends on the interval range defined by "interv" and the sign depends on "sc" and the source sign. + [getmant_note] + FOR j := 0 to 7 + i := j*32 + IF k[j] + dst[i+31:i] := GetNormalizedMantissa(a[i+31:i], sc, interv) + ELSE + dst[i+31:i] := 0 + FI +ENDFOR +dst[MAX:256] := 0 - +
immintrin.h
 - + Floating Point + AVX512VL AVX512F - Load - - Load 512-bits (composed of 16 packed single-precision (32-bit) floating-point elements) from memory into "dst". - "mem_addr" does not need to be aligned on any particular boundary. - -dst[511:0] := MEM[mem_addr+511:mem_addr] -dst[MAX:512] := 0 + Miscellaneous + + + + + Normalize the mantissas of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst". This intrinsic essentially calculates "±(2^k)*|x.significand|", where "k" depends on the interval range defined by "interv" and the sign depends on "sc" and the source sign. + [getmant_note] + FOR j := 0 to 3 + i := j*32 + dst[i+31:i] := GetNormalizedMantissa(a[i+31:i], sc, interv) +ENDFOR +dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point + AVX512VL AVX512F - Load - - - - Load packed single-precision (32-bit) floating-point elements from memory into "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - "mem_addr" does not need to be aligned on any particular boundary. - - -FOR j := 0 to 15 + Miscellaneous + + + + + + + Normalize the mantissas of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). This intrinsic essentially calculates "±(2^k)*|x.significand|", where "k" depends on the interval range defined by "interv" and the sign depends on "sc" and the source sign. + [getmant_note] + FOR j := 0 to 3 i := j*32 IF k[j] - dst[i+31:i] := MEM[mem_addr+i+31:mem_addr+i] + dst[i+31:i] := GetNormalizedMantissa(a[i+31:i], sc, interv) ELSE dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point + AVX512VL AVX512F - Store - - - - Store packed single-precision (32-bit) floating-point elements from "a" into memory using writemask "k". - "mem_addr" does not need to be aligned on any particular boundary. - + Miscellaneous + + + + + + Normalize the mantissas of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). This intrinsic essentially calculates "±(2^k)*|x.significand|", where "k" depends on the interval range defined by "interv" and the sign depends on "sc" and the source sign. + [getmant_note] + FOR j := 0 to 3 + i := j*32 + IF k[j] + dst[i+31:i] := GetNormalizedMantissa(a[i+31:i], sc, interv) + ELSE + dst[i+31:i] := 0 + FI +ENDFOR +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512VL + AVX512F + Miscellaneous + + + + + Copy "a" to "dst", then insert 128 bits (composed of 4 packed single-precision (32-bit) floating-point elements) from "b" into "dst" at the location specified by "imm8". -FOR j := 0 to 15 +dst[255:0] := a[255:0] +CASE (imm8[0]) OF +0: dst[127:0] := b[127:0] +1: dst[255:128] := b[127:0] +ESAC +dst[MAX:256] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512VL + AVX512F + Miscellaneous + + + + + + + Copy "a" to "tmp", then insert 128 bits (composed of 4 packed single-precision (32-bit) floating-point elements) from "b" into "tmp" at the location specified by "imm8". Store "tmp" to "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +tmp[255:0] := a[255:0] +CASE (imm8[0]) OF +0: tmp[127:0] := b[127:0] +1: tmp[255:128] := b[127:0] +ESAC +FOR j := 0 to 7 i := j*32 IF k[j] - MEM[mem_addr+i+31:mem_addr+i] := a[i+31:i] + dst[i+31:i] := tmp[i+31:i] + ELSE + dst[i+31:i] := src[i+31:i] FI ENDFOR +dst[MAX:256] := 0 - +
immintrin.h
 - + Floating Point + AVX512VL AVX512F - Load - - - Load packed single-precision (32-bit) floating-point elements from memory into "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - "mem_addr" does not need to be aligned on any particular boundary. - + Miscellaneous + + + + + + Copy "a" to "tmp", then insert 128 bits (composed of 4 packed single-precision (32-bit) floating-point elements) from "b" into "tmp" at the location specified by "imm8". Store "tmp" to "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 15 +tmp[255:0] := a[255:0] +CASE (imm8[0]) OF +0: tmp[127:0] := b[127:0] +1: tmp[255:128] := b[127:0] +ESAC +FOR j := 0 to 7 i := j*32 IF k[j] - dst[i+31:i] := MEM[mem_addr+i+31:mem_addr+i] + dst[i+31:i] := tmp[i+31:i] ELSE dst[i+31:i] := 0 FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - - Floating Point + + AVX512VL AVX512F - Store - - - Store 512-bits (composed of 16 packed single-precision (32-bit) floating-point elements) from "a" into memory. - "mem_addr" does not need to be aligned on any particular boundary. + Miscellaneous + + + + + Copy "a" to "dst", then insert 128 bits (composed of 4 packed 32-bit integers) from "b" into "dst" at the location specified by "imm8". -MEM[mem_addr+511:mem_addr] := a[511:0] +dst[255:0] := a[255:0] +CASE (imm8[0]) OF +0: dst[127:0] := b[127:0] +1: dst[255:128] := b[127:0] +ESAC +dst[MAX:256] := 0 - +
immintrin.h
 - - Floating Point - AVX512F/KNCNI - Arithmetic - - - - - Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). RM. + + AVX512VL + AVX512F + Miscellaneous + + + + + + + Copy "a" to "tmp", then insert 128 bits (composed of 4 packed 32-bit integers) from "b" into "tmp" at the location specified by "imm8". Store "tmp" to "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). +tmp[255:0] := a[255:0] +CASE (imm8[0]) OF +0: tmp[127:0] := b[127:0] +1: tmp[255:128] := b[127:0] +ESAC FOR j := 0 to 7 - i := j*64 + i := j*32 IF k[j] - dst[i+63:i] := a[i+63:i] * b[i+63:i] + dst[i+31:i] := tmp[i+31:i] ELSE - dst[i+63:i] := src[i+63:i] + dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - - Floating Point - AVX512F/KNCNI - Arithmetic - - - - - - Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - [round_note] + + AVX512VL + AVX512F + Miscellaneous + + + + + + Copy "a" to "tmp", then insert 128 bits (composed of 4 packed 32-bit integers) from "b" into "tmp" at the location specified by "imm8". Store "tmp" to "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). +tmp[255:0] := a[255:0] +CASE (imm8[0]) OF +0: tmp[127:0] := b[127:0] +1: tmp[255:128] := b[127:0] +ESAC FOR j := 0 to 7 - i := j*64 + i := j*32 IF k[j] - dst[i+63:i] := a[i+63:i] * b[i+63:i] + dst[i+31:i] := tmp[i+31:i] ELSE - dst[i+63:i] := src[i+63:i] + dst[i+31:i] := 0 FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Floating Point + AVX512VL AVX512F Arithmetic - - - - Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + + + + + Compare packed double-precision (64-bit) floating-point elements in "a" and "b", and store packed maximum values in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 7 +FOR j := 0 to 3 i := j*64 IF k[j] - dst[i+63:i] := a[i+63:i] * b[i+63:i] + dst[i+63:i] := MAX(a[i+63:i], b[i+63:i]) ELSE - dst[i+63:i] := 0 + dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Floating Point + AVX512VL AVX512F Arithmetic - - - - - Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - [round_note] + + + + + Compare packed double-precision (64-bit) floating-point elements in "a" and "b", and store packed maximum values in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 +FOR j := 0 to 3 i := j*64 IF k[j] - dst[i+63:i] := a[i+63:i] * b[i+63:i] + dst[i+63:i] := MAX(a[i+63:i], b[i+63:i]) ELSE dst[i+63:i] := 0 FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Floating Point - AVX512F/KNCNI + AVX512VL + AVX512F Arithmetic - - - Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", and store the results in "dst". - + + + + + + Compare packed double-precision (64-bit) floating-point elements in "a" and "b", and store packed maximum values in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 7 +FOR j := 0 to 1 i := j*64 - dst[i+63:i] := a[i+63:i] * b[i+63:i] + IF k[j] + dst[i+63:i] := MAX(a[i+63:i], b[i+63:i]) + ELSE + dst[i+63:i] := src[i+63:i] + FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point - AVX512F/KNCNI + AVX512VL + AVX512F Arithmetic - - - - Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", and store the results in "dst". - [round_note] - + + + + + Compare packed double-precision (64-bit) floating-point elements in "a" and "b", and store packed maximum values in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 +FOR j := 0 to 1 i := j*64 - dst[i+63:i] := a[i+63:i] * b[i+63:i] + IF k[j] + dst[i+63:i] := MAX(a[i+63:i], b[i+63:i]) + ELSE + dst[i+63:i] := 0 + FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point - AVX512F/KNCNI + AVX512VL + AVX512F Arithmetic - - - - - Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). RM. + + + + + + Compare packed single-precision (32-bit) floating-point elements in "a" and "b", and store packed maximum values in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 15 +FOR j := 0 to 7 i := j*32 IF k[j] - dst[i+31:i] := a[i+31:i] * b[i+31:i] + dst[i+31:i] := MAX(a[i+31:i], b[i+31:i]) ELSE dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Floating Point - AVX512F/KNCNI + AVX512VL + AVX512F Arithmetic - - - - - - Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - [round_note] + + + + + Compare packed single-precision (32-bit) floating-point elements in "a" and "b", and store packed maximum values in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 15 +FOR j := 0 to 7 i := j*32 IF k[j] - dst[i+31:i] := a[i+31:i] * b[i+31:i] + dst[i+31:i] := MAX(a[i+31:i], b[i+31:i]) ELSE - dst[i+31:i] := src[i+31:i] + dst[i+31:i] := 0 FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Floating Point + AVX512VL AVX512F Arithmetic - - - - Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + + + + + Compare packed single-precision (32-bit) floating-point elements in "a" and "b", and store packed maximum values in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 15 +FOR j := 0 to 3 i := j*32 IF k[j] - dst[i+31:i] := a[i+31:i] * b[i+31:i] + dst[i+31:i] := MAX(a[i+31:i], b[i+31:i]) ELSE - dst[i+31:i] := 0 + dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point + AVX512VL AVX512F Arithmetic - - - - - Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - [round_note] + + + + + Compare packed single-precision (32-bit) floating-point elements in "a" and "b", and store packed maximum values in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 15 +FOR j := 0 to 3 i := j*32 IF k[j] - dst[i+31:i] := a[i+31:i] * b[i+31:i] + dst[i+31:i] := MAX(a[i+31:i], b[i+31:i]) ELSE dst[i+31:i] := 0 FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point - AVX512F/KNCNI + AVX512VL + AVX512F Arithmetic - - - Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", and store the results in "dst". - + + + + + + Compare packed double-precision (64-bit) floating-point elements in "a" and "b", and store packed minimum values in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*32 - dst[i+31:i] := a[i+31:i] * b[i+31:i] +FOR j := 0 to 3 + i := j*64 + IF k[j] + dst[i+63:i] := MIN(a[i+63:i], b[i+63:i]) + ELSE + dst[i+63:i] := src[i+63:i] + FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Floating Point - AVX512F/KNCNI + AVX512VL + AVX512F Arithmetic - - - - Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", and store the results in "dst". - [round_note] - + + + + + Compare packed double-precision (64-bit) floating-point elements in "a" and "b", and store packed minimum values in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*32 - dst[i+31:i] := a[i+31:i] * b[i+31:i] +FOR j := 0 to 3 + i := j*64 + IF k[j] + dst[i+63:i] := MIN(a[i+63:i], b[i+63:i]) + ELSE + dst[i+63:i] := 0 + FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Floating Point + AVX512VL AVX512F Arithmetic - - - - - - Multiply the lower double-precision (64-bit) floating-point element in "a" and "b", store the result in the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". - [round_note] - + + + + + + Compare packed double-precision (64-bit) floating-point elements in "a" and "b", and store packed minimum values in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -IF k[0] - dst[63:0] := a[63:0] * b[63:0] -ELSE - dst[63:0] := src[63:0] -FI -dst[127:64] := a[127:64] +FOR j := 0 to 1 + i := j*64 + IF k[j] + dst[i+63:i] := MIN(a[i+63:i], b[i+63:i]) + ELSE + dst[i+63:i] := src[i+63:i] + FI +ENDFOR dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point + AVX512VL AVX512F Arithmetic - - - - - Multiply the lower double-precision (64-bit) floating-point element in "a" and "b", store the result in the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". + + + + + Compare packed double-precision (64-bit) floating-point elements in "a" and "b", and store packed minimum values in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -IF k[0] - dst[63:0] := a[63:0] * b[63:0] -ELSE - dst[63:0] := src[63:0] -FI -dst[127:64] := a[127:64] +FOR j := 0 to 1 + i := j*64 + IF k[j] + dst[i+63:i] := MIN(a[i+63:i], b[i+63:i]) + ELSE + dst[i+63:i] := 0 + FI +ENDFOR dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point + AVX512VL AVX512F Arithmetic - - - - - Multiply the lower double-precision (64-bit) floating-point element in "a" and "b", store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". - [round_note] - + + + + + + Compare packed single-precision (32-bit) floating-point elements in "a" and "b", and store packed minimum values in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -IF k[0] - dst[63:0] := a[63:0] * b[63:0] -ELSE - dst[63:0] := 0 -FI -dst[127:64] := a[127:64] -dst[MAX:128] := 0 +FOR j := 0 to 7 + i := j*32 + IF k[j] + dst[i+31:i] := MIN(a[i+31:i], b[i+31:i]) + ELSE + dst[i+31:i] := src[i+31:i] + FI +ENDFOR +dst[MAX:256] := 0 - +
immintrin.h
 - + Floating Point + AVX512VL AVX512F Arithmetic - - - - Multiply the lower double-precision (64-bit) floating-point element in "a" and "b", store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". + + + + + Compare packed single-precision (32-bit) floating-point elements in "a" and "b", and store packed minimum values in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -IF k[0] - dst[63:0] := a[63:0] * b[63:0] -ELSE - dst[63:0] := 0 -FI -dst[127:64] := a[127:64] -dst[MAX:128] := 0 +FOR j := 0 to 7 + i := j*32 + IF k[j] + dst[i+31:i] := MIN(a[i+31:i], b[i+31:i]) + ELSE + dst[i+31:i] := 0 + FI +ENDFOR +dst[MAX:256] := 0 - +
immintrin.h
 - + Floating Point + AVX512VL AVX512F Arithmetic - - - - Multiply the lower double-precision (64-bit) floating-point element in "a" and "b", store the result in the lower element of "dst", and copy the upper element from "a" to the upper element of "dst". - [round_note] - + + + + + + Compare packed single-precision (32-bit) floating-point elements in "a" and "b", and store packed minimum values in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -dst[63:0] := a[63:0] * b[63:0] -dst[127:64] := a[127:64] +FOR j := 0 to 3 + i := j*32 + IF k[j] + dst[i+31:i] := MIN(a[i+31:i], b[i+31:i]) + ELSE + dst[i+31:i] := src[i+31:i] + FI +ENDFOR dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point + AVX512VL AVX512F Arithmetic - - - - - - Multiply the lower single-precision (32-bit) floating-point element in "a" and "b", store the result in the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". - [round_note] - + + + + + Compare packed single-precision (32-bit) floating-point elements in "a" and "b", and store packed minimum values in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -IF k[0] - dst[31:0] := a[31:0] * b[31:0] -ELSE - dst[31:0] := src[31:0] -FI -dst[127:32] := a[127:32] +FOR j := 0 to 3 + i := j*32 + IF k[j] + dst[i+31:i] := MIN(a[i+31:i], b[i+31:i]) + ELSE + dst[i+31:i] := 0 + FI +ENDFOR dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point + AVX512VL AVX512F - Arithmetic - - - - - Multiply the lower single-precision (32-bit) floating-point element in "a" and "b", store the result in the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". + Load + + + + + Load packed double-precision (64-bit) floating-point elements from memory into "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). "mem_addr" must be aligned on a 32-byte boundary or a general-protection exception may be generated. -IF k[0] - dst[31:0] := a[31:0] * b[31:0] -ELSE - dst[31:0] := src[31:0] -FI -dst[127:32] := a[127:32] -dst[MAX:128] := 0 +FOR j := 0 to 3 + i := j*64 + IF k[j] + dst[i+63:i] := MEM[mem_addr+i+63:mem_addr+i] + ELSE + dst[i+63:i] := src[i+63:i] + FI +ENDFOR +dst[MAX:256] := 0 - +
immintrin.h
 - + Floating Point + AVX512VL AVX512F - Arithmetic - - - - - Multiply the lower single-precision (32-bit) floating-point element in "a" and "b", store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". - [round_note] - + Move + + + + + Move packed double-precision (64-bit) floating-point elements from "a" to "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -IF k[0] - dst[31:0] := a[31:0] * b[31:0] -ELSE - dst[31:0] := 0 -FI -dst[127:32] := a[127:32] -dst[MAX:128] := 0 +FOR j := 0 to 3 + i := j*64 + IF k[j] + dst[i+63:i] := a[i+63:i] + ELSE + dst[i+63:i] := src[i+63:i] + FI +ENDFOR +dst[MAX:256] := 0 - +
immintrin.h
 - + Floating Point + AVX512VL AVX512F - Arithmetic - - - - Multiply the lower single-precision (32-bit) floating-point element in "a" and "b", store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". + Store + + + + + Store packed double-precision (64-bit) floating-point elements from "a" into memory using writemask "k". + "mem_addr" must be aligned on a 32-byte boundary or a general-protection exception may be generated. -IF k[0] - dst[31:0] := a[31:0] * b[31:0] -ELSE - dst[31:0] := 0 -FI -dst[127:32] := a[127:32] -dst[MAX:128] := 0 +FOR j := 0 to 3 + i := j*64 + IF k[j] + MEM[mem_addr+i+63:mem_addr+i] := a[i+63:i] + FI +ENDFOR - +
immintrin.h
 - + Floating Point + AVX512VL AVX512F - Arithmetic - - - - Multiply the lower single-precision (32-bit) floating-point element in "a" and "b", store the result in the lower element of "dst", and copy the upper 3 packed elements from "a" to the upper elements of "dst". - [round_note] - + Load + + + + Load packed double-precision (64-bit) floating-point elements from memory into "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). "mem_addr" must be aligned on a 32-byte boundary or a general-protection exception may be generated. -dst[31:0] := a[31:0] * b[31:0] -dst[127:32] := a[127:32] -dst[MAX:128] := 0 +FOR j := 0 to 3 + i := j*64 + IF k[j] + dst[i+63:i] := MEM[mem_addr+i+63:mem_addr+i] + ELSE + dst[i+63:i] := 0 + FI +ENDFOR +dst[MAX:256] := 0 - +
immintrin.h
 - - Integer + + Floating Point + AVX512VL AVX512F - Special Math Functions - - Compute the absolute value of packed 32-bit integers in "a", and store the unsigned results in "dst". + Move + + + + Move packed double-precision (64-bit) floating-point elements from "a" into "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*32 - dst[i+31:i] := ABS(a[i+31:i]) +FOR j := 0 to 3 + i := j*64 + IF k[j] + dst[i+63:i] := a[i+63:i] + ELSE + dst[i+63:i] := 0 + FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - - Integer + + Floating Point + AVX512VL AVX512F - Special Math Functions - - - - Compute the absolute value of packed 32-bit integers in "a", and store the unsigned results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Load + + + + + Load packed double-precision (64-bit) floating-point elements from memory into "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). "mem_addr" must be aligned on a 16-byte boundary or a general-protection exception may be generated. -FOR j := 0 to 15 - i := j*32 +FOR j := 0 to 1 + i := j*64 IF k[j] - dst[i+31:i] := ABS(a[i+31:i]) + dst[i+63:i] := MEM[mem_addr+i+63:mem_addr+i] ELSE - dst[i+31:i] := src[i+31:i] + dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - - Integer + + Floating Point + AVX512VL AVX512F - Special Math Functions - - - Compute the absolute value of packed 32-bit integers in "a", and store the unsigned results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Move + + + + + Move packed double-precision (64-bit) floating-point elements from "a" to "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*32 +FOR j := 0 to 1 + i := j*64 IF k[j] - dst[i+31:i] := ABS(a[i+31:i]) + dst[i+63:i] := a[i+63:i] ELSE - dst[i+31:i] := 0 + dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - - Integer + + Floating Point + AVX512VL AVX512F - Special Math Functions - - Compute the absolute value of packed 64-bit integers in "a", and store the unsigned results in "dst". + Store + + + + + Store packed double-precision (64-bit) floating-point elements from "a" into memory using writemask "k". + "mem_addr" must be aligned on a 16-byte boundary or a general-protection exception may be generated. -FOR j := 0 to 7 +FOR j := 0 to 1 i := j*64 - dst[i+63:i] := ABS(a[i+63:i]) + IF k[j] + MEM[mem_addr+i+63:mem_addr+i] := a[i+63:i] + FI ENDFOR -dst[MAX:512] := 0 - +
immintrin.h
 - - Integer + + Floating Point + AVX512VL AVX512F - Special Math Functions - - - - Compute the absolute value of packed 64-bit integers in "a", and store the unsigned results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Load + + + + Load packed double-precision (64-bit) floating-point elements from memory into "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). "mem_addr" must be aligned on a 16-byte boundary or a general-protection exception may be generated. -FOR j := 0 to 7 +FOR j := 0 to 1 i := j*64 IF k[j] - dst[i+63:i] := ABS(a[i+63:i]) + dst[i+63:i] := MEM[mem_addr+i+63:mem_addr+i] ELSE - dst[i+63:i] := src[i+63:i] + dst[i+63:i] := 0 FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - - Integer + + Floating Point + AVX512VL AVX512F - Special Math Functions - - - Compute the absolute value of packed 64-bit integers in "a", and store the unsigned results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Move + + + + Move packed double-precision (64-bit) floating-point elements from "a" into "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 +FOR j := 0 to 1 i := j*64 IF k[j] - dst[i+63:i] := ABS(a[i+63:i]) + dst[i+63:i] := a[i+63:i] ELSE dst[i+63:i] := 0 FI ENDFOR -dst[MAX:512] := 0 - - -
immintrin.h
- - - Integer - AVX512F/KNCNI - Arithmetic - - - Add packed 32-bit integers in "a" and "b", and store the results in "dst". - -FOR j := 0 to 15 - i := j*32 - dst[i+31:i] := a[i+31:i] + b[i+31:i] -ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - - Integer - AVX512F/KNCNI - Arithmetic - - - - - Add packed 32-bit integers in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Floating Point + AVX512VL + AVX512F + Load + + + + + Load packed single-precision (32-bit) floating-point elements from memory into "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). "mem_addr" must be aligned on a 32-byte boundary or a general-protection exception may be generated. -FOR j := 0 to 15 +FOR j := 0 to 7 i := j*32 IF k[j] - dst[i+31:i] := a[i+31:i] + b[i+31:i] + dst[i+31:i] := MEM[mem_addr+i+31:mem_addr+i] ELSE dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - - Integer + + Floating Point + AVX512VL AVX512F - Arithmetic - - - - Add packed 32-bit integers in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Move + + + + + Move packed single-precision (32-bit) floating-point elements from "a" to "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 15 +FOR j := 0 to 7 i := j*32 IF k[j] - dst[i+31:i] := a[i+31:i] + b[i+31:i] + dst[i+31:i] := a[i+31:i] ELSE - dst[i+31:i] := 0 + dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - - Integer + + Floating Point + AVX512VL AVX512F - Arithmetic - - - Add packed 64-bit integers in "a" and "b", and store the results in "dst". + Store + + + + + Store packed single-precision (32-bit) floating-point elements from "a" into memory using writemask "k". + "mem_addr" must be aligned on a 32-byte boundary or a general-protection exception may be generated. FOR j := 0 to 7 - i := j*64 - dst[i+63:i] := a[i+63:i] + b[i+63:i] + i := j*32 + IF k[j] + MEM[mem_addr+i+31:mem_addr+i] := a[i+31:i] + FI ENDFOR -dst[MAX:512] := 0 - +
immintrin.h
 - - Integer + + Floating Point + AVX512VL AVX512F - Arithmetic - - - - - Add packed 64-bit integers in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Load + + + + Load packed single-precision (32-bit) floating-point elements from memory into "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). "mem_addr" must be aligned on a 32-byte boundary or a general-protection exception may be generated. FOR j := 0 to 7 - i := j*64 + i := j*32 IF k[j] - dst[i+63:i] := a[i+63:i] + b[i+63:i] + dst[i+31:i] := MEM[mem_addr+i+31:mem_addr+i] ELSE - dst[i+63:i] := src[i+63:i] + dst[i+31:i] := 0 FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - - Integer + + Floating Point + AVX512VL AVX512F - Arithmetic - - - - Add packed 64-bit integers in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Move + + + + Move packed single-precision (32-bit) floating-point elements from "a" into "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). FOR j := 0 to 7 - i := j*64 + i := j*32 IF k[j] - dst[i+63:i] := a[i+63:i] + b[i+63:i] + dst[i+31:i] := a[i+31:i] ELSE - dst[i+63:i] := 0 + dst[i+31:i] := 0 FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - - Integer - AVX512F/KNCNI - Logical - - - Compute the bitwise AND of packed 32-bit integers in "a" and "b", and store the results in "dst". + + Floating Point + AVX512VL + AVX512F + Load + + + + + Load packed single-precision (32-bit) floating-point elements from memory into "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). "mem_addr" must be aligned on a 16-byte boundary or a general-protection exception may be generated. -FOR j := 0 to 15 +FOR j := 0 to 3 i := j*32 - dst[i+31:i] := a[i+31:i] BITWISE AND b[i+31:i] + IF k[j] + dst[i+31:i] := MEM[mem_addr+i+31:mem_addr+i] + ELSE + dst[i+31:i] := src[i+31:i] + FI ENDFOR -dst[MAX:512] := 0 - - -
immintrin.h
- - - Integer - AVX512F/KNCNI - Logical - - - Compute the bitwise AND of 512 bits (representing integer data) in "a" and "b", and store the result in "dst". - -dst[511:0] := (a[511:0] AND b[511:0]) -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - - Integer + + Floating Point + AVX512VL AVX512F - Logical - - - - Compute the bitwise AND of packed 32-bit integers in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Move + + + + + Move packed single-precision (32-bit) floating-point elements from "a" to "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 15 +FOR j := 0 to 3 i := j*32 IF k[j] - dst[i+31:i] := a[i+31:i] AND b[i+31:i] + dst[i+31:i] := a[i+31:i] ELSE - dst[i+31:i] := 0 + dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - - Integer - AVX512F/KNCNI - Logical - - - Compute the bitwise NOT of packed 32-bit integers in "a" and then AND with "b", and store the results in "dst". + + Floating Point + AVX512VL + AVX512F + Store + + + + + Store packed single-precision (32-bit) floating-point elements from "a" into memory using writemask "k". + "mem_addr" must be aligned on a 16-byte boundary or a general-protection exception may be generated. -FOR j := 0 to 15 +FOR j := 0 to 3 i := j*32 - dst[i+31:i] := (NOT a[i+31:i]) AND b[i+31:i] + IF k[j] + MEM[mem_addr+i+31:mem_addr+i] := a[i+31:i] + FI ENDFOR -dst[MAX:512] := 0 - - -
immintrin.h
- - - Integer - AVX512F/KNCNI - Logical - - - Compute the bitwise NOT of 512 bits (representing integer data) in "a" and then AND with "b", and store the result in "dst". - -dst[511:0] := ((NOT a[511:0]) AND b[511:0]) -dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - AVX512F/KNCNI - Logical - - - - - Compute the bitwise NOT of packed 32-bit integers in "a" and then AND with "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Floating Point + AVX512VL + AVX512F + Load + + + + Load packed single-precision (32-bit) floating-point elements from memory into "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). "mem_addr" must be aligned on a 16-byte boundary or a general-protection exception may be generated. -FOR j := 0 to 15 +FOR j := 0 to 3 i := j*32 IF k[j] - dst[i+31:i] := ((NOT a[i+31:i]) AND b[i+31:i]) + dst[i+31:i] := MEM[mem_addr+i+31:mem_addr+i] ELSE - dst[i+31:i] := src[i+31:i] + dst[i+31:i] := 0 FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - - Integer + + Floating Point + AVX512VL AVX512F - Logical - - - - Compute the bitwise NOT of packed 32-bit integers in "a" and then AND with "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Move + + + + Move packed single-precision (32-bit) floating-point elements from "a" into "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 15 +FOR j := 0 to 3 i := j*32 IF k[j] - dst[i+31:i] := (NOT a[i+31:i]) AND b[i+31:i] + dst[i+31:i] := a[i+31:i] ELSE dst[i+31:i] := 0 FI ENDFOR -dst[MAX:512] := 0 - - -
immintrin.h
- - - Integer - AVX512F/KNCNI - Logical - - - Compute the bitwise NOT of 512 bits (composed of packed 64-bit integers) in "a" and then AND with "b", and store the results in "dst". - -dst[511:0] := ((NOT a[511:0]) AND b[511:0]) -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - - Integer - AVX512F/KNCNI - Logical - - - - - Compute the bitwise NOT of packed 64-bit integers in "a" and then AND with "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Floating Point + AVX512VL + AVX512F + Move + + + + + Duplicate even-indexed double-precision (64-bit) floating-point elements from "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 7 +tmp[63:0] := a[63:0] +tmp[127:64] := a[63:0] +tmp[191:128] := a[191:128] +tmp[255:192] := a[191:128] +FOR j := 0 to 3 i := j*64 IF k[j] - dst[i+63:i] := ((NOT a[i+63:i]) AND b[i+63:i]) + dst[i+63:i] := tmp[i+63:i] ELSE dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - - Integer + + Floating Point + AVX512VL AVX512F - Logical - - - - Compute the bitwise NOT of packed 64-bit integers in "a" and then AND with "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Move + + + + Duplicate even-indexed double-precision (64-bit) floating-point elements from "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 +tmp[63:0] := a[63:0] +tmp[127:64] := a[63:0] +tmp[191:128] := a[191:128] +tmp[255:192] := a[191:128] +FOR j := 0 to 3 i := j*64 IF k[j] - dst[i+63:i] := (NOT a[i+63:i]) AND b[i+63:i] + dst[i+63:i] := tmp[i+63:i] ELSE dst[i+63:i] := 0 FI ENDFOR -dst[MAX:512] := 0 - - -
immintrin.h
- - - Integer - AVX512F/KNCNI - Logical - - - Compute the bitwise AND of 512 bits (composed of packed 64-bit integers) in "a" and "b", and store the results in "dst". - -dst[511:0] := (a[511:0] AND b[511:0]) -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - - Integer - AVX512F/KNCNI - Logical - - - - - Compute the bitwise AND of packed 64-bit integers in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Floating Point + AVX512VL + AVX512F + Move + + + + + Duplicate even-indexed double-precision (64-bit) floating-point elements from "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 7 +tmp[63:0] := a[63:0] +tmp[127:64] := a[63:0] +FOR j := 0 to 1 i := j*64 IF k[j] - dst[i+63:i] := a[i+63:i] AND b[i+63:i] + dst[i+63:i] := tmp[i+63:i] ELSE dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - - Integer + + Floating Point + AVX512VL AVX512F - Logical - - - - Compute the bitwise AND of packed 64-bit integers in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Move + + + + Duplicate even-indexed double-precision (64-bit) floating-point elements from "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 +tmp[63:0] := a[63:0] +tmp[127:64] := a[63:0] +FOR j := 0 to 1 i := j*64 IF k[j] - dst[i+63:i] := a[i+63:i] AND b[i+63:i] + dst[i+63:i] := tmp[i+63:i] ELSE dst[i+63:i] := 0 FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - AVX512F/KNCNI - Swizzle - - - - Blend packed 32-bit integers from "a" and "b" using control mask "k", and store the results in "dst". + AVX512VL + AVX512F + Load + + + + + Load packed 32-bit integers from memory into "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + "mem_addr" must be aligned on a 32-byte boundary or a general-protection exception may be generated. -FOR j := 0 to 15 +FOR j := 0 to 7 i := j*32 IF k[j] - dst[i+31:i] := b[i+31:i] + dst[i+31:i] := MEM[mem_addr+i+31:mem_addr+i] ELSE - dst[i+31:i] := a[i+31:i] + dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - AVX512F/KNCNI - Swizzle - - - - Blend packed 64-bit integers from "a" and "b" using control mask "k", and store the results in "dst". + AVX512VL + AVX512F + Move + + + + + Move packed 32-bit integers from "a" to "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). FOR j := 0 to 7 - i := j*64 + i := j*32 IF k[j] - dst[i+63:i] := b[i+63:i] + dst[i+31:i] := a[i+31:i] ELSE - dst[i+63:i] := a[i+63:i] + dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer + AVX512VL AVX512F - Set - - Broadcast 8-bit integer "a" to all elements of "dst". + Store + + + + + Store packed 32-bit integers from "a" into memory using writemask "k". + "mem_addr" must be aligned on a 32-byte boundary or a general-protection exception may be generated. -FOR j := 0 to 63 - i := j*8 - dst[i+7:i] := a[7:0] +FOR j := 0 to 7 + i := j*32 + IF k[j] + MEM[mem_addr+i+31:mem_addr+i] := a[i+31:i] + FI ENDFOR -dst[MAX:512] := 0 - +
immintrin.h
 - + Integer + AVX512VL AVX512F - Swizzle - - Broadcast the low packed 32-bit integer from "a" to all elements of "dst". - -FOR j := 0 to 15 + Load + + + + Load packed 32-bit integers from memory into "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + "mem_addr" must be aligned on a 32-byte boundary or a general-protection exception may be generated. + +FOR j := 0 to 7 i := j*32 - dst[i+31:i] := a[31:0] + IF k[j] + dst[i+31:i] := MEM[mem_addr+i+31:mem_addr+i] + ELSE + dst[i+31:i] := 0 + FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer + AVX512VL AVX512F - Swizzle - - - - Broadcast the low packed 32-bit integer from "a" to all elements of "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Move + + + + Move packed 32-bit integers from "a" into "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 15 +FOR j := 0 to 7 i := j*32 IF k[j] - dst[i+31:i] := a[31:0] + dst[i+31:i] := a[i+31:i] ELSE - dst[i+31:i] := src[i+31:i] + dst[i+31:i] := 0 FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer + AVX512VL AVX512F - Set - - - - Broadcast 32-bit integer "a" to all elements of "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Load + + + + + Load packed 32-bit integers from memory into "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + "mem_addr" must be aligned on a 16-byte boundary or a general-protection exception may be generated. -FOR j := 0 to 15 +FOR j := 0 to 3 i := j*32 IF k[j] - dst[i+31:i] := a[31:0] + dst[i+31:i] := MEM[mem_addr+i+31:mem_addr+i] ELSE dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer + AVX512VL AVX512F - Swizzle - - - Broadcast the low packed 32-bit integer from "a" to all elements of "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Move + + + + + Move packed 32-bit integers from "a" to "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 15 +FOR j := 0 to 3 i := j*32 IF k[j] - dst[i+31:i] := a[31:0] + dst[i+31:i] := a[i+31:i] ELSE - dst[i+31:i] := 0 + dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer + AVX512VL AVX512F - Set - - - Broadcast 32-bit integer "a" to all elements of "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Store + + + + + Store packed 32-bit integers from "a" into memory using writemask "k". + "mem_addr" must be aligned on a 16-byte boundary or a general-protection exception may be generated. -FOR j := 0 to 15 +FOR j := 0 to 3 i := j*32 IF k[j] - dst[i+31:i] := a[31:0] - ELSE - dst[i+31:i] := 0 + MEM[mem_addr+i+31:mem_addr+i] := a[i+31:i] FI ENDFOR -dst[MAX:512] := 0 - +
immintrin.h
 - + Integer + AVX512VL AVX512F - Set - - Broadcast 32-bit integer "a" to all elements of "dst". + Load + + + + Load packed 32-bit integers from memory into "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + "mem_addr" must be aligned on a 16-byte boundary or a general-protection exception may be generated. -FOR j := 0 to 15 +FOR j := 0 to 3 i := j*32 - dst[i+31:i] := a[31:0] -ENDFOR -dst[MAX:512] := 0 - - -
immintrin.h
- - - Integer - AVX512CD - Swizzle - - Broadcast the low 8-bits from input mask "k" to all 64-bit elements of "dst". - -FOR j := 0 to 7 - i := j*64 - dst[i+63:i] := ZeroExtend(k[7:0]) + IF k[j] + dst[i+31:i] := MEM[mem_addr+i+31:mem_addr+i] + ELSE + dst[i+31:i] := 0 + FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - AVX512CD - Swizzle - - Broadcast the low 16-bits from input mask "k" to all 32-bit elements of "dst". + AVX512VL + AVX512F + Move + + + + Move packed 32-bit integers from "a" into "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 15 +FOR j := 0 to 3 i := j*32 - dst[i+31:i] := ZeroExtend(k[15:0]) + IF k[j] + dst[i+31:i] := a[i+31:i] + ELSE + dst[i+31:i] := 0 + FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer + AVX512VL AVX512F - Swizzle - - Broadcast the low packed 64-bit integer from "a" to all elements of "dst". + Load + + + + + Load packed 64-bit integers from memory into "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + "mem_addr" must be aligned on a 32-byte boundary or a general-protection exception may be generated. -FOR j := 0 to 7 +FOR j := 0 to 3 i := j*64 - dst[i+63:i] := a[63:0] + IF k[j] + dst[i+63:i] := MEM[mem_addr+i+63:mem_addr+i] + ELSE + dst[i+63:i] := src[i+63:i] + FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer + AVX512VL AVX512F - Swizzle - - - - Broadcast the low packed 64-bit integer from "a" to all elements of "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Move + + + + + Move packed 64-bit integers from "a" to "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 7 +FOR j := 0 to 3 i := j*64 IF k[j] - dst[i+63:i] := a[63:0] + dst[i+63:i] := a[i+63:i] ELSE dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer + AVX512VL AVX512F - Set - - - - Broadcast 64-bit integer "a" to all elements of "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Store + + + + + Store packed 64-bit integers from "a" into memory using writemask "k". + "mem_addr" must be aligned on a 32-byte boundary or a general-protection exception may be generated. -FOR j := 0 to 7 +FOR j := 0 to 3 i := j*64 IF k[j] - dst[i+63:i] := a[63:0] - ELSE - dst[i+63:i] := src[i+63:i] + MEM[mem_addr+i+63:mem_addr+i] := a[i+63:i] FI ENDFOR -dst[MAX:512] := 0 - +
immintrin.h
 - + Integer + AVX512VL AVX512F - Swizzle - - - Broadcast the low packed 64-bit integer from "a" to all elements of "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Load + + + + Load packed 64-bit integers from memory into "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + "mem_addr" must be aligned on a 32-byte boundary or a general-protection exception may be generated. -FOR j := 0 to 7 +FOR j := 0 to 3 i := j*64 IF k[j] - dst[i+63:i] := a[63:0] + dst[i+63:i] := MEM[mem_addr+i+63:mem_addr+i] ELSE dst[i+63:i] := 0 FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer + AVX512VL AVX512F - Set - - - Broadcast 64-bit integer "a" to all elements of "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Move + + + + Move packed 64-bit integers from "a" into "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 +FOR j := 0 to 3 i := j*64 IF k[j] - dst[i+63:i] := a[63:0] + dst[i+63:i] := a[i+63:i] ELSE dst[i+63:i] := 0 FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer + AVX512VL AVX512F - Set - - Broadcast 64-bit integer "a" to all elements of "dst". + Load + + + + + Load packed 64-bit integers from memory into "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + "mem_addr" must be aligned on a 16-byte boundary or a general-protection exception may be generated. -FOR j := 0 to 7 +FOR j := 0 to 1 i := j*64 - dst[i+63:i] := a[63:0] + IF k[j] + dst[i+63:i] := MEM[mem_addr+i+63:mem_addr+i] + ELSE + dst[i+63:i] := src[i+63:i] + FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer + AVX512VL AVX512F - Set - - Broadcast the low packed 16-bit integer from "a" to all all elements of "dst". - - -FOR j := 0 to 31 - i := j*16 - dst[i+15:i] := a[15:0] -ENDFOR -dst[MAX:512] := 0 - - -
immintrin.h
- - - Integer - Mask - AVX512F/KNCNI - Compare - - - - Compare packed 32-bit integers in "a" and "b" based on the comparison operand specified by "imm8", and store the results in mask vector "k". + Move + + + + + Move packed 64-bit integers from "a" to "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -CASE (imm8[7:0]) OF -0: OP := _MM_CMPINT_EQ -1: OP := _MM_CMPINT_LT -2: OP := _MM_CMPINT_LE -3: OP := _MM_CMPINT_FALSE -4: OP := _MM_CMPINT_NE -5: OP := _MM_CMPINT_NLT -6: OP := _MM_CMPINT_NLE -7: OP := _MM_CMPINT_TRUE -ESAC -FOR j := 0 to 15 - i := j*32 - k[j] := ( a[i+31:i] OP b[i+31:i] ) ? 1 : 0 +FOR j := 0 to 1 + i := j*64 + IF k[j] + dst[i+63:i] := a[i+63:i] + ELSE + dst[i+63:i] := src[i+63:i] + FI ENDFOR -k[MAX:16] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - Mask - AVX512F/KNCNI - Compare - - - Compare packed 32-bit integers in "a" and "b" for equality, and store the results in mask vector "k". + AVX512VL + AVX512F + Store + + + + + Store packed 64-bit integers from "a" into memory using writemask "k". + "mem_addr" must be aligned on a 16-byte boundary or a general-protection exception may be generated. -FOR j := 0 to 15 - i := j*32 - k[j] := ( a[i+31:i] == b[i+31:i] ) ? 1 : 0 +FOR j := 0 to 1 + i := j*64 + IF k[j] + MEM[mem_addr+i+63:mem_addr+i] := a[i+63:i] + FI ENDFOR -k[MAX:16] := 0 - +
immintrin.h
 - + Integer - Mask - AVX512F/KNCNI - Compare - - - Compare packed 32-bit integers in "a" and "b" for greater-than-or-equal, and store the results in mask vector "k". + AVX512VL + AVX512F + Load + + + + Load packed 64-bit integers from memory into "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + "mem_addr" must be aligned on a 16-byte boundary or a general-protection exception may be generated. -FOR j := 0 to 15 - i := j*32 - k[j] := ( a[i+31:i] >= b[i+31:i] ) ? 1 : 0 +FOR j := 0 to 1 + i := j*64 + IF k[j] + dst[i+63:i] := MEM[mem_addr+i+63:mem_addr+i] + ELSE + dst[i+63:i] := 0 + FI ENDFOR -k[MAX:16] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - Mask - AVX512F/KNCNI - Compare - - - Compare packed 32-bit integers in "a" and "b" for greater-than, and store the results in mask vector "k". + AVX512VL + AVX512F + Move + + + + Move packed 64-bit integers from "a" into "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*32 - k[j] := ( a[i+31:i] > b[i+31:i] ) ? 1 : 0 +FOR j := 0 to 1 + i := j*64 + IF k[j] + dst[i+63:i] := a[i+63:i] + ELSE + dst[i+63:i] := 0 + FI ENDFOR -k[MAX:16] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - Mask - AVX512F/KNCNI - Compare - - - Compare packed 32-bit integers in "a" and "b" for less-than-or-equal, and store the results in mask vector "k". + AVX512VL + AVX512F + Load + + + + + Load packed 32-bit integers from memory into "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + "mem_addr" does not need to be aligned on any particular boundary. -FOR j := 0 to 15 +FOR j := 0 to 7 i := j*32 - k[j] := ( a[i+31:i] <= b[i+31:i] ) ? 1 : 0 + IF k[j] + dst[i+31:i] := MEM[mem_addr+i+31:mem_addr+i] + ELSE + dst[i+31:i] := src[i+31:i] + FI ENDFOR -k[MAX:16] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - Mask + AVX512VL AVX512F - Compare - - - Compare packed 32-bit integers in "a" and "b" for less-than, and store the results in mask vector "k". + Store + + + + + Store packed 32-bit integers from "a" into memory using writemask "k". + "mem_addr" does not need to be aligned on any particular boundary. -FOR j := 0 to 15 +FOR j := 0 to 7 i := j*32 - k[j] := ( a[i+31:i] < b[i+31:i] ) ? 1 : 0 + IF k[j] + MEM[mem_addr+i+31:mem_addr+i] := a[i+31:i] + FI ENDFOR -k[MAX:16] := 0 - +
immintrin.h
 - + Integer - Mask - KNCNI - Compare - - - Compare packed 32-bit integers in "a" and "b" for less-than, and store the results in mask vector "k". + AVX512VL + AVX512F + Load + + + + Load packed 32-bit integers from memory into "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + "mem_addr" does not need to be aligned on any particular boundary. -FOR j := 0 to 15 +FOR j := 0 to 7 i := j*32 - k[j] := ( a[i+31:i] < b[i+31:i] ) ? 1 : 0 + IF k[j] + dst[i+31:i] := MEM[mem_addr+i+31:mem_addr+i] + ELSE + dst[i+31:i] := 0 + FI ENDFOR -k[MAX:16] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - Mask - AVX512F/KNCNI - Compare - - - Compare packed 32-bit integers in "a" and "b" for not-equal, and store the results in mask vector "k". + AVX512VL + AVX512F + Load + + + + + Load packed 32-bit integers from memory into "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + "mem_addr" does not need to be aligned on any particular boundary. -FOR j := 0 to 15 +FOR j := 0 to 3 i := j*32 - k[j] := ( a[i+31:i] != b[i+31:i] ) ? 1 : 0 + IF k[j] + dst[i+31:i] := MEM[mem_addr+i+31:mem_addr+i] + ELSE + dst[i+31:i] := src[i+31:i] + FI ENDFOR -k[MAX:16] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - Mask - AVX512F/KNCNI - Compare - - - - - Compare packed 32-bit integers in "a" and "b" based on the comparison operand specified by "imm8", and store the results in mask vector "k1" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + AVX512VL + AVX512F + Store + + + + + Store packed 32-bit integers from "a" into memory using writemask "k". + "mem_addr" does not need to be aligned on any particular boundary. -CASE (imm8[7:0]) OF -0: OP := _MM_CMPINT_EQ -1: OP := _MM_CMPINT_LT -2: OP := _MM_CMPINT_LE -3: OP := _MM_CMPINT_FALSE -4: OP := _MM_CMPINT_NE -5: OP := _MM_CMPINT_NLT -6: OP := _MM_CMPINT_NLE -7: OP := _MM_CMPINT_TRUE -ESAC -FOR j := 0 to 15 +FOR j := 0 to 3 i := j*32 - IF k1[j] - k[j] := ( a[i+31:i] OP b[i+31:i] ) ? 1 : 0 - ELSE - k[j] := 0 + IF k[j] + MEM[mem_addr+i+31:mem_addr+i] := a[i+31:i] FI ENDFOR -k[MAX:16] := 0 - +
immintrin.h
 - + Integer - Mask - AVX512F/KNCNI - Compare - - - - Compare packed 32-bit integers in "a" and "b" for equality, and store the results in mask vector "k1" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + AVX512VL + AVX512F + Load + + + + Load packed 32-bit integers from memory into "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + "mem_addr" does not need to be aligned on any particular boundary. -FOR j := 0 to 15 +FOR j := 0 to 3 i := j*32 - IF k1[j] - k[j] := ( a[i+31:i] == b[i+31:i] ) ? 1 : 0 - ELSE - k[j] := 0 + IF k[j] + dst[i+31:i] := MEM[mem_addr+i+31:mem_addr+i] + ELSE + dst[i+31:i] := 0 FI ENDFOR -k[MAX:16] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - Mask - AVX512F/KNCNI - Compare - - - - Compare packed 32-bit integers in "a" and "b" for greater-than-or-equal, and store the results in mask vector "k1" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + AVX512VL + AVX512F + Load + + + + + Load packed 64-bit integers from memory into "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + "mem_addr" does not need to be aligned on any particular boundary. -FOR j := 0 to 15 - i := j*32 - IF k1[j] - k[j] := ( a[i+31:i] >= b[i+31:i] ) ? 1 : 0 - ELSE - k[j] := 0 +FOR j := 0 to 3 + i := j*64 + IF k[j] + dst[i+63:i] := MEM[mem_addr+i+63:mem_addr+i] + ELSE + dst[i+63:i] := src[i+63:i] FI ENDFOR -k[MAX:16] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - Mask - AVX512F/KNCNI - Compare - - - - Compare packed 32-bit integers in "a" and "b" for greater-than, and store the results in mask vector "k1" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + AVX512VL + AVX512F + Store + + + + + Store packed 64-bit integers from "a" into memory using writemask "k". + "mem_addr" does not need to be aligned on any particular boundary. -FOR j := 0 to 15 - i := j*32 - IF k1[j] - k[j] := ( a[i+31:i] > b[i+31:i] ) ? 1 : 0 - ELSE - k[j] := 0 +FOR j := 0 to 3 + i := j*64 + IF k[j] + MEM[mem_addr+i+63:mem_addr+i] := a[i+63:i] FI ENDFOR -k[MAX:16] := 0 - +
immintrin.h
 - + Integer - Mask - AVX512F/KNCNI - Compare - - - - Compare packed 32-bit integers in "a" and "b" for less-than, and store the results in mask vector "k1" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + AVX512VL + AVX512F + Load + + + + Load packed 64-bit integers from memory into "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + "mem_addr" does not need to be aligned on any particular boundary. -FOR j := 0 to 15 - i := j*32 - IF k1[j] - k[j] := ( a[i+31:i] < b[i+31:i] ) ? 1 : 0 - ELSE - k[j] := 0 +FOR j := 0 to 3 + i := j*64 + IF k[j] + dst[i+63:i] := MEM[mem_addr+i+63:mem_addr+i] + ELSE + dst[i+63:i] := 0 FI ENDFOR -k[MAX:16] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - Mask + AVX512VL AVX512F - Compare - - - - Compare packed 32-bit integers in "a" and "b" for less-than-or-equal, and store the results in mask vector "k1" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Load + + + + + Load packed 64-bit integers from memory into "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + "mem_addr" does not need to be aligned on any particular boundary. -FOR j := 0 to 15 - i := j*32 - IF k1[j] - k[j] := ( a[i+31:i] <= b[i+31:i] ) ? 1 : 0 - ELSE - k[j] := 0 +FOR j := 0 to 1 + i := j*64 + IF k[j] + dst[i+63:i] := MEM[mem_addr+i+63:mem_addr+i] + ELSE + dst[i+63:i] := src[i+63:i] FI ENDFOR -k[MAX:16] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - Mask - KNCNI - Compare - - - - Compare packed 32-bit integers in "a" and "b" for less-than-or-equal, and store the results in mask vector "k1" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + AVX512VL + AVX512F + Store + + + + + Store packed 64-bit integers from "a" into memory using writemask "k". + "mem_addr" does not need to be aligned on any particular boundary. -FOR j := 0 to 15 - i := j*32 - IF k1[j] - k[j] := ( a[i+31:i] <= b[i+31:i] ) ? 1 : 0 - ELSE - k[j] := 0 +FOR j := 0 to 1 + i := j*64 + IF k[j] + MEM[mem_addr+i+63:mem_addr+i] := a[i+63:i] FI ENDFOR -k[MAX:16] := 0 - +
immintrin.h
 - + Integer - Mask - AVX512F/KNCNI - Compare - - - - Compare packed 32-bit integers in "a" and "b" for not-equal, and store the results in mask vector "k1" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + AVX512VL + AVX512F + Load + + + + Load packed 64-bit integers from memory into "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + "mem_addr" does not need to be aligned on any particular boundary. -FOR j := 0 to 15 - i := j*32 - IF k1[j] - k[j] := ( a[i+31:i] != b[i+31:i] ) ? 1 : 0 - ELSE - k[j] := 0 +FOR j := 0 to 1 + i := j*64 + IF k[j] + dst[i+63:i] := MEM[mem_addr+i+63:mem_addr+i] + ELSE + dst[i+63:i] := 0 FI ENDFOR -k[MAX:16] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - - Integer - Mask + + Floating Point + AVX512VL AVX512F - Compare - - - - Compare packed 64-bit integers in "a" and "b" based on the comparison operand specified by "imm8", and store the results in mask vector "k". + Move + + + + + Duplicate odd-indexed single-precision (32-bit) floating-point elements from "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -CASE (imm8[7:0]) OF -0: OP := _MM_CMPINT_EQ -1: OP := _MM_CMPINT_LT -2: OP := _MM_CMPINT_LE -3: OP := _MM_CMPINT_FALSE -4: OP := _MM_CMPINT_NE -5: OP := _MM_CMPINT_NLT -6: OP := _MM_CMPINT_NLE -7: OP := _MM_CMPINT_TRUE -ESAC +tmp[31:0] := a[63:32] +tmp[63:32] := a[63:32] +tmp[95:64] := a[127:96] +tmp[127:96] := a[127:96] +tmp[159:128] := a[191:160] +tmp[191:160] := a[191:160] +tmp[223:192] := a[255:224] +tmp[255:224] := a[255:224] FOR j := 0 to 7 - i := j*64 - k[j] := ( a[i+63:i] OP b[i+63:i] ) ? 1 : 0 + i := j*32 + IF k[j] + dst[i+31:i] := tmp[i+31:i] + ELSE + dst[i+31:i] := src[i+31:i] + FI ENDFOR -k[MAX:8] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - - Integer - Mask + + Floating Point + AVX512VL AVX512F - Compare - - - Compare packed 64-bit integers in "a" and "b" for equality, and store the results in mask vector "k". + Move + + + + Duplicate odd-indexed single-precision (32-bit) floating-point elements from "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). +tmp[31:0] := a[63:32] +tmp[63:32] := a[63:32] +tmp[95:64] := a[127:96] +tmp[127:96] := a[127:96] +tmp[159:128] := a[191:160] +tmp[191:160] := a[191:160] +tmp[223:192] := a[255:224] +tmp[255:224] := a[255:224] FOR j := 0 to 7 - i := j*64 - k[j] := ( a[i+63:i] == b[i+63:i] ) ? 1 : 0 + i := j*32 + IF k[j] + dst[i+31:i] := tmp[i+31:i] + ELSE + dst[i+31:i] := 0 + FI ENDFOR -k[MAX:8] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - - Integer - Mask + + Floating Point + AVX512VL AVX512F - Compare - - - Compare packed 64-bit integers in "a" and "b" for greater-than-or-equal, and store the results in mask vector "k". + Move + + + + + Duplicate odd-indexed single-precision (32-bit) floating-point elements from "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := j*64 - k[j] := ( a[i+63:i] >= b[i+63:i] ) ? 1 : 0 +tmp[31:0] := a[63:32] +tmp[63:32] := a[63:32] +tmp[95:64] := a[127:96] +tmp[127:96] := a[127:96] +FOR j := 0 to 3 + i := j*32 + IF k[j] + dst[i+31:i] := tmp[i+31:i] + ELSE + dst[i+31:i] := src[i+31:i] + FI ENDFOR -k[MAX:8] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - - Integer - Mask + + Floating Point + AVX512VL AVX512F - Compare - - - Compare packed 64-bit integers in "a" and "b" for greater-than, and store the results in mask vector "k". + Move + + + + Duplicate odd-indexed single-precision (32-bit) floating-point elements from "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := j*64 - k[j] := ( a[i+63:i] > b[i+63:i] ) ? 1 : 0 +tmp[31:0] := a[63:32] +tmp[63:32] := a[63:32] +tmp[95:64] := a[127:96] +tmp[127:96] := a[127:96] +FOR j := 0 to 3 + i := j*32 + IF k[j] + dst[i+31:i] := tmp[i+31:i] + ELSE + dst[i+31:i] := 0 + FI ENDFOR -k[MAX:8] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - - Integer - Mask + + Floating Point + AVX512VL AVX512F - Compare - - - Compare packed 64-bit integers in "a" and "b" for less-than-or-equal, and store the results in mask vector "k". + Move + + + + + Duplicate even-indexed single-precision (32-bit) floating-point elements from "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). +tmp[31:0] := a[31:0] +tmp[63:32] := a[31:0] +tmp[95:64] := a[95:64] +tmp[127:96] := a[95:64] +tmp[159:128] := a[159:128] +tmp[191:160] := a[159:128] +tmp[223:192] := a[223:192] +tmp[255:224] := a[223:192] FOR j := 0 to 7 - i := j*64 - k[j] := ( a[i+63:i] <= b[i+63:i] ) ? 1 : 0 -ENDFOR -k[MAX:8] := 0 + i := j*32 + IF k[j] + dst[i+31:i] := tmp[i+31:i] + ELSE + dst[i+31:i] := src[i+31:i] + FI +ENDFOR +dst[MAX:256] := 0 - +
immintrin.h
 - - Integer - Mask + + Floating Point + AVX512VL AVX512F - Compare - - - Compare packed 64-bit integers in "a" and "b" for less-than, and store the results in mask vector "k". + Move + + + + Duplicate even-indexed single-precision (32-bit) floating-point elements from "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). +tmp[31:0] := a[31:0] +tmp[63:32] := a[31:0] +tmp[95:64] := a[95:64] +tmp[127:96] := a[95:64] +tmp[159:128] := a[159:128] +tmp[191:160] := a[159:128] +tmp[223:192] := a[223:192] +tmp[255:224] := a[223:192] FOR j := 0 to 7 - i := j*64 - k[j] := ( a[i+63:i] < b[i+63:i] ) ? 1 : 0 + i := j*32 + IF k[j] + dst[i+31:i] := tmp[i+31:i] + ELSE + dst[i+31:i] := 0 + FI ENDFOR -k[MAX:8] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - - Integer - Mask + + Floating Point + AVX512VL AVX512F - Compare - - - Compare packed 64-bit integers in "a" and "b" for not-equal, and store the results in mask vector "k". + Move + + + + + Duplicate even-indexed single-precision (32-bit) floating-point elements from "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := j*64 - k[j] := ( a[i+63:i] != b[i+63:i] ) ? 1 : 0 -ENDFOR -k[MAX:8] := 0 +tmp[31:0] := a[31:0] +tmp[63:32] := a[31:0] +tmp[95:64] := a[95:64] +tmp[127:96] := a[95:64] +FOR j := 0 to 3 + i := j*32 + IF k[j] + dst[i+31:i] := tmp[i+31:i] + ELSE + dst[i+31:i] := src[i+31:i] + FI +ENDFOR +dst[MAX:128] := 0 - +
immintrin.h
 - - Integer - Mask + + Floating Point + AVX512VL AVX512F - Compare - - - - - Compare packed 64-bit integers in "a" and "b" based on the comparison operand specified by "imm8", and store the results in mask vector "k1" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Move + + + + Duplicate even-indexed single-precision (32-bit) floating-point elements from "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -CASE (imm8[7:0]) OF -0: OP := _MM_CMPINT_EQ -1: OP := _MM_CMPINT_LT -2: OP := _MM_CMPINT_LE -3: OP := _MM_CMPINT_FALSE -4: OP := _MM_CMPINT_NE -5: OP := _MM_CMPINT_NLT -6: OP := _MM_CMPINT_NLE -7: OP := _MM_CMPINT_TRUE -ESAC -FOR j := 0 to 7 - i := j*64 - IF k1[j] - k[j] := ( a[i+63:i] OP b[i+63:i] ) ? 1 : 0 - ELSE - k[j] := 0 +tmp[31:0] := a[31:0] +tmp[63:32] := a[31:0] +tmp[95:64] := a[95:64] +tmp[127:96] := a[95:64] +FOR j := 0 to 3 + i := j*32 + IF k[j] + dst[i+31:i] := tmp[i+31:i] + ELSE + dst[i+31:i] := 0 FI ENDFOR -k[MAX:8] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - - Integer - Mask + + Floating Point + AVX512VL AVX512F - Compare - - - - Compare packed 64-bit integers in "a" and "b" for equality, and store the results in mask vector "k1" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Load + + + + + Load packed double-precision (64-bit) floating-point elements from memoy into "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + "mem_addr" does not need to be aligned on any particular boundary. -FOR j := 0 to 7 +FOR j := 0 to 3 i := j*64 - IF k1[j] - k[j] := ( a[i+63:i] == b[i+63:i] ) ? 1 : 0 - ELSE - k[j] := 0 + IF k[j] + dst[i+63:i] := MEM[mem_addr+i+63:mem_addr+i] + ELSE + dst[i+63:i] := src[i+63:i] FI ENDFOR -k[MAX:8] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - - Integer - Mask + + Floating Point + AVX512VL AVX512F - Compare - - - - Compare packed 64-bit integers in "a" and "b" for greater-than-or-equal, and store the results in mask vector "k1" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Store + + + + + Store packed double-precision (64-bit) floating-point elements from "a" into memory using writemask "k". + "mem_addr" does not need to be aligned on any particular boundary. -FOR j := 0 to 7 +FOR j := 0 to 3 i := j*64 - IF k1[j] - k[j] := ( a[i+63:i] >= b[i+63:i] ) ? 1 : 0 - ELSE - k[j] := 0 + IF k[j] + MEM[mem_addr+i+63:mem_addr+i] := a[i+63:i] FI ENDFOR -k[MAX:8] := 0 - +
immintrin.h
 - - Integer - Mask + + Floating Point + AVX512VL AVX512F - Compare - - - - Compare packed 64-bit integers in "a" and "b" for greater-than, and store the results in mask vector "k1" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Load + + + + Load packed double-precision (64-bit) floating-point elements from memoy into "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + "mem_addr" does not need to be aligned on any particular boundary. -FOR j := 0 to 7 +FOR j := 0 to 3 i := j*64 - IF k1[j] - k[j] := ( a[i+63:i] > b[i+63:i] ) ? 1 : 0 - ELSE - k[j] := 0 + IF k[j] + dst[i+63:i] := MEM[mem_addr+i+63:mem_addr+i] + ELSE + dst[i+63:i] := 0 FI ENDFOR -k[MAX:8] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - - Integer - Mask + + Floating Point + AVX512VL AVX512F - Compare - - - - Compare packed 64-bit integers in "a" and "b" for less-than-or-equal, and store the results in mask vector "k1" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Load + + + + + Load packed double-precision (64-bit) floating-point elements from memoy into "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + "mem_addr" does not need to be aligned on any particular boundary. -FOR j := 0 to 7 +FOR j := 0 to 1 i := j*64 - IF k1[j] - k[j] := ( a[i+63:i] <= b[i+63:i] ) ? 1 : 0 - ELSE - k[j] := 0 + IF k[j] + dst[i+63:i] := MEM[mem_addr+i+63:mem_addr+i] + ELSE + dst[i+63:i] := src[i+63:i] FI ENDFOR -k[MAX:8] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - - Integer - Mask + + Floating Point + AVX512VL AVX512F - Compare - - - - Compare packed 64-bit integers in "a" and "b" for less-than, and store the results in mask vector "k1" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Store + + + + + Store packed double-precision (64-bit) floating-point elements from "a" into memory using writemask "k". + "mem_addr" does not need to be aligned on any particular boundary. -FOR j := 0 to 7 +FOR j := 0 to 1 i := j*64 - IF k1[j] - k[j] := ( a[i+63:i] < b[i+63:i] ) ? 1 : 0 - ELSE - k[j] := 0 + IF k[j] + MEM[mem_addr+i+63:mem_addr+i] := a[i+63:i] FI ENDFOR -k[MAX:8] := 0 - +
immintrin.h
 - - Integer - Mask + + Floating Point + AVX512VL AVX512F - Compare - - - - Compare packed 64-bit integers in "a" and "b" for not-equal, and store the results in mask vector "k1" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Load + + + + Load packed double-precision (64-bit) floating-point elements from memoy into "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + "mem_addr" does not need to be aligned on any particular boundary. -FOR j := 0 to 7 +FOR j := 0 to 1 i := j*64 - IF k1[j] - k[j] := ( a[i+63:i] != b[i+63:i] ) ? 1 : 0 - ELSE - k[j] := 0 + IF k[j] + dst[i+63:i] := MEM[mem_addr+i+63:mem_addr+i] + ELSE + dst[i+63:i] := 0 FI ENDFOR -k[MAX:8] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - - Integer - Mask - AVX512F/KNCNI - Compare - - - - Compare packed unsigned 32-bit integers in "a" and "b" based on the comparison operand specified by "imm8", and store the results in mask vector "k". + + Floating Point + AVX512VL + AVX512F + Load + + + + + Load packed single-precision (32-bit) floating-point elements from memory into "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + "mem_addr" does not need to be aligned on any particular boundary. -CASE (imm8[7:0]) OF -0: OP := _MM_CMPINT_EQ -1: OP := _MM_CMPINT_LT -2: OP := _MM_CMPINT_LE -3: OP := _MM_CMPINT_FALSE -4: OP := _MM_CMPINT_NE -5: OP := _MM_CMPINT_NLT -6: OP := _MM_CMPINT_NLE -7: OP := _MM_CMPINT_TRUE -ESAC -FOR j := 0 to 15 +FOR j := 0 to 7 i := j*32 - k[j] := ( a[i+31:i] OP b[i+31:i] ) ? 1 : 0 + IF k[j] + dst[i+31:i] := MEM[mem_addr+i+31:mem_addr+i] + ELSE + dst[i+31:i] := src[i+31:i] + FI ENDFOR -k[MAX:16] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - - Integer - Mask - AVX512F/KNCNI - Compare - - - Compare packed unsigned 32-bit integers in "a" and "b" for equality, and store the results in mask vector "k". + + Floating Point + AVX512VL + AVX512F + Store + + + + + Store packed single-precision (32-bit) floating-point elements from "a" into memory using writemask "k". + "mem_addr" does not need to be aligned on any particular boundary. -FOR j := 0 to 15 +FOR j := 0 to 7 i := j*32 - k[j] := ( a[i+31:i] == b[i+31:i] ) ? 1 : 0 + IF k[j] + MEM[mem_addr+i+31:mem_addr+i] := a[i+31:i] + FI ENDFOR -k[MAX:16] := 0 - +
immintrin.h
 - - Integer - Mask - AVX512F/KNCNI - Compare - - - Compare packed unsigned 32-bit integers in "a" and "b" for greater-than-or-equal, and store the results in mask vector "k". + + Floating Point + AVX512VL + AVX512F + Load + + + + Load packed single-precision (32-bit) floating-point elements from memory into "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + "mem_addr" does not need to be aligned on any particular boundary. -FOR j := 0 to 15 +FOR j := 0 to 7 i := j*32 - k[j] := ( a[i+31:i] >= b[i+31:i] ) ? 1 : 0 + IF k[j] + dst[i+31:i] := MEM[mem_addr+i+31:mem_addr+i] + ELSE + dst[i+31:i] := 0 + FI ENDFOR -k[MAX:16] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - - Integer - Mask - AVX512F/KNCNI - Compare - - - Compare packed unsigned 32-bit integers in "a" and "b" for greater-than, and store the results in mask vector "k". + + Floating Point + AVX512VL + AVX512F + Load + + + + + Load packed single-precision (32-bit) floating-point elements from memory into "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + "mem_addr" does not need to be aligned on any particular boundary. -FOR j := 0 to 15 +FOR j := 0 to 3 i := j*32 - k[j] := ( a[i+31:i] > b[i+31:i] ) ? 1 : 0 + IF k[j] + dst[i+31:i] := MEM[mem_addr+i+31:mem_addr+i] + ELSE + dst[i+31:i] := src[i+31:i] + FI ENDFOR -k[MAX:16] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - - Integer - Mask - AVX512F/KNCNI - Compare - - - Compare packed unsigned 32-bit integers in "a" and "b" for less-than-or-equal, and store the results in mask vector "k". + + Floating Point + AVX512VL + AVX512F + Store + + + + + Store packed single-precision (32-bit) floating-point elements from "a" into memory using writemask "k". + "mem_addr" does not need to be aligned on any particular boundary. -FOR j := 0 to 15 +FOR j := 0 to 3 i := j*32 - k[j] := ( a[i+31:i] <= b[i+31:i] ) ? 1 : 0 + IF k[j] + MEM[mem_addr+i+31:mem_addr+i] := a[i+31:i] + FI ENDFOR -k[MAX:16] := 0 - +
immintrin.h
 - - Integer - Mask - AVX512F/KNCNI - Compare - - - Compare packed unsigned 32-bit integers in "a" and "b" for less-than, and store the results in mask vector "k". + + Floating Point + AVX512VL + AVX512F + Load + + + + Load packed single-precision (32-bit) floating-point elements from memory into "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + "mem_addr" does not need to be aligned on any particular boundary. -FOR j := 0 to 15 +FOR j := 0 to 3 i := j*32 - k[j] := ( a[i+31:i] < b[i+31:i] ) ? 1 : 0 + IF k[j] + dst[i+31:i] := MEM[mem_addr+i+31:mem_addr+i] + ELSE + dst[i+31:i] := 0 + FI ENDFOR -k[MAX:16] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - - Integer - Mask - AVX512F/KNCNI - Compare - - - Compare packed unsigned 32-bit integers in "a" and "b" for not-equal, and store the results in mask vector "k". + + Floating Point + AVX512VL + AVX512F + Arithmetic + + + + + + Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*32 - k[j] := ( a[i+31:i] != b[i+31:i] ) ? 1 : 0 +FOR j := 0 to 3 + i := j*64 + IF k[j] + dst[i+63:i] := a[i+63:i] * b[i+63:i] + ELSE + dst[i+63:i] := src[i+63:i] + FI ENDFOR -k[MAX:16] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - - Integer - Mask - AVX512F/KNCNI - Compare - - - - - Compare packed unsigned 32-bit integers in "a" and "b" based on the comparison operand specified by "imm8", and store the results in mask vector "k1" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Floating Point + AVX512VL + AVX512F + Arithmetic + + + + + Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -CASE (imm8[7:0]) OF -0: OP := _MM_CMPINT_EQ -1: OP := _MM_CMPINT_LT -2: OP := _MM_CMPINT_LE -3: OP := _MM_CMPINT_FALSE -4: OP := _MM_CMPINT_NE -5: OP := _MM_CMPINT_NLT -6: OP := _MM_CMPINT_NLE -7: OP := _MM_CMPINT_TRUE -ESAC -FOR j := 0 to 15 - i := j*32 - IF k1[j] - k[j] := ( a[i+31:i] OP b[i+31:i] ) ? 1 : 0 - ELSE - k[j] := 0 +FOR j := 0 to 3 + i := j*64 + IF k[j] + dst[i+63:i] := a[i+63:i] * b[i+63:i] + ELSE + dst[i+63:i] := 0 FI ENDFOR -k[MAX:16] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - - Integer - Mask - AVX512F/KNCNI - Compare - - - - Compare packed unsigned 32-bit integers in "a" and "b" for equality, and store the results in mask vector "k1" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Floating Point + AVX512VL + AVX512F + Arithmetic + + + + + + Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*32 - IF k1[j] - k[j] := ( a[i+31:i] == b[i+31:i] ) ? 1 : 0 - ELSE - k[j] := 0 +FOR j := 0 to 1 + i := j*64 + IF k[j] + dst[i+63:i] := a[i+63:i] * b[i+63:i] + ELSE + dst[i+63:i] := src[i+63:i] FI ENDFOR -k[MAX:16] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - - Integer - Mask - AVX512F/KNCNI - Compare - - - - Compare packed unsigned 32-bit integers in "a" and "b" for greater-than-or-equal, and store the results in mask vector "k1" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Floating Point + AVX512VL + AVX512F + Arithmetic + + + + + Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*32 - IF k1[j] - k[j] := ( a[i+31:i] >= b[i+31:i] ) ? 1 : 0 - ELSE - k[j] := 0 +FOR j := 0 to 1 + i := j*64 + IF k[j] + dst[i+63:i] := a[i+63:i] * b[i+63:i] + ELSE + dst[i+63:i] := 0 FI ENDFOR -k[MAX:16] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - - Integer - Mask - AVX512F/KNCNI - Compare - - - - Compare packed unsigned 32-bit integers in "a" and "b" for greater-than, and store the results in mask vector "k1" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Floating Point + AVX512VL + AVX512F + Arithmetic + + + + + + Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). RM. -FOR j := 0 to 15 +FOR j := 0 to 7 i := j*32 - IF k1[j] - k[j] := ( a[i+31:i] > b[i+31:i] ) ? 1 : 0 - ELSE - k[j] := 0 + IF k[j] + dst[i+31:i] := a[i+31:i] * b[i+31:i] + ELSE + dst[i+31:i] := src[i+31:i] FI ENDFOR -k[MAX:16] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - - Integer - Mask - AVX512F/KNCNI - Compare - - - - Compare packed unsigned 32-bit integers in "a" and "b" for less-than, and store the results in mask vector "k1" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Floating Point + AVX512VL + AVX512F + Arithmetic + + + + + Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 15 +FOR j := 0 to 7 i := j*32 - IF k1[j] - k[j] := ( a[i+31:i] < b[i+31:i] ) ? 1 : 0 - ELSE - k[j] := 0 + IF k[j] + dst[i+31:i] := a[i+31:i] * b[i+31:i] + ELSE + dst[i+31:i] := 0 FI ENDFOR -k[MAX:16] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - - Integer - Mask - AVX512F/KNCNI - Compare - - - - Compare packed unsigned 32-bit integers in "a" and "b" for less-than-or-equal, and store the results in mask vector "k1" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Floating Point + AVX512VL + AVX512F + Arithmetic + + + + + + Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 15 +FOR j := 0 to 3 i := j*32 - IF k1[j] - k[j] := ( a[i+31:i] <= b[i+31:i] ) ? 1 : 0 - ELSE - k[j] := 0 + IF k[j] + dst[i+31:i] := a[i+31:i] * b[i+31:i] + ELSE + dst[i+31:i] := src[i+31:i] FI ENDFOR -k[MAX:16] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - - Integer - Mask - AVX512F/KNCNI - Compare - - - - Compare packed unsigned 32-bit integers in "a" and "b" for not-equal, and store the results in mask vector "k1" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Floating Point + AVX512VL + AVX512F + Arithmetic + + + + + Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 15 +FOR j := 0 to 3 i := j*32 - IF k1[j] - k[j] := ( a[i+31:i] != b[i+31:i] ) ? 1 : 0 - ELSE - k[j] := 0 + IF k[j] + dst[i+31:i] := a[i+31:i] * b[i+31:i] + ELSE + dst[i+31:i] := 0 FI ENDFOR -k[MAX:16] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - Mask + AVX512VL AVX512F - Compare - - - - Compare packed unsigned 64-bit integers in "a" and "b" based on the comparison operand specified by "imm8", and store the results in mask vector "k". + Arithmetic + + + + + Compute the absolute value of packed signed 32-bit integers in "a", and store the unsigned results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -CASE (imm8[7:0]) OF -0: OP := _MM_CMPINT_EQ -1: OP := _MM_CMPINT_LT -2: OP := _MM_CMPINT_LE -3: OP := _MM_CMPINT_FALSE -4: OP := _MM_CMPINT_NE -5: OP := _MM_CMPINT_NLT -6: OP := _MM_CMPINT_NLE -7: OP := _MM_CMPINT_TRUE -ESAC FOR j := 0 to 7 - i := j*64 - k[j] := ( a[i+63:i] OP b[i+63:i] ) ? 1 : 0 + i := j*32 + IF k[j] + dst[i+31:i] := ABS(a[i+31:i]) + ELSE + dst[i+31:i] := src[i+31:i] + FI ENDFOR -k[MAX:8] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - Mask + AVX512VL AVX512F - Compare - - - Compare packed unsigned 64-bit integers in "a" and "b" for equality, and store the results in mask vector "k". + Arithmetic + + + + Compute the absolute value of packed signed 32-bit integers in "a", and store the unsigned results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). FOR j := 0 to 7 - i := j*64 - k[j] := ( a[i+63:i] == b[i+63:i] ) ? 1 : 0 + i := j*32 + IF k[j] + dst[i+31:i] := ABS(a[i+31:i]) + ELSE + dst[i+31:i] := 0 + FI ENDFOR -k[MAX:8] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - Mask + AVX512VL AVX512F - Compare - - - Compare packed unsigned 64-bit integers in "a" and "b" for greater-than-or-equal, and store the results in mask vector "k". + Arithmetic + + + + + Compute the absolute value of packed signed 32-bit integers in "a", and store the unsigned results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := j*64 - k[j] := ( a[i+63:i] >= b[i+63:i] ) ? 1 : 0 +FOR j := 0 to 3 + i := j*32 + IF k[j] + dst[i+31:i] := ABS(a[i+31:i]) + ELSE + dst[i+31:i] := src[i+31:i] + FI ENDFOR -k[MAX:8] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - Mask + AVX512VL AVX512F - Compare - - - Compare packed unsigned 64-bit integers in "a" and "b" for greater-than, and store the results in mask vector "k". + Arithmetic + + + + Compute the absolute value of packed signed 32-bit integers in "a", and store the unsigned results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := j*64 - k[j] := ( a[i+63:i] > b[i+63:i] ) ? 1 : 0 +FOR j := 0 to 3 + i := j*32 + IF k[j] + dst[i+31:i] := ABS(a[i+31:i]) + ELSE + dst[i+31:i] := 0 + FI ENDFOR -k[MAX:8] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - Mask + AVX512VL AVX512F - Compare - - - Compare packed unsigned 64-bit integers in "a" and "b" for less-than-or-equal, and store the results in mask vector "k". + Arithmetic + + + Compute the absolute value of packed signed 64-bit integers in "a", and store the unsigned results in "dst". -FOR j := 0 to 7 +FOR j := 0 to 3 i := j*64 - k[j] := ( a[i+63:i] <= b[i+63:i] ) ? 1 : 0 + dst[i+63:i] := ABS(a[i+63:i]) ENDFOR -k[MAX:8] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - Mask + AVX512VL AVX512F - Compare - - - Compare packed unsigned 64-bit integers in "a" and "b" for less-than, and store the results in mask vector "k". + Arithmetic + + + + + Compute the absolute value of packed signed 64-bit integers in "a", and store the unsigned results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 7 +FOR j := 0 to 3 i := j*64 - k[j] := ( a[i+63:i] < b[i+63:i] ) ? 1 : 0 + IF k[j] + dst[i+63:i] := ABS(a[i+63:i]) + ELSE + dst[i+63:i] := src[i+63:i] + FI ENDFOR -k[MAX:8] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - Mask + AVX512VL AVX512F - Compare - - - Compare packed unsigned 64-bit integers in "a" and "b" for not-equal, and store the results in mask vector "k". + Arithmetic + + + + Compute the absolute value of packed signed 64-bit integers in "a", and store the unsigned results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 +FOR j := 0 to 3 i := j*64 - k[j] := ( a[i+63:i] != b[i+63:i] ) ? 1 : 0 + IF k[j] + dst[i+63:i] := ABS(a[i+63:i]) + ELSE + dst[i+63:i] := 0 + FI ENDFOR -k[MAX:8] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - Mask + AVX512VL AVX512F - Compare - - - - - Compare packed unsigned 64-bit integers in "a" and "b" based on the comparison operand specified by "imm8", and store the results in mask vector "k1" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Arithmetic + + + Compute the absolute value of packed signed 64-bit integers in "a", and store the unsigned results in "dst". -CASE (imm8[7:0]) OF -0: OP := _MM_CMPINT_EQ -1: OP := _MM_CMPINT_LT -2: OP := _MM_CMPINT_LE -3: OP := _MM_CMPINT_FALSE -4: OP := _MM_CMPINT_NE -5: OP := _MM_CMPINT_NLT -6: OP := _MM_CMPINT_NLE -7: OP := _MM_CMPINT_TRUE -ESAC -FOR j := 0 to 7 +FOR j := 0 to 1 i := j*64 - IF k1[j] - k[j] := ( a[i+63:i] OP b[i+63:i] ) ? 1 : 0 - ELSE - k[j] := 0 - FI + dst[i+63:i] := ABS(a[i+63:i]) ENDFOR -k[MAX:8] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - Mask + AVX512VL AVX512F - Compare - - - - Compare packed unsigned 64-bit integers in "a" and "b" for equality, and store the results in mask vector "k1" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Arithmetic + + + + + Compute the absolute value of packed signed 64-bit integers in "a", and store the unsigned results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 7 +FOR j := 0 to 1 i := j*64 - IF k1[j] - k[j] := ( a[i+63:i] == b[i+63:i] ) ? 1 : 0 - ELSE - k[j] := 0 + IF k[j] + dst[i+63:i] := ABS(a[i+63:i]) + ELSE + dst[i+63:i] := src[i+63:i] FI ENDFOR -k[MAX:8] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - Mask + AVX512VL AVX512F - Compare - - - - Compare packed unsigned 64-bit integers in "a" and "b" for greater-than-or-equal, and store the results in mask vector "k1" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Arithmetic + + + + Compute the absolute value of packed signed 64-bit integers in "a", and store the unsigned results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 +FOR j := 0 to 1 i := j*64 - IF k1[j] - k[j] := ( a[i+63:i] >= b[i+63:i] ) ? 1 : 0 - ELSE - k[j] := 0 + IF k[j] + dst[i+63:i] := ABS(a[i+63:i]) + ELSE + dst[i+63:i] := 0 FI ENDFOR -k[MAX:8] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - Mask + AVX512VL AVX512F - Compare - - - - Compare packed unsigned 64-bit integers in "a" and "b" for greater-than, and store the results in mask vector "k1" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Arithmetic + + + + + + Add packed 32-bit integers in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). FOR j := 0 to 7 - i := j*64 - IF k1[j] - k[j] := ( a[i+63:i] > b[i+63:i] ) ? 1 : 0 - ELSE - k[j] := 0 + i := j*32 + IF k[j] + dst[i+31:i] := a[i+31:i] + b[i+31:i] + ELSE + dst[i+31:i] := src[i+31:i] FI ENDFOR -k[MAX:8] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - Mask + AVX512VL AVX512F - Compare - - - - Compare packed unsigned 64-bit integers in "a" and "b" for less-than-or-equal, and store the results in mask vector "k1" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Arithmetic + + + + + Add packed 32-bit integers in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). FOR j := 0 to 7 - i := j*64 - IF k1[j] - k[j] := ( a[i+63:i] <= b[i+63:i] ) ? 1 : 0 - ELSE - k[j] := 0 + i := j*32 + IF k[j] + dst[i+31:i] := a[i+31:i] + b[i+31:i] + ELSE + dst[i+31:i] := 0 FI ENDFOR -k[MAX:8] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - Mask + AVX512VL AVX512F - Compare - - - - Compare packed unsigned 64-bit integers in "a" and "b" for less-than, and store the results in mask vector "k1" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Arithmetic + + + + + + Add packed 32-bit integers in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := j*64 - IF k1[j] - k[j] := ( a[i+63:i] < b[i+63:i] ) ? 1 : 0 - ELSE - k[j] := 0 +FOR j := 0 to 3 + i := j*32 + IF k[j] + dst[i+31:i] := a[i+31:i] + b[i+31:i] + ELSE + dst[i+31:i] := src[i+31:i] FI ENDFOR -k[MAX:8] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - Mask + AVX512VL AVX512F - Compare - - - - Compare packed unsigned 64-bit integers in "a" and "b" for not-equal, and store the results in mask vector "k1" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Arithmetic + + + + + Add packed 32-bit integers in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := j*64 - IF k1[j] - k[j] := ( a[i+63:i] != b[i+63:i] ) ? 1 : 0 - ELSE - k[j] := 0 +FOR j := 0 to 3 + i := j*32 + IF k[j] + dst[i+31:i] := a[i+31:i] + b[i+31:i] + ELSE + dst[i+31:i] := 0 FI ENDFOR -k[MAX:8] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer + AVX512VL AVX512F - Swizzle - - - - Contiguously store the active 32-bit integers in "a" (those with their respective bit set in writemask "k") to "dst", and pass through the remaining elements from "src". + Arithmetic + + + + + + Add packed 64-bit integers in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -size := 32 -m := 0 -FOR j := 0 to 15 - i := j*32 +FOR j := 0 to 3 + i := j*64 IF k[j] - dst[m+size-1:m] := a[i+31:i] - m := m + size + dst[i+63:i] := a[i+63:i] + b[i+63:i] + ELSE + dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[511:m] := src[511:m] -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer + AVX512VL AVX512F - Store - Swizzle - - - - Contiguously store the active 32-bit integers in "a" (those with their respective bit set in writemask "k") to unaligned memory at "base_addr". + Arithmetic + + + + + Add packed 64-bit integers in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -size := 32 -m := base_addr -FOR j := 0 to 15 - i := j*32 +FOR j := 0 to 3 + i := j*64 IF k[j] - MEM[m+size-1:m] := a[i+31:i] - m := m + size + dst[i+63:i] := a[i+63:i] + b[i+63:i] + ELSE + dst[i+63:i] :=0 FI ENDFOR +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer + AVX512VL AVX512F - Swizzle - - - Contiguously store the active 32-bit integers in "a" (those with their respective bit set in zeromask "k") to "dst", and set the remaining elements to zero. + Arithmetic + + + + + + Add packed 64-bit integers in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -size := 32 -m := 0 -FOR j := 0 to 15 - i := j*32 +FOR j := 0 to 1 + i := j*64 IF k[j] - dst[m+size-1:m] := a[i+31:i] - m := m + size + dst[i+63:i] := a[i+63:i] + b[i+63:i] + ELSE + dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[511:m] := 0 -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer + AVX512VL AVX512F - Swizzle - - - - Contiguously store the active 64-bit integers in "a" (those with their respective bit set in writemask "k") to "dst", and pass through the remaining elements from "src". + Arithmetic + + + + + Add packed 64-bit integers in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -size := 64 -m := 0 -FOR j := 0 to 7 +FOR j := 0 to 1 i := j*64 IF k[j] - dst[m+size-1:m] := a[i+63:i] - m := m + size + dst[i+63:i] := a[i+63:i] + b[i+63:i] + ELSE + dst[i+63:i] := 0 FI ENDFOR -dst[511:m] := src[511:m] -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer + AVX512VL AVX512F - Store - Swizzle - - - - Contiguously store the active 64-bit integers in "a" (those with their respective bit set in writemask "k") to unaligned memory at "base_addr". + Logical + + + + + + Compute the bitwise AND of packed 32-bit integers in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -size := 64 -m := base_addr FOR j := 0 to 7 - i := j*64 + i := j*32 IF k[j] - MEM[m+size-1:m] := a[i+63:i] - m := m + size + dst[i+31:i] := a[i+31:i] AND b[i+31:i] + ELSE + dst[i+31:i] := src[i+31:i] FI ENDFOR +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer + AVX512VL AVX512F - Swizzle - - - Contiguously store the active 64-bit integers in "a" (those with their respective bit set in zeromask "k") to "dst", and set the remaining elements to zero. + Logical + + + + + Compute the bitwise AND of packed 32-bit integers in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -size := 64 -m := 0 FOR j := 0 to 7 - i := j*64 + i := j*32 IF k[j] - dst[m+size-1:m] := a[i+63:i] - m := m + size + dst[i+31:i] := a[i+31:i] AND b[i+31:i] + ELSE + dst[i+31:i] := 0 FI ENDFOR -dst[511:m] := 0 -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - AVX512CD - Compare - - Test each 32-bit element of "a" for equality with all other elements in "a" closer to the least significant bit. Each element's comparison forms a zero extended bit vector in "dst". + AVX512VL + AVX512F + Logical + + + + + + Compute the bitwise AND of packed 32-bit integers in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 15 +FOR j := 0 to 3 i := j*32 - FOR k := 0 to j-1 - m := k*32 - dst[i+k] := (a[i+31:i] == a[m+31:m]) ? 1 : 0 - ENDFOR - dst[i+31:i+j] := 0 + IF k[j] + dst[i+31:i] := a[i+31:i] AND b[i+31:i] + ELSE + dst[i+31:i] := src[i+31:i] + FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - AVX512CD - Compare - - - - Test each 32-bit element of "a" for equality with all other elements in "a" closer to the least significant bit using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). Each element's comparison forms a zero extended bit vector in "dst". + AVX512VL + AVX512F + Logical + + + + + Compute the bitwise AND of packed 32-bit integers in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 15 +FOR j := 0 to 3 i := j*32 - IF k[i] - FOR l := 0 to j-1 - m := l*32 - dst[i+l] := (a[i+31:i] == a[m+31:m]) ? 1 : 0 - ENDFOR - dst[i+31:i+j] := 0 + IF k[j] + dst[i+31:i] := a[i+31:i] AND b[i+31:i] ELSE - dst[i+31:i] := src[i+31:i] + dst[i+31:i] := 0 FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - AVX512CD - Compare - - - Test each 32-bit element of "a" for equality with all other elements in "a" closer to the least significant bit using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). Each element's comparison forms a zero extended bit vector in "dst". - -FOR j := 0 to 15 + AVX512VL + AVX512F + Logical + + + + + + Compute the bitwise NOT of packed 32-bit integers in "a" and then AND with "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +FOR j := 0 to 7 i := j*32 - IF k[i] - FOR l := 0 to j-1 - m := l*32 - dst[i+l] := (a[i+31:i] == a[m+31:m]) ? 1 : 0 - ENDFOR - dst[i+31:i+j] := 0 + IF k[j] + dst[i+31:i] := ((NOT a[i+31:i]) AND b[i+31:i]) ELSE - dst[i+31:i] := 0 + dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - AVX512CD - Compare - - Test each 64-bit element of "a" for equality with all other elements in "a" closer to the least significant bit. Each element's comparison forms a zero extended bit vector in "dst". + AVX512VL + AVX512F + Logical + + + + + Compute the bitwise NOT of packed 32-bit integers in "a" and then AND with "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). FOR j := 0 to 7 - i := j*64 - FOR k := 0 to j-1 - m := k*64 - dst[i+k] := (a[i+63:i] == a[m+63:m]) ? 1 : 0 - ENDFOR - dst[i+63:i+j] := 0 + i := j*32 + IF k[j] + dst[i+31:i] := (NOT a[i+31:i]) AND b[i+31:i] + ELSE + dst[i+31:i] := 0 + FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - AVX512CD - Compare - - - - Test each 64-bit element of "a" for equality with all other elements in "a" closer to the least significant bit using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). Each element's comparison forms a zero extended bit vector in "dst". + AVX512VL + AVX512F + Logical + + + + + + Compute the bitwise NOT of packed 32-bit integers in "a" and then AND with "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 7 +FOR j := 0 to 3 + i := j*32 + IF k[j] + dst[i+31:i] := ((NOT a[i+31:i]) AND b[i+31:i]) + ELSE + dst[i+31:i] := src[i+31:i] + FI +ENDFOR +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Integer + AVX512VL + AVX512F + Logical + + + + + Compute the bitwise NOT of packed 32-bit integers in "a" and then AND with "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + +FOR j := 0 to 3 + i := j*32 + IF k[j] + dst[i+31:i] := (NOT a[i+31:i]) AND b[i+31:i] + ELSE + dst[i+31:i] := 0 + FI +ENDFOR +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Integer + AVX512VL + AVX512F + Logical + + + + + + Compute the bitwise NOT of packed 64-bit integers in "a" and then AND with "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +FOR j := 0 to 3 i := j*64 IF k[j] - FOR l := 0 to j-1 - m := l*64 - dst[i+l] := (a[i+63:i] == a[m+63:m]) ? 1 : 0 - ENDFOR - dst[i+63:i+j] := 0 + dst[i+63:i] := ((NOT a[i+63:i]) AND b[i+63:i]) ELSE dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - AVX512CD - Compare - - - Test each 64-bit element of "a" for equality with all other elements in "a" closer to the least significant bit using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). Each element's comparison forms a zero extended bit vector in "dst". + AVX512VL + AVX512F + Logical + + + + + Compute the bitwise NOT of packed 64-bit integers in "a" and then AND with "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 +FOR j := 0 to 3 i := j*64 IF k[j] - FOR l := 0 to j-1 - m := l*64 - dst[i+l] := (a[i+63:i] == a[m+63:m]) ? 1 : 0 - ENDFOR - dst[i+63:i+j] := 0 + dst[i+63:i] := (NOT a[i+63:i]) AND b[i+63:i] ELSE dst[i+63:i] := 0 FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - AVX512F/KNCNI - Swizzle - - - - - Shuffle 32-bit integers in "a" across lanes using the corresponding index in "idx", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). Note that this intrinsic shuffles across 128-bit lanes, unlike past intrinsics that use the "permutevar" name. This intrinsic is identical to "_mm512_mask_permutexvar_epi32", and it is recommended that you use that intrinsic name. + AVX512VL + AVX512F + Logical + + + + + + Compute the bitwise NOT of packed 64-bit integers in "a" and then AND with "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*32 - id := idx[i+3:i]*32 +FOR j := 0 to 1 + i := j*64 IF k[j] - dst[i+31:i] := a[id+31:id] + dst[i+63:i] := ((NOT a[i+63:i]) AND b[i+63:i]) ELSE - dst[i+31:i] := src[i+31:i] + dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer + AVX512VL AVX512F - Swizzle - - - - - Shuffle 32-bit integers in "a" across lanes using the corresponding index in "idx", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Logical + + + + + Compute the bitwise NOT of packed 64-bit integers in "a" and then AND with "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*32 - id := idx[i+3:i]*32 +FOR j := 0 to 1 + i := j*64 IF k[j] - dst[i+31:i] := a[id+31:id] + dst[i+63:i] := (NOT a[i+63:i]) AND b[i+63:i] ELSE - dst[i+31:i] := src[i+31:i] + dst[i+63:i] := 0 FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - - + Integer + AVX512VL AVX512F - Swizzle - - - - Shuffle 32-bit integers in "a" across lanes using the corresponding index in "idx", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Logical + + + + + + Compute the bitwise AND of packed 64-bit integers in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*32 - id := idx[i+3:i]*32 +FOR j := 0 to 3 + i := j*64 IF k[j] - dst[i+31:i] := a[id+31:id] + dst[i+63:i] := a[i+63:i] AND b[i+63:i] ELSE - dst[i+31:i] := 0 + dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - AVX512F/KNCNI - Swizzle - - - Shuffle 32-bit integers in "a" across lanes using the corresponding index in "idx", and store the results in "dst". Note that this intrinsic shuffles across 128-bit lanes, unlike past intrinsics that use the "permutevar" name. This intrinsic is identical to "_mm512_permutexvar_epi32", and it is recommended that you use that intrinsic name. - + AVX512VL + AVX512F + Logical + + + + + Compute the bitwise AND of packed 64-bit integers in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*32 - id := idx[i+3:i]*32 - dst[i+31:i] := a[id+31:id] +FOR j := 0 to 3 + i := j*64 + IF k[j] + dst[i+63:i] := a[i+63:i] AND b[i+63:i] + ELSE + dst[i+63:i] := 0 + FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer + AVX512VL AVX512F - Swizzle - - - Shuffle 32-bit integers in "a" across lanes using the corresponding index in "idx", and store the results in "dst". - + Logical + + + + + + Compute the bitwise AND of packed 64-bit integers in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*32 - id := idx[i+3:i]*32 - dst[i+31:i] := a[id+31:id] +FOR j := 0 to 1 + i := j*64 + IF k[j] + dst[i+63:i] := a[i+63:i] AND b[i+63:i] + ELSE + dst[i+63:i] := src[i+63:i] + FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer + AVX512VL AVX512F - Swizzle - - - - - Shuffle 32-bit integers in "a" and "b" across lanes using the corresponding selector and index in "idx", and store the results in "dst" using writemask "k" (elements are copied from "idx" when the corresponding mask bit is not set). + Logical + + + + + Compute the bitwise AND of packed 64-bit integers in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*32 - off := idx[i+3:i]*32 +FOR j := 0 to 1 + i := j*64 IF k[j] - dst[i+31:i] := idx[i+4] ? b[off+31:off] : a[off+31:off] + dst[i+63:i] := a[i+63:i] AND b[i+63:i] ELSE - dst[i+31:i] := idx[i+31:i] + dst[i+63:i] := 0 FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer + AVX512VL AVX512F - Swizzle - - - - - Shuffle 32-bit integers in "a" and "b" across lanes using the corresponding selector and index in "idx", and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). + Miscellaneous + + + + + Blend packed 32-bit integers from "a" and "b" using control mask "k", and store the results in "dst". -FOR j := 0 to 15 +FOR j := 0 to 7 i := j*32 - off := idx[i+3:i]*32 IF k[j] - dst[i+31:i] := idx[i+4] ? b[off+31:off] : a[off+31:off] + dst[i+31:i] := b[i+31:i] ELSE dst[i+31:i] := a[i+31:i] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer + AVX512VL AVX512F - Swizzle - - - - - Shuffle 32-bit integers in "a" and "b" across lanes using the corresponding selector and index in "idx", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Miscellaneous + + + + + Blend packed 32-bit integers from "a" and "b" using control mask "k", and store the results in "dst". -FOR j := 0 to 15 +FOR j := 0 to 3 i := j*32 - off := idx[i+3:i]*32 IF k[j] - dst[i+31:i] := (idx[i+4]) ? b[off+31:off] : a[off+31:off] + dst[i+31:i] := b[i+31:i] ELSE - dst[i+31:i] := 0 + dst[i+31:i] := a[i+31:i] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - - +
immintrin.h
 - + Integer + AVX512VL AVX512F - Swizzle - - - - Shuffle 32-bit integers in "a" and "b" across lanes using the corresponding selector and index in "idx", and store the results in "dst". - + Miscellaneous + + + + + Blend packed 64-bit integers from "a" and "b" using control mask "k", and store the results in "dst". -FOR j := 0 to 15 - i := j*32 - off := idx[i+3:i]*32 - dst[i+31:i] := idx[i+4] ? b[off+31:off] : a[off+31:off] +FOR j := 0 to 3 + i := j*64 + IF k[j] + dst[i+63:i] := b[i+63:i] + ELSE + dst[i+63:i] := a[i+63:i] + FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Swizzle - - - - - Shuffle double-precision (64-bit) floating-point elements in "a" and "b" across lanes using the corresponding selector and index in "idx", and store the results in "dst" using writemask "k" (elements are copied from "idx" when the corresponding mask bit is not set) + Miscellaneous + + + + + Blend packed 64-bit integers from "a" and "b" using control mask "k", and store the results in "dst". -FOR j := 0 to 7 +FOR j := 0 to 1 i := j*64 - off := idx[i+2:i]*64 IF k[j] - dst[i+63:i] := idx[i+3] ? b[off+63:off] : a[off+63:off] + dst[i+63:i] := b[i+63:i] ELSE - dst[i+63:i] := idx[i+63:i] + dst[i+63:i] := a[i+63:i] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Swizzle - - - - - Shuffle double-precision (64-bit) floating-point elements in "a" and "b" across lanes using the corresponding selector and index in "idx", and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). + Miscellaneous + + + + + Broadcast the low packed 32-bit integer from "a" to all elements of "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). FOR j := 0 to 7 - i := j*64 - off := idx[i+2:i]*64 + i := j*32 IF k[j] - dst[i+63:i] := idx[i+3] ? b[off+63:off] : a[off+63:off] + dst[i+31:i] := a[31:0] ELSE - dst[i+63:i] := a[i+63:i] + dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Swizzle - - - - - Shuffle double-precision (64-bit) floating-point elements in "a" and "b" across lanes using the corresponding selector and index in "idx", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Set + + + + + Broadcast 32-bit integer "a" to all elements of "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). FOR j := 0 to 7 - i := j*64 - off := idx[i+2:i]*64 + i := j*32 IF k[j] - dst[i+63:i] := (idx[i+3]) ? b[off+63:off] : a[off+63:off] + dst[i+31:i] := a[31:0] ELSE - dst[i+63:i] := 0 + dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Swizzle - - - - Shuffle double-precision (64-bit) floating-point elements in "a" and "b" across lanes using the corresponding selector and index in "idx", and store the results in "dst". + Miscellaneous + + + + Broadcast the low packed 32-bit integer from "a" to all elements of "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). FOR j := 0 to 7 - i := j*64 - off := idx[i+2:i]*64 - dst[i+63:i] := idx[i+3] ? b[off+63:off] : a[off+63:off] + i := j*32 + IF k[j] + dst[i+31:i] := a[31:0] + ELSE + dst[i+31:i] := 0 + FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Swizzle - - - - - Shuffle single-precision (32-bit) floating-point elements in "a" and "b" across lanes using the corresponding selector and index in "idx", and store the results in "dst" using writemask "k" (elements are copied from "idx" when the corresponding mask bit is not set). + Set + + + + Broadcast 32-bit integer "a" to all elements of "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 15 +FOR j := 0 to 7 i := j*32 - off := idx[i+3:i]*32 IF k[j] - dst[i+31:i] := idx[i+4] ? b[off+31:off] : a[off+31:off] + dst[i+31:i] := a[31:0] ELSE - dst[i+31:i] := idx[i+31:i] + dst[i+31:i] := 0 FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Swizzle - - - - - Shuffle single-precision (32-bit) floating-point elements in "a" and "b" across lanes using the corresponding selector and index in "idx", and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). + Miscellaneous + + + + + Broadcast the low packed 32-bit integer from "a" to all elements of "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 15 +FOR j := 0 to 3 i := j*32 - off := idx[i+3:i]*32 IF k[j] - dst[i+31:i] := idx[i+4] ? b[off+31:off] : a[off+31:off] + dst[i+31:i] := a[31:0] ELSE - dst[i+31:i] := a[i+31:i] + dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Swizzle - - - - - Shuffle single-precision (32-bit) floating-point elements in "a" and "b" across lanes using the corresponding selector and index in "idx", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Set + + + + + Broadcast 32-bit integer "a" to all elements of "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 15 +FOR j := 0 to 3 i := j*32 - off := idx[i+3:i]*32 IF k[j] - dst[i+31:i] := (idx[i+4]) ? b[off+31:off] : a[off+31:off] + dst[i+31:i] := a[31:0] ELSE - dst[i+31:i] := 0 + dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Swizzle - - - - Shuffle single-precision (32-bit) floating-point elements in "a" and "b" across lanes using the corresponding selector and index in "idx", and store the results in "dst". + Miscellaneous + + + + Broadcast the low packed 32-bit integer from "a" to all elements of "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 15 +FOR j := 0 to 3 i := j*32 - off := idx[i+3:i]*32 - dst[i+31:i] := idx[i+4] ? b[off+31:off] : a[off+31:off] + IF k[j] + dst[i+31:i] := a[31:0] + ELSE + dst[i+31:i] := 0 + FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - - +
immintrin.h
 - + Integer + AVX512VL AVX512F - Swizzle - - - - - Shuffle 64-bit integers in "a" and "b" across lanes using the corresponding selector and index in "idx", and store the results in "dst" using writemask "k" (elements are copied from "idx" when the corresponding mask bit is not set). + Set + + + + Broadcast 32-bit integer "a" to all elements of "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := j*64 - off := idx[i+2:i]*64 +FOR j := 0 to 3 + i := j*32 IF k[j] - dst[i+63:i] := idx[i+3] ? b[off+63:off] : a[off+63:off] + dst[i+31:i] := a[31:0] ELSE - dst[i+63:i] := idx[i+63:i] + dst[i+31:i] := 0 FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer + AVX512VL AVX512F - Swizzle - - - - - Shuffle 64-bit integers in "a" and "b" across lanes using the corresponding selector and index in "idx", and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). + Miscellaneous + + + + + Broadcast the low packed 64-bit integer from "a" to all elements of "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 7 +FOR j := 0 to 3 i := j*64 - off := idx[i+2:i]*64 IF k[j] - dst[i+63:i] := idx[i+3] ? b[off+63:off] : a[off+63:off] + dst[i+63:i] := a[63:0] ELSE - dst[i+63:i] := a[i+63:i] + dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer + AVX512VL AVX512F - Swizzle - - - - - Shuffle 64-bit integers in "a" and "b" across lanes using the corresponding selector and index in "idx", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Set + + + + + Broadcast 64-bit integer "a" to all elements of "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 7 +FOR j := 0 to 3 i := j*64 - off := idx[i+2:i]*64 IF k[j] - dst[i+63:i] := (idx[i+3]) ? b[off+63:off] : a[off+63:off] + dst[i+63:i] := a[63:0] ELSE - dst[i+63:i] := 0 + dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - - +
immintrin.h
 - + Integer + AVX512VL AVX512F - Swizzle - - - - Shuffle 64-bit integers in "a" and "b" across lanes using the corresponding selector and index in "idx", and store the results in "dst". - + Miscellaneous + + + + Broadcast the low packed 64-bit integer from "a" to all elements of "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 +FOR j := 0 to 3 i := j*64 - off := idx[i+2:i]*64 - dst[i+63:i] := idx[i+3] ? b[off+63:off] : a[off+63:off] + IF k[j] + dst[i+63:i] := a[63:0] + ELSE + dst[i+63:i] := 0 + FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Swizzle - - - - - Shuffle double-precision (64-bit) floating-point elements in "a" within 128-bit lanes using the control in "imm8", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Set + + + + Broadcast 64-bit integer "a" to all elements of "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -IF (imm8[0] == 0) tmp_dst[63:0] := a[63:0] -IF (imm8[0] == 1) tmp_dst[63:0] := a[127:64] -IF (imm8[1] == 0) tmp_dst[127:64] := a[63:0] -IF (imm8[1] == 1) tmp_dst[127:64] := a[127:64] -IF (imm8[2] == 0) tmp_dst[191:128] := a[191:128] -IF (imm8[2] == 1) tmp_dst[191:128] := a[255:192] -IF (imm8[3] == 0) tmp_dst[255:192] := a[191:128] -IF (imm8[3] == 1) tmp_dst[255:192] := a[255:192] -IF (imm8[4] == 0) tmp_dst[319:256] := a[319:256] -IF (imm8[4] == 1) tmp_dst[319:256] := a[383:320] -IF (imm8[5] == 0) tmp_dst[383:320] := a[319:256] -IF (imm8[5] == 1) tmp_dst[383:320] := a[383:320] -IF (imm8[6] == 0) tmp_dst[447:384] := a[447:384] -IF (imm8[6] == 1) tmp_dst[447:384] := a[511:448] -IF (imm8[7] == 0) tmp_dst[511:448] := a[447:384] -IF (imm8[7] == 1) tmp_dst[511:448] := a[511:448] -FOR j := 0 to 7 +FOR j := 0 to 3 i := j*64 IF k[j] - dst[i+63:i] := tmp_dst[i+63:i] + dst[i+63:i] := a[63:0] ELSE - dst[i+63:i] := src[i+63:i] + dst[i+63:i] := 0 FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Swizzle - - - - - Shuffle double-precision (64-bit) floating-point elements in "a" within 128-bit lanes using the control in "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Miscellaneous + + + + + Broadcast the low packed 64-bit integer from "a" to all elements of "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -IF (b[1] == 0) tmp_dst[63:0] := a[63:0] -IF (b[1] == 1) tmp_dst[63:0] := a[127:64] -IF (b[65] == 0) tmp_dst[127:64] := a[63:0] -IF (b[65] == 1) tmp_dst[127:64] := a[127:64] -IF (b[129] == 0) tmp_dst[191:128] := a[191:128] -IF (b[129] == 1) tmp_dst[191:128] := a[255:192] -IF (b[193] == 0) tmp_dst[255:192] := a[191:128] -IF (b[193] == 1) tmp_dst[255:192] := a[255:192] -IF (b[257] == 0) tmp_dst[319:256] := a[319:256] -IF (b[257] == 1) tmp_dst[319:256] := a[383:320] -IF (b[321] == 0) tmp_dst[383:320] := a[319:256] -IF (b[321] == 1) tmp_dst[383:320] := a[383:320] -IF (b[385] == 0) tmp_dst[447:384] := a[447:384] -IF (b[385] == 1) tmp_dst[447:384] := a[511:448] -IF (b[449] == 0) tmp_dst[511:448] := a[447:384] -IF (b[449] == 1) tmp_dst[511:448] := a[511:448] -FOR j := 0 to 7 +FOR j := 0 to 1 i := j*64 IF k[j] - dst[i+63:i] := tmp_dst[i+63:i] + dst[i+63:i] := a[63:0] ELSE dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Swizzle - - - - Shuffle double-precision (64-bit) floating-point elements in "a" within 128-bit lanes using the control in "imm8", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Set + + + + + Broadcast 64-bit integer "a" to all elements of "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -IF (imm8[0] == 0) tmp_dst[63:0] := a[63:0] -IF (imm8[0] == 1) tmp_dst[63:0] := a[127:64] -IF (imm8[1] == 0) tmp_dst[127:64] := a[63:0] -IF (imm8[1] == 1) tmp_dst[127:64] := a[127:64] -IF (imm8[2] == 0) tmp_dst[191:128] := a[191:128] -IF (imm8[2] == 1) tmp_dst[191:128] := a[255:192] -IF (imm8[3] == 0) tmp_dst[255:192] := a[191:128] -IF (imm8[3] == 1) tmp_dst[255:192] := a[255:192] -IF (imm8[4] == 0) tmp_dst[319:256] := a[319:256] -IF (imm8[4] == 1) tmp_dst[319:256] := a[383:320] -IF (imm8[5] == 0) tmp_dst[383:320] := a[319:256] -IF (imm8[5] == 1) tmp_dst[383:320] := a[383:320] -IF (imm8[6] == 0) tmp_dst[447:384] := a[447:384] -IF (imm8[6] == 1) tmp_dst[447:384] := a[511:448] -IF (imm8[7] == 0) tmp_dst[511:448] := a[447:384] -IF (imm8[7] == 1) tmp_dst[511:448] := a[511:448] -FOR j := 0 to 7 +FOR j := 0 to 1 i := j*64 IF k[j] - dst[i+63:i] := tmp_dst[i+63:i] + dst[i+63:i] := a[63:0] ELSE - dst[i+63:i] := 0 + dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Swizzle - - - - Shuffle double-precision (64-bit) floating-point elements in "a" within 128-bit lanes using the control in "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Miscellaneous + + + + Broadcast the low packed 64-bit integer from "a" to all elements of "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -IF (b[1] == 0) tmp_dst[63:0] := a[63:0] -IF (b[1] == 1) tmp_dst[63:0] := a[127:64] -IF (b[65] == 0) tmp_dst[127:64] := a[63:0] -IF (b[65] == 1) tmp_dst[127:64] := a[127:64] -IF (b[129] == 0) tmp_dst[191:128] := a[191:128] -IF (b[129] == 1) tmp_dst[191:128] := a[255:192] -IF (b[193] == 0) tmp_dst[255:192] := a[191:128] -IF (b[193] == 1) tmp_dst[255:192] := a[255:192] -IF (b[257] == 0) tmp_dst[319:256] := a[319:256] -IF (b[257] == 1) tmp_dst[319:256] := a[383:320] -IF (b[321] == 0) tmp_dst[383:320] := a[319:256] -IF (b[321] == 1) tmp_dst[383:320] := a[383:320] -IF (b[385] == 0) tmp_dst[447:384] := a[447:384] -IF (b[385] == 1) tmp_dst[447:384] := a[511:448] -IF (b[449] == 0) tmp_dst[511:448] := a[447:384] -IF (b[449] == 1) tmp_dst[511:448] := a[511:448] -FOR j := 0 to 7 +FOR j := 0 to 1 i := j*64 IF k[j] - dst[i+63:i] := tmp_dst[i+63:i] + dst[i+63:i] := a[63:0] ELSE dst[i+63:i] := 0 FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Swizzle - - - Shuffle double-precision (64-bit) floating-point elements in "a" within 128-bit lanes using the control in "imm8", and store the results in "dst". + Set + + + + Broadcast 64-bit integer "a" to all elements of "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -IF (imm8[0] == 0) dst[63:0] := a[63:0] -IF (imm8[0] == 1) dst[63:0] := a[127:64] -IF (imm8[1] == 0) dst[127:64] := a[63:0] -IF (imm8[1] == 1) dst[127:64] := a[127:64] -IF (imm8[2] == 0) dst[191:128] := a[191:128] -IF (imm8[2] == 1) dst[191:128] := a[255:192] -IF (imm8[3] == 0) dst[255:192] := a[191:128] -IF (imm8[3] == 1) dst[255:192] := a[255:192] -IF (imm8[4] == 0) dst[319:256] := a[319:256] -IF (imm8[4] == 1) dst[319:256] := a[383:320] -IF (imm8[5] == 0) dst[383:320] := a[319:256] -IF (imm8[5] == 1) dst[383:320] := a[383:320] -IF (imm8[6] == 0) dst[447:384] := a[447:384] -IF (imm8[6] == 1) dst[447:384] := a[511:448] -IF (imm8[7] == 0) dst[511:448] := a[447:384] -IF (imm8[7] == 1) dst[511:448] := a[511:448] -dst[MAX:512] := 0 +FOR j := 0 to 1 + i := j*64 + IF k[j] + dst[i+63:i] := a[63:0] + ELSE + dst[i+63:i] := 0 + FI +ENDFOR +dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point + + Integer + Mask + AVX512VL AVX512F - Swizzle - - - Shuffle double-precision (64-bit) floating-point elements in "a" within 128-bit lanes using the control in "b", and store the results in "dst". - -IF (b[1] == 0) dst[63:0] := a[63:0] -IF (b[1] == 1) dst[63:0] := a[127:64] -IF (b[65] == 0) dst[127:64] := a[63:0] -IF (b[65] == 1) dst[127:64] := a[127:64] -IF (b[129] == 0) dst[191:128] := a[191:128] -IF (b[129] == 1) dst[191:128] := a[255:192] -IF (b[193] == 0) dst[255:192] := a[191:128] -IF (b[193] == 1) dst[255:192] := a[255:192] -IF (b[257] == 0) dst[319:256] := a[319:256] -IF (b[257] == 1) dst[319:256] := a[383:320] -IF (b[321] == 0) dst[383:320] := a[319:256] -IF (b[321] == 1) dst[383:320] := a[383:320] -IF (b[385] == 0) dst[447:384] := a[447:384] -IF (b[385] == 1) dst[447:384] := a[511:448] -IF (b[449] == 0) dst[511:448] := a[447:384] -IF (b[449] == 1) dst[511:448] := a[511:448] -dst[MAX:512] := 0 - - -
immintrin.h
- - - Floating Point - AVX512F - Swizzle - - - - - Shuffle single-precision (32-bit) floating-point elements in "a" within 128-bit lanes using the control in "imm8", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -SELECT4(src, control){ - CASE(control[1:0]) - 0: tmp[31:0] := src[31:0] - 1: tmp[31:0] := src[63:32] - 2: tmp[31:0] := src[95:64] - 3: tmp[31:0] := src[127:96] - ESAC - RETURN tmp[31:0] -} - -tmp_dst[31:0] := SELECT4(a[127:0], imm8[1:0]) -tmp_dst[63:32] := SELECT4(a[127:0], imm8[3:2]) -tmp_dst[95:64] := SELECT4(a[127:0], imm8[5:4]) -tmp_dst[127:96] := SELECT4(a[127:0], imm8[7:6]) -tmp_dst[159:128] := SELECT4(a[255:128], imm8[1:0]) -tmp_dst[191:160] := SELECT4(a[255:128], imm8[3:2]) -tmp_dst[223:192] := SELECT4(a[255:128], imm8[5:4]) -tmp_dst[255:224] := SELECT4(a[255:128], imm8[7:6]) -tmp_dst[287:256] := SELECT4(a[383:256], imm8[1:0]) -tmp_dst[319:288] := SELECT4(a[383:256], imm8[3:2]) -tmp_dst[351:320] := SELECT4(a[383:256], imm8[5:4]) -tmp_dst[383:352] := SELECT4(a[383:256], imm8[7:6]) -tmp_dst[415:384] := SELECT4(a[511:384], imm8[1:0]) -tmp_dst[447:416] := SELECT4(a[511:384], imm8[3:2]) -tmp_dst[479:448] := SELECT4(a[511:384], imm8[5:4]) -tmp_dst[511:480] := SELECT4(a[511:384], imm8[7:6]) -FOR j := 0 to 15 + Compare + + + + + Compare packed signed 32-bit integers in "a" and "b" based on the comparison operand specified by "imm8", and store the results in mask vector "k". + CASE (imm8[2:0]) OF +0: OP := _MM_CMPINT_EQ +1: OP := _MM_CMPINT_LT +2: OP := _MM_CMPINT_LE +3: OP := _MM_CMPINT_FALSE +4: OP := _MM_CMPINT_NE +5: OP := _MM_CMPINT_NLT +6: OP := _MM_CMPINT_NLE +7: OP := _MM_CMPINT_TRUE +ESAC +FOR j := 0 to 7 i := j*32 - IF k[j] - dst[i+31:i] := tmp_dst[i+31:i] - ELSE - dst[i+31:i] := src[i+31:i] - FI + k[j] := ( a[i+31:i] OP b[i+31:i] ) ? 1 : 0 ENDFOR -dst[MAX:512] := 0 +k[MAX:8] := 0 - +
immintrin.h
 - - Floating Point + + Integer + Mask + AVX512VL AVX512F - Swizzle - - - - - Shuffle single-precision (32-bit) floating-point elements in "a" within 128-bit lanes using the control in "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Compare + + + + Compare packed signed 32-bit integers in "a" and "b" for equality, and store the results in mask vector "k". -SELECT4(src, control){ - CASE(control[1:0]) - 0: tmp[31:0] := src[31:0] - 1: tmp[31:0] := src[63:32] - 2: tmp[31:0] := src[95:64] - 3: tmp[31:0] := src[127:96] - ESAC - RETURN tmp[31:0] -} - -tmp_dst[31:0] := SELECT4(a[127:0], b[1:0]) -tmp_dst[63:32] := SELECT4(a[127:0], b[33:32]) -tmp_dst[95:64] := SELECT4(a[127:0], b[65:64]) -tmp_dst[127:96] := SELECT4(a[127:0], b[97:96]) -tmp_dst[159:128] := SELECT4(a[255:128], b[129:128]) -tmp_dst[191:160] := SELECT4(a[255:128], b[161:160]) -tmp_dst[223:192] := SELECT4(a[255:128], b[193:192]) -tmp_dst[255:224] := SELECT4(a[255:128], b[225:224]) -tmp_dst[287:256] := SELECT4(a[383:256], b[257:256]) -tmp_dst[319:288] := SELECT4(a[383:256], b[289:288]) -tmp_dst[351:320] := SELECT4(a[383:256], b[321:320]) -tmp_dst[383:352] := SELECT4(a[383:256], b[353:352]) -tmp_dst[415:384] := SELECT4(a[511:384], b[385:384]) -tmp_dst[447:416] := SELECT4(a[511:384], b[417:416]) -tmp_dst[479:448] := SELECT4(a[511:384], b[449:448]) -tmp_dst[511:480] := SELECT4(a[511:384], b[481:480]) -FOR j := 0 to 15 +FOR j := 0 to 7 i := j*32 - IF k[j] - dst[i+31:i] := tmp_dst[i+31:i] - ELSE - dst[i+31:i] := src[i+31:i] - FI + k[j] := ( a[i+31:i] == b[i+31:i] ) ? 1 : 0 ENDFOR -dst[MAX:512] := 0 +k[MAX:8] := 0 - +
immintrin.h
 - - Floating Point + + Integer + Mask + AVX512VL AVX512F - Swizzle - - - - Shuffle single-precision (32-bit) floating-point elements in "a" within 128-bit lanes using the control in "imm8", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Compare + + + + Compare packed signed 32-bit integers in "a" and "b" for greater-than-or-equal, and store the results in mask vector "k". -SELECT4(src, control){ - CASE(control[1:0]) - 0: tmp[31:0] := src[31:0] - 1: tmp[31:0] := src[63:32] - 2: tmp[31:0] := src[95:64] - 3: tmp[31:0] := src[127:96] - ESAC - RETURN tmp[31:0] -} - -tmp_dst[31:0] := SELECT4(a[127:0], imm8[1:0]) -tmp_dst[63:32] := SELECT4(a[127:0], imm8[3:2]) -tmp_dst[95:64] := SELECT4(a[127:0], imm8[5:4]) -tmp_dst[127:96] := SELECT4(a[127:0], imm8[7:6]) -tmp_dst[159:128] := SELECT4(a[255:128], imm8[1:0]) -tmp_dst[191:160] := SELECT4(a[255:128], imm8[3:2]) -tmp_dst[223:192] := SELECT4(a[255:128], imm8[5:4]) -tmp_dst[255:224] := SELECT4(a[255:128], imm8[7:6]) -tmp_dst[287:256] := SELECT4(a[383:256], imm8[1:0]) -tmp_dst[319:288] := SELECT4(a[383:256], imm8[3:2]) -tmp_dst[351:320] := SELECT4(a[383:256], imm8[5:4]) -tmp_dst[383:352] := SELECT4(a[383:256], imm8[7:6]) -tmp_dst[415:384] := SELECT4(a[511:384], imm8[1:0]) -tmp_dst[447:416] := SELECT4(a[511:384], imm8[3:2]) -tmp_dst[479:448] := SELECT4(a[511:384], imm8[5:4]) -tmp_dst[511:480] := SELECT4(a[511:384], imm8[7:6]) -FOR j := 0 to 15 +FOR j := 0 to 7 i := j*32 - IF k[j] - dst[i+31:i] := tmp_dst[i+31:i] - ELSE - dst[i+31:i] := 0 - FI + k[j] := ( a[i+31:i] >= b[i+31:i] ) ? 1 : 0 ENDFOR -dst[MAX:512] := 0 +k[MAX:8] := 0 - +
immintrin.h
 - - Floating Point + + Integer + Mask + AVX512VL AVX512F - Swizzle - - - - Shuffle single-precision (32-bit) floating-point elements in "a" within 128-bit lanes using the control in "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Compare + + + + Compare packed signed 32-bit integers in "a" and "b" for greater-than, and store the results in mask vector "k". -SELECT4(src, control){ - CASE(control[1:0]) - 0: tmp[31:0] := src[31:0] - 1: tmp[31:0] := src[63:32] - 2: tmp[31:0] := src[95:64] - 3: tmp[31:0] := src[127:96] - ESAC - RETURN tmp[31:0] -} - -tmp_dst[31:0] := SELECT4(a[127:0], b[1:0]) -tmp_dst[63:32] := SELECT4(a[127:0], b[33:32]) -tmp_dst[95:64] := SELECT4(a[127:0], b[65:64]) -tmp_dst[127:96] := SELECT4(a[127:0], b[97:96]) -tmp_dst[159:128] := SELECT4(a[255:128], b[129:128]) -tmp_dst[191:160] := SELECT4(a[255:128], b[161:160]) -tmp_dst[223:192] := SELECT4(a[255:128], b[193:192]) -tmp_dst[255:224] := SELECT4(a[255:128], b[225:224]) -tmp_dst[287:256] := SELECT4(a[383:256], b[257:256]) -tmp_dst[319:288] := SELECT4(a[383:256], b[289:288]) -tmp_dst[351:320] := SELECT4(a[383:256], b[321:320]) -tmp_dst[383:352] := SELECT4(a[383:256], b[353:352]) -tmp_dst[415:384] := SELECT4(a[511:384], b[385:384]) -tmp_dst[447:416] := SELECT4(a[511:384], b[417:416]) -tmp_dst[479:448] := SELECT4(a[511:384], b[449:448]) -tmp_dst[511:480] := SELECT4(a[511:384], b[481:480]) -FOR j := 0 to 15 +FOR j := 0 to 7 i := j*32 - IF k[j] - dst[i+31:i] := tmp_dst[i+31:i] - ELSE - dst[i+31:i] := 0 - FI + k[j] := ( a[i+31:i] > b[i+31:i] ) ? 1 : 0 ENDFOR -dst[MAX:512] := 0 +k[MAX:8] := 0 - +
immintrin.h
 - - Floating Point + + Integer + Mask + AVX512VL AVX512F - Swizzle - - - Shuffle single-precision (32-bit) floating-point elements in "a" within 128-bit lanes using the control in "imm8", and store the results in "dst". + Compare + + + + Compare packed signed 32-bit integers in "a" and "b" for less-than-or-equal, and store the results in mask vector "k". -SELECT4(src, control){ - CASE(control[1:0]) - 0: tmp[31:0] := src[31:0] - 1: tmp[31:0] := src[63:32] - 2: tmp[31:0] := src[95:64] - 3: tmp[31:0] := src[127:96] - ESAC - RETURN tmp[31:0] -} - -dst[31:0] := SELECT4(a[127:0], imm8[1:0]) -dst[63:32] := SELECT4(a[127:0], imm8[3:2]) -dst[95:64] := SELECT4(a[127:0], imm8[5:4]) -dst[127:96] := SELECT4(a[127:0], imm8[7:6]) -dst[159:128] := SELECT4(a[255:128], imm8[1:0]) -dst[191:160] := SELECT4(a[255:128], imm8[3:2]) -dst[223:192] := SELECT4(a[255:128], imm8[5:4]) -dst[255:224] := SELECT4(a[255:128], imm8[7:6]) -dst[287:256] := SELECT4(a[383:256], imm8[1:0]) -dst[319:288] := SELECT4(a[383:256], imm8[3:2]) -dst[351:320] := SELECT4(a[383:256], imm8[5:4]) -dst[383:352] := SELECT4(a[383:256], imm8[7:6]) -dst[415:384] := SELECT4(a[511:384], imm8[1:0]) -dst[447:416] := SELECT4(a[511:384], imm8[3:2]) -dst[479:448] := SELECT4(a[511:384], imm8[5:4]) -dst[511:480] := SELECT4(a[511:384], imm8[7:6]) -dst[MAX:512] := 0 +FOR j := 0 to 7 + i := j*32 + k[j] := ( a[i+31:i] <= b[i+31:i] ) ? 1 : 0 +ENDFOR +k[MAX:8] := 0 - +
immintrin.h
 - - Floating Point + + Integer + Mask + AVX512VL AVX512F - Swizzle - - - Shuffle single-precision (32-bit) floating-point elements in "a" within 128-bit lanes using the control in "b", and store the results in "dst". + Compare + + + + Compare packed signed 32-bit integers in "a" and "b" for less-than, and store the results in mask vector "k". -SELECT4(src, control){ - CASE(control[1:0]) - 0: tmp[31:0] := src[31:0] - 1: tmp[31:0] := src[63:32] - 2: tmp[31:0] := src[95:64] - 3: tmp[31:0] := src[127:96] - ESAC - RETURN tmp[31:0] -} - -dst[31:0] := SELECT4(a[127:0], b[1:0]) -dst[63:32] := SELECT4(a[127:0], b[33:32]) -dst[95:64] := SELECT4(a[127:0], b[65:64]) -dst[127:96] := SELECT4(a[127:0], b[97:96]) -dst[159:128] := SELECT4(a[255:128], b[129:128]) -dst[191:160] := SELECT4(a[255:128], b[161:160]) -dst[223:192] := SELECT4(a[255:128], b[193:192]) -dst[255:224] := SELECT4(a[255:128], b[225:224]) -dst[287:256] := SELECT4(a[383:256], b[257:256]) -dst[319:288] := SELECT4(a[383:256], b[289:288]) -dst[351:320] := SELECT4(a[383:256], b[321:320]) -dst[383:352] := SELECT4(a[383:256], b[353:352]) -dst[415:384] := SELECT4(a[511:384], b[385:384]) -dst[447:416] := SELECT4(a[511:384], b[417:416]) -dst[479:448] := SELECT4(a[511:384], b[449:448]) -dst[511:480] := SELECT4(a[511:384], b[481:480]) -dst[MAX:512] := 0 +FOR j := 0 to 7 + i := j*32 + k[j] := ( a[i+31:i] < b[i+31:i] ) ? 1 : 0 +ENDFOR +k[MAX:8] := 0 - +
immintrin.h
 - - Floating Point + + Integer + Mask + AVX512VL AVX512F - Swizzle - - - - - Shuffle double-precision (64-bit) floating-point elements in "a" within 256-bit lanes using the control in "imm8", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Compare + + + + Compare packed signed 32-bit integers in "a" and "b" for not-equal, and store the results in mask vector "k". -SELECT4(src, control){ - CASE(control[1:0]) - 0: tmp[63:0] := src[63:0] - 1: tmp[63:0] := src[127:64] - 2: tmp[63:0] := src[191:128] - 3: tmp[63:0] := src[255:192] - ESAC - RETURN tmp[63:0] -} - -tmp_dst[63:0] := SELECT4(a[255:0], imm8[1:0]) -tmp_dst[127:64] := SELECT4(a[255:0], imm8[3:2]) -tmp_dst[191:128] := SELECT4(a[255:0], imm8[5:4]) -tmp_dst[255:192] := SELECT4(a[255:0], imm8[7:6]) -tmp_dst[319:256] := SELECT4(a[511:256], imm8[1:0]) -tmp_dst[383:320] := SELECT4(a[511:256], imm8[3:2]) -tmp_dst[447:384] := SELECT4(a[511:256], imm8[5:4]) -tmp_dst[511:448] := SELECT4(a[511:256], imm8[7:6]) FOR j := 0 to 7 - i := j*64 - IF k[j] - dst[i+63:i] := tmp_dst[i+63:i] - ELSE - dst[i+63:i] := src[i+63:i] - FI + i := j*32 + k[j] := ( a[i+31:i] != b[i+31:i] ) ? 1 : 0 ENDFOR -dst[MAX:512] := 0 +k[MAX:8] := 0 - +
immintrin.h
 - - Floating Point + + Integer + Mask + AVX512VL AVX512F - Swizzle - - - - - Shuffle double-precision (64-bit) floating-point elements in "a" across lanes using the corresponding index in "idx", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - + Compare + + + + + + Compare packed signed 32-bit integers in "a" and "b" based on the comparison operand specified by "imm8", and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). + CASE (imm8[2:0]) OF +0: OP := _MM_CMPINT_EQ +1: OP := _MM_CMPINT_LT +2: OP := _MM_CMPINT_LE +3: OP := _MM_CMPINT_FALSE +4: OP := _MM_CMPINT_NE +5: OP := _MM_CMPINT_NLT +6: OP := _MM_CMPINT_NLE +7: OP := _MM_CMPINT_TRUE +ESAC FOR j := 0 to 7 - i := j*64 - id := idx[i+2:i]*64 - IF k[j] - dst[i+63:i] := a[id+63:id] - ELSE - dst[i+63:i] := src[i+63:i] + i := j*32 + IF k1[j] + k[j] := ( a[i+31:i] OP b[i+31:i] ) ? 1 : 0 + ELSE + k[j] := 0 FI ENDFOR -dst[MAX:512] := 0 +k[MAX:8] := 0 - +
immintrin.h
 - - Floating Point + + Integer + Mask + AVX512VL AVX512F - Swizzle - - - - Shuffle double-precision (64-bit) floating-point elements in "a" within 256-bit lanes using the control in "imm8", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Compare + + + + + Compare packed signed 32-bit integers in "a" and "b" for equality, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). -SELECT4(src, control){ - CASE(control[1:0]) - 0: tmp[63:0] := src[63:0] - 1: tmp[63:0] := src[127:64] - 2: tmp[63:0] := src[191:128] - 3: tmp[63:0] := src[255:192] - ESAC - RETURN tmp[63:0] -} - -tmp_dst[63:0] := SELECT4(a[255:0], imm8[1:0]) -tmp_dst[127:64] := SELECT4(a[255:0], imm8[3:2]) -tmp_dst[191:128] := SELECT4(a[255:0], imm8[5:4]) -tmp_dst[255:192] := SELECT4(a[255:0], imm8[7:6]) -tmp_dst[319:256] := SELECT4(a[511:256], imm8[1:0]) -tmp_dst[383:320] := SELECT4(a[511:256], imm8[3:2]) -tmp_dst[447:384] := SELECT4(a[511:256], imm8[5:4]) -tmp_dst[511:448] := SELECT4(a[511:256], imm8[7:6]) FOR j := 0 to 7 - i := j*64 - IF k[j] - dst[i+63:i] := tmp_dst[i+63:i] - ELSE - dst[i+63:i] := 0 + i := j*32 + IF k1[j] + k[j] := ( a[i+31:i] == b[i+31:i] ) ? 1 : 0 + ELSE + k[j] := 0 FI ENDFOR -dst[MAX:512] := 0 +k[MAX:8] := 0 - +
immintrin.h
 - - Floating Point + + Integer + Mask + AVX512VL AVX512F - Swizzle - - - - Shuffle double-precision (64-bit) floating-point elements in "a" across lanes using the corresponding index in "idx", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Compare + + + + + Compare packed signed 32-bit integers in "a" and "b" for greater-than-or-equal, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). FOR j := 0 to 7 - i := j*64 - id := idx[i+2:i]*64 - IF k[j] - dst[i+63:i] := a[id+63:id] - ELSE - dst[i+63:i] := 0 + i := j*32 + IF k1[j] + k[j] := ( a[i+31:i] >= b[i+31:i] ) ? 1 : 0 + ELSE + k[j] := 0 FI ENDFOR -dst[MAX:512] := 0 +k[MAX:8] := 0 - +
immintrin.h
 - - Floating Point + + Integer + Mask + AVX512VL AVX512F - Swizzle - - - Shuffle double-precision (64-bit) floating-point elements in "a" within 256-bit lanes using the control in "imm8", and store the results in "dst". + Compare + + + + + Compare packed signed 32-bit integers in "a" and "b" for greater-than, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). -SELECT4(src, control){ - CASE(control[1:0]) - 0: tmp[63:0] := src[63:0] - 1: tmp[63:0] := src[127:64] - 2: tmp[63:0] := src[191:128] - 3: tmp[63:0] := src[255:192] - ESAC - RETURN tmp[63:0] -} - -dst[63:0] := SELECT4(a[255:0], imm8[1:0]) -dst[127:64] := SELECT4(a[255:0], imm8[3:2]) -dst[191:128] := SELECT4(a[255:0], imm8[5:4]) -dst[255:192] := SELECT4(a[255:0], imm8[7:6]) -dst[319:256] := SELECT4(a[511:256], imm8[1:0]) -dst[383:320] := SELECT4(a[511:256], imm8[3:2]) -dst[447:384] := SELECT4(a[511:256], imm8[5:4]) -dst[511:448] := SELECT4(a[511:256], imm8[7:6]) -dst[MAX:512] := 0 +FOR j := 0 to 7 + i := j*32 + IF k1[j] + k[j] := ( a[i+31:i] > b[i+31:i] ) ? 1 : 0 + ELSE + k[j] := 0 + FI +ENDFOR +k[MAX:8] := 0 - +
immintrin.h
 - - Floating Point + + Integer + Mask + AVX512VL AVX512F - Swizzle - - - Shuffle double-precision (64-bit) floating-point elements in "a" across lanes using the corresponding index in "idx", and store the results in "dst". + Compare + + + + + Compare packed signed 32-bit integers in "a" and "b" for less-than-or-equal, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). FOR j := 0 to 7 - i := j*64 - id := idx[i+2:i]*64 - dst[i+63:i] := a[id+63:id] + i := j*32 + IF k1[j] + k[j] := ( a[i+31:i] <= b[i+31:i] ) ? 1 : 0 + ELSE + k[j] := 0 + FI ENDFOR -dst[MAX:512] := 0 +k[MAX:8] := 0 - +
immintrin.h
 - - Floating Point + + Integer + Mask + AVX512VL AVX512F - Swizzle - - - - - Shuffle single-precision (32-bit) floating-point elements in "a" across lanes using the corresponding index in "idx", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Compare + + + + + Compare packed signed 32-bit integers in "a" and "b" for less-than, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 15 +FOR j := 0 to 7 i := j*32 - id := idx[i+3:i]*32 - IF k[j] - dst[i+31:i] := a[id+31:id] - ELSE - dst[i+31:i] := src[i+31:i] + IF k1[j] + k[j] := ( a[i+31:i] < b[i+31:i] ) ? 1 : 0 + ELSE + k[j] := 0 FI ENDFOR -dst[MAX:512] := 0 +k[MAX:8] := 0 - +
immintrin.h
 - - Floating Point + + Integer + Mask + AVX512VL AVX512F - Swizzle - - - - Shuffle single-precision (32-bit) floating-point elements in "a" across lanes using the corresponding index in "idx", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Compare + + + + + Compare packed signed 32-bit integers in "a" and "b" for not-equal, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 15 +FOR j := 0 to 7 i := j*32 - id := idx[i+3:i]*32 - IF k[j] - dst[i+31:i] := a[id+31:id] - ELSE - dst[i+31:i] := 0 + IF k1[j] + k[j] := ( a[i+31:i] != b[i+31:i] ) ? 1 : 0 + ELSE + k[j] := 0 FI ENDFOR -dst[MAX:512] := 0 +k[MAX:8] := 0 - +
immintrin.h
 - - Floating Point + + Integer + Mask + AVX512VL AVX512F - Swizzle - - - Shuffle single-precision (32-bit) floating-point elements in "a" across lanes using the corresponding index in "idx". - -FOR j := 0 to 15 + Compare + + + + + Compare packed signed 32-bit integers in "a" and "b" based on the comparison operand specified by "imm8", and store the results in mask vector "k". + CASE (imm8[2:0]) OF +0: OP := _MM_CMPINT_EQ +1: OP := _MM_CMPINT_LT +2: OP := _MM_CMPINT_LE +3: OP := _MM_CMPINT_FALSE +4: OP := _MM_CMPINT_NE +5: OP := _MM_CMPINT_NLT +6: OP := _MM_CMPINT_NLE +7: OP := _MM_CMPINT_TRUE +ESAC +FOR j := 0 to 3 i := j*32 - id := idx[i+3:i]*32 - dst[i+31:i] := a[id+31:id] + k[j] := ( a[i+31:i] OP b[i+31:i] ) ? 1 : 0 ENDFOR -dst[MAX:512] := 0 +k[MAX:4] := 0 - +
immintrin.h
 - + Integer + Mask + AVX512VL AVX512F - Swizzle - - - - - Shuffle 64-bit integers in "a" within 256-bit lanes using the control in "imm8", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Compare + + + + Compare packed signed 32-bit integers in "a" and "b" for equality, and store the results in mask vector "k". -SELECT4(src, control){ - CASE(control[1:0]) - 0: tmp[63:0] := src[63:0] - 1: tmp[63:0] := src[127:64] - 2: tmp[63:0] := src[191:128] - 3: tmp[63:0] := src[255:192] - ESAC - RETURN tmp[63:0] -} - -tmp_dst[63:0] := SELECT4(a[255:0], imm8[1:0]) -tmp_dst[127:64] := SELECT4(a[255:0], imm8[3:2]) -tmp_dst[191:128] := SELECT4(a[255:0], imm8[5:4]) -tmp_dst[255:192] := SELECT4(a[255:0], imm8[7:6]) -tmp_dst[319:256] := SELECT4(a[511:256], imm8[1:0]) -tmp_dst[383:320] := SELECT4(a[511:256], imm8[3:2]) -tmp_dst[447:384] := SELECT4(a[511:256], imm8[5:4]) -tmp_dst[511:448] := SELECT4(a[511:256], imm8[7:6]) -FOR j := 0 to 7 - i := j*64 - IF k[j] - dst[i+63:i] := tmp_dst[i+63:i] - ELSE - dst[i+63:i] := src[i+63:i] - FI +FOR j := 0 to 3 + i := j*32 + k[j] := ( a[i+31:i] == b[i+31:i] ) ? 1 : 0 ENDFOR -dst[MAX:512] := 0 +k[MAX:4] := 0 - +
immintrin.h
 - + Integer + Mask + AVX512VL AVX512F - Swizzle - - - - - Shuffle 64-bit integers in "a" across lanes using the corresponding index in "idx", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Compare + + + + Compare packed signed 32-bit integers in "a" and "b" for greater-than-or-equal, and store the results in mask vector "k". -FOR j := 0 to 7 - i := j*64 - id := idx[i+2:i]*64 - IF k[j] - dst[i+63:i] := a[id+63:id] - ELSE - dst[i+63:i] := src[i+63:i] - FI +FOR j := 0 to 3 + i := j*32 + k[j] := ( a[i+31:i] >= b[i+31:i] ) ? 1 : 0 ENDFOR -dst[MAX:512] := 0 +k[MAX:4] := 0 - +
immintrin.h
 - + Integer + Mask + AVX512VL AVX512F - Swizzle - - - - Shuffle 64-bit integers in "a" within 256-bit lanes using the control in "imm8", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Compare + + + + Compare packed signed 32-bit integers in "a" and "b" for greater-than, and store the results in mask vector "k". -SELECT4(src, control){ - CASE(control[1:0]) - 0: tmp[63:0] := src[63:0] - 1: tmp[63:0] := src[127:64] - 2: tmp[63:0] := src[191:128] - 3: tmp[63:0] := src[255:192] - ESAC - RETURN tmp[63:0] -} - -tmp_dst[63:0] := SELECT4(a[255:0], imm8[1:0]) -tmp_dst[127:64] := SELECT4(a[255:0], imm8[3:2]) -tmp_dst[191:128] := SELECT4(a[255:0], imm8[5:4]) -tmp_dst[255:192] := SELECT4(a[255:0], imm8[7:6]) -tmp_dst[319:256] := SELECT4(a[511:256], imm8[1:0]) -tmp_dst[383:320] := SELECT4(a[511:256], imm8[3:2]) -tmp_dst[447:384] := SELECT4(a[511:256], imm8[5:4]) -tmp_dst[511:448] := SELECT4(a[511:256], imm8[7:6]) -FOR j := 0 to 7 - i := j*64 - IF k[j] - dst[i+63:i] := tmp_dst[i+63:i] - ELSE - dst[i+63:i] := 0 - FI +FOR j := 0 to 3 + i := j*32 + k[j] := ( a[i+31:i] > b[i+31:i] ) ? 1 : 0 ENDFOR -dst[MAX:512] := 0 +k[MAX:4] := 0 - +
immintrin.h
 - + Integer + Mask + AVX512VL AVX512F - Swizzle - - - - Shuffle 64-bit integers in "a" across lanes using the corresponding index in "idx", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Compare + + + + Compare packed signed 32-bit integers in "a" and "b" for less-than-or-equal, and store the results in mask vector "k". -FOR j := 0 to 7 - i := j*64 - id := idx[i+2:i]*64 - IF k[j] - dst[i+63:i] := a[id+63:id] - ELSE - dst[i+63:i] := 0 - FI +FOR j := 0 to 3 + i := j*32 + k[j] := ( a[i+31:i] <= b[i+31:i] ) ? 1 : 0 ENDFOR -dst[MAX:512] := 0 +k[MAX:4] := 0 - +
immintrin.h
 - + Integer + Mask + AVX512VL AVX512F - Swizzle - - - Shuffle 64-bit integers in "a" within 256-bit lanes using the control in "imm8", and store the results in "dst". + Compare + + + + Compare packed signed 32-bit integers in "a" and "b" for less-than, and store the results in mask vector "k". -SELECT4(src, control){ - CASE(control[1:0]) - 0: tmp[63:0] := src[63:0] - 1: tmp[63:0] := src[127:64] - 2: tmp[63:0] := src[191:128] - 3: tmp[63:0] := src[255:192] - ESAC - RETURN tmp[63:0] -} - -dst[63:0] := SELECT4(a[255:0], imm8[1:0]) -dst[127:64] := SELECT4(a[255:0], imm8[3:2]) -dst[191:128] := SELECT4(a[255:0], imm8[5:4]) -dst[255:192] := SELECT4(a[255:0], imm8[7:6]) -dst[319:256] := SELECT4(a[511:256], imm8[1:0]) -dst[383:320] := SELECT4(a[511:256], imm8[3:2]) -dst[447:384] := SELECT4(a[511:256], imm8[5:4]) -dst[511:448] := SELECT4(a[511:256], imm8[7:6]) -dst[MAX:512] := 0 +FOR j := 0 to 3 + i := j*32 + k[j] := ( a[i+31:i] < b[i+31:i] ) ? 1 : 0 +ENDFOR +k[MAX:4] := 0 - +
immintrin.h
 - + Integer + Mask + AVX512VL AVX512F - Swizzle - - - Shuffle 64-bit integers in "a" across lanes using the corresponding index in "idx", and store the results in "dst". + Compare + + + + Compare packed signed 32-bit integers in "a" and "b" for not-equal, and store the results in mask vector "k". -FOR j := 0 to 7 - i := j*64 - id := idx[i+2:i]*64 - dst[i+63:i] := a[id+63:id] +FOR j := 0 to 3 + i := j*32 + k[j] := ( a[i+31:i] != b[i+31:i] ) ? 1 : 0 ENDFOR -dst[MAX:512] := 0 +k[MAX:4] := 0 - +
immintrin.h
 - + Integer + Mask + AVX512VL AVX512F - Swizzle - - - - Load contiguous active 32-bit integers from "a" (those with their respective bit set in mask "k"), and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -m := 0 -FOR j := 0 to 15 + Compare + + + + + + Compare packed signed 32-bit integers in "a" and "b" based on the comparison operand specified by "imm8", and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). + CASE (imm8[2:0]) OF +0: OP := _MM_CMPINT_EQ +1: OP := _MM_CMPINT_LT +2: OP := _MM_CMPINT_LE +3: OP := _MM_CMPINT_FALSE +4: OP := _MM_CMPINT_NE +5: OP := _MM_CMPINT_NLT +6: OP := _MM_CMPINT_NLE +7: OP := _MM_CMPINT_TRUE +ESAC +FOR j := 0 to 3 i := j*32 - IF k[j] - dst[i+31:i] := a[m+31:m] - m := m + 32 - ELSE - dst[i+31:i] := src[i+31:i] + IF k1[j] + k[j] := ( a[i+31:i] OP b[i+31:i] ) ? 1 : 0 + ELSE + k[j] := 0 FI ENDFOR -dst[MAX:512] := 0 +k[MAX:4] := 0 - +
immintrin.h
 - + Integer + Mask + AVX512VL AVX512F - Load - Swizzle - - - - Load contiguous active 32-bit integers from unaligned memory at "mem_addr" (those with their respective bit set in mask "k"), and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Compare + + + + + Compare packed signed 32-bit integers in "a" and "b" for equality, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). -m := 0 -FOR j := 0 to 15 +FOR j := 0 to 3 i := j*32 - IF k[j] - dst[i+31:i] := MEM[mem_addr+m+31:mem_addr+m] - m := m + 32 - ELSE - dst[i+31:i] := src[i+31:i] + IF k1[j] + k[j] := ( a[i+31:i] == b[i+31:i] ) ? 1 : 0 + ELSE + k[j] := 0 FI ENDFOR -dst[MAX:512] := 0 +k[MAX:4] := 0 - +
immintrin.h
 - + Integer + Mask + AVX512VL AVX512F - Swizzle - - - Load contiguous active 32-bit integers from "a" (those with their respective bit set in mask "k"), and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Compare + + + + + Compare packed signed 32-bit integers in "a" and "b" for greater-than-or-equal, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). -m := 0 -FOR j := 0 to 15 +FOR j := 0 to 3 i := j*32 - IF k[j] - dst[i+31:i] := a[m+31:m] - m := m + 32 - ELSE - dst[i+31:i] := 0 + IF k1[j] + k[j] := ( a[i+31:i] >= b[i+31:i] ) ? 1 : 0 + ELSE + k[j] := 0 FI ENDFOR -dst[MAX:512] := 0 +k[MAX:4] := 0 - +
immintrin.h
 - + Integer + Mask + AVX512VL AVX512F - Load - Swizzle - - - Load contiguous active 32-bit integers from unaligned memory at "mem_addr" (those with their respective bit set in mask "k"), and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Compare + + + + + Compare packed signed 32-bit integers in "a" and "b" for greater-than, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). -m := 0 -FOR j := 0 to 15 +FOR j := 0 to 3 i := j*32 - IF k[j] - dst[i+31:i] := MEM[mem_addr+m+31:mem_addr+m] - m := m + 32 - ELSE - dst[i+31:i] := 0 + IF k1[j] + k[j] := ( a[i+31:i] > b[i+31:i] ) ? 1 : 0 + ELSE + k[j] := 0 FI ENDFOR -dst[MAX:512] := 0 +k[MAX:4] := 0 - +
immintrin.h
 - + Integer + Mask + AVX512VL AVX512F - Swizzle - - - - Load contiguous active 64-bit integers from "a" (those with their respective bit set in mask "k"), and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Compare + + + + + Compare packed signed 32-bit integers in "a" and "b" for less-than-or-equal, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). -m := 0 -FOR j := 0 to 7 - i := j*64 - IF k[j] - dst[i+63:i] := a[m+63:m] - m := m + 64 - ELSE - dst[i+63:i] := src[i+63:i] +FOR j := 0 to 3 + i := j*32 + IF k1[j] + k[j] := ( a[i+31:i] <= b[i+31:i] ) ? 1 : 0 + ELSE + k[j] := 0 FI ENDFOR -dst[MAX:512] := 0 +k[MAX:4] := 0 - +
immintrin.h
 - + Integer + Mask + AVX512VL AVX512F - Load - Swizzle - - - - Load contiguous active 64-bit integers from unaligned memory at "mem_addr" (those with their respective bit set in mask "k"), and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Compare + + + + + Compare packed signed 32-bit integers in "a" and "b" for less-than, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). -m := 0 -FOR j := 0 to 7 - i := j*64 - IF k[j] - dst[i+63:i] := MEM[mem_addr+m+63:mem_addr+m] - m := m + 64 - ELSE - dst[i+63:i] := src[i+63:i] +FOR j := 0 to 3 + i := j*32 + IF k1[j] + k[j] := ( a[i+31:i] < b[i+31:i] ) ? 1 : 0 + ELSE + k[j] := 0 FI ENDFOR -dst[MAX:512] := 0 +k[MAX:4] := 0 - +
immintrin.h
 - + Integer + Mask + AVX512VL AVX512F - Swizzle - - - Load contiguous active 64-bit integers from "a" (those with their respective bit set in mask "k"), and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Compare + + + + + Compare packed signed 32-bit integers in "a" and "b" for not-equal, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). -m := 0 -FOR j := 0 to 7 - i := j*64 - IF k[j] - dst[i+63:i] := a[m+63:m] - m := m + 64 - ELSE - dst[i+63:i] := 0 +FOR j := 0 to 3 + i := j*32 + IF k1[j] + k[j] := ( a[i+31:i] != b[i+31:i] ) ? 1 : 0 + ELSE + k[j] := 0 FI ENDFOR -dst[MAX:512] := 0 +k[MAX:4] := 0 - +
immintrin.h
 - + Integer + Mask + AVX512VL AVX512F - Load - Swizzle - - - Load contiguous active 64-bit integers from unaligned memory at "mem_addr" (those with their respective bit set in mask "k"), and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - -m := 0 -FOR j := 0 to 7 + Compare + + + + + Compare packed signed 64-bit integers in "a" and "b" based on the comparison operand specified by "imm8", and store the results in mask vector "k". + CASE (imm8[2:0]) OF +0: OP := _MM_CMPINT_EQ +1: OP := _MM_CMPINT_LT +2: OP := _MM_CMPINT_LE +3: OP := _MM_CMPINT_FALSE +4: OP := _MM_CMPINT_NE +5: OP := _MM_CMPINT_NLT +6: OP := _MM_CMPINT_NLE +7: OP := _MM_CMPINT_TRUE +ESAC +FOR j := 0 to 3 i := j*64 - IF k[j] - dst[i+63:i] := MEM[mem_addr+m+63:mem_addr+m] - m := m + 64 - ELSE - dst[i+63:i] := 0 - FI + k[j] := ( a[i+63:i] OP b[i+63:i] ) ? 1 : 0 ENDFOR -dst[MAX:512] := 0 +k[MAX:4] := 0 - +
immintrin.h
 - + Integer - AVX512F/KNCNI - Load - - - - - Gather 32-bit integers from memory using 32-bit indices. 32-bit elements are loaded from addresses starting at "base_addr" and offset by each 32-bit element in "vindex" (each index is scaled by the factor in "scale"). Gathered elements are merged into "dst". "scale" should be 1, 2, 4 or 8. - + Mask + AVX512VL + AVX512F + Compare + + + + Compare packed signed 64-bit integers in "a" and "b" for equality, and store the results in mask vector "k". -FOR j := 0 to 15 - i := j*32 - dst[i+31:i] := MEM[base_addr + SignExtend(vindex[i+31:i])*scale] +FOR j := 0 to 3 + i := j*64 + k[j] := ( a[i+63:i] == b[i+63:i] ) ? 1 : 0 ENDFOR -dst[MAX:512] := 0 +k[MAX:4] := 0 - +
immintrin.h
 - + Integer - AVX512F/KNCNI - Load - - - - - - - Gather 32-bit integers from memory using 32-bit indices. 32-bit elements are loaded from addresses starting at "base_addr" and offset by each 32-bit element in "vindex" (each index is scaled by the factor in "scale"). Gathered elements are merged into "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). "scale" should be 1, 2, 4 or 8. - + Mask + AVX512VL + AVX512F + Compare + + + + Compare packed signed 64-bit integers in "a" and "b" for greater-than-or-equal, and store the results in mask vector "k". -FOR j := 0 to 15 - i := j*32 - IF k[j] - dst[i+31:i] := MEM[base_addr + SignExtend(vindex[i+31:i])*scale] - k[j] := 0 - ELSE - dst[i+31:i] := src[i+31:i] - FI +FOR j := 0 to 3 + i := j*64 + k[j] := ( a[i+63:i] >= b[i+63:i] ) ? 1 : 0 ENDFOR -k[MAX:16] := 0 -dst[MAX:512] := 0 +k[MAX:4] := 0 - +
immintrin.h
 - + Integer + Mask + AVX512VL AVX512F - Load - - - - Gather 64-bit integers from memory using 32-bit indices. 64-bit elements are loaded from addresses starting at "base_addr" and offset by each 32-bit element in "vindex" (each index is scaled by the factor in "scale"). Gathered elements are merged into "dst". "scale" should be 1, 2, 4 or 8. - + Compare + + + + Compare packed signed 64-bit integers in "a" and "b" for greater-than, and store the results in mask vector "k". -FOR j := 0 to 7 +FOR j := 0 to 3 i := j*64 - m := j*32 - dst[i+63:i] := MEM[base_addr + SignExtend(vindex[m+31:m])*scale] + k[j] := ( a[i+63:i] > b[i+63:i] ) ? 1 : 0 ENDFOR -dst[MAX:512] := 0 +k[MAX:4] := 0 - +
immintrin.h
 - + Integer + Mask + AVX512VL AVX512F - Load - - - - - - Gather 64-bit integers from memory using 32-bit indices. 64-bit elements are loaded from addresses starting at "base_addr" and offset by each 32-bit element in "vindex" (each index is scaled by the factor in "scale"). Gathered elements are merged into "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). "scale" should be 1, 2, 4 or 8. - + Compare + + + + Compare packed signed 64-bit integers in "a" and "b" for less-than-or-equal, and store the results in mask vector "k". -FOR j := 0 to 7 +FOR j := 0 to 3 i := j*64 - m := j*32 - IF k[j] - dst[i+63:i] := MEM[base_addr + SignExtend(vindex[m+31:m])*scale] - k[j] := 0 - ELSE - dst[i+63:i] := src[i+63:i] - FI + k[j] := ( a[i+63:i] <= b[i+63:i] ) ? 1 : 0 ENDFOR -k[MAX:8] := 0 -dst[MAX:512] := 0 +k[MAX:4] := 0 - +
immintrin.h
 - + Integer + Mask + AVX512VL AVX512F - Load - - - - Gather 32-bit integers from memory using 64-bit indices. 32-bit elements are loaded from addresses starting at "base_addr" and offset by each 64-bit element in "vindex" (each index is scaled by the factor in "scale"). Gathered elements are merged into "dst". "scale" should be 1, 2, 4 or 8. - + Compare + + + + Compare packed signed 64-bit integers in "a" and "b" for less-than, and store the results in mask vector "k". -FOR j := 0 to 7 - i := j*32 - m := j*64 - dst[i+31:i] := MEM[base_addr + SignExtend(vindex[i+63:i])*scale] +FOR j := 0 to 3 + i := j*64 + k[j] := ( a[i+63:i] < b[i+63:i] ) ? 1 : 0 ENDFOR -dst[MAX:256] := 0 +k[MAX:4] := 0 - +
immintrin.h
 - + Integer + Mask + AVX512VL AVX512F - Load - - - - - - Gather 32-bit integers from memory using 64-bit indices. 32-bit elements are loaded from addresses starting at "base_addr" and offset by each 64-bit element in "vindex" (each index is scaled by the factor in "scale"). Gathered elements are merged into "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). "scale" should be 1, 2, 4 or 8. - + Compare + + + + Compare packed signed 64-bit integers in "a" and "b" for not-equal, and store the results in mask vector "k". -FOR j := 0 to 7 - i := j*32 - m := j*64 - IF k[j] - dst[i+31:i] := MEM[base_addr + SignExtend(vindex[i+63:i])*scale] +FOR j := 0 to 3 + i := j*64 + k[j] := ( a[i+63:i] != b[i+63:i] ) ? 1 : 0 +ENDFOR +k[MAX:4] := 0 + + +
immintrin.h
+ + + Integer + Mask + AVX512VL + AVX512F + Compare + + + + + + Compare packed signed 64-bit integers in "a" and "b" based on the comparison operand specified by "imm8", and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). + CASE (imm8[2:0]) OF +0: OP := _MM_CMPINT_EQ +1: OP := _MM_CMPINT_LT +2: OP := _MM_CMPINT_LE +3: OP := _MM_CMPINT_FALSE +4: OP := _MM_CMPINT_NE +5: OP := _MM_CMPINT_NLT +6: OP := _MM_CMPINT_NLE +7: OP := _MM_CMPINT_TRUE +ESAC +FOR j := 0 to 3 + i := j*64 + IF k1[j] + k[j] := ( a[i+63:i] OP b[i+63:i] ) ? 1 : 0 + ELSE k[j] := 0 - ELSE - dst[i+31:i] := src[i+31:i] FI ENDFOR -k[MAX:8] := 0 -dst[MAX:256] := 0 +k[MAX:4] := 0 - +
immintrin.h
 - + Integer + Mask + AVX512VL AVX512F - Load - - - - Gather 64-bit integers from memory using 64-bit indices. 64-bit elements are loaded from addresses starting at "base_addr" and offset by each 64-bit element in "vindex" (each index is scaled by the factor in "scale"). Gathered elements are merged into "dst". "scale" should be 1, 2, 4 or 8. - + Compare + + + + + Compare packed signed 64-bit integers in "a" and "b" for equality, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 +FOR j := 0 to 3 i := j*64 - dst[i+63:i] := MEM[base_addr + SignExtend(vindex[i+63:i])*scale] + IF k1[j] + k[j] := ( a[i+63:i] == b[i+63:i] ) ? 1 : 0 + ELSE + k[j] := 0 + FI ENDFOR -dst[MAX:512] := 0 +k[MAX:4] := 0 - +
immintrin.h
 - + Integer + Mask + AVX512VL AVX512F - Load - - - - - - Gather 64-bit integers from memory using 64-bit indices. 64-bit elements are loaded from addresses starting at "base_addr" and offset by each 64-bit element in "vindex" (each index is scaled by the factor in "scale"). Gathered elements are merged into "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). "scale" should be 1, 2, 4 or 8. - + Compare + + + + + Compare packed signed 64-bit integers in "a" and "b" for greater-than-or-equal, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 +FOR j := 0 to 3 i := j*64 - IF k[j] - dst[i+63:i] := MEM[base_addr + SignExtend(vindex[i+63:i])*scale] + IF k1[j] + k[j] := ( a[i+63:i] >= b[i+63:i] ) ? 1 : 0 + ELSE k[j] := 0 - ELSE - dst[i+63:i] := src[i+63:i] FI ENDFOR -k[MAX:8] := 0 -dst[MAX:512] := 0 +k[MAX:4] := 0 - +
immintrin.h
 - + Integer - AVX512CD - Bit Manipulation - - Counts the number of leading zero bits in each packed 32-bit integer in "a", and store the results in "dst". + Mask + AVX512VL + AVX512F + Compare + + + + + Compare packed signed 64-bit integers in "a" and "b" for greater-than, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*32 - tmp := 31 - dst[i+31:i] := 0 - DO WHILE (tmp >= 0 AND a[i+tmp] == 0) - tmp := tmp - 1 - dst[i+31:i] := dst[i+31:i] + 1 - OD +FOR j := 0 to 3 + i := j*64 + IF k1[j] + k[j] := ( a[i+63:i] > b[i+63:i] ) ? 1 : 0 + ELSE + k[j] := 0 + FI ENDFOR -dst[MAX:512] := 0 +k[MAX:4] := 0 - +
immintrin.h
 - + Integer - AVX512CD - Bit Manipulation - - - - Counts the number of leading zero bits in each packed 32-bit integer in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Mask + AVX512VL + AVX512F + Compare + + + + + Compare packed signed 64-bit integers in "a" and "b" for less-than-or-equal, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*32 - IF k[j] - tmp := 31 - dst[i+31:i] := 0 - DO WHILE (tmp >= 0 AND a[i+tmp] == 0) - tmp := tmp - 1 - dst[i+31:i] := dst[i+31:i] + 1 - OD - ELSE - dst[i+31:i] := src[i+31:i] +FOR j := 0 to 3 + i := j*64 + IF k1[j] + k[j] := ( a[i+63:i] <= b[i+63:i] ) ? 1 : 0 + ELSE + k[j] := 0 FI ENDFOR -dst[MAX:512] := 0 +k[MAX:4] := 0 - +
immintrin.h
 - + Integer - AVX512CD - Bit Manipulation - - - Counts the number of leading zero bits in each packed 32-bit integer in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Mask + AVX512VL + AVX512F + Compare + + + + + Compare packed signed 64-bit integers in "a" and "b" for less-than, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*32 - IF k[j] - tmp := 31 - dst[i+31:i] := 0 - DO WHILE (tmp >= 0 AND a[i+tmp] == 0) - tmp := tmp - 1 - dst[i+31:i] := dst[i+31:i] + 1 - OD - ELSE - dst[i+31:i] := 0 +FOR j := 0 to 3 + i := j*64 + IF k1[j] + k[j] := ( a[i+63:i] < b[i+63:i] ) ? 1 : 0 + ELSE + k[j] := 0 FI ENDFOR -dst[MAX:512] := 0 +k[MAX:4] := 0 - +
immintrin.h
 - + Integer - AVX512CD - Bit Manipulation - - Counts the number of leading zero bits in each packed 64-bit integer in "a", and store the results in "dst". + Mask + AVX512VL + AVX512F + Compare + + + + + Compare packed signed 64-bit integers in "a" and "b" for not-equal, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 +FOR j := 0 to 3 i := j*64 - tmp := 63 - dst[i+63:i] := 0 - DO WHILE (tmp >= 0 AND a[i+tmp] == 0) - tmp := tmp - 1 - dst[i+63:i] := dst[i+63:i] + 1 - OD + IF k1[j] + k[j] := ( a[i+63:i] != b[i+63:i] ) ? 1 : 0 + ELSE + k[j] := 0 + FI ENDFOR -dst[MAX:512] := 0 +k[MAX:4] := 0 - +
immintrin.h
 - + Integer - AVX512CD - Bit Manipulation - - - - Counts the number of leading zero bits in each packed 64-bit integer in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -FOR j := 0 to 7 + Mask + AVX512VL + AVX512F + Compare + + + + + Compare packed signed 64-bit integers in "a" and "b" based on the comparison operand specified by "imm8", and store the results in mask vector "k". + CASE (imm8[2:0]) OF +0: OP := _MM_CMPINT_EQ +1: OP := _MM_CMPINT_LT +2: OP := _MM_CMPINT_LE +3: OP := _MM_CMPINT_FALSE +4: OP := _MM_CMPINT_NE +5: OP := _MM_CMPINT_NLT +6: OP := _MM_CMPINT_NLE +7: OP := _MM_CMPINT_TRUE +ESAC +FOR j := 0 to 1 i := j*64 - IF k[j] - tmp := 63 - dst[i+63:i] := 0 - DO WHILE (tmp >= 0 AND a[i+tmp] == 0) - tmp := tmp - 1 - dst[i+63:i] := dst[i+63:i] + 1 - OD - ELSE - dst[i+63:i] := src[i+63:i] - FI + k[j] := ( a[i+63:i] OP b[i+63:i] ) ? 1 : 0 ENDFOR -dst[MAX:512] := 0 +k[MAX:2] := 0 - +
immintrin.h
 - + Integer - AVX512CD - Bit Manipulation - - - Counts the number of leading zero bits in each packed 64-bit integer in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Mask + AVX512VL + AVX512F + Compare + + + + Compare packed signed 64-bit integers in "a" and "b" for equality, and store the results in mask vector "k". -FOR j := 0 to 7 +FOR j := 0 to 1 i := j*64 - IF k[j] - tmp := 63 - dst[i+63:i] := 0 - DO WHILE (tmp >= 0 AND a[i+tmp] == 0) - tmp := tmp - 1 - dst[i+63:i] := dst[i+63:i] + 1 - OD - ELSE - dst[i+63:i] := 0 - FI + k[j] := ( a[i+63:i] == b[i+63:i] ) ? 1 : 0 ENDFOR -dst[MAX:512] := 0 +k[MAX:2] := 0 - +
immintrin.h
 - + Integer - AVX512F/KNCNI - Special Math Functions - - - - - Compare packed 32-bit integers in "a" and "b", and store packed maximum values in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Mask + AVX512VL + AVX512F + Compare + + + + Compare packed signed 64-bit integers in "a" and "b" for greater-than-or-equal, and store the results in mask vector "k". -FOR j := 0 to 15 - i := j*32 - IF k[j] - IF a[i+31:i] > b[i+31:i] - dst[i+31:i] := a[i+31:i] - ELSE - dst[i+31:i] := b[i+31:i] - FI - ELSE - dst[i+31:i] := src[i+31:i] - FI +FOR j := 0 to 1 + i := j*64 + k[j] := ( a[i+63:i] >= b[i+63:i] ) ? 1 : 0 ENDFOR -dst[MAX:512] := 0 +k[MAX:2] := 0 - +
immintrin.h
 - + Integer + Mask + AVX512VL AVX512F - Special Math Functions - - - - Compare packed 32-bit integers in "a" and "b", and store packed maximum values in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Compare + + + + Compare packed signed 64-bit integers in "a" and "b" for greater-than, and store the results in mask vector "k". -FOR j := 0 to 15 - i := j*32 - IF k[j] - IF a[i+31:i] > b[i+31:i] - dst[i+31:i] := a[i+31:i] - ELSE - dst[i+31:i] := b[i+31:i] - FI - ELSE - dst[i+31:i] := 0 - FI +FOR j := 0 to 1 + i := j*64 + k[j] := ( a[i+63:i] > b[i+63:i] ) ? 1 : 0 ENDFOR -dst[MAX:512] := 0 +k[MAX:2] := 0 - +
immintrin.h
 - + Integer - AVX512F/KNCNI - Special Math Functions - - - Compare packed 32-bit integers in "a" and "b", and store packed maximum values in "dst". + Mask + AVX512VL + AVX512F + Compare + + + + Compare packed signed 64-bit integers in "a" and "b" for less-than-or-equal, and store the results in mask vector "k". -FOR j := 0 to 15 - i := j*32 - IF a[i+31:i] > b[i+31:i] - dst[i+31:i] := a[i+31:i] - ELSE - dst[i+31:i] := b[i+31:i] - FI +FOR j := 0 to 1 + i := j*64 + k[j] := ( a[i+63:i] <= b[i+63:i] ) ? 1 : 0 ENDFOR -dst[MAX:512] := 0 +k[MAX:2] := 0 - +
immintrin.h
 - + Integer + Mask + AVX512VL AVX512F - Special Math Functions - - - - - Compare packed 64-bit integers in "a" and "b", and store packed maximum values in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Compare + + + + Compare packed signed 64-bit integers in "a" and "b" for less-than, and store the results in mask vector "k". -FOR j := 0 to 7 +FOR j := 0 to 1 i := j*64 - IF k[j] - IF a[i+63:i] > b[i+63:i] - dst[i+63:i] := a[i+63:i] - ELSE - dst[i+63:i] := b[i+63:i] - FI - ELSE - dst[i+63:i] := src[i+63:i] - FI + k[j] := ( a[i+63:i] < b[i+63:i] ) ? 1 : 0 ENDFOR -dst[MAX:512] := 0 +k[MAX:2] := 0 - +
immintrin.h
 - + Integer + Mask + AVX512VL AVX512F - Special Math Functions - - - - Compare packed 64-bit integers in "a" and "b", and store packed maximum values in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Compare + + + + Compare packed signed 64-bit integers in "a" and "b" for not-equal, and store the results in mask vector "k". -FOR j := 0 to 7 +FOR j := 0 to 1 i := j*64 - IF k[j] - IF a[i+63:i] > b[i+63:i] - dst[i+63:i] := a[i+63:i] - ELSE - dst[i+63:i] := b[i+63:i] - FI - ELSE - dst[i+63:i] := 0 - FI + k[j] := ( a[i+63:i] != b[i+63:i] ) ? 1 : 0 ENDFOR -dst[MAX:512] := 0 +k[MAX:2] := 0 - +
immintrin.h
 - + Integer + Mask + AVX512VL AVX512F - Special Math Functions - - - Compare packed 64-bit integers in "a" and "b", and store packed maximum values in "dst". - -FOR j := 0 to 7 + Compare + + + + + + Compare packed signed 64-bit integers in "a" and "b" based on the comparison operand specified by "imm8", and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). + CASE (imm8[2:0]) OF +0: OP := _MM_CMPINT_EQ +1: OP := _MM_CMPINT_LT +2: OP := _MM_CMPINT_LE +3: OP := _MM_CMPINT_FALSE +4: OP := _MM_CMPINT_NE +5: OP := _MM_CMPINT_NLT +6: OP := _MM_CMPINT_NLE +7: OP := _MM_CMPINT_TRUE +ESAC +FOR j := 0 to 1 i := j*64 - IF a[i+63:i] > b[i+63:i] - dst[i+63:i] := a[i+63:i] - ELSE - dst[i+63:i] := b[i+63:i] + IF k1[j] + k[j] := ( a[i+63:i] OP b[i+63:i] ) ? 1 : 0 + ELSE + k[j] := 0 FI ENDFOR -dst[MAX:512] := 0 +k[MAX:2] := 0 - +
immintrin.h
 - + Integer - AVX512F/KNCNI - Special Math Functions - - - - - Compare packed unsigned 32-bit integers in "a" and "b", and store packed maximum values in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Mask + AVX512VL + AVX512F + Compare + + + + + Compare packed signed 64-bit integers in "a" and "b" for equality, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*32 - IF k[j] - IF a[i+31:i] > b[i+31:i] - dst[i+31:i] := a[i+31:i] - ELSE - dst[i+31:i] := b[i+31:i] - FI - ELSE - dst[i+31:i] := src[i+31:i] +FOR j := 0 to 1 + i := j*64 + IF k1[j] + k[j] := ( a[i+63:i] == b[i+63:i] ) ? 1 : 0 + ELSE + k[j] := 0 FI ENDFOR -dst[MAX:512] := 0 +k[MAX:2] := 0 - +
immintrin.h
 - + Integer + Mask + AVX512VL AVX512F - Special Math Functions - - - - Compare packed unsigned 32-bit integers in "a" and "b", and store packed maximum values in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Compare + + + + + Compare packed signed 64-bit integers in "a" and "b" for greater-than-or-equal, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*32 - IF k[j] - IF a[i+31:i] > b[i+31:i] - dst[i+31:i] := a[i+31:i] - ELSE - dst[i+31:i] := b[i+31:i] - FI - ELSE - dst[i+31:i] := 0 +FOR j := 0 to 1 + i := j*64 + IF k1[j] + k[j] := ( a[i+63:i] >= b[i+63:i] ) ? 1 : 0 + ELSE + k[j] := 0 FI ENDFOR -dst[MAX:512] := 0 +k[MAX:2] := 0 - +
immintrin.h
 - + Integer - AVX512F/KNCNI - Special Math Functions - - - Compare packed unsigned 32-bit integers in "a" and "b", and store packed maximum values in "dst". + Mask + AVX512VL + AVX512F + Compare + + + + + Compare packed signed 64-bit integers in "a" and "b" for greater-than, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*32 - IF a[i+31:i] > b[i+31:i] - dst[i+31:i] := a[i+31:i] - ELSE - dst[i+31:i] := b[i+31:i] +FOR j := 0 to 1 + i := j*64 + IF k1[j] + k[j] := ( a[i+63:i] > b[i+63:i] ) ? 1 : 0 + ELSE + k[j] := 0 FI ENDFOR -dst[MAX:512] := 0 +k[MAX:2] := 0 - +
immintrin.h
 - + Integer + Mask + AVX512VL AVX512F - Special Math Functions - - - - - Compare packed unsigned 64-bit integers in "a" and "b", and store packed maximum values in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Compare + + + + + Compare packed signed 64-bit integers in "a" and "b" for less-than-or-equal, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 +FOR j := 0 to 1 i := j*64 - IF k[j] - IF a[i+63:i] > b[i+63:i] - dst[i+63:i] := a[i+63:i] - ELSE - dst[i+63:i] := b[i+63:i] - FI - ELSE - dst[i+63:i] := src[i+63:i] + IF k1[j] + k[j] := ( a[i+63:i] <= b[i+63:i] ) ? 1 : 0 + ELSE + k[j] := 0 FI ENDFOR -dst[MAX:512] := 0 +k[MAX:2] := 0 - +
immintrin.h
 - + Integer + Mask + AVX512VL AVX512F - Special Math Functions - - - - Compare packed unsigned 64-bit integers in "a" and "b", and store packed maximum values in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Compare + + + + + Compare packed signed 64-bit integers in "a" and "b" for less-than, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 +FOR j := 0 to 1 i := j*64 - IF k[j] - IF a[i+63:i] > b[i+63:i] - dst[i+63:i] := a[i+63:i] - ELSE - dst[i+63:i] := b[i+63:i] - FI - ELSE - dst[i+63:i] := 0 + IF k1[j] + k[j] := ( a[i+63:i] < b[i+63:i] ) ? 1 : 0 + ELSE + k[j] := 0 FI ENDFOR -dst[MAX:512] := 0 +k[MAX:2] := 0 - +
immintrin.h
 - + Integer + Mask + AVX512VL AVX512F - Special Math Functions - - - Compare packed unsigned 64-bit integers in "a" and "b", and store packed maximum values in "dst". + Compare + + + + + Compare packed signed 64-bit integers in "a" and "b" for not-equal, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 +FOR j := 0 to 1 i := j*64 - IF a[i+63:i] > b[i+63:i] - dst[i+63:i] := a[i+63:i] - ELSE - dst[i+63:i] := b[i+63:i] + IF k1[j] + k[j] := ( a[i+63:i] != b[i+63:i] ) ? 1 : 0 + ELSE + k[j] := 0 FI ENDFOR -dst[MAX:512] := 0 +k[MAX:2] := 0 - +
immintrin.h
 - + Integer - AVX512F/KNCNI - Special Math Functions - - - - - Compare packed 32-bit integers in "a" and "b", and store packed minimum values in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -FOR j := 0 to 15 + Mask + AVX512VL + AVX512F + Compare + + + + + Compare packed unsigned 32-bit integers in "a" and "b" based on the comparison operand specified by "imm8", and store the results in mask vector "k". + CASE (imm8[2:0]) OF +0: OP := _MM_CMPINT_EQ +1: OP := _MM_CMPINT_LT +2: OP := _MM_CMPINT_LE +3: OP := _MM_CMPINT_FALSE +4: OP := _MM_CMPINT_NE +5: OP := _MM_CMPINT_NLT +6: OP := _MM_CMPINT_NLE +7: OP := _MM_CMPINT_TRUE +ESAC +FOR j := 0 to 7 i := j*32 - IF k[j] - IF a[i+31:i] < b[i+31:i] - dst[i+31:i] := a[i+31:i] - ELSE - dst[i+31:i] := b[i+31:i] - FI - ELSE - dst[i+31:i] := src[i+31:i] - FI + k[j] := ( a[i+31:i] OP b[i+31:i] ) ? 1 : 0 ENDFOR -dst[MAX:512] := 0 +k[MAX:8] := 0 - +
immintrin.h
 - + Integer + Mask + AVX512VL AVX512F - Special Math Functions - - - - Compare packed 32-bit integers in "a" and "b", and store packed minimum values in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Compare + + + + Compare packed unsigned 32-bit integers in "a" and "b" for equality, and store the results in mask vector "k". -FOR j := 0 to 15 +FOR j := 0 to 7 i := j*32 - IF k[j] - IF a[i+31:i] < b[i+31:i] - dst[i+31:i] := a[i+31:i] - ELSE - dst[i+31:i] := b[i+31:i] - FI - ELSE - dst[i+31:i] := 0 - FI + k[j] := ( a[i+31:i] == b[i+31:i] ) ? 1 : 0 ENDFOR -dst[MAX:512] := 0 +k[MAX:8] := 0 - +
immintrin.h
 - + Integer - AVX512F/KNCNI - Special Math Functions - - - Compare packed 32-bit integers in "a" and "b", and store packed minimum values in "dst". - + Mask + AVX512VL + AVX512F + Compare + + + + Compare packed unsigned 32-bit integers in "a" and "b" for greater-than-or-equal, and store the results in mask vector "k". -FOR j := 0 to 15 +FOR j := 0 to 7 i := j*32 - IF a[i+31:i] < b[i+31:i] - dst[i+31:i] := a[i+31:i] - ELSE - dst[i+31:i] := b[i+31:i] - FI + k[j] := ( a[i+31:i] >= b[i+31:i] ) ? 1 : 0 ENDFOR -dst[MAX:512] := 0 +k[MAX:8] := 0 - +
immintrin.h
 - + Integer + Mask + AVX512VL AVX512F - Special Math Functions - - - - - Compare packed 64-bit integers in "a" and "b", and store packed minimum values in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Compare + + + + Compare packed unsigned 32-bit integers in "a" and "b" for greater-than, and store the results in mask vector "k". FOR j := 0 to 7 - i := j*64 - IF k[j] - IF a[i+63:i] < b[i+63:i] - dst[i+63:i] := a[i+63:i] - ELSE - dst[i+63:i] := b[i+63:i] - FI - ELSE - dst[i+63:i] := src[i+63:i] - FI + i := j*32 + k[j] := ( a[i+31:i] > b[i+31:i] ) ? 1 : 0 ENDFOR -dst[MAX:512] := 0 +k[MAX:8] := 0 - +
immintrin.h
 - + Integer + Mask + AVX512VL AVX512F - Special Math Functions - - - - Compare packed 64-bit integers in "a" and "b", and store packed minimum values in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Compare + + + + Compare packed unsigned 32-bit integers in "a" and "b" for less-than-or-equal, and store the results in mask vector "k". FOR j := 0 to 7 - i := j*64 - IF k[j] - IF a[i+63:i] < b[i+63:i] - dst[i+63:i] := a[i+63:i] - ELSE - dst[i+63:i] := b[i+63:i] - FI - ELSE - dst[i+63:i] := 0 - FI + i := j*32 + k[j] := ( a[i+31:i] <= b[i+31:i] ) ? 1 : 0 ENDFOR -dst[MAX:512] := 0 +k[MAX:8] := 0 - +
immintrin.h
 - + Integer + Mask + AVX512VL AVX512F - Special Math Functions - - - Compare packed 64-bit integers in "a" and "b", and store packed minimum values in "dst". + Compare + + + + Compare packed unsigned 32-bit integers in "a" and "b" for less-than, and store the results in mask vector "k". FOR j := 0 to 7 - i := j*64 - IF a[i+63:i] < b[i+63:i] - dst[i+63:i] := a[i+63:i] - ELSE - dst[i+63:i] := b[i+63:i] - FI + i := j*32 + k[j] := ( a[i+31:i] < b[i+31:i] ) ? 1 : 0 ENDFOR -dst[MAX:512] := 0 +k[MAX:8] := 0 - +
immintrin.h
 - + Integer - AVX512F/KNCNI - Special Math Functions - - - - - Compare packed unsigned 32-bit integers in "a" and "b", and store packed minimum values in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Mask + AVX512VL + AVX512F + Compare + + + + Compare packed unsigned 32-bit integers in "a" and "b" for not-equal, and store the results in mask vector "k". -FOR j := 0 to 15 +FOR j := 0 to 7 i := j*32 - IF k[j] - IF a[i+31:i] < b[i+31:i] - dst[i+31:i] := a[i+31:i] - ELSE - dst[i+31:i] := b[i+31:i] - FI - ELSE - dst[i+31:i] := src[i+31:i] - FI + k[j] := ( a[i+31:i] != b[i+31:i] ) ? 1 : 0 ENDFOR -dst[MAX:512] := 0 +k[MAX:8] := 0 - +
immintrin.h
 - + Integer + Mask + AVX512VL AVX512F - Special Math Functions - - - - Compare packed unsigned 32-bit integers in "a" and "b", and store packed minimum values in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - -FOR j := 0 to 15 + Compare + + + + + + Compare packed unsigned 32-bit integers in "a" and "b" based on the comparison operand specified by "imm8", and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). + CASE (imm8[2:0]) OF +0: OP := _MM_CMPINT_EQ +1: OP := _MM_CMPINT_LT +2: OP := _MM_CMPINT_LE +3: OP := _MM_CMPINT_FALSE +4: OP := _MM_CMPINT_NE +5: OP := _MM_CMPINT_NLT +6: OP := _MM_CMPINT_NLE +7: OP := _MM_CMPINT_TRUE +ESAC +FOR j := 0 to 7 i := j*32 - IF k[j] - IF a[i+31:i] < b[i+31:i] - dst[i+31:i] := a[i+31:i] - ELSE - dst[i+31:i] := b[i+31:i] - FI - ELSE - dst[i+31:i] := 0 + IF k1[j] + k[j] := ( a[i+31:i] OP b[i+31:i] ) ? 1 : 0 + ELSE + k[j] := 0 FI ENDFOR -dst[MAX:512] := 0 +k[MAX:8] := 0 - +
immintrin.h
 - + Integer - AVX512F/KNCNI - Special Math Functions - - - Compare packed unsigned 32-bit integers in "a" and "b", and store packed minimum values in "dst". - + Mask + AVX512VL + AVX512F + Compare + + + + + Compare packed unsigned 32-bit integers in "a" and "b" for equality, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 15 +FOR j := 0 to 7 i := j*32 - IF a[i+31:i] < b[i+31:i] - dst[i+31:i] := a[i+31:i] - ELSE - dst[i+31:i] := b[i+31:i] + IF k1[j] + k[j] := ( a[i+31:i] == b[i+31:i] ) ? 1 : 0 + ELSE + k[j] := 0 FI ENDFOR -dst[MAX:512] := 0 +k[MAX:8] := 0 - +
immintrin.h
 - + Integer + Mask + AVX512VL AVX512F - Special Math Functions - - - - - Compare packed unsigned 64-bit integers in "a" and "b", and store packed minimum values in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Compare + + + + + Compare packed unsigned 32-bit integers in "a" and "b" for greater-than-or-equal, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). FOR j := 0 to 7 - i := j*64 - IF k[j] - IF a[i+63:i] < b[i+63:i] - dst[i+63:i] := a[i+63:i] - ELSE - dst[i+63:i] := b[i+63:i] - FI - ELSE - dst[i+63:i] := src[i+63:i] + i := j*32 + IF k1[j] + k[j] := ( a[i+31:i] >= b[i+31:i] ) ? 1 : 0 + ELSE + k[j] := 0 FI ENDFOR -dst[MAX:512] := 0 +k[MAX:8] := 0 - +
immintrin.h
 - + Integer + Mask + AVX512VL AVX512F - Special Math Functions - - - - Compare packed unsigned 64-bit integers in "a" and "b", and store packed minimum values in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Compare + + + + + Compare packed unsigned 32-bit integers in "a" and "b" for greater-than, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). FOR j := 0 to 7 - i := j*64 - IF k[j] - IF a[i+63:i] < b[i+63:i] - dst[i+63:i] := a[i+63:i] - ELSE - dst[i+63:i] := b[i+63:i] - FI - ELSE - dst[i+63:i] := 0 + i := j*32 + IF k1[j] + k[j] := ( a[i+31:i] > b[i+31:i] ) ? 1 : 0 + ELSE + k[j] := 0 FI ENDFOR -dst[MAX:512] := 0 +k[MAX:8] := 0 - +
immintrin.h
 - + Integer + Mask + AVX512VL AVX512F - Special Math Functions - - - Compare packed unsigned 64-bit integers in "a" and "b", and store packed minimum values in "dst". + Compare + + + + + Compare packed unsigned 32-bit integers in "a" and "b" for less-than-or-equal, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). FOR j := 0 to 7 - i := j*64 - IF a[i+63:i] < b[i+63:i] - dst[i+63:i] := a[i+63:i] - ELSE - dst[i+63:i] := b[i+63:i] + i := j*32 + IF k1[j] + k[j] := ( a[i+31:i] <= b[i+31:i] ) ? 1 : 0 + ELSE + k[j] := 0 FI ENDFOR -dst[MAX:512] := 0 +k[MAX:8] := 0 - +
immintrin.h
 - + Integer + Mask + AVX512VL AVX512F - Convert - - Convert packed 32-bit integers in "a" to packed 8-bit integers with truncation, and store the results in "dst". + Compare + + + + + Compare packed unsigned 32-bit integers in "a" and "b" for less-than, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := 32*j - k := 8*j - dst[k+7:k] := Truncate_Int32_To_Int8(a[i+31:i]) +FOR j := 0 to 7 + i := j*32 + IF k1[j] + k[j] := ( a[i+31:i] < b[i+31:i] ) ? 1 : 0 + ELSE + k[j] := 0 + FI ENDFOR -dst[MAX:128] := 0 +k[MAX:8] := 0 - +
immintrin.h
 - + Integer + Mask + AVX512VL AVX512F - Convert - - - - Convert packed 32-bit integers in "a" to packed 8-bit integers with truncation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Compare + + + + + Compare packed unsigned 32-bit integers in "a" and "b" for not-equal, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := 32*j - l := 8*j - IF k[j] - dst[l+7:l] := Truncate_Int32_To_Int8(a[i+31:i]) - ELSE - dst[l+7:l] := src[l+7:l] +FOR j := 0 to 7 + i := j*32 + IF k1[j] + k[j] := ( a[i+31:i] != b[i+31:i] ) ? 1 : 0 + ELSE + k[j] := 0 FI ENDFOR -dst[MAX:128] := 0 +k[MAX:8] := 0 - +
immintrin.h
 - + Integer + Mask + AVX512VL AVX512F - Convert - Store - - - - Convert packed 32-bit integers in "a" to packed 8-bit integers with truncation, and store the active results (those with their respective bit set in writemask "k") to unaligned memory at "base_addr". - -FOR j := 0 to 15 - i := 32*j - l := 8*j - IF k[j] - MEM[base_addr+l+7:base_addr+l] := Truncate_Int32_To_Int8(a[i+31:i]) - FI + Compare + + + + + Compare packed unsigned 32-bit integers in "a" and "b" based on the comparison operand specified by "imm8", and store the results in mask vector "k". + CASE (imm8[2:0]) OF +0: OP := _MM_CMPINT_EQ +1: OP := _MM_CMPINT_LT +2: OP := _MM_CMPINT_LE +3: OP := _MM_CMPINT_FALSE +4: OP := _MM_CMPINT_NE +5: OP := _MM_CMPINT_NLT +6: OP := _MM_CMPINT_NLE +7: OP := _MM_CMPINT_TRUE +ESAC +FOR j := 0 to 3 + i := j*32 + k[j] := ( a[i+31:i] OP b[i+31:i] ) ? 1 : 0 ENDFOR +k[MAX:4] := 0 - +
immintrin.h
 - + Integer + Mask + AVX512VL AVX512F - Convert - - - Convert packed 32-bit integers in "a" to packed 8-bit integers with truncation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Compare + + + + Compare packed unsigned 32-bit integers in "a" and "b" for equality, and store the results in mask vector "k". -FOR j := 0 to 15 - i := 32*j - l := 8*j - IF k[j] - dst[l+7:l] := Truncate_Int32_To_Int8(a[i+31:i]) - ELSE - dst[l+7:l] := 0 - FI +FOR j := 0 to 3 + i := j*32 + k[j] := ( a[i+31:i] == b[i+31:i] ) ? 1 : 0 ENDFOR -dst[MAX:128] := 0 +k[MAX:4] := 0 - +
immintrin.h
 - + Integer + Mask + AVX512VL AVX512F - Convert - - Convert packed 32-bit integers in "a" to packed 16-bit integers with truncation, and store the results in "dst". + Compare + + + + Compare packed unsigned 32-bit integers in "a" and "b" for greater-than-or-equal, and store the results in mask vector "k". -FOR j := 0 to 15 - i := 32*j - k := 16*j - dst[k+15:k] := Truncate_Int32_To_Int16(a[i+31:i]) +FOR j := 0 to 3 + i := j*32 + k[j] := ( a[i+31:i] >= b[i+31:i] ) ? 1 : 0 ENDFOR -dst[MAX:256] := 0 +k[MAX:4] := 0 - +
immintrin.h
 - + Integer + Mask + AVX512VL AVX512F - Convert - - - - Convert packed 32-bit integers in "a" to packed 16-bit integers with truncation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Compare + + + + Compare packed unsigned 32-bit integers in "a" and "b" for greater-than, and store the results in mask vector "k". -FOR j := 0 to 15 - i := 32*j - l := 16*j - IF k[j] - dst[l+15:l] := Truncate_Int32_To_Int16(a[i+31:i]) - ELSE - dst[l+15:l] := src[l+15:l] - FI +FOR j := 0 to 3 + i := j*32 + k[j] := ( a[i+31:i] > b[i+31:i] ) ? 1 : 0 ENDFOR -dst[MAX:256] := 0 +k[MAX:4] := 0 - +
immintrin.h
 - + Integer + Mask + AVX512VL AVX512F - Convert - Store - - - - Convert packed 32-bit integers in "a" to packed 16-bit integers with truncation, and store the active results (those with their respective bit set in writemask "k") to unaligned memory at "base_addr". + Compare + + + + Compare packed unsigned 32-bit integers in "a" and "b" for less-than-or-equal, and store the results in mask vector "k". -FOR j := 0 to 15 - i := 32*j - l := 16*j - IF k[j] - MEM[base_addr+l+15:base_addr+l] := Truncate_Int32_To_Int16(a[i+31:i]) - FI +FOR j := 0 to 3 + i := j*32 + k[j] := ( a[i+31:i] <= b[i+31:i] ) ? 1 : 0 ENDFOR +k[MAX:4] := 0 - +
immintrin.h
 - + Integer + Mask + AVX512VL AVX512F - Convert - - - Convert packed 32-bit integers in "a" to packed 16-bit integers with truncation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Compare + + + + Compare packed unsigned 32-bit integers in "a" and "b" for less-than, and store the results in mask vector "k". -FOR j := 0 to 15 - i := 32*j - l := 16*j - IF k[j] - dst[l+15:l] := Truncate_Int32_To_Int16(a[i+31:i]) - ELSE - dst[l+15:l] := 0 - FI +FOR j := 0 to 3 + i := j*32 + k[j] := ( a[i+31:i] < b[i+31:i] ) ? 1 : 0 ENDFOR -dst[MAX:256] := 0 +k[MAX:4] := 0 - +
immintrin.h
 - + Integer + Mask + AVX512VL AVX512F - Convert - - Convert packed 64-bit integers in "a" to packed 8-bit integers with truncation, and store the results in "dst". + Compare + + + + Compare packed unsigned 32-bit integers in "a" and "b" for not-equal, and store the results in mask vector "k". -FOR j := 0 to 7 - i := 64*j - k := 8*j - dst[k+7:k] := Truncate_Int64_To_Int8(a[i+63:i]) +FOR j := 0 to 3 + i := j*32 + k[j] := ( a[i+31:i] != b[i+31:i] ) ? 1 : 0 ENDFOR -dst[MAX:128] := 0 +k[MAX:4] := 0 - +
immintrin.h
 - + Integer + Mask + AVX512VL AVX512F - Convert - - - - Convert packed 64-bit integers in "a" to packed 8-bit integers with truncation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -FOR j := 0 to 7 - i := 64*j - l := 8*j - IF k[j] - dst[l+7:l] := Truncate_Int64_To_Int8(a[i+63:i]) - ELSE - dst[l+7:l] := src[l+7:l] + Compare + + + + + + Compare packed unsigned 32-bit integers in "a" and "b" based on the comparison operand specified by "imm8", and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). + CASE (imm8[2:0]) OF +0: OP := _MM_CMPINT_EQ +1: OP := _MM_CMPINT_LT +2: OP := _MM_CMPINT_LE +3: OP := _MM_CMPINT_FALSE +4: OP := _MM_CMPINT_NE +5: OP := _MM_CMPINT_NLT +6: OP := _MM_CMPINT_NLE +7: OP := _MM_CMPINT_TRUE +ESAC +FOR j := 0 to 3 + i := j*32 + IF k1[j] + k[j] := ( a[i+31:i] OP b[i+31:i] ) ? 1 : 0 + ELSE + k[j] := 0 FI ENDFOR -dst[MAX:128] := 0 +k[MAX:4] := 0 - +
immintrin.h
 - + Integer + Mask + AVX512VL AVX512F - Convert - Store - - - - Convert packed 64-bit integers in "a" to packed 8-bit integers with truncation, and store the active results (those with their respective bit set in writemask "k") to unaligned memory at "base_addr". + Compare + + + + + Compare packed unsigned 32-bit integers in "a" and "b" for equality, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := 64*j - l := 8*j - IF k[j] - MEM[base_addr+l+7:base_addr+l] := Truncate_Int64_To_Int8(a[i+63:i]) +FOR j := 0 to 3 + i := j*32 + IF k1[j] + k[j] := ( a[i+31:i] == b[i+31:i] ) ? 1 : 0 + ELSE + k[j] := 0 FI ENDFOR +k[MAX:4] := 0 - +
immintrin.h
 - + Integer + Mask + AVX512VL AVX512F - Convert - - - Convert packed 64-bit integers in "a" to packed 8-bit integers with truncation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Compare + + + + + Compare packed unsigned 32-bit integers in "a" and "b" for greater-than-or-equal, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := 64*j - l := 8*j - IF k[j] - dst[l+7:l] := Truncate_Int64_To_Int8(a[i+63:i]) - ELSE - dst[l+7:l] := 0 +FOR j := 0 to 3 + i := j*32 + IF k1[j] + k[j] := ( a[i+31:i] >= b[i+31:i] ) ? 1 : 0 + ELSE + k[j] := 0 FI ENDFOR -dst[MAX:128] := 0 +k[MAX:4] := 0 - +
immintrin.h
 - + Integer + Mask + AVX512VL AVX512F - Convert - - Convert packed 64-bit integers in "a" to packed 32-bit integers with truncation, and store the results in "dst". + Compare + + + + + Compare packed unsigned 32-bit integers in "a" and "b" for greater-than, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := 64*j - k := 32*j - dst[k+31:k] := Truncate_Int64_To_Int32(a[i+63:i]) +FOR j := 0 to 3 + i := j*32 + IF k1[j] + k[j] := ( a[i+31:i] > b[i+31:i] ) ? 1 : 0 + ELSE + k[j] := 0 + FI ENDFOR -dst[MAX:256] := 0 +k[MAX:4] := 0 - +
immintrin.h
 - + Integer + Mask + AVX512VL AVX512F - Convert - - - - Convert packed 64-bit integers in "a" to packed 32-bit integers with truncation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Compare + + + + + Compare packed unsigned 32-bit integers in "a" and "b" for less-than-or-equal, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := 64*j - l := 32*j - IF k[j] - dst[l+31:l] := Truncate_Int64_To_Int32(a[i+63:i]) - ELSE - dst[l+31:l] := src[l+31:l] +FOR j := 0 to 3 + i := j*32 + IF k1[j] + k[j] := ( a[i+31:i] <= b[i+31:i] ) ? 1 : 0 + ELSE + k[j] := 0 FI ENDFOR -dst[MAX:256] := 0 +k[MAX:4] := 0 - +
immintrin.h
 - + Integer + Mask + AVX512VL AVX512F - Convert - Store - - - - Convert packed 64-bit integers in "a" to packed 32-bit integers with truncation, and store the active results (those with their respective bit set in writemask "k") to unaligned memory at "base_addr". + Compare + + + + + Compare packed unsigned 32-bit integers in "a" and "b" for less-than, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := 64*j - l := 32*j - IF k[j] - MEM[base_addr+l+31:base_addr+l] := Truncate_Int64_To_Int32(a[i+63:i]) +FOR j := 0 to 3 + i := j*32 + IF k1[j] + k[j] := ( a[i+31:i] < b[i+31:i] ) ? 1 : 0 + ELSE + k[j] := 0 FI ENDFOR +k[MAX:4] := 0 - +
immintrin.h
 - + Integer + Mask + AVX512VL AVX512F - Convert - - - Convert packed 64-bit integers in "a" to packed 32-bit integers with truncation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Compare + + + + + Compare packed unsigned 32-bit integers in "a" and "b" for not-equal, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := 64*j - l := 32*j - IF k[j] - dst[l+31:l] := Truncate_Int64_To_Int32(a[i+63:i]) - ELSE - dst[l+31:l] := 0 +FOR j := 0 to 3 + i := j*32 + IF k1[j] + k[j] := ( a[i+31:i] != b[i+31:i] ) ? 1 : 0 + ELSE + k[j] := 0 FI ENDFOR -dst[MAX:256] := 0 +k[MAX:4] := 0 - +
immintrin.h
 - + Integer + Mask + AVX512VL AVX512F - Convert - - Convert packed 64-bit integers in "a" to packed 16-bit integers with truncation, and store the results in "dst". - -FOR j := 0 to 7 - i := 64*j - k := 16*j - dst[k+15:k] := Truncate_Int64_To_Int16(a[i+63:i]) + Compare + + + + + Compare packed unsigned 64-bit integers in "a" and "b" based on the comparison operand specified by "imm8", and store the results in mask vector "k". + CASE (imm8[2:0]) OF +0: OP := _MM_CMPINT_EQ +1: OP := _MM_CMPINT_LT +2: OP := _MM_CMPINT_LE +3: OP := _MM_CMPINT_FALSE +4: OP := _MM_CMPINT_NE +5: OP := _MM_CMPINT_NLT +6: OP := _MM_CMPINT_NLE +7: OP := _MM_CMPINT_TRUE +ESAC +FOR j := 0 to 3 + i := j*64 + k[j] := ( a[i+63:i] OP b[i+63:i] ) ? 1 : 0 ENDFOR -dst[MAX:128] := 0 +k[MAX:4] := 0 - +
immintrin.h
 - + Integer + Mask + AVX512VL AVX512F - Convert - - - - Convert packed 64-bit integers in "a" to packed 16-bit integers with truncation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Compare + + + + Compare packed unsigned 64-bit integers in "a" and "b" for equality, and store the results in mask vector "k". -FOR j := 0 to 7 - i := 64*j - l := 16*j - IF k[j] - dst[l+15:l] := Truncate_Int64_To_Int16(a[i+63:i]) - ELSE - dst[l+15:l] := src[l+15:l] - FI +FOR j := 0 to 3 + i := j*64 + k[j] := ( a[i+63:i] == b[i+63:i] ) ? 1 : 0 ENDFOR -dst[MAX:128] := 0 +k[MAX:4] := 0 - +
immintrin.h
 - + Integer + Mask + AVX512VL AVX512F - Convert - Store - - - - Convert packed 64-bit integers in "a" to packed 16-bit integers with truncation, and store the active results (those with their respective bit set in writemask "k") to unaligned memory at "base_addr". + Compare + + + + Compare packed unsigned 64-bit integers in "a" and "b" for greater-than-or-equal, and store the results in mask vector "k". -FOR j := 0 to 7 - i := 64*j - l := 16*j - IF k[j] - MEM[base_addr+l+15:base_addr+l] := Truncate_Int64_To_Int16(a[i+63:i]) - FI +FOR j := 0 to 3 + i := j*64 + k[j] := ( a[i+63:i] >= b[i+63:i] ) ? 1 : 0 ENDFOR +k[MAX:4] := 0 - +
immintrin.h
 - + Integer + Mask + AVX512VL AVX512F - Convert - - - Convert packed 64-bit integers in "a" to packed 16-bit integers with truncation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Compare + + + + Compare packed unsigned 64-bit integers in "a" and "b" for greater-than, and store the results in mask vector "k". -FOR j := 0 to 7 - i := 64*j - l := 16*j - IF k[j] - dst[l+15:l] := Truncate_Int64_To_Int16(a[i+63:i]) - ELSE - dst[l+15:l] := 0 - FI +FOR j := 0 to 3 + i := j*64 + k[j] := ( a[i+63:i] > b[i+63:i] ) ? 1 : 0 ENDFOR -dst[MAX:128] := 0 +k[MAX:4] := 0 - +
immintrin.h
 - + Integer + Mask + AVX512VL AVX512F - Convert - - Convert packed 32-bit integers in "a" to packed 8-bit integers with signed saturation, and store the results in "dst". + Compare + + + + Compare packed unsigned 64-bit integers in "a" and "b" for less-than-or-equal, and store the results in mask vector "k". -FOR j := 0 to 15 - i := 32*j - k := 8*j - dst[k+7:k] := Saturate_Int32_To_Int8(a[i+31:i]) +FOR j := 0 to 3 + i := j*64 + k[j] := ( a[i+63:i] <= b[i+63:i] ) ? 1 : 0 ENDFOR -dst[MAX:128] := 0 +k[MAX:4] := 0 - +
immintrin.h
 - + Integer + Mask + AVX512VL AVX512F - Convert - - - - Convert packed 32-bit integers in "a" to packed 8-bit integers with signed saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Compare + + + + Compare packed unsigned 64-bit integers in "a" and "b" for less-than, and store the results in mask vector "k". -FOR j := 0 to 15 - i := 32*j - l := 8*j - IF k[j] - dst[l+7:l] := Saturate_Int32_To_Int8(a[i+31:i]) - ELSE - dst[l+7:l] := src[l+7:l] - FI +FOR j := 0 to 3 + i := j*64 + k[j] := ( a[i+63:i] < b[i+63:i] ) ? 1 : 0 ENDFOR -dst[MAX:128] := 0 +k[MAX:4] := 0 - +
immintrin.h
 - + Integer + Mask + AVX512VL AVX512F - Convert - Store - - - - Convert packed 32-bit integers in "a" to packed 8-bit integers with signed saturation, and store the active results (those with their respective bit set in writemask "k") to unaligned memory at "base_addr". + Compare + + + + Compare packed unsigned 64-bit integers in "a" and "b" for not-equal, and store the results in mask vector "k". -FOR j := 0 to 15 - i := 32*j - l := 8*j - IF k[j] - MEM[base_addr+l+7:base_addr+l] := Saturate_Int32_To_Int8(a[i+31:i]) - FI +FOR j := 0 to 3 + i := j*64 + k[j] := ( a[i+63:i] != b[i+63:i] ) ? 1 : 0 ENDFOR +k[MAX:4] := 0 - +
immintrin.h
 - + Integer + Mask + AVX512VL AVX512F - Convert - - - Convert packed 32-bit integers in "a" to packed 8-bit integers with signed saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - -FOR j := 0 to 15 - i := 32*j - l := 8*j - IF k[j] - dst[l+7:l] := Saturate_Int32_To_Int8(a[i+31:i]) - ELSE - dst[l+7:l] := 0 + Compare + + + + + + Compare packed unsigned 64-bit integers in "a" and "b" based on the comparison operand specified by "imm8", and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). + CASE (imm8[2:0]) OF +0: OP := _MM_CMPINT_EQ +1: OP := _MM_CMPINT_LT +2: OP := _MM_CMPINT_LE +3: OP := _MM_CMPINT_FALSE +4: OP := _MM_CMPINT_NE +5: OP := _MM_CMPINT_NLT +6: OP := _MM_CMPINT_NLE +7: OP := _MM_CMPINT_TRUE +ESAC +FOR j := 0 to 3 + i := j*64 + IF k1[j] + k[j] := ( a[i+63:i] OP b[i+63:i] ) ? 1 : 0 + ELSE + k[j] := 0 FI ENDFOR -dst[MAX:128] := 0 +k[MAX:4] := 0 - +
immintrin.h
 - + Integer + Mask + AVX512VL AVX512F - Convert - - Convert packed 32-bit integers in "a" to packed 16-bit integers with signed saturation, and store the results in "dst". + Compare + + + + + Compare packed unsigned 64-bit integers in "a" and "b" for equality, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := 32*j - k := 16*j - dst[k+15:k] := Saturate_Int32_To_Int16(a[i+31:i]) +FOR j := 0 to 3 + i := j*64 + IF k1[j] + k[j] := ( a[i+63:i] == b[i+63:i] ) ? 1 : 0 + ELSE + k[j] := 0 + FI ENDFOR -dst[MAX:256] := 0 +k[MAX:4] := 0 - +
immintrin.h
 - + Integer + Mask + AVX512VL AVX512F - Convert - - - - Convert packed 32-bit integers in "a" to packed 16-bit integers with signed saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Compare + + + + + Compare packed unsigned 64-bit integers in "a" and "b" for greater-than-or-equal, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := 32*j - l := 16*j - IF k[j] - dst[l+15:l] := Saturate_Int32_To_Int16(a[i+31:i]) - ELSE - dst[l+15:l] := src[l+15:l] +FOR j := 0 to 3 + i := j*64 + IF k1[j] + k[j] := ( a[i+63:i] >= b[i+63:i] ) ? 1 : 0 + ELSE + k[j] := 0 FI ENDFOR -dst[MAX:256] := 0 +k[MAX:4] := 0 - +
immintrin.h
 - + Integer + Mask + AVX512VL AVX512F - Convert - Store - - - - Convert packed 32-bit integers in "a" to packed 16-bit integers with signed saturation, and store the active results (those with their respective bit set in writemask "k") to unaligned memory at "base_addr". + Compare + + + + + Compare packed unsigned 64-bit integers in "a" and "b" for greater-than, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := 32*j - l := 16*j - IF k[j] - MEM[base_addr+l+15:base_addr+l] := Saturate_Int32_To_Int16(a[i+31:i]) +FOR j := 0 to 3 + i := j*64 + IF k1[j] + k[j] := ( a[i+63:i] > b[i+63:i] ) ? 1 : 0 + ELSE + k[j] := 0 FI ENDFOR +k[MAX:4] := 0 - +
immintrin.h
 - + Integer + Mask + AVX512VL AVX512F - Convert - - - Convert packed 32-bit integers in "a" to packed 16-bit integers with signed saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Compare + + + + + Compare packed unsigned 64-bit integers in "a" and "b" for less-than-or-equal, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := 32*j - l := 16*j - IF k[j] - dst[l+15:l] := Saturate_Int32_To_Int16(a[i+31:i]) - ELSE - dst[l+15:l] := 0 +FOR j := 0 to 3 + i := j*64 + IF k1[j] + k[j] := ( a[i+63:i] <= b[i+63:i] ) ? 1 : 0 + ELSE + k[j] := 0 FI ENDFOR -dst[MAX:256] := 0 +k[MAX:4] := 0 - +
immintrin.h
 - + Integer + Mask + AVX512VL AVX512F - Convert - - Convert packed 64-bit integers in "a" to packed 8-bit integers with signed saturation, and store the results in "dst". + Compare + + + + + Compare packed unsigned 64-bit integers in "a" and "b" for less-than, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := 64*j - k := 8*j - dst[k+7:k] := Saturate_Int64_To_Int8(a[i+63:i]) +FOR j := 0 to 3 + i := j*64 + IF k1[j] + k[j] := ( a[i+63:i] < b[i+63:i] ) ? 1 : 0 + ELSE + k[j] := 0 + FI ENDFOR -dst[MAX:64] := 0 +k[MAX:4] := 0 - +
immintrin.h
 - + Integer + Mask + AVX512VL AVX512F - Convert - - - - Convert packed 64-bit integers in "a" to packed 8-bit integers with signed saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Compare + + + + + Compare packed unsigned 64-bit integers in "a" and "b" for not-equal, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := 64*j - l := 8*j - IF k[j] - dst[l+7:l] := Saturate_Int64_To_Int8(a[i+63:i]) - ELSE - dst[l+7:l] := src[l+7:l] - FI -ENDFOR -dst[MAX:64] := 0 - - -
immintrin.h
- - - Integer - AVX512F - Convert - Store - - - - Convert packed 64-bit integers in "a" to packed 8-bit integers with signed saturation, and store the active results (those with their respective bit set in writemask "k") to unaligned memory at "base_addr". - -FOR j := 0 to 7 - i := 64*j - l := 8*j - IF k[j] - MEM[base_addr+l+7:base_addr+l] := Saturate_Int64_To_Int8(a[i+63:i]) +FOR j := 0 to 3 + i := j*64 + IF k1[j] + k[j] := ( a[i+63:i] != b[i+63:i] ) ? 1 : 0 + ELSE + k[j] := 0 FI ENDFOR +k[MAX:4] := 0 - +
immintrin.h
 - + Integer + Mask + AVX512VL AVX512F - Convert - - - Convert packed 64-bit integers in "a" to packed 8-bit integers with signed saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - -FOR j := 0 to 7 - i := 64*j - l := 8*j - IF k[j] - dst[l+7:l] := Saturate_Int64_To_Int8(a[i+63:i]) - ELSE - dst[l+7:l] := 0 - FI + Compare + + + + + Compare packed unsigned 64-bit integers in "a" and "b" based on the comparison operand specified by "imm8", and store the results in mask vector "k". + CASE (imm8[2:0]) OF +0: OP := _MM_CMPINT_EQ +1: OP := _MM_CMPINT_LT +2: OP := _MM_CMPINT_LE +3: OP := _MM_CMPINT_FALSE +4: OP := _MM_CMPINT_NE +5: OP := _MM_CMPINT_NLT +6: OP := _MM_CMPINT_NLE +7: OP := _MM_CMPINT_TRUE +ESAC +FOR j := 0 to 1 + i := j*64 + k[j] := ( a[i+63:i] OP b[i+63:i] ) ? 1 : 0 ENDFOR -dst[MAX:64] := 0 +k[MAX:2] := 0 - +
immintrin.h
 - + Integer + Mask + AVX512VL AVX512F - Convert - - Convert packed 64-bit integers in "a" to packed 32-bit integers with signed saturation, and store the results in "dst". + Compare + + + + Compare packed unsigned 64-bit integers in "a" and "b" for equality, and store the results in mask vector "k". -FOR j := 0 to 7 - i := 64*j - k := 32*j - dst[k+31:k] := Saturate_Int64_To_Int32(a[i+63:i]) +FOR j := 0 to 1 + i := j*64 + k[j] := ( a[i+63:i] == b[i+63:i] ) ? 1 : 0 ENDFOR -dst[MAX:256] := 0 +k[MAX:2] := 0 - +
immintrin.h
 - + Integer + Mask + AVX512VL AVX512F - Convert - - - - Convert packed 64-bit integers in "a" to packed 32-bit integers with signed saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Compare + + + + Compare packed unsigned 64-bit integers in "a" and "b" for greater-than-or-equal, and store the results in mask vector "k". -FOR j := 0 to 7 - i := 64*j - l := 32*j - IF k[j] - dst[l+31:l] := Saturate_Int64_To_Int32(a[i+63:i]) - ELSE - dst[l+31:l] := src[l+31:l] - FI +FOR j := 0 to 1 + i := j*64 + k[j] := ( a[i+63:i] >= b[i+63:i] ) ? 1 : 0 ENDFOR -dst[MAX:256] := 0 +k[MAX:2] := 0 - +
immintrin.h
 - + Integer + Mask + AVX512VL AVX512F - Convert - Store - - - - Convert packed 64-bit integers in "a" to packed 32-bit integers with signed saturation, and store the active results (those with their respective bit set in writemask "k") to unaligned memory at "base_addr". + Compare + + + + Compare packed unsigned 64-bit integers in "a" and "b" for greater-than, and store the results in mask vector "k". -FOR j := 0 to 7 - i := 64*j - l := 32*j - IF k[j] - MEM[base_addr+l+31:base_addr+l] := Saturate_Int64_To_Int32(a[i+63:i]) - FI +FOR j := 0 to 1 + i := j*64 + k[j] := ( a[i+63:i] > b[i+63:i] ) ? 1 : 0 ENDFOR +k[MAX:2] := 0 - +
immintrin.h
 - + Integer + Mask + AVX512VL AVX512F - Convert - - - Convert packed 64-bit integers in "a" to packed 32-bit integers with signed saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Compare + + + + Compare packed unsigned 64-bit integers in "a" and "b" for less-than-or-equal, and store the results in mask vector "k". -FOR j := 0 to 7 - i := 64*j - l := 32*j - IF k[j] - dst[l+31:l] := Saturate_Int64_To_Int32(a[i+63:i]) - ELSE - dst[l+31:l] := 0 - FI +FOR j := 0 to 1 + i := j*64 + k[j] := ( a[i+63:i] <= b[i+63:i] ) ? 1 : 0 ENDFOR -dst[MAX:256] := 0 +k[MAX:2] := 0 - +
immintrin.h
 - + Integer + Mask + AVX512VL AVX512F - Convert - - Convert packed 64-bit integers in "a" to packed 16-bit integers with signed saturation, and store the results in "dst". + Compare + + + + Compare packed unsigned 64-bit integers in "a" and "b" for less-than, and store the results in mask vector "k". -FOR j := 0 to 7 - i := 64*j - k := 16*j - dst[k+15:k] := Saturate_Int64_To_Int16(a[i+63:i]) +FOR j := 0 to 1 + i := j*64 + k[j] := ( a[i+63:i] < b[i+63:i] ) ? 1 : 0 ENDFOR -dst[MAX:128] := 0 +k[MAX:2] := 0 - +
immintrin.h
 - + Integer + Mask + AVX512VL AVX512F - Convert - - - - Convert packed 64-bit integers in "a" to packed 16-bit integers with signed saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Compare + + + + Compare packed unsigned 64-bit integers in "a" and "b" for not-equal, and store the results in mask vector "k". -FOR j := 0 to 7 - i := 64*j - l := 16*j - IF k[j] - dst[l+15:l] := Saturate_Int64_To_Int16(a[i+63:i]) - ELSE - dst[l+15:l] := src[l+15:l] - FI +FOR j := 0 to 1 + i := j*64 + k[j] := ( a[i+63:i] != b[i+63:i] ) ? 1 : 0 ENDFOR -dst[MAX:128] := 0 +k[MAX:2] := 0 - +
immintrin.h
 - + Integer + Mask + AVX512VL AVX512F - Convert - Store - - - - Convert packed 64-bit integers in "a" to packed 16-bit integers with signed saturation, and store the active results (those with their respective bit set in writemask "k") to unaligned memory at "base_addr". - -FOR j := 0 to 7 - i := 64*j - l := 16*j - IF k[j] - MEM[base_addr+l+15:base_addr+l] := Saturate_Int64_To_Int16(a[i+63:i]) + Compare + + + + + + Compare packed unsigned 64-bit integers in "a" and "b" based on the comparison operand specified by "imm8", and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). + CASE (imm8[2:0]) OF +0: OP := _MM_CMPINT_EQ +1: OP := _MM_CMPINT_LT +2: OP := _MM_CMPINT_LE +3: OP := _MM_CMPINT_FALSE +4: OP := _MM_CMPINT_NE +5: OP := _MM_CMPINT_NLT +6: OP := _MM_CMPINT_NLE +7: OP := _MM_CMPINT_TRUE +ESAC +FOR j := 0 to 1 + i := j*64 + IF k1[j] + k[j] := ( a[i+63:i] OP b[i+63:i] ) ? 1 : 0 + ELSE + k[j] := 0 FI ENDFOR +k[MAX:2] := 0 - +
immintrin.h
 - + Integer + Mask + AVX512VL AVX512F - Convert - - - Convert packed 64-bit integers in "a" to packed 16-bit integers with signed saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Compare + + + + + Compare packed unsigned 64-bit integers in "a" and "b" for equality, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := 64*j - l := 16*j - IF k[j] - dst[l+15:l] := Saturate_Int64_To_Int16(a[i+63:i]) - ELSE - dst[l+15:l] := 0 +FOR j := 0 to 1 + i := j*64 + IF k1[j] + k[j] := ( a[i+63:i] == b[i+63:i] ) ? 1 : 0 + ELSE + k[j] := 0 FI ENDFOR -dst[MAX:128] := 0 +k[MAX:2] := 0 - +
immintrin.h
 - + Integer + Mask + AVX512VL AVX512F - Convert - - Sign extend packed 8-bit integers in "a" to packed 32-bit integers, and store the results in "dst". + Compare + + + + + Compare packed unsigned 64-bit integers in "a" and "b" for greater-than-or-equal, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := 32*j - k := 8*j - dst[i+31:i] := SignExtend(a[k+7:k]) +FOR j := 0 to 1 + i := j*64 + IF k1[j] + k[j] := ( a[i+63:i] >= b[i+63:i] ) ? 1 : 0 + ELSE + k[j] := 0 + FI ENDFOR -dst[MAX:512] := 0 +k[MAX:2] := 0 - +
immintrin.h
 - + Integer + Mask + AVX512VL AVX512F - Convert - - - - Sign extend packed 8-bit integers in "a" to packed 32-bit integers, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Compare + + + + + Compare packed unsigned 64-bit integers in "a" and "b" for greater-than, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := 32*j - l := 8*j - IF k[j] - dst[i+31:i] := SignExtend(a[l+7:l]) - ELSE - dst[i+31:i] := src[i+31:i] +FOR j := 0 to 1 + i := j*64 + IF k1[j] + k[j] := ( a[i+63:i] > b[i+63:i] ) ? 1 : 0 + ELSE + k[j] := 0 FI ENDFOR -dst[MAX:512] := 0 +k[MAX:2] := 0 - +
immintrin.h
 - + Integer + Mask + AVX512VL AVX512F - Convert - - - Sign extend packed 8-bit integers in "a" to packed 32-bit integers, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Compare + + + + + Compare packed unsigned 64-bit integers in "a" and "b" for less-than-or-equal, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := 32*j - l := 8*j - IF k[j] - dst[i+31:i] := SignExtend(a[l+7:l]) - ELSE - dst[i+31:i] := 0 +FOR j := 0 to 1 + i := j*64 + IF k1[j] + k[j] := ( a[i+63:i] <= b[i+63:i] ) ? 1 : 0 + ELSE + k[j] := 0 FI ENDFOR -dst[MAX:512] := 0 +k[MAX:2] := 0 - +
immintrin.h
 - + Integer + Mask + AVX512VL AVX512F - Convert - - Sign extend packed 8-bit integers in the low 8 bytes of "a" to packed 64-bit integers, and store the results in "dst". + Compare + + + + + Compare packed unsigned 64-bit integers in "a" and "b" for less-than, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := 64*j - k := 8*j - dst[i+63:i] := SignExtend(a[k+7:k]) +FOR j := 0 to 1 + i := j*64 + IF k1[j] + k[j] := ( a[i+63:i] < b[i+63:i] ) ? 1 : 0 + ELSE + k[j] := 0 + FI ENDFOR -dst[MAX:512] := 0 +k[MAX:2] := 0 - +
immintrin.h
 - + Integer + Mask + AVX512VL AVX512F - Convert - - - - Sign extend packed 8-bit integers in the low 8 bytes of "a" to packed 64-bit integers, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Compare + + + + + Compare packed unsigned 64-bit integers in "a" and "b" for not-equal, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := 64*j - l := 8*j - IF k[j] - dst[i+63:i] := SignExtend(a[l+7:l]) - ELSE - dst[i+63:i] := src[i+63:i] +FOR j := 0 to 1 + i := j*64 + IF k1[j] + k[j] := ( a[i+63:i] != b[i+63:i] ) ? 1 : 0 + ELSE + k[j] := 0 FI ENDFOR -dst[MAX:512] := 0 +k[MAX:2] := 0 - +
immintrin.h
 - + Integer + AVX512VL AVX512F - Convert - - - Sign extend packed 8-bit integers in the low 8 bytes of "a" to packed 64-bit integers, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Miscellaneous + + + + + Contiguously store the active 32-bit integers in "a" (those with their respective bit set in writemask "k") to "dst", and pass through the remaining elements from "src". +size := 32 +m := 0 FOR j := 0 to 7 - i := 64*j - l := 8*j + i := j*32 IF k[j] - dst[i+63:i] := SignExtend(a[l+7:l]) - ELSE - dst[i+63:i] := 0 + dst[m+size-1:m] := a[i+31:i] + m := m + size FI ENDFOR -dst[MAX:512] := 0 +dst[255:m] := src[255:m] +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer + AVX512VL AVX512F - Convert - - Sign extend packed 32-bit integers in "a" to packed 64-bit integers, and store the results in "dst". + Store + + + + + Contiguously store the active 32-bit integers in "a" (those with their respective bit set in writemask "k") to unaligned memory at "base_addr". +size := 32 +m := base_addr FOR j := 0 to 7 - i := 64*j - k := 32*j - dst[i+63:i] := SignExtend(a[k+31:k]) + i := j*32 + IF k[j] + MEM[m+size-1:m] := a[i+31:i] + m := m + size + FI ENDFOR -dst[MAX:512] := 0 - +
immintrin.h
 - + Integer + AVX512VL AVX512F - Convert - - - - Sign extend packed 32-bit integers in "a" to packed 64-bit integers, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Miscellaneous + + + + Contiguously store the active 32-bit integers in "a" (those with their respective bit set in zeromask "k") to "dst", and set the remaining elements to zero. +size := 32 +m := 0 FOR j := 0 to 7 - i := 64*j - l := 32*j + i := j*32 IF k[j] - dst[i+63:i] := SignExtend(a[l+31:l]) - ELSE - dst[i+63:i] := src[i+63:i] + dst[m+size-1:m] := a[i+31:i] + m := m + size FI ENDFOR -dst[MAX:512] := 0 +dst[255:m] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer + AVX512VL AVX512F - Convert - - - Sign extend packed 32-bit integers in "a" to packed 64-bit integers, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Miscellaneous + + + + + Contiguously store the active 32-bit integers in "a" (those with their respective bit set in writemask "k") to "dst", and pass through the remaining elements from "src". -FOR j := 0 to 7 - i := 64*j - l := 32*j +size := 32 +m := 0 +FOR j := 0 to 3 + i := j*32 IF k[j] - dst[i+63:i] := SignExtend(a[l+31:l]) - ELSE - dst[i+63:i] := 0 + dst[m+size-1:m] := a[i+31:i] + m := m + size FI ENDFOR -dst[MAX:512] := 0 +dst[127:m] := src[127:m] +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer + AVX512VL AVX512F - Convert - - Sign extend packed 16-bit integers in "a" to packed 32-bit integers, and store the results in "dst". + Store + + + + + Contiguously store the active 32-bit integers in "a" (those with their respective bit set in writemask "k") to unaligned memory at "base_addr". -FOR j := 0 to 15 - i := 32*j - k := 16*j - dst[i+31:i] := SignExtend(a[k+15:k]) +size := 32 +m := base_addr +FOR j := 0 to 3 + i := j*32 + IF k[j] + MEM[m+size-1:m] := a[i+31:i] + m := m + size + FI ENDFOR -dst[MAX:512] := 0 - +
immintrin.h
 - + Integer + AVX512VL AVX512F - Convert - - - - Sign extend packed 16-bit integers in "a" to packed 32-bit integers, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Miscellaneous + + + + Contiguously store the active 32-bit integers in "a" (those with their respective bit set in zeromask "k") to "dst", and set the remaining elements to zero. -FOR j := 0 to 15 +size := 32 +m := 0 +FOR j := 0 to 3 i := j*32 - l := j*16 IF k[j] - dst[i+31:i] := SignExtend(a[l+15:l]) - ELSE - dst[i+31:i] := src[i+31:i] + dst[m+size-1:m] := a[i+31:i] + m := m + size FI ENDFOR -dst[MAX:512] := 0 +dst[127:m] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer + AVX512VL AVX512F - Convert - - - Sign extend packed 16-bit integers in "a" to packed 32-bit integers, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Miscellaneous + + + + + Contiguously store the active 64-bit integers in "a" (those with their respective bit set in writemask "k") to "dst", and pass through the remaining elements from "src". -FOR j := 0 to 15 - i := 32*j - l := 16*j +size := 64 +m := 0 +FOR j := 0 to 3 + i := j*64 IF k[j] - dst[i+31:i] := SignExtend(a[l+15:l]) - ELSE - dst[i+31:i] := 0 + dst[m+size-1:m] := a[i+63:i] + m := m + size FI ENDFOR -dst[MAX:512] := 0 +dst[255:m] := src[255:m] +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer + AVX512VL AVX512F - Convert - - Sign extend packed 16-bit integers in "a" to packed 64-bit integers, and store the results in "dst". + Store + + + + + Contiguously store the active 64-bit integers in "a" (those with their respective bit set in writemask "k") to unaligned memory at "base_addr". -FOR j := 0 to 7 - i := 64*j - k := 16*j - dst[i+63:i] := SignExtend(a[k+15:k]) +size := 64 +m := base_addr +FOR j := 0 to 3 + i := j*64 + IF k[j] + MEM[m+size-1:m] := a[i+63:i] + m := m + size + FI ENDFOR -dst[MAX:512] := 0 - +
immintrin.h
 - + Integer + AVX512VL AVX512F - Convert - - - - Sign extend packed 16-bit integers in "a" to packed 64-bit integers, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Miscellaneous + + + + Contiguously store the active 64-bit integers in "a" (those with their respective bit set in zeromask "k") to "dst", and set the remaining elements to zero. -FOR j := 0 to 7 - i := 64*j - l := 16*j +size := 64 +m := 0 +FOR j := 0 to 3 + i := j*64 IF k[j] - dst[i+63:i] := SignExtend(a[l+15:l]) - ELSE - dst[i+63:i] := src[i+63:i] + dst[m+size-1:m] := a[i+63:i] + m := m + size FI ENDFOR -dst[MAX:512] := 0 +dst[255:m] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer + AVX512VL AVX512F - Convert - - - Sign extend packed 16-bit integers in "a" to packed 64-bit integers, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Miscellaneous + + + + + Contiguously store the active 64-bit integers in "a" (those with their respective bit set in writemask "k") to "dst", and pass through the remaining elements from "src". -FOR j := 0 to 7 - i := 64*j - l := 16*j +size := 64 +m := 0 +FOR j := 0 to 1 + i := j*64 IF k[j] - dst[i+63:i] := SignExtend(a[l+15:l]) - ELSE - dst[i+63:i] := 0 + dst[m+size-1:m] := a[i+63:i] + m := m + size FI ENDFOR -dst[MAX:512] := 0 +dst[127:m] := src[127:m] +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer + AVX512VL AVX512F - Convert - - Convert packed unsigned 32-bit integers in "a" to packed unsigned 8-bit integers with unsigned saturation, and store the results in "dst". + Store + + + + + Contiguously store the active 64-bit integers in "a" (those with their respective bit set in writemask "k") to unaligned memory at "base_addr". -FOR j := 0 to 15 - i := 32*j - k := 8*j - dst[k+7:k] := Saturate_UnsignedInt32_To_Int8(a[i+31:i]) +size := 64 +m := base_addr +FOR j := 0 to 1 + i := j*64 + IF k[j] + MEM[m+size-1:m] := a[i+63:i] + m := m + size + FI ENDFOR -dst[MAX:128] := 0 - +
immintrin.h
 - + Integer + AVX512VL AVX512F - Convert - - - - Convert packed unsigned 32-bit integers in "a" to packed unsigned 8-bit integers with unsigned saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Miscellaneous + + + + Contiguously store the active 64-bit integers in "a" (those with their respective bit set in zeromask "k") to "dst", and set the remaining elements to zero. -FOR j := 0 to 15 - i := 32*j - l := 8*j +size := 64 +m := 0 +FOR j := 0 to 1 + i := j*64 IF k[j] - dst[l+7:l] := Saturate_UnsignedInt32_To_Int8(a[i+31:i]) - ELSE - dst[l+7:l] := src[l+7:l] + dst[m+size-1:m] := a[i+63:i] + m := m + size FI ENDFOR +dst[127:m] := 0 dst[MAX:128] := 0 - +
immintrin.h
 - + Integer + AVX512VL AVX512F - Convert - Store - - - - Convert packed unsigned 32-bit integers in "a" to packed 8-bit integers with unsigned saturation, and store the active results (those with their respective bit set in writemask "k") to unaligned memory at "base_addr". + Miscellaneous + + + + + + Shuffle 32-bit integers in "a" across lanes using the corresponding index in "idx", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := 32*j - l := 8*j +FOR j := 0 to 7 + i := j*32 + id := idx[i+2:i]*32 IF k[j] - MEM[base_addr+l+7:base_addr+l] := Saturate_UnsignedInt32_To_Int8(a[i+31:i]) + dst[i+31:i] := a[id+31:id] + ELSE + dst[i+31:i] := src[i+31:i] FI ENDFOR +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer + AVX512VL AVX512F - Convert - - - Convert packed unsigned 32-bit integers in "a" to packed unsigned 8-bit integers with unsigned saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Miscellaneous + + + + + Shuffle 32-bit integers in "a" across lanes using the corresponding index in "idx", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := 32*j - l := 8*j +FOR j := 0 to 7 + i := j*32 + id := idx[i+2:i]*32 IF k[j] - dst[l+7:l] := Saturate_UnsignedInt32_To_Int8(a[i+31:i]) + dst[i+31:i] := a[id+31:id] ELSE - dst[l+7:l] := 0 + dst[i+31:i] := 0 FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer + AVX512VL AVX512F - Convert - - Convert packed unsigned 32-bit integers in "a" to packed unsigned 16-bit integers with unsigned saturation, and store the results in "dst". + Miscellaneous + + + + Shuffle 32-bit integers in "a" across lanes using the corresponding index in "idx", and store the results in "dst". -FOR j := 0 to 15 - i := 32*j - k := 16*j - dst[k+15:k] := Saturate_UnsignedInt32_To_Int16(a[i+31:i]) -ENDFOR -dst[MAX:256] := 0 +FOR j := 0 to 7 + i := j*32 + id := idx[i+2:i]*32 + dst[i+31:i] := a[id+31:id] +ENDFOR +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer + AVX512VL AVX512F - Convert - - - - Convert packed unsigned 32-bit integers in "a" to packed unsigned 16-bit integers with unsigned saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Miscellaneous + + + + + + Shuffle 32-bit integers in "a" and "b" across lanes using the corresponding selector and index in "idx", and store the results in "dst" using writemask "k" (elements are copied from "idx" when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := 32*j - l := 16*j +FOR j := 0 to 7 + i := j*32 + off := idx[i+2:i]*32 IF k[j] - dst[l+15:l] := Saturate_UnsignedInt32_To_Int16(a[i+31:i]) + dst[i+31:i] := idx[i+3] ? b[off+31:off] : a[off+31:off] ELSE - dst[l+15:l] := src[l+15:l] + dst[i+31:i] := idx[i+31:i] FI ENDFOR dst[MAX:256] := 0 - +
immintrin.h
 - + Integer + AVX512VL AVX512F - Convert - Store - - - - Convert packed unsigned 32-bit integers in "a" to packed 16-bit integers with unsigned saturation, and store the active results (those with their respective bit set in writemask "k") to unaligned memory at "base_addr". + Miscellaneous + + + + + + Shuffle 32-bit integers in "a" and "b" across lanes using the corresponding selector and index in "idx", and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := 32*j - l := 16*j +FOR j := 0 to 7 + i := j*32 + off := idx[i+2:i]*32 IF k[j] - MEM[base_addr+l+15:base_addr+l] := Saturate_UnsignedInt32_To_Int16(a[i+31:i]) + dst[i+31:i] := idx[i+3] ? b[off+31:off] : a[off+31:off] + ELSE + dst[i+31:i] := a[i+31:i] FI ENDFOR +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer + AVX512VL AVX512F - Convert - - - Convert packed unsigned 32-bit integers in "a" to packed unsigned 16-bit integers with unsigned saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Miscellaneous + + + + + + Shuffle 32-bit integers in "a" and "b" across lanes using the corresponding selector and index in "idx", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := 32*j - l := 16*j +FOR j := 0 to 7 + i := j*32 + off := idx[i+2:i]*32 IF k[j] - dst[l+15:l] := Saturate_UnsignedInt32_To_Int16(a[i+31:i]) + dst[i+31:i] := (idx[i+3]) ? b[off+31:off] : a[off+31:off] ELSE - dst[l+15:l] := 0 + dst[i+31:i] := 0 FI ENDFOR dst[MAX:256] := 0 - + +
immintrin.h
 - + Integer + AVX512VL AVX512F - Convert - - Convert packed unsigned 64-bit integers in "a" to packed unsigned 8-bit integers with unsigned saturation, and store the results in "dst". + Miscellaneous + + + + + Shuffle 32-bit integers in "a" and "b" across lanes using the corresponding selector and index in "idx", and store the results in "dst". FOR j := 0 to 7 - i := 64*j - k := 8*j - dst[k+7:k] := Saturate_UnsignedInt64_To_Int8(a[i+63:i]) + i := j*32 + off := idx[i+2:i]*32 + dst[i+31:i] := idx[i+3] ? b[off+31:off] : a[off+31:off] ENDFOR -dst[MAX:64] := 0 +dst[MAX:256] := 0 - + +
immintrin.h
 - + Integer + AVX512VL AVX512F - Convert - - - - Convert packed unsigned 64-bit integers in "a" to packed unsigned 8-bit integers with unsigned saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Miscellaneous + + + + + + Shuffle 32-bit integers in "a" and "b" using the corresponding selector and index in "idx", and store the results in "dst" using writemask "k" (elements are copied from "idx" when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := 64*j - l := 8*j +FOR j := 0 to 3 + i := j*32 + off := idx[i+1:i]*32 IF k[j] - dst[l+7:l] := Saturate_UnsignedInt64_To_Int8(a[i+63:i]) + dst[i+31:i] := idx[i+2] ? b[off+31:off] : a[off+31:off] ELSE - dst[l+7:l] := src[l+7:l] + dst[i+31:i] := idx[i+31:i] FI ENDFOR -dst[MAX:64] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer + AVX512VL AVX512F - Convert - Store - - - - Convert packed unsigned 64-bit integers in "a" to packed 8-bit integers with unsigned saturation, and store the active results (those with their respective bit set in writemask "k") to unaligned memory at "base_addr". + Miscellaneous + + + + + + Shuffle 32-bit integers in "a" and "b" using the corresponding selector and index in "idx", and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := 64*j - l := 8*j +FOR j := 0 to 3 + i := j*32 + off := idx[i+1:i]*32 IF k[j] - MEM[base_addr+l+7:base_addr+l] := Saturate_UnsignedInt64_To_Int8(a[i+63:i]) + dst[i+31:i] := idx[i+2] ? b[off+31:off] : a[off+31:off] + ELSE + dst[i+31:i] := a[i+31:i] FI ENDFOR +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer + AVX512VL AVX512F - Convert - - - Convert packed unsigned 64-bit integers in "a" to packed unsigned 8-bit integers with unsigned saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Miscellaneous + + + + + + Shuffle 32-bit integers in "a" and "b" using the corresponding selector and index in "idx", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := 64*j - l := 8*j +FOR j := 0 to 3 + i := j*32 + off := idx[i+1:i]*32 IF k[j] - dst[l+7:l] := Saturate_UnsignedInt64_To_Int8(a[i+63:i]) + dst[i+31:i] := (idx[i+2]) ? b[off+31:off] : a[off+31:off] ELSE - dst[l+7:l] := 0 + dst[i+31:i] := 0 FI ENDFOR -dst[MAX:64] := 0 +dst[MAX:128] := 0 - + +
immintrin.h
 - + Integer + AVX512VL AVX512F - Convert - - Convert packed unsigned 64-bit integers in "a" to packed unsigned 32-bit integers with unsigned saturation, and store the results in "dst". + Miscellaneous + + + + + Shuffle 32-bit integers in "a" and "b" using the corresponding selector and index in "idx", and store the results in "dst". -FOR j := 0 to 7 - i := 64*j - k := 32*j - dst[k+31:k] := Saturate_UnsignedInt64_To_Int32(a[i+63:i]) +FOR j := 0 to 3 + i := j*32 + off := idx[i+1:i]*32 + dst[i+31:i] := idx[i+2] ? b[off+31:off] : a[off+31:off] ENDFOR -dst[MAX:256] := 0 +dst[MAX:128] := 0 - + +
immintrin.h
 - - Integer + + Floating Point + AVX512VL AVX512F - Convert - - - - Convert packed unsigned 64-bit integers in "a" to packed unsigned 32-bit integers with unsigned saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Miscellaneous + + + + + + Shuffle double-precision (64-bit) floating-point elements in "a" and "b" across lanes using the corresponding selector and index in "idx", and store the results in "dst" using writemask "k" (elements are copied from "idx" when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := 64*j - l := 32*j +FOR j := 0 to 3 + i := j*64 + off := idx[i+1:i]*64 IF k[j] - dst[l+31:l] := Saturate_UnsignedInt64_To_Int32(a[i+63:i]) + dst[i+63:i] := idx[i+2] ? b[off+63:off] : a[off+63:off] ELSE - dst[l+31:l] := src[l+31:l] + dst[i+63:i] := idx[i+63:i] FI ENDFOR dst[MAX:256] := 0 - +
immintrin.h
 - - Integer + + Floating Point + AVX512VL AVX512F - Convert - Store - - - - Convert packed unsigned 64-bit integers in "a" to packed 32-bit integers with unsigned saturation, and store the active results (those with their respective bit set in writemask "k") to unaligned memory at "base_addr". + Miscellaneous + + + + + + Shuffle double-precision (64-bit) floating-point elements in "a" and "b" across lanes using the corresponding selector and index in "idx", and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := 64*j - l := 32*j +FOR j := 0 to 3 + i := j*64 + off := idx[i+1:i]*64 IF k[j] - MEM[base_addr+l+31:base_addr+l] := Saturate_UnsignedInt64_To_Int32(a[i+63:i]) + dst[i+63:i] := idx[i+2] ? b[off+63:off] : a[off+63:off] + ELSE + dst[i+63:i] := a[i+63:i] FI ENDFOR +dst[MAX:256] := 0 - +
immintrin.h
 - - Integer + + Floating Point + AVX512VL AVX512F - Convert - - - Convert packed unsigned 64-bit integers in "a" to packed unsigned 32-bit integers with unsigned saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Miscellaneous + + + + + + Shuffle double-precision (64-bit) floating-point elements in "a" and "b" across lanes using the corresponding selector and index in "idx", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := 64*j - l := 32*j +FOR j := 0 to 3 + i := j*64 + off := idx[i+1:i]*64 IF k[j] - dst[l+31:l] := Saturate_UnsignedInt64_To_Int32(a[i+63:i]) + dst[i+63:i] := (idx[i+2]) ? b[off+63:off] : a[off+63:off] ELSE - dst[l+31:l] := 0 + dst[i+63:i] := 0 FI ENDFOR dst[MAX:256] := 0 - + +
immintrin.h
 - - Integer + + Floating Point + AVX512VL AVX512F - Convert - - Convert packed unsigned 64-bit integers in "a" to packed unsigned 16-bit integers with unsigned saturation, and store the results in "dst". + Miscellaneous + + + + + Shuffle double-precision (64-bit) floating-point elements in "a" and "b" across lanes using the corresponding selector and index in "idx", and store the results in "dst". -FOR j := 0 to 7 - i := 64*j - k := 16*j - dst[k+15:k] := Saturate_UnsignedInt64_To_Int16(a[i+63:i]) +FOR j := 0 to 3 + i := j*64 + off := idx[i+1:i]*64 + dst[i+63:i] := idx[i+2] ? b[off+63:off] : a[off+63:off] ENDFOR -dst[MAX:128] := 0 +dst[MAX:256] := 0 - + +
immintrin.h
 - - Integer + + Floating Point + AVX512VL AVX512F - Convert - - - - Convert packed unsigned 64-bit integers in "a" to packed unsigned 16-bit integers with unsigned saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Miscellaneous + + + + + + Shuffle double-precision (64-bit) floating-point elements in "a" and "b" using the corresponding selector and index in "idx", and store the results in "dst" using writemask "k" (elements are copied from "idx" when the corresponding mask bit is not set) -FOR j := 0 to 7 - i := 64*j - l := 16*j +FOR j := 0 to 1 + i := j*64 + off := idx[i]*64 IF k[j] - dst[l+15:l] := Saturate_UnsignedInt64_To_Int16(a[i+63:i]) + dst[i+63:i] := idx[i+1] ? b[off+63:off] : a[off+63:off] ELSE - dst[l+15:l] := src[l+15:l] + dst[i+63:i] := idx[i+63:i] FI ENDFOR dst[MAX:128] := 0 - +
immintrin.h
 - - Integer + + Floating Point + AVX512VL AVX512F - Convert - Store - - - - Convert packed unsigned 64-bit integers in "a" to packed 16-bit integers with unsigned saturation, and store the active results (those with their respective bit set in writemask "k") to unaligned memory at "base_addr". + Miscellaneous + + + + + + Shuffle double-precision (64-bit) floating-point elements in "a" and "b" using the corresponding selector and index in "idx", and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := 64*j - l := 16*j +FOR j := 0 to 1 + i := j*64 + off := idx[i]*64 IF k[j] - MEM[base_addr+l+15:base_addr+l] := Saturate_UnsignedInt64_To_Int16(a[i+63:i]) + dst[i+63:i] := idx[i+1] ? b[off+63:off] : a[off+63:off] + ELSE + dst[i+63:i] := a[i+63:i] FI ENDFOR +dst[MAX:128] := 0 - +
immintrin.h
 - - Integer + + Floating Point + AVX512VL AVX512F - Convert - - - Convert packed unsigned 64-bit integers in "a" to packed unsigned 16-bit integers with unsigned saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Miscellaneous + + + + + + Shuffle double-precision (64-bit) floating-point elements in "a" and "b" using the corresponding selector and index in "idx", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := 64*j - l := 16*j +FOR j := 0 to 1 + i := j*64 + off := idx[i]*64 IF k[j] - dst[l+15:l] := Saturate_UnsignedInt64_To_Int16(a[i+63:i]) + dst[i+63:i] := (idx[i+1]) ? b[off+63:off] : a[off+63:off] ELSE - dst[l+15:l] := 0 + dst[i+63:i] := 0 FI ENDFOR dst[MAX:128] := 0 - + +
immintrin.h
 - - Integer + + Floating Point + AVX512VL AVX512F - Convert - - Zero extend packed unsigned 8-bit integers in "a" to packed 32-bit integers, and store the results in "dst". + Miscellaneous + + + + + Shuffle double-precision (64-bit) floating-point elements in "a" and "b" using the corresponding selector and index in "idx", and store the results in "dst". -FOR j := 0 to 15 - i := 32*j - k := 8*j - dst[i+31:i] := ZeroExtend(a[k+7:k]) +FOR j := 0 to 1 + i := j*64 + off := idx[i]*64 + dst[i+63:i] := idx[i+1] ? b[off+63:off] : a[off+63:off] ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - + +
immintrin.h
 - - Integer + + Floating Point + AVX512VL AVX512F - Convert - - - - Zero extend packed unsigned 8-bit integers in "a" to packed 32-bit integers, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Miscellaneous + + + + + + Shuffle single-precision (32-bit) floating-point elements in "a" and "b" across lanes using the corresponding selector and index in "idx", and store the results in "dst" using writemask "k" (elements are copied from "idx" when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := 32*j - l := 8*j +FOR j := 0 to 7 + i := j*32 + off := idx[i+2:i]*32 IF k[j] - dst[i+31:i] := ZeroExtend(a[l+7:l]) + dst[i+31:i] := idx[i+3] ? b[off+31:off] : a[off+31:off] ELSE - dst[i+31:i] := src[i+31:i] + dst[i+31:i] := idx[i+31:i] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - - Integer + + Floating Point + AVX512VL AVX512F - Convert - - - Zero extend packed unsigned 8-bit integers in "a" to packed 32-bit integers, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Miscellaneous + + + + + + Shuffle single-precision (32-bit) floating-point elements in "a" and "b" across lanes using the corresponding selector and index in "idx", and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := 32*j - l := 8*j +FOR j := 0 to 7 + i := j*32 + off := idx[i+2:i]*32 IF k[j] - dst[i+31:i] := ZeroExtend(a[l+7:l]) + dst[i+31:i] := idx[i+3] ? b[off+31:off] : a[off+31:off] ELSE - dst[i+31:i] := 0 + dst[i+31:i] := a[i+31:i] FI ENDFOR -dst[MAX:512] := 0 - - -
immintrin.h
- - - Integer - AVX512F - Convert - - Zero extend packed unsigned 8-bit integers in the low 8 byte sof "a" to packed 64-bit integers, and store the results in "dst". - -FOR j := 0 to 7 - i := 64*j - k := 8*j - dst[i+63:i] := ZeroExtend(a[k+7:k]) -ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - - Integer + + Floating Point + AVX512VL AVX512F - Convert - - - - Zero extend packed unsigned 8-bit integers in the low 8 bytes of "a" to packed 64-bit integers, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Miscellaneous + + + + + + Shuffle single-precision (32-bit) floating-point elements in "a" and "b" across lanes using the corresponding selector and index in "idx", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). FOR j := 0 to 7 - i := 64*j - l := 8*j + i := j*32 + off := idx[i+2:i]*32 IF k[j] - dst[i+63:i] := ZeroExtend(a[l+7:l]) + dst[i+31:i] := (idx[i+3]) ? b[off+31:off] : a[off+31:off] ELSE - dst[i+63:i] := src[i+63:i] + dst[i+31:i] := 0 FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - + +
immintrin.h
 - - Integer + + Floating Point + AVX512VL AVX512F - Convert - - - Zero extend packed unsigned 8-bit integers in the low 8 bytes of "a" to packed 64-bit integers, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Miscellaneous + + + + + Shuffle single-precision (32-bit) floating-point elements in "a" and "b" across lanes using the corresponding selector and index in "idx", and store the results in "dst". FOR j := 0 to 7 - i := 64*j - l := 8*j - IF k[j] - dst[i+63:i] := ZeroExtend(a[l+7:l]) - ELSE - dst[i+63:i] := 0 - FI + i := j*32 + off := idx[i+2:i]*32 + dst[i+31:i] := idx[i+3] ? b[off+31:off] : a[off+31:off] ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - + +
immintrin.h
 - - Integer + + Floating Point + AVX512VL AVX512F - Convert - - Zero extend packed unsigned 32-bit integers in "a" to packed 64-bit integers, and store the results in "dst". + Miscellaneous + + + + + + Shuffle single-precision (32-bit) floating-point elements in "a" and "b" using the corresponding selector and index in "idx", and store the results in "dst" using writemask "k" (elements are copied from "idx" when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := 64*j - k := 32*j - dst[i+63:i] := ZeroExtend(a[k+31:k]) +FOR j := 0 to 3 + i := j*32 + off := idx[i+1:i]*32 + IF k[j] + dst[i+31:i] := idx[i+2] ? b[off+31:off] : a[off+31:off] + ELSE + dst[i+31:i] := idx[i+31:i] + FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - - Integer + + Floating Point + AVX512VL AVX512F - Convert - - - - Zero extend packed unsigned 32-bit integers in "a" to packed 64-bit integers, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Miscellaneous + + + + + + Shuffle single-precision (32-bit) floating-point elements in "a" and "b" using the corresponding selector and index in "idx", and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := 64*j - l := 32*j +FOR j := 0 to 3 + i := j*32 + off := idx[i+1:i]*32 IF k[j] - dst[i+63:i] := ZeroExtend(a[l+31:l]) + dst[i+31:i] := idx[i+2] ? b[off+31:off] : a[off+31:off] ELSE - dst[i+63:i] := src[i+63:i] + dst[i+31:i] := a[i+31:i] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - - Integer + + Floating Point + AVX512VL AVX512F - Convert - - - Zero extend packed unsigned 32-bit integers in "a" to packed 64-bit integers, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Miscellaneous + + + + + + Shuffle single-precision (32-bit) floating-point elements in "a" and "b" using the corresponding selector and index in "idx", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := 64*j - l := 32*j +FOR j := 0 to 3 + i := j*32 + off := idx[i+1:i]*32 IF k[j] - dst[i+63:i] := ZeroExtend(a[l+31:l]) - ELSE - dst[i+63:i] := 0 + dst[i+31:i] := (idx[i+2]) ? b[off+31:off] : a[off+31:off] + ELSE + dst[i+31:i] := 0 FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - + +
immintrin.h
 - - Integer + + Floating Point + AVX512VL AVX512F - Convert - - Zero extend packed unsigned 16-bit integers in "a" to packed 32-bit integers, and store the results in "dst". + Miscellaneous + + + + + Shuffle single-precision (32-bit) floating-point elements in "a" and "b" using the corresponding selector and index in "idx", and store the results in "dst". -FOR j := 0 to 15 - i := 32*j - k := 16*j - dst[i+31:i] := ZeroExtend(a[k+15:k]) +FOR j := 0 to 3 + i := j*32 + off := idx[i+1:i]*32 + dst[i+31:i] := idx[i+2] ? b[off+31:off] : a[off+31:off] ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - + +
immintrin.h
 - + Integer + AVX512VL AVX512F - Convert - - - - Zero extend packed unsigned 16-bit integers in "a" to packed 32-bit integers, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Miscellaneous + + + + + + Shuffle 64-bit integers in "a" and "b" across lanes using the corresponding selector and index in "idx", and store the results in "dst" using writemask "k" (elements are copied from "idx" when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := 32*j - l := 16*j +FOR j := 0 to 3 + i := j*64 + off := idx[i+1:i]*64 IF k[j] - dst[i+31:i] := ZeroExtend(a[l+15:l]) + dst[i+63:i] := idx[i+2] ? b[off+63:off] : a[off+63:off] ELSE - dst[i+31:i] := src[i+31:i] + dst[i+63:i] := idx[i+63:i] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer + AVX512VL AVX512F - Convert - - - Zero extend packed unsigned 16-bit integers in "a" to packed 32-bit integers, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Miscellaneous + + + + + + Shuffle 64-bit integers in "a" and "b" across lanes using the corresponding selector and index in "idx", and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := 32*j - l := 16*j +FOR j := 0 to 3 + i := j*64 + off := idx[i+1:i]*64 IF k[j] - dst[i+31:i] := ZeroExtend(a[l+15:l]) + dst[i+63:i] := idx[i+2] ? b[off+63:off] : a[off+63:off] ELSE - dst[i+31:i] := 0 + dst[i+63:i] := a[i+63:i] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer + AVX512VL AVX512F - Convert - - Zero extend packed unsigned 16-bit integers in "a" to packed 64-bit integers, and store the results in "dst". + Miscellaneous + + + + + + Shuffle 64-bit integers in "a" and "b" across lanes using the corresponding selector and index in "idx", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := 64*j - k := 16*j - dst[i+63:i] := ZeroExtend(a[k+15:k]) +FOR j := 0 to 3 + i := j*64 + off := idx[i+1:i]*64 + IF k[j] + dst[i+63:i] := (idx[i+2]) ? b[off+63:off] : a[off+63:off] + ELSE + dst[i+63:i] := 0 + FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - + +
immintrin.h
 - + Integer + AVX512VL AVX512F - Convert - - - - Zero extend packed unsigned 16-bit integers in "a" to packed 64-bit integers, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Miscellaneous + + + + + Shuffle 64-bit integers in "a" and "b" across lanes using the corresponding selector and index in "idx", and store the results in "dst". -FOR j := 0 to 7 - i := 64*j - l := 16*j - IF k[j] - dst[i+63:i] := ZeroExtend(a[l+15:l]) - ELSE - dst[i+63:i] := src[i+63:i] - FI +FOR j := 0 to 3 + i := j*64 + off := idx[i+1:i]*64 + dst[i+63:i] := idx[i+2] ? b[off+63:off] : a[off+63:off] ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - + +
immintrin.h
 - + Integer + AVX512VL AVX512F - Convert - - - Zero extend packed unsigned 16-bit integers in "a" to packed 64-bit integers, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Miscellaneous + + + + + + Shuffle 64-bit integers in "a" and "b" using the corresponding selector and index in "idx", and store the results in "dst" using writemask "k" (elements are copied from "idx" when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := 64*j - l := 16*j +FOR j := 0 to 1 + i := j*64 + off := idx[i]*64 IF k[j] - dst[i+63:i] := ZeroExtend(a[l+15:l]) + dst[i+63:i] := idx[i+1] ? b[off+63:off] : a[off+63:off] ELSE - dst[i+63:i] := 0 + dst[i+63:i] := idx[i+63:i] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer + AVX512VL AVX512F - Arithmetic - - - - - Multiply the low 32-bit integers from each packed 64-bit element in "a" and "b", and store the signed 64-bit results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Miscellaneous + + + + + + Shuffle 64-bit integers in "a" and "b" using the corresponding selector and index in "idx", and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). -FOR j := 0 to 7 +FOR j := 0 to 1 i := j*64 + off := idx[i]*64 IF k[j] - dst[i+63:i] := a[i+31:i] * b[i+31:i] + dst[i+63:i] := idx[i+1] ? b[off+63:off] : a[off+63:off] ELSE - dst[i+63:i] := src[i+63:i] + dst[i+63:i] := a[i+63:i] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer + AVX512VL AVX512F - Arithmetic - - - - Multiply the low 32-bit integers from each packed 64-bit element in "a" and "b", and store the signed 64-bit results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Miscellaneous + + + + + + Shuffle 64-bit integers in "a" and "b" using the corresponding selector and index in "idx", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 +FOR j := 0 to 1 i := j*64 + off := idx[i]*64 IF k[j] - dst[i+63:i] := a[i+31:i] * b[i+31:i] + dst[i+63:i] := (idx[i+1]) ? b[off+63:off] : a[off+63:off] ELSE dst[i+63:i] := 0 FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - + +
immintrin.h
 - + Integer + AVX512VL AVX512F - Arithmetic - - - Multiply the low 32-bit integers from each packed 64-bit element in "a" and "b", and store the signed 64-bit results in "dst". + Miscellaneous + + + + + Shuffle 64-bit integers in "a" and "b" using the corresponding selector and index in "idx", and store the results in "dst". -FOR j := 0 to 7 +FOR j := 0 to 1 i := j*64 - dst[i+63:i] := a[i+31:i] * b[i+31:i] + off := idx[i]*64 + dst[i+63:i] := idx[i+1] ? b[off+63:off] : a[off+63:off] ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - + +
immintrin.h
 - - Integer - AVX512F/KNCNI - Arithmetic - - - - - Multiply the packed 32-bit integers in "a" and "b", producing intermediate 64-bit integers, and store the low 32 bits of the intermediate integers in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -FOR j := 0 to 15 - i := j*32 + + Floating Point + AVX512VL + AVX512F + Miscellaneous + + + + + + Shuffle double-precision (64-bit) floating-point elements in "a" within 128-bit lanes using the control in "imm8", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +IF (imm8[0] == 0) tmp_dst[63:0] := a[63:0]; FI +IF (imm8[0] == 1) tmp_dst[63:0] := a[127:64]; FI +IF (imm8[1] == 0) tmp_dst[127:64] := a[63:0]; FI +IF (imm8[1] == 1) tmp_dst[127:64] := a[127:64]; FI +IF (imm8[2] == 0) tmp_dst[191:128] := a[191:128]; FI +IF (imm8[2] == 1) tmp_dst[191:128] := a[255:192]; FI +IF (imm8[3] == 0) tmp_dst[255:192] := a[191:128]; FI +IF (imm8[3] == 1) tmp_dst[255:192] := a[255:192]; FI +FOR j := 0 to 3 + i := j*64 IF k[j] - tmp[63:0] := a[i+31:i] * b[i+31:i] - dst[i+31:i] := tmp[31:0] + dst[i+63:i] := tmp_dst[i+63:i] ELSE - dst[i+31:i] := src[i+31:i] + dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:512] := 0 - - -
immintrin.h
- - - Integer - AVX512F/KNCNI - Arithmetic - - - Multiply the packed 32-bit integers in "a" and "b", producing intermediate 64-bit integers, and store the low 32 bits of the intermediate integers in "dst". - -FOR j := 0 to 15 - i := j*32 - tmp[63:0] := a[i+31:i] * b[i+31:i] - dst[i+31:i] := tmp[31:0] -ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - - Integer + + Floating Point + AVX512VL AVX512F - Arithmetic - - - - - Multiply the low unsigned 32-bit integers from each packed 64-bit element in "a" and "b", and store the unsigned 64-bit results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Miscellaneous + + + + + + Shuffle double-precision (64-bit) floating-point elements in "a" within 128-bit lanes using the control in "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 7 +IF (b[1] == 0) tmp_dst[63:0] := a[63:0]; FI +IF (b[1] == 1) tmp_dst[63:0] := a[127:64]; FI +IF (b[65] == 0) tmp_dst[127:64] := a[63:0]; FI +IF (b[65] == 1) tmp_dst[127:64] := a[127:64]; FI +IF (b[129] == 0) tmp_dst[191:128] := a[191:128]; FI +IF (b[129] == 1) tmp_dst[191:128] := a[255:192]; FI +IF (b[193] == 0) tmp_dst[255:192] := a[191:128]; FI +IF (b[193] == 1) tmp_dst[255:192] := a[255:192]; FI +FOR j := 0 to 3 i := j*64 IF k[j] - dst[i+63:i] := a[i+31:i] * b[i+31:i] + dst[i+63:i] := tmp_dst[i+63:i] ELSE dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - - Integer + + Floating Point + AVX512VL AVX512F - Arithmetic - - - - Multiply the low unsigned 32-bit integers from each packed 64-bit element in "a" and "b", and store the unsigned 64-bit results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Miscellaneous + + + + + Shuffle double-precision (64-bit) floating-point elements in "a" within 128-bit lanes using the control in "imm8", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 +IF (imm8[0] == 0) tmp_dst[63:0] := a[63:0]; FI +IF (imm8[0] == 1) tmp_dst[63:0] := a[127:64]; FI +IF (imm8[1] == 0) tmp_dst[127:64] := a[63:0]; FI +IF (imm8[1] == 1) tmp_dst[127:64] := a[127:64]; FI +IF (imm8[2] == 0) tmp_dst[191:128] := a[191:128]; FI +IF (imm8[2] == 1) tmp_dst[191:128] := a[255:192]; FI +IF (imm8[3] == 0) tmp_dst[255:192] := a[191:128]; FI +IF (imm8[3] == 1) tmp_dst[255:192] := a[255:192]; FI +FOR j := 0 to 3 i := j*64 IF k[j] - dst[i+63:i] := a[i+31:i] * b[i+31:i] + dst[i+63:i] := tmp_dst[i+63:i] ELSE dst[i+63:i] := 0 FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - - Integer + + Floating Point + AVX512VL AVX512F - Arithmetic - - - Multiply the low unsigned 32-bit integers from each packed 64-bit element in "a" and "b", and store the unsigned 64-bit results in "dst". + Miscellaneous + + + + + Shuffle double-precision (64-bit) floating-point elements in "a" within 128-bit lanes using the control in "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 +IF (b[1] == 0) tmp_dst[63:0] := a[63:0]; FI +IF (b[1] == 1) tmp_dst[63:0] := a[127:64]; FI +IF (b[65] == 0) tmp_dst[127:64] := a[63:0]; FI +IF (b[65] == 1) tmp_dst[127:64] := a[127:64]; FI +IF (b[129] == 0) tmp_dst[191:128] := a[191:128]; FI +IF (b[129] == 1) tmp_dst[191:128] := a[255:192]; FI +IF (b[193] == 0) tmp_dst[255:192] := a[191:128]; FI +IF (b[193] == 1) tmp_dst[255:192] := a[255:192]; FI +FOR j := 0 to 3 i := j*64 - dst[i+63:i] := a[i+31:i] * b[i+31:i] -ENDFOR -dst[MAX:512] := 0 - - -
immintrin.h
- - - Integer - AVX512F/KNCNI - Logical - - - - - Compute the bitwise OR of packed 32-bit integers in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -FOR j := 0 to 15 - i := j*32 IF k[j] - dst[i+31:i] := a[i+31:i] OR b[i+31:i] + dst[i+63:i] := tmp_dst[i+63:i] ELSE - dst[i+31:i] := src[i+31:i] + dst[i+63:i] := 0 FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - - Integer + + Floating Point + AVX512VL AVX512F - Logical - - - - Compute the bitwise OR of packed 32-bit integers in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Miscellaneous + + + + + + Shuffle double-precision (64-bit) floating-point elements in "a" using the control in "imm8", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*32 +IF (imm8[0] == 0) tmp_dst[63:0] := a[63:0]; FI +IF (imm8[0] == 1) tmp_dst[63:0] := a[127:64]; FI +IF (imm8[1] == 0) tmp_dst[127:64] := a[63:0]; FI +IF (imm8[1] == 1) tmp_dst[127:64] := a[127:64]; FI +FOR j := 0 to 1 + i := j*64 IF k[j] - dst[i+31:i] := a[i+31:i] OR b[i+31:i] + dst[i+63:i] := tmp_dst[i+63:i] ELSE - dst[i+31:i] := 0 + dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:512] := 0 - - -
immintrin.h
- - - Integer - AVX512F/KNCNI - Logical - - - Compute the bitwise OR of packed 32-bit integers in "a" and "b", and store the results in "dst". - -FOR j := 0 to 15 - i := j*32 - dst[i+31:i] := a[i+31:i] OR b[i+31:i] -ENDFOR -dst[MAX:512] := 0 - - -
immintrin.h
- - - Integer - AVX512F/KNCNI - Logical - - - Compute the bitwise OR of 512 bits (representing integer data) in "a" and "b", and store the result in "dst". - -dst[511:0] := (a[511:0] OR b[511:0]) -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - - Integer - AVX512F/KNCNI - Logical - - - - - Compute the bitwise OR of packed 64-bit integers in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Floating Point + AVX512VL + AVX512F + Miscellaneous + + + + + + Shuffle double-precision (64-bit) floating-point elements in "a" using the control in "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 7 +IF (b[1] == 0) tmp_dst[63:0] := a[63:0]; FI +IF (b[1] == 1) tmp_dst[63:0] := a[127:64]; FI +IF (b[65] == 0) tmp_dst[127:64] := a[63:0]; FI +IF (b[65] == 1) tmp_dst[127:64] := a[127:64]; FI +FOR j := 0 to 1 i := j*64 IF k[j] - dst[i+63:i] := a[i+63:i] OR b[i+63:i] + dst[i+63:i] := tmp_dst[i+63:i] ELSE dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - - Integer + + Floating Point + AVX512VL AVX512F - Logical - - - - Compute the bitwise OR of packed 64-bit integers in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Miscellaneous + + + + + Shuffle double-precision (64-bit) floating-point elements in "a" using the control in "imm8", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 +IF (imm8[0] == 0) tmp_dst[63:0] := a[63:0]; FI +IF (imm8[0] == 1) tmp_dst[63:0] := a[127:64]; FI +IF (imm8[1] == 0) tmp_dst[127:64] := a[63:0]; FI +IF (imm8[1] == 1) tmp_dst[127:64] := a[127:64]; FI +FOR j := 0 to 1 i := j*64 IF k[j] - dst[i+63:i] := a[i+63:i] OR b[i+63:i] + dst[i+63:i] := tmp_dst[i+63:i] ELSE dst[i+63:i] := 0 FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - - Integer - AVX512F/KNCNI - Logical - - - Compute the bitwise OR of packed 64-bit integers in "a" and "b", and store the resut in "dst". - + + Floating Point + AVX512VL + AVX512F + Miscellaneous + + + + + Shuffle double-precision (64-bit) floating-point elements in "a" using the control in "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 +IF (b[1] == 0) tmp_dst[63:0] := a[63:0]; FI +IF (b[1] == 1) tmp_dst[63:0] := a[127:64]; FI +IF (b[65] == 0) tmp_dst[127:64] := a[63:0]; FI +IF (b[65] == 1) tmp_dst[127:64] := a[127:64]; FI +FOR j := 0 to 1 i := j*64 - dst[i+63:i] := a[i+63:i] OR b[i+63:i] + IF k[j] + dst[i+63:i] := tmp_dst[i+63:i] + ELSE + dst[i+63:i] := 0 + FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - - Integer + + Floating Point + AVX512VL AVX512F - Shift - - - - - Rotate the bits in each packed 32-bit integer in "a" to the left by the number of bits specified in "imm8", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -LEFT_ROTATE_DWORDS(src, count_src){ - count := count_src modulo 32 - RETURN (src << count) OR (src >> (32 - count)) + Miscellaneous + + + + + + Shuffle single-precision (32-bit) floating-point elements in "a" within 128-bit lanes using the control in "imm8", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +DEFINE SELECT4(src, control) { + CASE(control[1:0]) OF + 0: tmp[31:0] := src[31:0] + 1: tmp[31:0] := src[63:32] + 2: tmp[31:0] := src[95:64] + 3: tmp[31:0] := src[127:96] + ESAC + RETURN tmp[31:0] } -FOR j := 0 to 15 +tmp_dst[31:0] := SELECT4(a[127:0], imm8[1:0]) +tmp_dst[63:32] := SELECT4(a[127:0], imm8[3:2]) +tmp_dst[95:64] := SELECT4(a[127:0], imm8[5:4]) +tmp_dst[127:96] := SELECT4(a[127:0], imm8[7:6]) +tmp_dst[159:128] := SELECT4(a[255:128], imm8[1:0]) +tmp_dst[191:160] := SELECT4(a[255:128], imm8[3:2]) +tmp_dst[223:192] := SELECT4(a[255:128], imm8[5:4]) +tmp_dst[255:224] := SELECT4(a[255:128], imm8[7:6]) +FOR j := 0 to 7 i := j*32 IF k[j] - dst[i+31:i] := LEFT_ROTATE_DWORDS(a[i+31:i], imm8[7:0]) + dst[i+31:i] := tmp_dst[i+31:i] ELSE dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - - Integer + + Floating Point + AVX512VL AVX512F - Shift - - - - Rotate the bits in each packed 32-bit integer in "a" to the left by the number of bits specified in "imm8", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Miscellaneous + + + + + + Shuffle single-precision (32-bit) floating-point elements in "a" within 128-bit lanes using the control in "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -LEFT_ROTATE_DWORDS(src, count_src){ - count := count_src modulo 32 - RETURN (src << count) OR (src >> (32 - count)) +DEFINE SELECT4(src, control) { + CASE(control[1:0]) OF + 0: tmp[31:0] := src[31:0] + 1: tmp[31:0] := src[63:32] + 2: tmp[31:0] := src[95:64] + 3: tmp[31:0] := src[127:96] + ESAC + RETURN tmp[31:0] } -FOR j := 0 to 15 +tmp_dst[31:0] := SELECT4(a[127:0], b[1:0]) +tmp_dst[63:32] := SELECT4(a[127:0], b[33:32]) +tmp_dst[95:64] := SELECT4(a[127:0], b[65:64]) +tmp_dst[127:96] := SELECT4(a[127:0], b[97:96]) +tmp_dst[159:128] := SELECT4(a[255:128], b[129:128]) +tmp_dst[191:160] := SELECT4(a[255:128], b[161:160]) +tmp_dst[223:192] := SELECT4(a[255:128], b[193:192]) +tmp_dst[255:224] := SELECT4(a[255:128], b[225:224]) +FOR j := 0 to 7 i := j*32 IF k[j] - dst[i+31:i] := LEFT_ROTATE_DWORDS(a[i+31:i], imm8[7:0]) + dst[i+31:i] := tmp_dst[i+31:i] ELSE - dst[i+31:i] := 0 + dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - - Integer + + Floating Point + AVX512VL AVX512F - Shift - - - Rotate the bits in each packed 32-bit integer in "a" to the left by the number of bits specified in "imm8", and store the results in "dst". + Miscellaneous + + + + + Shuffle single-precision (32-bit) floating-point elements in "a" within 128-bit lanes using the control in "imm8", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -LEFT_ROTATE_DWORDS(src, count_src){ - count := count_src modulo 32 - RETURN (src << count) OR (src >> (32 - count)) -} -FOR j := 0 to 15 - i := j*32 - dst[i+31:i] := LEFT_ROTATE_DWORDS(a[i+31:i], imm8[7:0]) -ENDFOR -dst[MAX:512] := 0 - - -
immintrin.h
- - - Integer - AVX512F - Shift - - - - - Rotate the bits in each packed 64-bit integer in "a" to the left by the number of bits specified in "imm8", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -LEFT_ROTATE_QWORDS(src, count_src){ - count := count_src modulo 64 - RETURN (src << count) OR (src >> (64 - count)) +DEFINE SELECT4(src, control) { + CASE(control[1:0]) OF + 0: tmp[31:0] := src[31:0] + 1: tmp[31:0] := src[63:32] + 2: tmp[31:0] := src[95:64] + 3: tmp[31:0] := src[127:96] + ESAC + RETURN tmp[31:0] } +tmp_dst[31:0] := SELECT4(a[127:0], imm8[1:0]) +tmp_dst[63:32] := SELECT4(a[127:0], imm8[3:2]) +tmp_dst[95:64] := SELECT4(a[127:0], imm8[5:4]) +tmp_dst[127:96] := SELECT4(a[127:0], imm8[7:6]) +tmp_dst[159:128] := SELECT4(a[255:128], imm8[1:0]) +tmp_dst[191:160] := SELECT4(a[255:128], imm8[3:2]) +tmp_dst[223:192] := SELECT4(a[255:128], imm8[5:4]) +tmp_dst[255:224] := SELECT4(a[255:128], imm8[7:6]) FOR j := 0 to 7 - i := j*64 + i := j*32 IF k[j] - dst[i+63:i] := LEFT_ROTATE_QWORDS(a[i+63:i], imm8[7:0]) + dst[i+31:i] := tmp_dst[i+31:i] ELSE - dst[i+63:i] := src[i+63:i] + dst[i+31:i] := 0 FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - - Integer + + Floating Point + AVX512VL AVX512F - Shift - - - - Rotate the bits in each packed 64-bit integer in "a" to the left by the number of bits specified in "imm8", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Miscellaneous + + + + + Shuffle single-precision (32-bit) floating-point elements in "a" within 128-bit lanes using the control in "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -LEFT_ROTATE_QWORDS(src, count_src){ - count := count_src modulo 64 - RETURN (src << count) OR (src >> (64 - count)) +DEFINE SELECT4(src, control) { + CASE(control[1:0]) OF + 0: tmp[31:0] := src[31:0] + 1: tmp[31:0] := src[63:32] + 2: tmp[31:0] := src[95:64] + 3: tmp[31:0] := src[127:96] + ESAC + RETURN tmp[31:0] } +tmp_dst[31:0] := SELECT4(a[127:0], b[1:0]) +tmp_dst[63:32] := SELECT4(a[127:0], b[33:32]) +tmp_dst[95:64] := SELECT4(a[127:0], b[65:64]) +tmp_dst[127:96] := SELECT4(a[127:0], b[97:96]) +tmp_dst[159:128] := SELECT4(a[255:128], b[129:128]) +tmp_dst[191:160] := SELECT4(a[255:128], b[161:160]) +tmp_dst[223:192] := SELECT4(a[255:128], b[193:192]) +tmp_dst[255:224] := SELECT4(a[255:128], b[225:224]) FOR j := 0 to 7 - i := j*64 + i := j*32 IF k[j] - dst[i+63:i] := LEFT_ROTATE_QWORDS(a[i+63:i], imm8[7:0]) + dst[i+31:i] := tmp_dst[i+31:i] ELSE - dst[i+63:i] := 0 + dst[i+31:i] := 0 FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - - Integer + + Floating Point + AVX512VL AVX512F - Shift - - - Rotate the bits in each packed 64-bit integer in "a" to the left by the number of bits specified in "imm8", and store the results in "dst". + Miscellaneous + + + + + + Shuffle single-precision (32-bit) floating-point elements in "a" using the control in "imm8", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -LEFT_ROTATE_QWORDS(src, count_src){ - count := count_src modulo 64 - RETURN (src << count) OR (src >> (64 - count)) +DEFINE SELECT4(src, control) { + CASE(control[1:0]) OF + 0: tmp[31:0] := src[31:0] + 1: tmp[31:0] := src[63:32] + 2: tmp[31:0] := src[95:64] + 3: tmp[31:0] := src[127:96] + ESAC + RETURN tmp[31:0] } -FOR j := 0 to 7 - i := j*64 - dst[i+63:i] := LEFT_ROTATE_QWORDS(a[i+63:i], imm8[7:0]) +tmp_dst[31:0] := SELECT4(a[127:0], imm8[1:0]) +tmp_dst[63:32] := SELECT4(a[127:0], imm8[3:2]) +tmp_dst[95:64] := SELECT4(a[127:0], imm8[5:4]) +tmp_dst[127:96] := SELECT4(a[127:0], imm8[7:6]) +FOR j := 0 to 3 + i := j*32 + IF k[j] + dst[i+31:i] := tmp_dst[i+31:i] + ELSE + dst[i+31:i] := src[i+31:i] + FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - - Integer + + Floating Point + AVX512VL AVX512F - Shift - - - - - Rotate the bits in each packed 32-bit integer in "a" to the left by the number of bits specified in the corresponding element of "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -LEFT_ROTATE_DWORDS(src, count_src){ - count := count_src modulo 32 - RETURN (src << count) OR (src >> (32 - count)) + Miscellaneous + + + + + + Shuffle single-precision (32-bit) floating-point elements in "a" using the control in "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +DEFINE SELECT4(src, control) { + CASE(control[1:0]) OF + 0: tmp[31:0] := src[31:0] + 1: tmp[31:0] := src[63:32] + 2: tmp[31:0] := src[95:64] + 3: tmp[31:0] := src[127:96] + ESAC + RETURN tmp[31:0] } -FOR j := 0 to 15 +tmp_dst[31:0] := SELECT4(a[127:0], b[1:0]) +tmp_dst[63:32] := SELECT4(a[127:0], b[33:32]) +tmp_dst[95:64] := SELECT4(a[127:0], b[65:64]) +tmp_dst[127:96] := SELECT4(a[127:0], b[97:96]) +FOR j := 0 to 3 i := j*32 IF k[j] - dst[i+31:i] := LEFT_ROTATE_DWORDS(a[i+31:i], b[i+31:i]) + dst[i+31:i] := tmp_dst[i+31:i] ELSE dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - - Integer + + Floating Point + AVX512VL AVX512F - Shift - - - - Rotate the bits in each packed 32-bit integer in "a" to the left by the number of bits specified in the corresponding element of "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Miscellaneous + + + + + Shuffle single-precision (32-bit) floating-point elements in "a" using the control in "imm8", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -LEFT_ROTATE_DWORDS(src, count_src){ - count := count_src modulo 32 - RETURN (src << count) OR (src >> (32 - count)) +DEFINE SELECT4(src, control) { + CASE(control[1:0]) OF + 0: tmp[31:0] := src[31:0] + 1: tmp[31:0] := src[63:32] + 2: tmp[31:0] := src[95:64] + 3: tmp[31:0] := src[127:96] + ESAC + RETURN tmp[31:0] } -FOR j := 0 to 15 +tmp_dst[31:0] := SELECT4(a[127:0], imm8[1:0]) +tmp_dst[63:32] := SELECT4(a[127:0], imm8[3:2]) +tmp_dst[95:64] := SELECT4(a[127:0], imm8[5:4]) +tmp_dst[127:96] := SELECT4(a[127:0], imm8[7:6]) +FOR j := 0 to 3 i := j*32 IF k[j] - dst[i+31:i] := LEFT_ROTATE_DWORDS(a[i+31:i], b[i+31:i]) + dst[i+31:i] := tmp_dst[i+31:i] ELSE dst[i+31:i] := 0 FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - - Integer + + Floating Point + AVX512VL AVX512F - Shift - - - Rotate the bits in each packed 32-bit integer in "a" to the left by the number of bits specified in the corresponding element of "b", and store the results in "dst". + Miscellaneous + + + + + Shuffle single-precision (32-bit) floating-point elements in "a" using the control in "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -LEFT_ROTATE_DWORDS(src, count_src){ - count := count_src modulo 32 - RETURN (src << count) OR (src >> (32 - count)) +DEFINE SELECT4(src, control) { + CASE(control[1:0]) OF + 0: tmp[31:0] := src[31:0] + 1: tmp[31:0] := src[63:32] + 2: tmp[31:0] := src[95:64] + 3: tmp[31:0] := src[127:96] + ESAC + RETURN tmp[31:0] } -FOR j := 0 to 15 +tmp_dst[31:0] := SELECT4(a[127:0], b[1:0]) +tmp_dst[63:32] := SELECT4(a[127:0], b[33:32]) +tmp_dst[95:64] := SELECT4(a[127:0], b[65:64]) +tmp_dst[127:96] := SELECT4(a[127:0], b[97:96]) +FOR j := 0 to 3 i := j*32 - dst[i+31:i] := LEFT_ROTATE_DWORDS(a[i+31:i], b[i+31:i]) + IF k[j] + dst[i+31:i] := tmp_dst[i+31:i] + ELSE + dst[i+31:i] := 0 + FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - - Integer + + Floating Point + AVX512VL AVX512F - Shift - - - - - Rotate the bits in each packed 64-bit integer in "a" to the left by the number of bits specified in the corresponding element of "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -LEFT_ROTATE_QWORDS(src, count_src){ - count := count_src modulo 64 - RETURN (src << count) OR (src >> (64 - count)) + Miscellaneous + + + + + + Shuffle double-precision (64-bit) floating-point elements in "a" across lanes using the control in "imm8", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +DEFINE SELECT4(src, control) { + CASE(control[1:0]) OF + 0: tmp[63:0] := src[63:0] + 1: tmp[63:0] := src[127:64] + 2: tmp[63:0] := src[191:128] + 3: tmp[63:0] := src[255:192] + ESAC + RETURN tmp[63:0] } -FOR j := 0 to 7 +tmp_dst[63:0] := SELECT4(a[255:0], imm8[1:0]) +tmp_dst[127:64] := SELECT4(a[255:0], imm8[3:2]) +tmp_dst[191:128] := SELECT4(a[255:0], imm8[5:4]) +tmp_dst[255:192] := SELECT4(a[255:0], imm8[7:6]) +FOR j := 0 to 3 i := j*64 IF k[j] - dst[i+63:i] := LEFT_ROTATE_QWORDS(a[i+63:i], b[i+63:i]) + dst[i+63:i] := tmp_dst[i+63:i] ELSE dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - - Integer + + Floating Point + AVX512VL AVX512F - Shift - - - - Rotate the bits in each packed 64-bit integer in "a" to the left by the number of bits specified in the corresponding element of "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Miscellaneous + + + + + + Shuffle double-precision (64-bit) floating-point elements in "a" across lanes using the corresponding index in "idx", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -LEFT_ROTATE_QWORDS(src, count_src){ - count := count_src modulo 64 - RETURN (src << count) OR (src >> (64 - count)) -} -FOR j := 0 to 7 +FOR j := 0 to 3 i := j*64 + id := idx[i+1:i]*64 IF k[j] - dst[i+63:i] := LEFT_ROTATE_QWORDS(a[i+63:i], b[i+63:i]) + dst[i+63:i] := a[id+63:id] ELSE - dst[i+63:i] := 0 + dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - - Integer + + Floating Point + AVX512VL AVX512F - Shift - - - Rotate the bits in each packed 64-bit integer in "a" to the left by the number of bits specified in the corresponding element of "b", and store the results in "dst". + Miscellaneous + + + + + Shuffle double-precision (64-bit) floating-point elements in "a" across lanes using the control in "imm8", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -LEFT_ROTATE_QWORDS(src, count_src){ - count := count_src modulo 64 - RETURN (src << count) OR (src >> (64 - count)) +DEFINE SELECT4(src, control) { + CASE(control[1:0]) OF + 0: tmp[63:0] := src[63:0] + 1: tmp[63:0] := src[127:64] + 2: tmp[63:0] := src[191:128] + 3: tmp[63:0] := src[255:192] + ESAC + RETURN tmp[63:0] } -FOR j := 0 to 7 +tmp_dst[63:0] := SELECT4(a[255:0], imm8[1:0]) +tmp_dst[127:64] := SELECT4(a[255:0], imm8[3:2]) +tmp_dst[191:128] := SELECT4(a[255:0], imm8[5:4]) +tmp_dst[255:192] := SELECT4(a[255:0], imm8[7:6]) +FOR j := 0 to 3 i := j*64 - dst[i+63:i] := LEFT_ROTATE_QWORDS(a[i+63:i], b[i+63:i]) -ENDFOR -dst[MAX:512] := 0 - - -
immintrin.h
- - - Integer - AVX512F - Shift - - - - - Rotate the bits in each packed 32-bit integer in "a" to the right by the number of bits specified in "imm8", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -RIGHT_ROTATE_DWORDS(src, count_src){ - count := count_src modulo 32 - RETURN (src >>count) OR (src << (32 - count)) -} -FOR j := 0 to 15 - i := j*32 IF k[j] - dst[i+31:i] := RIGHT_ROTATE_DWORDS(a[i+31:i], imm8[7:0]) + dst[i+63:i] := tmp_dst[i+63:i] ELSE - dst[i+31:i] := src[i+31:i] + dst[i+63:i] := 0 FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - - Integer + + Floating Point + AVX512VL AVX512F - Shift - - - - Rotate the bits in each packed 32-bit integer in "a" to the right by the number of bits specified in "imm8", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Miscellaneous + + + + + Shuffle double-precision (64-bit) floating-point elements in "a" across lanes using the corresponding index in "idx", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -RIGHT_ROTATE_DWORDS(src, count_src){ - count := count_src modulo 32 - RETURN (src >>count) OR (src << (32 - count)) -} -FOR j := 0 to 15 - i := j*32 +FOR j := 0 to 3 + i := j*64 + id := idx[i+1:i]*64 IF k[j] - dst[i+31:i] := RIGHT_ROTATE_DWORDS(a[i+31:i], imm8[7:0]) + dst[i+63:i] := a[id+63:id] ELSE - dst[i+31:i] := 0 + dst[i+63:i] := 0 FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - - Integer + + Floating Point + AVX512VL AVX512F - Shift - - - Rotate the bits in each packed 32-bit integer in "a" to the right by the number of bits specified in "imm8", and store the results in "dst". + Miscellaneous + + + + Shuffle double-precision (64-bit) floating-point elements in "a" across lanes using the control in "imm8", and store the results in "dst". -RIGHT_ROTATE_DWORDS(src, count_src){ - count := count_src modulo 32 - RETURN (src >>count) OR (src << (32 - count)) +DEFINE SELECT4(src, control) { + CASE(control[1:0]) OF + 0: tmp[63:0] := src[63:0] + 1: tmp[63:0] := src[127:64] + 2: tmp[63:0] := src[191:128] + 3: tmp[63:0] := src[255:192] + ESAC + RETURN tmp[63:0] } -FOR j := 0 to 15 - i := j*32 - dst[i+31:i] := RIGHT_ROTATE_DWORDS(a[i+31:i], imm8[7:0]) -ENDFOR -dst[MAX:512] := 0 - - -
immintrin.h
- - - Integer - AVX512F - Shift - - - - - Rotate the bits in each packed 64-bit integer in "a" to the right by the number of bits specified in "imm8", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -RIGHT_ROTATE_QWORDS(src, count_src){ - count := count_src modulo 64 - RETURN (src >> count) OR (src << (64 - count)) -} -FOR j := 0 to 7 - i := j*64 - IF k[j] - dst[i+63:i] := RIGHT_ROTATE_QWORDS(a[i+63:i], imm8[7:0]) - ELSE - dst[i+63:i] := src[i+63:i] - FI -ENDFOR -dst[MAX:512] := 0 +dst[63:0] := SELECT4(a[255:0], imm8[1:0]) +dst[127:64] := SELECT4(a[255:0], imm8[3:2]) +dst[191:128] := SELECT4(a[255:0], imm8[5:4]) +dst[255:192] := SELECT4(a[255:0], imm8[7:6]) +dst[MAX:256] := 0 - +
immintrin.h
 - - Integer + + Floating Point + AVX512VL AVX512F - Shift - - - - Rotate the bits in each packed 64-bit integer in "a" to the right by the number of bits specified in "imm8", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Miscellaneous + + + + Shuffle double-precision (64-bit) floating-point elements in "a" across lanes using the corresponding index in "idx", and store the results in "dst". -RIGHT_ROTATE_QWORDS(src, count_src){ - count := count_src modulo 64 - RETURN (src >> count) OR (src << (64 - count)) -} -FOR j := 0 to 7 +FOR j := 0 to 3 i := j*64 - IF k[j] - dst[i+63:i] := RIGHT_ROTATE_QWORDS(a[i+63:i], imm8[7:0]) - ELSE - dst[i+63:i] := 0 - FI + id := idx[i+1:i]*64 + dst[i+63:i] := a[id+63:id] ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - - Integer + + Floating Point + AVX512VL AVX512F - Shift - - - Rotate the bits in each packed 64-bit integer in "a" to the right by the number of bits specified in "imm8", and store the results in "dst". + Miscellaneous + + + + + + Shuffle single-precision (32-bit) floating-point elements in "a" across lanes using the corresponding index in "idx", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -RIGHT_ROTATE_QWORDS(src, count_src){ - count := count_src modulo 64 - RETURN (src >> count) OR (src << (64 - count)) -} FOR j := 0 to 7 - i := j*64 - dst[i+63:i] := RIGHT_ROTATE_QWORDS(a[i+63:i], imm8[7:0]) -ENDFOR -dst[MAX:512] := 0 - - -
immintrin.h
- - - Integer - AVX512F - Shift - - - - - Rotate the bits in each packed 32-bit integer in "a" to the right by the number of bits specified in the corresponding element of "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -RIGHT_ROTATE_DWORDS(src, count_src){ - count := count_src modulo 32 - RETURN (src >>count) OR (src << (32 - count)) -} -FOR j := 0 to 15 i := j*32 + id := idx[i+2:i]*32 IF k[j] - dst[i+31:i] := RIGHT_ROTATE_DWORDS(a[i+31:i], b[i+31:i]) + dst[i+31:i] := a[id+31:id] ELSE dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - - Integer + + Floating Point + AVX512VL AVX512F Miscellaneous - - - - Rotate the bits in each packed 32-bit integer in "a" to the right by the number of bits specified in the corresponding element of "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + + + + Shuffle single-precision (32-bit) floating-point elements in "a" across lanes using the corresponding index in "idx", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -RIGHT_ROTATE_DWORDS(src, count_src){ - count := count_src modulo 32 - RETURN (src >>count) OR (src << (32 - count)) -} -FOR j := 0 to 15 +FOR j := 0 to 7 i := j*32 + id := idx[i+2:i]*32 IF k[j] - dst[i+31:i] := RIGHT_ROTATE_DWORDS(a[i+31:i], b[i+31:i]) + dst[i+31:i] := a[id+31:id] ELSE dst[i+31:i] := 0 FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - - Integer + + Floating Point + AVX512VL AVX512F - Shift - - - Rotate the bits in each packed 32-bit integer in "a" to the right by the number of bits specified in the corresponding element of "b", and store the results in "dst". + Miscellaneous + + + + Shuffle single-precision (32-bit) floating-point elements in "a" across lanes using the corresponding index in "idx". -RIGHT_ROTATE_DWORDS(src, count_src){ - count := count_src modulo 32 - RETURN (src >>count) OR (src << (32 - count)) -} -FOR j := 0 to 15 +FOR j := 0 to 7 i := j*32 - dst[i+31:i] := RIGHT_ROTATE_DWORDS(a[i+31:i], b[i+31:i]) + id := idx[i+2:i]*32 + dst[i+31:i] := a[id+31:id] ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer + AVX512VL AVX512F - Shift - - - - - Rotate the bits in each packed 64-bit integer in "a" to the right by the number of bits specified in the corresponding element of "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -RIGHT_ROTATE_QWORDS(src, count_src){ - count := count_src modulo 64 - RETURN (src >> count) OR (src << (64 - count)) + Miscellaneous + + + + + + Shuffle 64-bit integers in "a" across lanes lanes using the control in "imm8", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +DEFINE SELECT4(src, control) { + CASE(control[1:0]) OF + 0: tmp[63:0] := src[63:0] + 1: tmp[63:0] := src[127:64] + 2: tmp[63:0] := src[191:128] + 3: tmp[63:0] := src[255:192] + ESAC + RETURN tmp[63:0] } -FOR j := 0 to 7 +tmp_dst[63:0] := SELECT4(a[255:0], imm8[1:0]) +tmp_dst[127:64] := SELECT4(a[255:0], imm8[3:2]) +tmp_dst[191:128] := SELECT4(a[255:0], imm8[5:4]) +tmp_dst[255:192] := SELECT4(a[255:0], imm8[7:6]) +FOR j := 0 to 3 i := j*64 IF k[j] - dst[i+63:i] := RIGHT_ROTATE_QWORDS(a[i+63:i], b[i+63:i]) + dst[i+63:i] := tmp_dst[i+63:i] ELSE dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer + AVX512VL AVX512F - Shift - - - - Rotate the bits in each packed 64-bit integer in "a" to the right by the number of bits specified in the corresponding element of "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Miscellaneous + + + + + + Shuffle 64-bit integers in "a" across lanes using the corresponding index in "idx", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -RIGHT_ROTATE_QWORDS(src, count_src){ - count := count_src modulo 64 - RETURN (src >> count) OR (src << (64 - count)) -} -FOR j := 0 to 7 +FOR j := 0 to 3 i := j*64 + id := idx[i+1:i]*64 IF k[j] - dst[i+63:i] := RIGHT_ROTATE_QWORDS(a[i+63:i], b[i+63:i]) + dst[i+63:i] := a[id+63:id] ELSE - dst[i+63:i] := 0 + dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer + AVX512VL AVX512F - Shift - - - Rotate the bits in each packed 64-bit integer in "a" to the right by the number of bits specified in the corresponding element of "b", and store the results in "dst". + Miscellaneous + + + + + Shuffle 64-bit integers in "a" across lanes using the control in "imm8", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -RIGHT_ROTATE_QWORDS(src, count_src){ - count := count_src modulo 64 - RETURN (src >> count) OR (src << (64 - count)) +DEFINE SELECT4(src, control) { + CASE(control[1:0]) OF + 0: tmp[63:0] := src[63:0] + 1: tmp[63:0] := src[127:64] + 2: tmp[63:0] := src[191:128] + 3: tmp[63:0] := src[255:192] + ESAC + RETURN tmp[63:0] } -FOR j := 0 to 7 +tmp_dst[63:0] := SELECT4(a[255:0], imm8[1:0]) +tmp_dst[127:64] := SELECT4(a[255:0], imm8[3:2]) +tmp_dst[191:128] := SELECT4(a[255:0], imm8[5:4]) +tmp_dst[255:192] := SELECT4(a[255:0], imm8[7:6]) +FOR j := 0 to 3 i := j*64 - dst[i+63:i] := RIGHT_ROTATE_QWORDS(a[i+63:i], b[i+63:i]) -ENDFOR -dst[MAX:512] := 0 - - -
immintrin.h
- - - Integer - AVX512F/KNCNI - Store - - - - - Scatter 32-bit integers from "a" into memory using 32-bit indices. 32-bit elements are stored at addresses starting at "base_addr" and offset by each 32-bit element in "vindex" (each index is scaled by the factor in "scale"). "scale" should be 1, 2, 4 or 8. - -FOR j := 0 to 15 - i := j*32 - MEM[base_addr + SignExtend(vindex[i+31:i])*scale] := a[i+31:i] -ENDFOR - - -
immintrin.h
- - - Integer - AVX512F/KNCNI - Store - - - - - - Scatter 32-bit integers from "a" into memory using 32-bit indices. 32-bit elements are stored at addresses starting at "base_addr" and offset by each 32-bit element in "vindex" (each index is scaled by the factor in "scale") subject to mask "k" (elements are not stored when the corresponding mask bit is not set). "scale" should be 1, 2, 4 or 8. - - -FOR j := 0 to 15 - i := j*32 IF k[j] - MEM[base_addr + SignExtend(vindex[i+31:i])*scale] := a[i+31:i] - k[j] := 0 + dst[i+63:i] := tmp_dst[i+63:i] + ELSE + dst[i+63:i] := 0 FI ENDFOR -k[MAX:16] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer + AVX512VL AVX512F - Store - - - - - Scatter 64-bit integers from "a" into memory using 32-bit indices. 64-bit elements are stored at addresses starting at "base_addr" and offset by each 32-bit element in "vindex" (each index is scaled by the factor in "scale"). "scale" should be 1, 2, 4 or 8. - + Miscellaneous + + + + + Shuffle 64-bit integers in "a" across lanes using the corresponding index in "idx", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 +FOR j := 0 to 3 i := j*64 - l := j*32 - MEM[base_addr + SignExtend(vindex[l+31:l])*scale] := a[i+63:i] + id := idx[i+1:i]*64 + IF k[j] + dst[i+63:i] := a[id+63:id] + ELSE + dst[i+63:i] := 0 + FI ENDFOR +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer + AVX512VL AVX512F - Store - - - - - - Scatter 64-bit integers from "a" into memory using 32-bit indices. 64-bit elements are stored at addresses starting at "base_addr" and offset by each 32-bit element in "vindex" (each index is scaled by the factor in "scale") subject to mask "k" (elements are not stored when the corresponding mask bit is not set). "scale" should be 1, 2, 4 or 8. - + Miscellaneous + + + + Shuffle 64-bit integers in "a" across lanes using the control in "imm8", and store the results in "dst". -FOR j := 0 to 7 - i := j*64 - l := j*32 - IF k[j] - MEM[base_addr + SignExtend(vindex[l+31:l])*scale] := a[i+63:i] - k[j] := 0 - FI -ENDFOR -k[MAX:8] := 0 +DEFINE SELECT4(src, control) { + CASE(control[1:0]) OF + 0: tmp[63:0] := src[63:0] + 1: tmp[63:0] := src[127:64] + 2: tmp[63:0] := src[191:128] + 3: tmp[63:0] := src[255:192] + ESAC + RETURN tmp[63:0] +} +dst[63:0] := SELECT4(a[255:0], imm8[1:0]) +dst[127:64] := SELECT4(a[255:0], imm8[3:2]) +dst[191:128] := SELECT4(a[255:0], imm8[5:4]) +dst[255:192] := SELECT4(a[255:0], imm8[7:6]) +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer + AVX512VL AVX512F - Store - - - - - Scatter 32-bit integers from "a" into memory using 64-bit indices. 32-bit elements are stored at addresses starting at "base_addr" and offset by each 64-bit element in "vindex" (each index is scaled by the factor in "scale"). "scale" should be 1, 2, 4 or 8. - + Miscellaneous + + + + Shuffle 64-bit integers in "a" across lanes using the corresponding index in "idx", and store the results in "dst". -FOR j := 0 to 7 - i := j*32 - l := j*64 - MEM[base_addr + SignExtend(vindex[l+63:l])*scale] := a[i+31:i] +FOR j := 0 to 3 + i := j*64 + id := idx[i+1:i]*64 + dst[i+63:i] := a[id+63:id] ENDFOR +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer + AVX512VL AVX512F - Store - - - - - - Scatter 32-bit integers from "a" into memory using 64-bit indices. 32-bit elements are stored at addresses starting at "base_addr" and offset by each 64-bit element in "vindex" (each index is scaled by the factor in "scale") subject to mask "k" (elements are not stored when the corresponding mask bit is not set). "scale" should be 1, 2, 4 or 8. - + Miscellaneous + + + + + Load contiguous active 32-bit integers from "a" (those with their respective bit set in mask "k"), and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). +m := 0 FOR j := 0 to 7 i := j*32 - l := j*64 IF k[j] - MEM[base_addr + SignExtend(vindex[l+63:l])*scale] := a[i+31:i] - k[j] := 0 + dst[i+31:i] := a[m+31:m] + m := m + 32 + ELSE + dst[i+31:i] := src[i+31:i] FI ENDFOR -k[MAX:8] := 0 - - -
immintrin.h
- - - Integer - AVX512F - Store - - - - - Scatter 64-bit integers from "a" into memory using 64-bit indices. 64-bit elements are stored at addresses starting at "base_addr" and offset by each 64-bit element in "vindex" (each index is scaled by the factor in "scale"). "scale" should be 1, 2, 4 or 8. - - -FOR j := 0 to 7 - i := j*64 - MEM[base_addr + SignExtend(vindex[i+63:i])*scale] := a[i+63:i] -ENDFOR +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer + AVX512VL AVX512F - Store - - - - - - Scatter 64-bit integers from "a" into memory using 64-bit indices. 64-bit elements are stored at addresses starting at "base_addr" and offset by each 64-bit element in "vindex" (each index is scaled by the factor in "scale") subject to mask "k" (elements are not stored when the corresponding mask bit is not set). "scale" should be 1, 2, 4 or 8. - + Load + + + + + Load contiguous active 32-bit integers from unaligned memory at "mem_addr" (those with their respective bit set in mask "k"), and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). +m := 0 FOR j := 0 to 7 - i := j*64 + i := j*32 IF k[j] - MEM[base_addr + SignExtend(vindex[i+63:i])*scale] := a[i+63:i] - k[j] := 0 + dst[i+31:i] := MEM[mem_addr+m+31:mem_addr+m] + m := m + 32 + ELSE + dst[i+31:i] := src[i+31:i] FI ENDFOR -k[MAX:8] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - AVX512F/KNCNI - Swizzle - - - - - Shuffle 32-bit integers in "a" within 128-bit lanes using the control in "imm8", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + AVX512VL + AVX512F + Miscellaneous + + + + Load contiguous active 32-bit integers from "a" (those with their respective bit set in mask "k"), and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -SELECT4(src, control){ - CASE(control[1:0]) - 0: tmp[31:0] := src[31:0] - 1: tmp[31:0] := src[63:32] - 2: tmp[31:0] := src[95:64] - 3: tmp[31:0] := src[127:96] - ESAC - RETURN tmp[31:0] -} - -tmp_dst[31:0] := SELECT4(a[127:0], imm8[1:0]) -tmp_dst[63:32] := SELECT4(a[127:0], imm8[3:2]) -tmp_dst[95:64] := SELECT4(a[127:0], imm8[5:4]) -tmp_dst[127:96] := SELECT4(a[127:0], imm8[7:6]) -tmp_dst[159:128] := SELECT4(a[255:128], imm8[1:0]) -tmp_dst[191:160] := SELECT4(a[255:128], imm8[3:2]) -tmp_dst[223:192] := SELECT4(a[255:128], imm8[5:4]) -tmp_dst[255:224] := SELECT4(a[255:128], imm8[7:6]) -tmp_dst[287:256] := SELECT4(a[383:256], imm8[1:0]) -tmp_dst[319:288] := SELECT4(a[383:256], imm8[3:2]) -tmp_dst[351:320] := SELECT4(a[383:256], imm8[5:4]) -tmp_dst[383:352] := SELECT4(a[383:256], imm8[7:6]) -tmp_dst[415:384] := SELECT4(a[511:384], imm8[1:0]) -tmp_dst[447:416] := SELECT4(a[511:384], imm8[3:2]) -tmp_dst[479:448] := SELECT4(a[511:384], imm8[5:4]) -tmp_dst[511:480] := SELECT4(a[511:384], imm8[7:6]) -FOR j := 0 to 15 +m := 0 +FOR j := 0 to 7 i := j*32 IF k[j] - dst[i+31:i] := tmp_dst[i+31:i] + dst[i+31:i] := a[m+31:m] + m := m + 32 ELSE - dst[i+31:i] := src[i+31:i] + dst[i+31:i] := 0 FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer + AVX512VL AVX512F - Swizzle - - - - Shuffle 32-bit integers in "a" within 128-bit lanes using the control in "imm8", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Load + + + + Load contiguous active 32-bit integers from unaligned memory at "mem_addr" (those with their respective bit set in mask "k"), and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -SELECT4(src, control){ - CASE(control[1:0]) - 0: tmp[31:0] := src[31:0] - 1: tmp[31:0] := src[63:32] - 2: tmp[31:0] := src[95:64] - 3: tmp[31:0] := src[127:96] - ESAC - RETURN tmp[31:0] -} - -tmp_dst[31:0] := SELECT4(a[127:0], imm8[1:0]) -tmp_dst[63:32] := SELECT4(a[127:0], imm8[3:2]) -tmp_dst[95:64] := SELECT4(a[127:0], imm8[5:4]) -tmp_dst[127:96] := SELECT4(a[127:0], imm8[7:6]) -tmp_dst[159:128] := SELECT4(a[255:128], imm8[1:0]) -tmp_dst[191:160] := SELECT4(a[255:128], imm8[3:2]) -tmp_dst[223:192] := SELECT4(a[255:128], imm8[5:4]) -tmp_dst[255:224] := SELECT4(a[255:128], imm8[7:6]) -tmp_dst[287:256] := SELECT4(a[383:256], imm8[1:0]) -tmp_dst[319:288] := SELECT4(a[383:256], imm8[3:2]) -tmp_dst[351:320] := SELECT4(a[383:256], imm8[5:4]) -tmp_dst[383:352] := SELECT4(a[383:256], imm8[7:6]) -tmp_dst[415:384] := SELECT4(a[511:384], imm8[1:0]) -tmp_dst[447:416] := SELECT4(a[511:384], imm8[3:2]) -tmp_dst[479:448] := SELECT4(a[511:384], imm8[5:4]) -tmp_dst[511:480] := SELECT4(a[511:384], imm8[7:6]) -FOR j := 0 to 15 +m := 0 +FOR j := 0 to 7 i := j*32 IF k[j] - dst[i+31:i] := tmp_dst[i+31:i] + dst[i+31:i] := MEM[mem_addr+m+31:mem_addr+m] + m := m + 32 ELSE dst[i+31:i] := 0 FI ENDFOR -dst[MAX:512] := 0 - - -
immintrin.h
- - - Integer - AVX512F/KNCNI - Swizzle - - - Shuffle 32-bit integers in "a" within 128-bit lanes using the control in "imm8", and store the results in "dst". - -SELECT4(src, control){ - CASE(control[1:0]) - 0: tmp[31:0] := src[31:0] - 1: tmp[31:0] := src[63:32] - 2: tmp[31:0] := src[95:64] - 3: tmp[31:0] := src[127:96] - ESAC - RETURN tmp[31:0] -} - -dst[31:0] := SELECT4(a[127:0], imm8[1:0]) -dst[63:32] := SELECT4(a[127:0], imm8[3:2]) -dst[95:64] := SELECT4(a[127:0], imm8[5:4]) -dst[127:96] := SELECT4(a[127:0], imm8[7:6]) -dst[159:128] := SELECT4(a[255:128], imm8[1:0]) -dst[191:160] := SELECT4(a[255:128], imm8[3:2]) -dst[223:192] := SELECT4(a[255:128], imm8[5:4]) -dst[255:224] := SELECT4(a[255:128], imm8[7:6]) -dst[287:256] := SELECT4(a[383:256], imm8[1:0]) -dst[319:288] := SELECT4(a[383:256], imm8[3:2]) -dst[351:320] := SELECT4(a[383:256], imm8[5:4]) -dst[383:352] := SELECT4(a[383:256], imm8[7:6]) -dst[415:384] := SELECT4(a[511:384], imm8[1:0]) -dst[447:416] := SELECT4(a[511:384], imm8[3:2]) -dst[479:448] := SELECT4(a[511:384], imm8[5:4]) -dst[511:480] := SELECT4(a[511:384], imm8[7:6]) -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer + AVX512VL AVX512F - Shift - - - - - Shift packed 32-bit integers in "a" left by "count" while shifting in zeros, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Miscellaneous + + + + + Load contiguous active 32-bit integers from "a" (those with their respective bit set in mask "k"), and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 15 +m := 0 +FOR j := 0 to 3 i := j*32 IF k[j] - IF count[63:0] > 31 - dst[i+31:i] := 0 - ELSE - dst[i+31:i] := ZeroExtend(a[i+31:i] << count[63:0]) - FI + dst[i+31:i] := a[m+31:m] + m := m + 32 ELSE dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - AVX512F/KNCNI - Shift - - - - - Shift packed 32-bit integers in "a" left by "imm8" while shifting in zeros, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + AVX512VL + AVX512F + Load + + + + + Load contiguous active 32-bit integers from unaligned memory at "mem_addr" (those with their respective bit set in mask "k"), and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 15 +m := 0 +FOR j := 0 to 3 i := j*32 IF k[j] - IF imm8[7:0] > 31 - dst[i+31:i] := 0 - ELSE - dst[i+31:i] := ZeroExtend(a[i+31:i] << imm8[7:0]) - FI + dst[i+31:i] := MEM[mem_addr+m+31:mem_addr+m] + m := m + 32 ELSE dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer + AVX512VL AVX512F - Shift - - - - Shift packed 32-bit integers in "a" left by "count" while shifting in zeros, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Miscellaneous + + + + Load contiguous active 32-bit integers from "a" (those with their respective bit set in mask "k"), and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 15 +m := 0 +FOR j := 0 to 3 i := j*32 IF k[j] - IF count[63:0] > 31 - dst[i+31:i] := 0 - ELSE - dst[i+31:i] := ZeroExtend(a[i+31:i] << count[63:0]) - FI + dst[i+31:i] := a[m+31:m] + m := m + 32 ELSE dst[i+31:i] := 0 FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer + AVX512VL AVX512F - Shift - - - - Shift packed 32-bit integers in "a" left by "imm8" while shifting in zeros, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Load + + + + Load contiguous active 32-bit integers from unaligned memory at "mem_addr" (those with their respective bit set in mask "k"), and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 15 +m := 0 +FOR j := 0 to 3 i := j*32 IF k[j] - IF imm8[7:0] > 31 - dst[i+31:i] := 0 - ELSE - dst[i+31:i] := ZeroExtend(a[i+31:i] << imm8[7:0]) - FI + dst[i+31:i] := MEM[mem_addr+m+31:mem_addr+m] + m := m + 32 ELSE dst[i+31:i] := 0 FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer + AVX512VL AVX512F - Shift - - - Shift packed 32-bit integers in "a" left by "count" while shifting in zeros, and store the results in "dst". + Miscellaneous + + + + + Load contiguous active 64-bit integers from "a" (those with their respective bit set in mask "k"), and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*32 - IF count[63:0] > 31 - dst[i+31:i] := 0 +m := 0 +FOR j := 0 to 3 + i := j*64 + IF k[j] + dst[i+63:i] := a[m+63:m] + m := m + 64 ELSE - dst[i+31:i] := ZeroExtend(a[i+31:i] << count[63:0]) + dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - AVX512F/KNCNI - Shift - - - Shift packed 32-bit integers in "a" left by "imm8" while shifting in zeros, and store the results in "dst". + AVX512VL + AVX512F + Load + + + + + Load contiguous active 64-bit integers from unaligned memory at "mem_addr" (those with their respective bit set in mask "k"), and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*32 - IF imm8[7:0] > 31 - dst[i+31:i] := 0 +m := 0 +FOR j := 0 to 3 + i := j*64 + IF k[j] + dst[i+63:i] := MEM[mem_addr+m+63:mem_addr+m] + m := m + 64 ELSE - dst[i+31:i] := ZeroExtend(a[i+31:i] << imm8[7:0]) + dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer + AVX512VL AVX512F - Shift - - - - - Shift packed 64-bit integers in "a" left by "count" while shifting in zeros, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Miscellaneous + + + + Load contiguous active 64-bit integers from "a" (those with their respective bit set in mask "k"), and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 +m := 0 +FOR j := 0 to 3 i := j*64 IF k[j] - IF count[63:0] > 63 - dst[i+63:i] := 0 - ELSE - dst[i+63:i] := ZeroExtend(a[i+63:i] << count[63:0]) - FI + dst[i+63:i] := a[m+63:m] + m := m + 64 ELSE - dst[i+63:i] := src[i+63:i] + dst[i+63:i] := 0 FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer + AVX512VL AVX512F - Shift - - - - - Shift packed 64-bit integers in "a" left by "imm8" while shifting in zeros, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Load + + + + Load contiguous active 64-bit integers from unaligned memory at "mem_addr" (those with their respective bit set in mask "k"), and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 +m := 0 +FOR j := 0 to 3 i := j*64 IF k[j] - IF imm8[7:0] > 63 - dst[i+63:i] := 0 - ELSE - dst[i+63:i] := ZeroExtend(a[i+63:i] << imm8[7:0]) - FI + dst[i+63:i] := MEM[mem_addr+m+63:mem_addr+m] + m := m + 64 ELSE - dst[i+63:i] := src[i+63:i] + dst[i+63:i] := 0 FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer + AVX512VL AVX512F - Shift - - - - Shift packed 64-bit integers in "a" left by "count" while shifting in zeros, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Miscellaneous + + + + + Load contiguous active 64-bit integers from "a" (those with their respective bit set in mask "k"), and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 7 +m := 0 +FOR j := 0 to 1 i := j*64 IF k[j] - IF count[63:0] > 63 - dst[i+63:i] := 0 - ELSE - dst[i+63:i] := ZeroExtend(a[i+63:i] << count[63:0]) - FI + dst[i+63:i] := a[m+63:m] + m := m + 64 ELSE - dst[i+63:i] := 0 + dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer + AVX512VL AVX512F - Shift - - - - Shift packed 64-bit integers in "a" left by "imm8" while shifting in zeros, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Load + + + + + Load contiguous active 64-bit integers from unaligned memory at "mem_addr" (those with their respective bit set in mask "k"), and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 7 +m := 0 +FOR j := 0 to 1 i := j*64 IF k[j] - IF imm8[7:0] > 63 - dst[i+63:i] := 0 - ELSE - dst[i+63:i] := ZeroExtend(a[i+63:i] << imm8[7:0]) - FI + dst[i+63:i] := MEM[mem_addr+m+63:mem_addr+m] + m := m + 64 ELSE - dst[i+63:i] := 0 + dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer + AVX512VL AVX512F - Shift - - - Shift packed 64-bit integers in "a" left by "count" while shifting in zeros, and store the results in "dst". + Miscellaneous + + + + Load contiguous active 64-bit integers from "a" (those with their respective bit set in mask "k"), and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 +m := 0 +FOR j := 0 to 1 i := j*64 - IF count[63:0] > 63 - dst[i+63:i] := 0 + IF k[j] + dst[i+63:i] := a[m+63:m] + m := m + 64 ELSE - dst[i+63:i] := ZeroExtend(a[i+63:i] << count[63:0]) + dst[i+63:i] := 0 FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer + AVX512VL AVX512F - Shift - - - Shift packed 64-bit integers in "a" left by "imm8" while shifting in zeros, and store the results in "dst". + Load + + + + Load contiguous active 64-bit integers from unaligned memory at "mem_addr" (those with their respective bit set in mask "k"), and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 +m := 0 +FOR j := 0 to 1 i := j*64 - IF imm8[7:0] > 63 - dst[i+63:i] := 0 + IF k[j] + dst[i+63:i] := MEM[mem_addr+m+63:mem_addr+m] + m := m + 64 ELSE - dst[i+63:i] := ZeroExtend(a[i+63:i] << imm8[7:0]) + dst[i+63:i] := 0 FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - AVX512F/KNCNI - Shift - - - - - Shift packed 32-bit integers in "a" left by the amount specified by the corresponding element in "count" while shifting in zeros, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + AVX512VL + AVX512F + Load + + + + + + + Gather 32-bit integers from memory using 32-bit indices. 32-bit elements are loaded from addresses starting at "base_addr" and offset by each 32-bit element in "vindex" (each index is scaled by the factor in "scale"). Gathered elements are merged into "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). "scale" should be 1, 2, 4 or 8. -FOR j := 0 to 15 +FOR j := 0 to 7 i := j*32 + m := j*32 IF k[j] - dst[i+31:i] := ZeroExtend(a[i+31:i] << count[i+31:i]) + addr := base_addr + SignExtend64(vindex[m+31:m]) * ZeroExtend64(scale) * 8 + dst[i+31:i] := MEM[addr+31:addr] ELSE dst[i+31:i] := src[i+31:i] - FI + FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer + AVX512VL AVX512F - Shift - - - - Shift packed 32-bit integers in "a" left by the amount specified by the corresponding element in "count" while shifting in zeros, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Load + + + + + + + Gather 32-bit integers from memory using 32-bit indices. 32-bit elements are loaded from addresses starting at "base_addr" and offset by each 32-bit element in "vindex" (each index is scaled by the factor in "scale"). Gathered elements are merged into "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). "scale" should be 1, 2, 4 or 8. -FOR j := 0 to 15 +FOR j := 0 to 3 i := j*32 + m := j*32 IF k[j] - dst[i+31:i] := ZeroExtend(a[i+31:i] << count[i+31:i]) + addr := base_addr + SignExtend64(vindex[m+31:m]) * ZeroExtend64(scale) * 8 + dst[i+31:i] := MEM[addr+31:addr] ELSE - dst[i+31:i] := 0 - FI + dst[i+31:i] := src[i+31:i] + FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - AVX512F/KNCNI - Shift - - - Shift packed 32-bit integers in "a" left by the amount specified by the corresponding element in "count" while shifting in zeros, and store the results in "dst". + AVX512VL + AVX512F + Load + + + + + + + Gather 64-bit integers from memory using 32-bit indices. 64-bit elements are loaded from addresses starting at "base_addr" and offset by each 32-bit element in "vindex" (each index is scaled by the factor in "scale"). Gathered elements are merged into "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). "scale" should be 1, 2, 4 or 8. -FOR j := 0 to 15 - i := j*32 - dst[i+31:i] := ZeroExtend(a[i+31:i] << count[i+31:i]) +FOR j := 0 to 3 + i := j*64 + m := j*32 + IF k[j] + addr := base_addr + SignExtend64(vindex[m+31:m]) * ZeroExtend64(scale) * 8 + dst[i+63:i] := MEM[addr+63:addr] + ELSE + dst[i+63:i] := src[i+63:i] + FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer + AVX512VL AVX512F - Shift - - - - - Shift packed 64-bit integers in "a" left by the amount specified by the corresponding element in "count" while shifting in zeros, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Load + + + + + + + Gather 64-bit integers from memory using 32-bit indices. 64-bit elements are loaded from addresses starting at "base_addr" and offset by each 32-bit element in "vindex" (each index is scaled by the factor in "scale"). Gathered elements are merged into "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). "scale" should be 1, 2, 4 or 8. -FOR j := 0 to 7 +FOR j := 0 to 1 i := j*64 + m := j*32 IF k[j] - dst[i+63:i] := ZeroExtend(a[i+63:i] << count[i+63:i]) + addr := base_addr + SignExtend64(vindex[m+31:m]) * ZeroExtend64(scale) * 8 + dst[i+63:i] := MEM[addr+63:addr] ELSE dst[i+63:i] := src[i+63:i] - FI + FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer + AVX512VL AVX512F - Shift - - - - Shift packed 64-bit integers in "a" left by the amount specified by the corresponding element in "count" while shifting in zeros, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Load + + + + + + + Gather 32-bit integers from memory using 64-bit indices. 32-bit elements are loaded from addresses starting at "base_addr" and offset by each 64-bit element in "vindex" (each index is scaled by the factor in "scale"). Gathered elements are merged into "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). "scale" should be 1, 2, 4 or 8. -FOR j := 0 to 7 - i := j*64 +FOR j := 0 to 3 + i := j*32 + m := j*64 IF k[j] - dst[i+63:i] := ZeroExtend(a[i+63:i] << count[i+63:i]) + addr := base_addr + vindex[m+63:m] * ZeroExtend64(scale) * 8 + dst[i+31:i] := MEM[addr+31:addr] ELSE - dst[i+63:i] := 0 - FI + dst[i+31:i] := src[i+31:i] + FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer + AVX512VL AVX512F - Shift - - - Shift packed 64-bit integers in "a" left by the amount specified by the corresponding element in "count" while shifting in zeros, and store the results in "dst". + Load + + + + + + + Gather 32-bit integers from memory using 64-bit indices. 32-bit elements are loaded from addresses starting at "base_addr" and offset by each 64-bit element in "vindex" (each index is scaled by the factor in "scale"). Gathered elements are merged into "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). "scale" should be 1, 2, 4 or 8. -FOR j := 0 to 7 - i := j*64 - dst[i+63:i] := ZeroExtend(a[i+63:i] << count[i+63:i]) +FOR j := 0 to 1 + i := j*32 + m := j*64 + IF k[j] + addr := base_addr + vindex[m+63:m] * ZeroExtend64(scale) * 8 + dst[i+31:i] := MEM[addr+31:addr] + ELSE + dst[i+31:i] := src[i+31:i] + FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:64] := 0 - +
immintrin.h
 - + Integer + AVX512VL AVX512F - Shift - - - - - Shift packed 32-bit integers in "a" right by "count" while shifting in sign bits, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Load + + + + + + + Gather 64-bit integers from memory using 64-bit indices. 64-bit elements are loaded from addresses starting at "base_addr" and offset by each 64-bit element in "vindex" (each index is scaled by the factor in "scale"). Gathered elements are merged into "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). "scale" should be 1, 2, 4 or 8. -FOR j := 0 to 15 - i := j*32 +FOR j := 0 to 3 + i := j*64 + m := j*64 IF k[j] - IF count[63:0] > 31 - dst[i+31:i] := SignBit - ELSE - dst[i+31:i] := SignExtend(a[i+31:i] >> count[63:0]) - FI + addr := base_addr + vindex[m+63:m] * ZeroExtend64(scale) * 8 + dst[i+63:i] := MEM[addr+63:addr] ELSE - dst[i+31:i] := src[i+31:i] + dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - AVX512F/KNCNI - Shift - - - - - Shift packed 32-bit integers in "a" right by "imm8" while shifting in sign bits, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + AVX512VL + AVX512F + Load + + + + + + + Gather 64-bit integers from memory using 64-bit indices. 64-bit elements are loaded from addresses starting at "base_addr" and offset by each 64-bit element in "vindex" (each index is scaled by the factor in "scale"). Gathered elements are merged into "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). "scale" should be 1, 2, 4 or 8. -FOR j := 0 to 15 - i := j*32 +FOR j := 0 to 1 + i := j*64 + m := j*64 IF k[j] - IF imm8[7:0] > 31 - dst[i+31:i] := SignBit - ELSE - dst[i+31:i] := SignExtend(a[i+31:i] >> imm8[7:0]) - FI + addr := base_addr + vindex[m+63:m] * ZeroExtend64(scale) * 8 + dst[i+63:i] := MEM[addr+63:addr] ELSE - dst[i+31:i] := src[i+31:i] + dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer + AVX512VL AVX512F - Shift - - - - Shift packed 32-bit integers in "a" right by "count" while shifting in sign bits, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Arithmetic + + + + + + Compare packed signed 32-bit integers in "a" and "b", and store packed maximum values in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 15 +FOR j := 0 to 7 i := j*32 IF k[j] - IF count[63:0] > 31 - dst[i+31:i] := SignBit - ELSE - dst[i+31:i] := SignExtend(a[i+31:i] >> count[63:0]) - FI + dst[i+31:i] := MAX(a[i+31:i], b[i+31:i]) ELSE - dst[i+31:i] := 0 + dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer + AVX512VL AVX512F - Shift - - - - Shift packed 32-bit integers in "a" right by "imm8" while shifting in sign bits, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Arithmetic + + + + + Compare packed signed 32-bit integers in "a" and "b", and store packed maximum values in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 15 +FOR j := 0 to 7 i := j*32 IF k[j] - IF imm8[7:0] > 31 - dst[i+31:i] := SignBit - ELSE - dst[i+31:i] := SignExtend(a[i+31:i] >> imm8[7:0]) - FI + dst[i+31:i] := MAX(a[i+31:i], b[i+31:i]) ELSE dst[i+31:i] := 0 FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer + AVX512VL AVX512F - Shift - - - Shift packed 32-bit integers in "a" right by "count" while shifting in sign bits, and store the results in "dst". + Arithmetic + + + + + + Compare packed signed 32-bit integers in "a" and "b", and store packed maximum values in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 15 +FOR j := 0 to 3 i := j*32 - IF count[63:0] > 31 - dst[i+31:i] := SignBit + IF k[j] + dst[i+31:i] := MAX(a[i+31:i], b[i+31:i]) ELSE - dst[i+31:i] := SignExtend(a[i+31:i] >> count[63:0]) + dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - AVX512F/KNCNI - Shift - - - Shift packed 32-bit integers in "a" right by "imm8" while shifting in sign bits, and store the results in "dst". + AVX512VL + AVX512F + Arithmetic + + + + + Compare packed signed 32-bit integers in "a" and "b", and store packed maximum values in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 15 +FOR j := 0 to 3 i := j*32 - IF imm8[7:0] > 31 - dst[i+31:i] := SignBit + IF k[j] + dst[i+31:i] := MAX(a[i+31:i], b[i+31:i]) ELSE - dst[i+31:i] := SignExtend(a[i+31:i] >> imm8[7:0]) + dst[i+31:i] := 0 FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer + AVX512VL AVX512F - Shift - - - - - Shift packed 64-bit integers in "a" right by "count" while shifting in sign bits, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Arithmetic + + + + + + Compare packed signed 64-bit integers in "a" and "b", and store packed maximum values in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 7 +FOR j := 0 to 3 i := j*64 IF k[j] - IF count[63:0] > 63 - dst[i+63:i] := SignBit - ELSE - dst[i+63:i] := SignExtend(a[i+63:i] >> count[63:0]) - FI + dst[i+63:i] := MAX(a[i+63:i], b[i+63:i]) ELSE dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer + AVX512VL AVX512F - Shift - - - - - Shift packed 64-bit integers in "a" right by "imm8" while shifting in sign bits, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Arithmetic + + + + + Compare packed signed 64-bit integers in "a" and "b", and store packed maximum values in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 +FOR j := 0 to 3 i := j*64 IF k[j] - IF imm8[7:0] > 63 - dst[i+63:i] := SignBit - ELSE - dst[i+63:i] := SignExtend(a[i+63:i] << imm8[7:0]) - FI + dst[i+63:i] := MAX(a[i+63:i], b[i+63:i]) ELSE - dst[i+63:i] := src[i+63:i] + dst[i+63:i] := 0 FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer + AVX512VL AVX512F - Shift - - - - Shift packed 64-bit integers in "a" right by "count" while shifting in sign bits, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Arithmetic + + + + Compare packed signed 64-bit integers in "a" and "b", and store packed maximum values in "dst". -FOR j := 0 to 7 +FOR j := 0 to 3 + i := j*64 + dst[i+63:i] := MAX(a[i+63:i], b[i+63:i]) +ENDFOR +dst[MAX:256] := 0 + + +
immintrin.h
+ + + Integer + AVX512VL + AVX512F + Arithmetic + + + + + + Compare packed signed 64-bit integers in "a" and "b", and store packed maximum values in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +FOR j := 0 to 1 i := j*64 IF k[j] - IF count[63:0] > 63 - dst[i+63:i] := SignBit - ELSE - dst[i+63:i] := SignExtend(a[i+63:i] >> count[63:0]) - FI + dst[i+63:i] := MAX(a[i+63:i], b[i+63:i]) ELSE - dst[i+63:i] := 0 + dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer + AVX512VL AVX512F - Shift - - - - Shift packed 64-bit integers in "a" right by "imm8" while shifting in sign bits, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Arithmetic + + + + + Compare packed signed 64-bit integers in "a" and "b", and store packed maximum values in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 +FOR j := 0 to 1 i := j*64 IF k[j] - IF imm8[7:0] > 63 - dst[i+63:i] := SignBit - ELSE - dst[i+63:i] := SignExtend(a[i+63:i] << imm8[7:0]) - FI + dst[i+63:i] := MAX(a[i+63:i], b[i+63:i]) ELSE dst[i+63:i] := 0 FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer + AVX512VL AVX512F - Shift - - - Shift packed 64-bit integers in "a" right by "count" while shifting in sign bits, and store the results in "dst". + Arithmetic + + + + Compare packed signed 64-bit integers in "a" and "b", and store packed maximum values in "dst". -FOR j := 0 to 7 +FOR j := 0 to 1 i := j*64 - IF count[63:0] > 63 - dst[i+63:i] := SignBit - ELSE - dst[i+63:i] := SignExtend(a[i+63:i] >> count[63:0]) - FI + dst[i+63:i] := MAX(a[i+63:i], b[i+63:i]) ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer + AVX512VL AVX512F - Shift - - - Shift packed 64-bit integers in "a" right by "imm8" while shifting in sign bits, and store the results in "dst". + Arithmetic + + + + + + Compare packed unsigned 32-bit integers in "a" and "b", and store packed maximum values in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). FOR j := 0 to 7 - i := j*64 - IF imm8[7:0] > 63 - dst[i+63:i] := SignBit + i := j*32 + IF k[j] + dst[i+31:i] := MAX(a[i+31:i], b[i+31:i]) ELSE - dst[i+63:i] := SignExtend(a[i+63:i] << imm8[7:0]) + dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - AVX512F/KNCNI - Shift - - - - - Shift packed 32-bit integers in "a" right by the amount specified by the corresponding element in "count" while shifting in sign bits, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + AVX512VL + AVX512F + Arithmetic + + + + + Compare packed unsigned 32-bit integers in "a" and "b", and store packed maximum values in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 15 +FOR j := 0 to 7 i := j*32 IF k[j] - dst[i+31:i] := SignExtend(a[i+31:i] >> count[i+31:i]) + dst[i+31:i] := MAX(a[i+31:i], b[i+31:i]) ELSE - dst[i+31:i] := src[i+31:i] - FI + dst[i+31:i] := 0 + FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer + AVX512VL AVX512F - Shift - - - - Shift packed 32-bit integers in "a" right by the amount specified by the corresponding element in "count" while shifting in sign bits, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Arithmetic + + + + + + Compare packed unsigned 32-bit integers in "a" and "b", and store packed maximum values in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 15 +FOR j := 0 to 3 i := j*32 IF k[j] - dst[i+31:i] := SignExtend(a[i+31:i] >> count[i+31:i]) + dst[i+31:i] := MAX(a[i+31:i], b[i+31:i]) ELSE - dst[i+31:i] := 0 - FI + dst[i+31:i] := src[i+31:i] + FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - AVX512F/KNCNI - Shift - - - Shift packed 32-bit integers in "a" right by the amount specified by the corresponding element in "count" while shifting in sign bits, and store the results in "dst". + AVX512VL + AVX512F + Arithmetic + + + + + Compare packed unsigned 32-bit integers in "a" and "b", and store packed maximum values in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 15 +FOR j := 0 to 3 i := j*32 - dst[i+31:i] := SignExtend(a[i+31:i] >> count[i+31:i]) + IF k[j] + dst[i+31:i] := MAX(a[i+31:i], b[i+31:i]) + ELSE + dst[i+31:i] := 0 + FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer + AVX512VL AVX512F - Shift - - - - - Shift packed 64-bit integers in "a" right by the amount specified by the corresponding element in "count" while shifting in sign bits, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Arithmetic + + + + + + Compare packed unsigned 64-bit integers in "a" and "b", and store packed maximum values in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 7 +FOR j := 0 to 3 i := j*64 IF k[j] - dst[i+63:i] := SignExtend(a[i+63:i] >> count[i+63:i]) + dst[i+63:i] := MAX(a[i+63:i], b[i+63:i]) ELSE dst[i+63:i] := src[i+63:i] - FI + FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer + AVX512VL AVX512F - Shift - - - - Shift packed 64-bit integers in "a" right by the amount specified by the corresponding element in "count" while shifting in sign bits, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Arithmetic + + + + + Compare packed unsigned 64-bit integers in "a" and "b", and store packed maximum values in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 +FOR j := 0 to 3 i := j*64 IF k[j] - dst[i+63:i] := SignExtend(a[i+63:i] >> count[i+63:i]) + dst[i+63:i] := MAX(a[i+63:i], b[i+63:i]) ELSE dst[i+63:i] := 0 - FI + FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer + AVX512VL AVX512F - Shift - - - Shift packed 64-bit integers in "a" right by the amount specified by the corresponding element in "count" while shifting in sign bits, and store the results in "dst". + Arithmetic + + + + Compare packed unsigned 64-bit integers in "a" and "b", and store packed maximum values in "dst". -FOR j := 0 to 7 +FOR j := 0 to 3 i := j*64 - dst[i+63:i] := SignExtend(a[i+63:i] >> count[i+63:i]) + dst[i+63:i] := MAX(a[i+63:i], b[i+63:i]) ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer + AVX512VL AVX512F - Shift - - - - - Shift packed 32-bit integers in "a" right by "count" while shifting in zeros, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Arithmetic + + + + + + Compare packed unsigned 64-bit integers in "a" and "b", and store packed maximum values in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*32 +FOR j := 0 to 1 + i := j*64 IF k[j] - IF count[63:0] > 31 - dst[i+31:i] := 0 - ELSE - dst[i+31:i] := ZeroExtend(a[i+31:i] >> count[63:0]) - FI + dst[i+63:i] := MAX(a[i+63:i], b[i+63:i]) ELSE - dst[i+31:i] := src[i+31:i] + dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - AVX512F/KNCNI - Shift - - - - - Shift packed 32-bit integers in "a" right by "imm8" while shifting in zeros, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + AVX512VL + AVX512F + Arithmetic + + + + + Compare packed unsigned 64-bit integers in "a" and "b", and store packed maximum values in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*32 +FOR j := 0 to 1 + i := j*64 IF k[j] - IF imm8[7:0] > 31 - dst[i+31:i] := 0 - ELSE - dst[i+31:i] := ZeroExtend(a[i+31:i] >> imm8[7:0]) - FI + dst[i+63:i] := MAX(a[i+63:i], b[i+63:i]) ELSE - dst[i+31:i] := src[i+31:i] + dst[i+63:i] := 0 FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer + AVX512VL AVX512F - Shift - - - - Shift packed 32-bit integers in "a" right by "count" while shifting in zeros, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Arithmetic + + + + Compare packed unsigned 64-bit integers in "a" and "b", and store packed maximum values in "dst". -FOR j := 0 to 15 - i := j*32 - IF k[j] - IF count[63:0] > 31 - dst[i+31:i] := 0 - ELSE - dst[i+31:i] := ZeroExtend(a[i+31:i] >> count[63:0]) - FI - ELSE - dst[i+31:i] := 0 - FI +FOR j := 0 to 1 + i := j*64 + dst[i+63:i] := MAX(a[i+63:i], b[i+63:i]) ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer + AVX512VL AVX512F - Shift - - - - Shift packed 32-bit integers in "a" right by "imm8" while shifting in zeros, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Arithmetic + + + + + + Compare packed signed 32-bit integers in "a" and "b", and store packed minimum values in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 15 +FOR j := 0 to 7 i := j*32 IF k[j] - IF imm8[7:0] > 31 - dst[i+31:i] := 0 - ELSE - dst[i+31:i] := ZeroExtend(a[i+31:i] >> imm8[7:0]) - FI + dst[i+31:i] := MIN(a[i+31:i], b[i+31:i]) ELSE - dst[i+31:i] := 0 + dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer + AVX512VL AVX512F - Shift - - - Shift packed 32-bit integers in "a" right by "count" while shifting in zeros, and store the results in "dst". + Arithmetic + + + + + Compare packed signed 32-bit integers in "a" and "b", and store packed minimum values in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 15 +FOR j := 0 to 7 i := j*32 - IF count[63:0] > 31 - dst[i+31:i] := 0 + IF k[j] + dst[i+31:i] := MIN(a[i+31:i], b[i+31:i]) ELSE - dst[i+31:i] := ZeroExtend(a[i+31:i] >> count[63:0]) + dst[i+31:i] := 0 FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - AVX512F/KNCNI - Shift - - - Shift packed 32-bit integers in "a" right by "imm8" while shifting in zeros, and store the results in "dst". + AVX512VL + AVX512F + Arithmetic + + + + + + Compare packed signed 32-bit integers in "a" and "b", and store packed minimum values in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 15 +FOR j := 0 to 3 i := j*32 - IF imm8[7:0] > 31 - dst[i+31:i] := 0 + IF k[j] + dst[i+31:i] := MIN(a[i+31:i], b[i+31:i]) ELSE - dst[i+31:i] := ZeroExtend(a[i+31:i] >> imm8[7:0]) + dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer + AVX512VL AVX512F - Shift - - - - - Shift packed 64-bit integers in "a" right by "count" while shifting in zeros, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Arithmetic + + + + + Compare packed signed 32-bit integers in "a" and "b", and store packed minimum values in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := j*64 +FOR j := 0 to 3 + i := j*32 IF k[j] - IF count[63:0] > 63 - dst[i+63:i] := 0 - ELSE - dst[i+63:i] := ZeroExtend(a[i+63:i] >> count[63:0]) - FI + dst[i+31:i] := MIN(a[i+31:i], b[i+31:i]) ELSE - dst[i+63:i] := src[i+63:i] + dst[i+31:i] := 0 FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer + AVX512VL AVX512F - Shift - - - - - Shift packed 64-bit integers in "a" right by "imm8" while shifting in zeros, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Arithmetic + + + + + + Compare packed signed 64-bit integers in "a" and "b", and store packed minimum values in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 7 +FOR j := 0 to 3 i := j*64 IF k[j] - IF imm8[7:0] > 63 - dst[i+63:i] := 0 - ELSE - dst[i+63:i] := ZeroExtend(a[i+63:i] >> imm8[7:0]) - FI + dst[i+63:i] := MIN(a[i+63:i], b[i+63:i]) ELSE dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer + AVX512VL AVX512F - Shift - - - - Shift packed 64-bit integers in "a" right by "count" while shifting in zeros, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Arithmetic + + + + + Compare packed signed 64-bit integers in "a" and "b", and store packed minimum values in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 +FOR j := 0 to 3 i := j*64 IF k[j] - IF count[63:0] > 63 - dst[i+63:i] := 0 - ELSE - dst[i+63:i] := ZeroExtend(a[i+63:i] >> count[63:0]) - FI + dst[i+63:i] := MIN(a[i+63:i], b[i+63:i]) ELSE dst[i+63:i] := 0 FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer + AVX512VL AVX512F - Shift - - - - Shift packed 64-bit integers in "a" right by "imm8" while shifting in zeros, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Arithmetic + + + + Compare packed signed 64-bit integers in "a" and "b", and store packed minimum values in "dst". -FOR j := 0 to 7 +FOR j := 0 to 3 i := j*64 - IF k[j] - IF imm8[7:0] > 63 - dst[i+63:i] := 0 - ELSE - dst[i+63:i] := ZeroExtend(a[i+63:i] >> imm8[7:0]) - FI - ELSE - dst[i+63:i] := 0 - FI + dst[i+63:i] := MIN(a[i+63:i], b[i+63:i]) ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer + AVX512VL AVX512F - Shift - - - Shift packed 64-bit integers in "a" right by "count" while shifting in zeros, and store the results in "dst". + Arithmetic + + + + + + Compare packed signed 64-bit integers in "a" and "b", and store packed minimum values in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 7 +FOR j := 0 to 1 i := j*64 - IF count[63:0] > 63 - dst[i+63:i] := 0 + IF k[j] + dst[i+63:i] := MIN(a[i+63:i], b[i+63:i]) ELSE - dst[i+63:i] := ZeroExtend(a[i+63:i] >> count[63:0]) + dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer + AVX512VL AVX512F - Shift - - - Shift packed 64-bit integers in "a" right by "imm8" while shifting in zeros, and store the results in "dst". + Arithmetic + + + + + Compare packed signed 64-bit integers in "a" and "b", and store packed minimum values in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 +FOR j := 0 to 1 i := j*64 - IF imm8[7:0] > 63 - dst[i+63:i] := 0 + IF k[j] + dst[i+63:i] := MIN(a[i+63:i], b[i+63:i]) ELSE - dst[i+63:i] := ZeroExtend(a[i+63:i] >> imm8[7:0]) + dst[i+63:i] := 0 FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - AVX512F/KNCNI - Shift - - - - - Shift packed 32-bit integers in "a" right by the amount specified by the corresponding element in "count" while shifting in zeros, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + AVX512VL + AVX512F + Arithmetic + + + + Compare packed signed 64-bit integers in "a" and "b", and store packed minimum values in "dst". -FOR j := 0 to 15 +FOR j := 0 to 1 + i := j*64 + dst[i+63:i] := MIN(a[i+63:i], b[i+63:i]) +ENDFOR +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Integer + AVX512VL + AVX512F + Arithmetic + + + + + + Compare packed unsigned 32-bit integers in "a" and "b", and store packed minimum values in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +FOR j := 0 to 7 i := j*32 IF k[j] - dst[i+31:i] := ZeroExtend(a[i+31:i] >> count[i+31:i]) + dst[i+31:i] := MIN(a[i+31:i], b[i+31:i]) ELSE dst[i+31:i] := src[i+31:i] - FI + FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer + AVX512VL AVX512F - Shift - - - - Shift packed 32-bit integers in "a" right by the amount specified by the corresponding element in "count" while shifting in zeros, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Arithmetic + + + + + Compare packed unsigned 32-bit integers in "a" and "b", and store packed minimum values in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 15 +FOR j := 0 to 7 i := j*32 IF k[j] - dst[i+31:i] := ZeroExtend(a[i+31:i] >> count[i+31:i]) + dst[i+31:i] := MIN(a[i+31:i], b[i+31:i]) ELSE dst[i+31:i] := 0 - FI + FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - AVX512F/KNCNI - Shift - - - Shift packed 32-bit integers in "a" right by the amount specified by the corresponding element in "count" while shifting in zeros, and store the results in "dst". + AVX512VL + AVX512F + Arithmetic + + + + + + Compare packed unsigned 32-bit integers in "a" and "b", and store packed minimum values in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 15 +FOR j := 0 to 3 i := j*32 - dst[i+31:i] := ZeroExtend(a[i+31:i] >> count[i+31:i]) + IF k[j] + dst[i+31:i] := MIN(a[i+31:i], b[i+31:i]) + ELSE + dst[i+31:i] := src[i+31:i] + FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer + AVX512VL AVX512F - Shift - - - - - Shift packed 64-bit integers in "a" right by the amount specified by the corresponding element in "count" while shifting in zeros, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Arithmetic + + + + + Compare packed unsigned 32-bit integers in "a" and "b", and store packed minimum values in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 +FOR j := 0 to 3 + i := j*32 + IF k[j] + dst[i+31:i] := MIN(a[i+31:i], b[i+31:i]) + ELSE + dst[i+31:i] := 0 + FI +ENDFOR +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Integer + AVX512VL + AVX512F + Arithmetic + + + + + + Compare packed unsigned 64-bit integers in "a" and "b", and store packed minimum values in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +FOR j := 0 to 3 i := j*64 IF k[j] - dst[i+63:i] := ZeroExtend(a[i+63:i] >> count[i+63:i]) + dst[i+63:i] := MIN(a[i+63:i], b[i+63:i]) ELSE dst[i+63:i] := src[i+63:i] - FI + FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer + AVX512VL AVX512F - Shift - - - - Shift packed 64-bit integers in "a" right by the amount specified by the corresponding element in "count" while shifting in zeros, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Arithmetic + + + + + Compare packed unsigned 64-bit integers in "a" and "b", and store packed minimum values in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 +FOR j := 0 to 3 i := j*64 IF k[j] - dst[i+63:i] := ZeroExtend(a[i+63:i] >> count[i+63:i]) + dst[i+63:i] := MIN(a[i+63:i], b[i+63:i]) ELSE dst[i+63:i] := 0 - FI + FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer + AVX512VL AVX512F - Shift - - - Shift packed 64-bit integers in "a" right by the amount specified by the corresponding element in "count" while shifting in zeros, and store the results in "dst". + Arithmetic + + + + Compare packed unsigned 64-bit integers in "a" and "b", and store packed minimum values in "dst". -FOR j := 0 to 7 +FOR j := 0 to 3 i := j*64 - dst[i+63:i] := ZeroExtend(a[i+63:i] >> count[i+63:i]) + dst[i+63:i] := MIN(a[i+63:i], b[i+63:i]) ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - AVX512F/KNCNI + AVX512VL + AVX512F Arithmetic - - - - - Subtract packed 32-bit integers in "b" from packed 32-bit integers in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + + + + + Compare packed unsigned 64-bit integers in "a" and "b", and store packed minimum values in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*32 +FOR j := 0 to 1 + i := j*64 IF k[j] - dst[i+31:i] := a[i+31:i] - b[i+31:i] + dst[i+63:i] := MIN(a[i+63:i], b[i+63:i]) ELSE - dst[i+31:i] := src[i+31:i] - FI + dst[i+63:i] := src[i+63:i] + FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer + AVX512VL AVX512F Arithmetic - - - - Subtract packed 32-bit integers in "b" from packed 32-bit integers in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + + + + Compare packed unsigned 64-bit integers in "a" and "b", and store packed minimum values in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*32 +FOR j := 0 to 1 + i := j*64 IF k[j] - dst[i+31:i] := a[i+31:i] - b[i+31:i] + dst[i+63:i] := MIN(a[i+63:i], b[i+63:i]) ELSE - dst[i+31:i] := 0 - FI + dst[i+63:i] := 0 + FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - AVX512F/KNCNI + AVX512VL + AVX512F Arithmetic - - - Subtract packed 32-bit integers in "b" from packed 32-bit integers in "a", and store the results in "dst". + + + + Compare packed unsigned 64-bit integers in "a" and "b", and store packed minimum values in "dst". -FOR j := 0 to 15 - i := j*32 - dst[i+31:i] := a[i+31:i] - b[i+31:i] +FOR j := 0 to 1 + i := j*64 + dst[i+63:i] := MIN(a[i+63:i], b[i+63:i]) ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer + AVX512VL AVX512F - Arithmetic - - - - - Subtract packed 64-bit integers in "b" from packed 64-bit integers in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Convert + + + Convert packed 32-bit integers in "a" to packed 8-bit integers with truncation, and store the results in "dst". FOR j := 0 to 7 - i := j*64 - IF k[j] - dst[i+63:i] := a[i+63:i] - b[i+63:i] - ELSE - dst[i+63:i] := src[i+63:i] - FI + i := 32*j + k := 8*j + dst[k+7:k] := Truncate8(a[i+31:i]) ENDFOR -dst[MAX:512] := 0 +dst[MAX:64] := 0 - +
immintrin.h
 - + Integer + AVX512VL AVX512F - Arithmetic - - - - Subtract packed 64-bit integers in "b" from packed 64-bit integers in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Convert + + + + + Convert packed 32-bit integers in "a" to packed 8-bit integers with truncation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). FOR j := 0 to 7 - i := j*64 + i := 32*j + l := 8*j IF k[j] - dst[i+63:i] := a[i+63:i] - b[i+63:i] + dst[l+7:l] := Truncate8(a[i+31:i]) ELSE - dst[i+63:i] := 0 - FI + dst[l+7:l] := src[l+7:l] + FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:64] := 0 - +
immintrin.h
 - + Integer + AVX512VL AVX512F - Arithmetic - - - Subtract packed 64-bit integers in "b" from packed 64-bit integers in "a", and store the results in "dst". + Convert + Store + + + + + Convert packed 32-bit integers in "a" to packed 8-bit integers with truncation, and store the active results (those with their respective bit set in writemask "k") to unaligned memory at "base_addr". FOR j := 0 to 7 - i := j*64 - dst[i+63:i] := a[i+63:i] - b[i+63:i] + i := 32*j + l := 8*j + IF k[j] + MEM[base_addr+l+7:base_addr+l] := Truncate8(a[i+31:i]) + FI ENDFOR -dst[MAX:512] := 0 - +
immintrin.h
 - + Integer + AVX512VL AVX512F - Logical - - - - - - Bitwise ternary logic that provides the capability to implement any three-operand binary function; the specific binary function is specified by value in "imm8". For each bit in each packed 32-bit integer, the corresponding bit from "src", "a", and "b" are used to form a 3 bit index into "imm8", and the value at that bit in "imm8" is written to the corresponding bit in "dst" using writemask "k" at 32-bit granularity (32-bit elements are copied from "src" when the corresponding mask bit is not set). + Convert + + + + Convert packed 32-bit integers in "a" to packed 8-bit integers with truncation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*32 +FOR j := 0 to 7 + i := 32*j + l := 8*j IF k[j] - FOR h := 0 to 31 - index[2:0] := (src[i+h] << 2) OR (a[i+h] << 1) OR b[i+h] - dst[i+h] := imm8[index[2:0]] - ENDFOR + dst[l+7:l] := Truncate8(a[i+31:i]) ELSE - dst[i+31:i] := src[i+31:i] + dst[l+7:l] := 0 FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:64] := 0 - +
immintrin.h
 - + Integer + AVX512VL AVX512F - Logical - - - - - - Bitwise ternary logic that provides the capability to implement any three-operand binary function; the specific binary function is specified by value in "imm8". For each bit in each packed 32-bit integer, the corresponding bit from "a", "b", and "c" are used to form a 3 bit index into "imm8", and the value at that bit in "imm8" is written to the corresponding bit in "dst" using zeromask "k" at 32-bit granularity (32-bit elements are zeroed out when the corresponding mask bit is not set). + Convert + + + Convert packed 32-bit integers in "a" to packed 8-bit integers with truncation, and store the results in "dst". -FOR j := 0 to 15 - i := j*32 - IF k[j] - FOR h := 0 to 31 - index[2:0] := (a[i+h] << 2) OR (b[i+h] << 1) OR c[i+h] - dst[i+h] := imm8[index[2:0]] - ENDFOR - ELSE - dst[i+31:i] := 0 - FI +FOR j := 0 to 3 + i := 32*j + k := 8*j + dst[k+7:k] := Truncate8(a[i+31:i]) ENDFOR -dst[MAX:512] := 0 +dst[MAX:32] := 0 - +
immintrin.h
 - + Integer + AVX512VL AVX512F - Logical - - - - - Bitwise ternary logic that provides the capability to implement any three-operand binary function; the specific binary function is specified by value in "imm8". For each bit in each packed 32-bit integer, the corresponding bit from "a", "b", and "c" are used to form a 3 bit index into "imm8", and the value at that bit in "imm8" is written to the corresponding bit in "dst". + Convert + + + + + Convert packed 32-bit integers in "a" to packed 8-bit integers with truncation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*32 - FOR h := 0 to 31 - index[2:0] := (a[i+h] << 2) OR (b[i+h] << 1) OR c[i+h] - dst[i+h] := imm8[index[2:0]] - ENDFOR +FOR j := 0 to 3 + i := 32*j + l := 8*j + IF k[j] + dst[l+7:l] := Truncate8(a[i+31:i]) + ELSE + dst[l+7:l] := src[l+7:l] + FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:32] := 0 - +
immintrin.h
 - + Integer + AVX512VL AVX512F - Logical - - - - - - Bitwise ternary logic that provides the capability to implement any three-operand binary function; the specific binary function is specified by value in "imm8". For each bit in each packed 64-bit integer, the corresponding bit from "src", "a", and "b" are used to form a 3 bit index into "imm8", and the value at that bit in "imm8" is written to the corresponding bit in "dst" using writemask "k" at 64-bit granularity (64-bit elements are copied from "src" when the corresponding mask bit is not set). + Convert + Store + + + + + Convert packed 32-bit integers in "a" to packed 8-bit integers with truncation, and store the active results (those with their respective bit set in writemask "k") to unaligned memory at "base_addr". -FOR j := 0 to 7 - i := j*64 +FOR j := 0 to 3 + i := 32*j + l := 8*j IF k[j] - FOR h := 0 to 63 - index[2:0] := (src[i+h] << 2) OR (a[i+h] << 1) OR b[i+h] - dst[i+h] := imm8[index[2:0]] - ENDFOR - ELSE - dst[i+63:i] := src[i+63:i] + MEM[base_addr+l+7:base_addr+l] := Truncate8(a[i+31:i]) FI ENDFOR -dst[MAX:512] := 0 - +
immintrin.h
 - + Integer + AVX512VL AVX512F - Logical - - - - - - Bitwise ternary logic that provides the capability to implement any three-operand binary function; the specific binary function is specified by value in "imm8". For each bit in each packed 64-bit integer, the corresponding bit from "a", "b", and "c" are used to form a 3 bit index into "imm8", and the value at that bit in "imm8" is written to the corresponding bit in "dst" using zeromask "k" at 64-bit granularity (64-bit elements are zeroed out when the corresponding mask bit is not set). + Convert + + + + Convert packed 32-bit integers in "a" to packed 8-bit integers with truncation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := j*64 +FOR j := 0 to 3 + i := 32*j + l := 8*j IF k[j] - FOR h := 0 to 63 - index[2:0] := (a[i+h] << 2) OR (b[i+h] << 1) OR c[i+h] - dst[i+h] := imm8[index[2:0]] - ENDFOR + dst[l+7:l] := Truncate8(a[i+31:i]) ELSE - dst[i+63:i] := 0 + dst[l+7:l] := 0 FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:32] := 0 - +
immintrin.h
 - + Integer + AVX512VL AVX512F - Logical - - - - - Bitwise ternary logic that provides the capability to implement any three-operand binary function; the specific binary function is specified by value in "imm8". For each bit in each packed 64-bit integer, the corresponding bit from "a", "b", and "c" are used to form a 3 bit index into "imm8", and the value at that bit in "imm8" is written to the corresponding bit in "dst". + Convert + + + Convert packed 32-bit integers in "a" to packed 16-bit integers with truncation, and store the results in "dst". FOR j := 0 to 7 - i := j*64 - FOR h := 0 to 63 - index[2:0] := (a[i+h] << 2) OR (b[i+h] << 1) OR c[i+h] - dst[i+h] := imm8[index[2:0]] - ENDFOR + i := 32*j + k := 16*j + dst[k+15:k] := Truncate16(a[i+31:i]) ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - Mask - AVX512F/KNCNI - Logical - - - - Compute the bitwise AND of packed 32-bit integers in "a" and "b", producing intermediate 32-bit values, and set the corresponding bit in result mask "k" (subject to writemask "k") if the intermediate value is non-zero. + AVX512VL + AVX512F + Convert + + + + + Convert packed 32-bit integers in "a" to packed 16-bit integers with truncation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*32 - IF k1[j] - k[j] := ((a[i+31:i] AND b[i+31:i]) != 0) ? 1 : 0 +FOR j := 0 to 7 + i := 32*j + l := 16*j + IF k[j] + dst[l+15:l] := Truncate16(a[i+31:i]) ELSE - k[j] := 0 + dst[l+15:l] := src[l+15:l] FI ENDFOR -k[MAX:16] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - Mask - AVX512F/KNCNI - Logical - - - Compute the bitwise AND of packed 32-bit integers in "a" and "b", producing intermediate 32-bit values, and set the corresponding bit in result mask "k" if the intermediate value is non-zero. + AVX512VL + AVX512F + Convert + Store + + + + + Convert packed 32-bit integers in "a" to packed 16-bit integers with truncation, and store the active results (those with their respective bit set in writemask "k") to unaligned memory at "base_addr". -FOR j := 0 to 15 - i := j*32 - k[j] := ((a[i+31:i] AND b[i+31:i]) != 0) ? 1 : 0 +FOR j := 0 to 7 + i := 32*j + l := 16*j + IF k[j] + MEM[base_addr+l+15:base_addr+l] := Truncate16(a[i+31:i]) + FI ENDFOR -k[MAX:16] := 0 - +
immintrin.h
 - + Integer - Mask + AVX512VL AVX512F - Logical - - - - Compute the bitwise AND of packed 64-bit integers in "a" and "b", producing intermediate 64-bit values, and set the corresponding bit in result mask "k" (subject to writemask "k") if the intermediate value is non-zero. + Convert + + + + Convert packed 32-bit integers in "a" to packed 16-bit integers with truncation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). FOR j := 0 to 7 - i := j*64 - IF k1[j] - k[j] := ((a[i+63:i] AND b[i+63:i]) != 0) ? 1 : 0 + i := 32*j + l := 16*j + IF k[j] + dst[l+15:l] := Truncate16(a[i+31:i]) ELSE - k[j] := 0 + dst[l+15:l] := 0 FI ENDFOR -k[MAX:8] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - Mask + AVX512VL AVX512F - Logical - - - Compute the bitwise AND of packed 64-bit integers in "a" and "b", producing intermediate 64-bit values, and set the corresponding bit in result mask "k" if the intermediate value is non-zero. + Convert + + + Convert packed 32-bit integers in "a" to packed 16-bit integers with truncation, and store the results in "dst". -FOR j := 0 to 7 - i := j*64 - k[j] := ((a[i+63:i] AND b[i+63:i]) != 0) ? 1 : 0 +FOR j := 0 to 3 + i := 32*j + k := 16*j + dst[k+15:k] := Truncate16(a[i+31:i]) ENDFOR -k[MAX:8] := 0 +dst[MAX:64] := 0 - +
immintrin.h
 - + Integer - Mask + AVX512VL AVX512F - Logical - - - - Compute the bitwise NAND of packed 32-bit integers in "a" and "b", producing intermediate 32-bit values, and set the corresponding bit in result mask "k" (subject to writemask "k") if the intermediate value is zero. + Convert + + + + + Convert packed 32-bit integers in "a" to packed 16-bit integers with truncation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*32 - IF k1[j] - k[j] := ((a[i+31:i] AND b[i+31:i]) == 0) ? 1 : 0 +FOR j := 0 to 3 + i := 32*j + l := 16*j + IF k[j] + dst[l+15:l] := Truncate16(a[i+31:i]) ELSE - k[j] := 0 + dst[l+15:l] := src[l+15:l] FI ENDFOR -k[MAX:16] := 0 +dst[MAX:64] := 0 - +
immintrin.h
 - + Integer - Mask + AVX512VL AVX512F - Logical - - - Compute the bitwise NAND of packed 32-bit integers in "a" and "b", producing intermediate 32-bit values, and set the corresponding bit in result mask "k" if the intermediate value is zero. + Convert + Store + + + + + Convert packed 32-bit integers in "a" to packed 16-bit integers with truncation, and store the active results (those with their respective bit set in writemask "k") to unaligned memory at "base_addr". -FOR j := 0 to 15 - i := j*32 - k[j] := ((a[i+31:i] AND b[i+31:i]) == 0) ? 1 : 0 +FOR j := 0 to 3 + i := 32*j + l := 16*j + IF k[j] + MEM[base_addr+l+15:base_addr+l] := Truncate16(a[i+31:i]) + FI ENDFOR -k[MAX:16] := 0 - +
immintrin.h
 - + Integer - Mask + AVX512VL AVX512F - Logical - - - - Compute the bitwise NAND of packed 64-bit integers in "a" and "b", producing intermediate 64-bit values, and set the corresponding bit in result mask "k" (subject to writemask "k") if the intermediate value is zero. + Convert + + + + Convert packed 32-bit integers in "a" to packed 16-bit integers with truncation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := j*64 - IF k1[j] - k[j] := ((a[i+63:i] AND b[i+63:i]) == 0) ? 1 : 0 +FOR j := 0 to 3 + i := 32*j + l := 16*j + IF k[j] + dst[l+15:l] := Truncate16(a[i+31:i]) ELSE - k[j] := 0 + dst[l+15:l] := 0 FI ENDFOR -k[MAX:8] := 0 +dst[MAX:64] := 0 - +
immintrin.h
 - + Integer - Mask + AVX512VL AVX512F - Logical - - - Compute the bitwise NAND of packed 64-bit integers in "a" and "b", producing intermediate 64-bit values, and set the corresponding bit in result mask "k" if the intermediate value is zero. + Convert + + + Convert packed 64-bit integers in "a" to packed 8-bit integers with truncation, and store the results in "dst". -FOR j := 0 to 7 - i := j*64 - k[j] := ((a[i+63:i] AND b[i+63:i]) == 0) ? 1 : 0 +FOR j := 0 to 3 + i := 64*j + k := 8*j + dst[k+7:k] := Truncate8(a[i+63:i]) ENDFOR -k[MAX:8] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer + AVX512VL AVX512F - Swizzle - - - - - Unpack and interleave 32-bit integers from the high half of each 128-bit lane in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Convert + + + + + Convert packed 64-bit integers in "a" to packed 8-bit integers with truncation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -INTERLEAVE_HIGH_DWORDS(src1[127:0], src2[127:0]){ - dst[31:0] := src1[95:64] - dst[63:32] := src2[95:64] - dst[95:64] := src1[127:96] - dst[127:96] := src2[127:96] - RETURN dst[127:0] -} - -tmp_dst[127:0] := INTERLEAVE_HIGH_DWORDS(a[127:0], b[127:0]) -tmp_dst[255:128] := INTERLEAVE_HIGH_DWORDS(a[255:128], b[255:128]) -tmp_dst[383:256] := INTERLEAVE_HIGH_DWORDS(a[383:256], b[383:256]) -tmp_dst[511:384] := INTERLEAVE_HIGH_DWORDS(a[511:384], b[511:384]) - -FOR j := 0 to 15 - i := j*32 +FOR j := 0 to 3 + i := 64*j + l := 8*j IF k[j] - dst[i+31:i] := tmp_dst[i+31:i] + dst[l+7:l] := Truncate8(a[i+63:i]) ELSE - dst[i+31:i] := src[i+31:i] + dst[l+7:l] := src[l+7:l] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer + AVX512VL AVX512F - Swizzle - - - - Unpack and interleave 32-bit integers from the high half of each 128-bit lane in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Convert + Store + + + + + Convert packed 64-bit integers in "a" to packed 8-bit integers with truncation, and store the active results (those with their respective bit set in writemask "k") to unaligned memory at "base_addr". -INTERLEAVE_HIGH_DWORDS(src1[127:0], src2[127:0]){ - dst[31:0] := src1[95:64] - dst[63:32] := src2[95:64] - dst[95:64] := src1[127:96] - dst[127:96] := src2[127:96] - RETURN dst[127:0] -} - -tmp_dst[127:0] := INTERLEAVE_HIGH_DWORDS(a[127:0], b[127:0]) -tmp_dst[255:128] := INTERLEAVE_HIGH_DWORDS(a[255:128], b[255:128]) -tmp_dst[383:256] := INTERLEAVE_HIGH_DWORDS(a[383:256], b[383:256]) -tmp_dst[511:384] := INTERLEAVE_HIGH_DWORDS(a[511:384], b[511:384]) - -FOR j := 0 to 15 - i := j*32 +FOR j := 0 to 3 + i := 64*j + l := 8*j IF k[j] - dst[i+31:i] := tmp_dst[i+31:i] - ELSE - dst[i+31:i] := 0 + MEM[base_addr+l+7:base_addr+l] := Truncate8(a[i+63:i]) FI ENDFOR -dst[MAX:512] := 0 - +
immintrin.h
 - + Integer + AVX512VL AVX512F - Swizzle - - - Unpack and interleave 32-bit integers from the high half of each 128-bit lane in "a" and "b", and store the results in "dst". + Convert + + + + Convert packed 64-bit integers in "a" to packed 8-bit integers with truncation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -INTERLEAVE_HIGH_DWORDS(src1[127:0], src2[127:0]){ - dst[31:0] := src1[95:64] - dst[63:32] := src2[95:64] - dst[95:64] := src1[127:96] - dst[127:96] := src2[127:96] - RETURN dst[127:0] -} - -dst[127:0] := INTERLEAVE_HIGH_DWORDS(a[127:0], b[127:0]) -dst[255:128] := INTERLEAVE_HIGH_DWORDS(a[255:128], b[255:128]) -dst[383:256] := INTERLEAVE_HIGH_DWORDS(a[383:256], b[383:256]) -dst[511:384] := INTERLEAVE_HIGH_DWORDS(a[511:384], b[511:384]) -dst[MAX:512] := 0 +FOR j := 0 to 3 + i := 64*j + l := 8*j + IF k[j] + dst[l+7:l] := Truncate8(a[i+63:i]) + ELSE + dst[l+7:l] := 0 + FI +ENDFOR +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer + AVX512VL AVX512F - Swizzle - - - - - Unpack and interleave 64-bit integers from the high half of each 128-bit lane in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Convert + + + Convert packed 64-bit integers in "a" to packed 8-bit integers with truncation, and store the results in "dst". -INTERLEAVE_HIGH_QWORDS(src1[127:0], src2[127:0]){ - dst[63:0] := src1[127:64] - dst[127:64] := src2[127:64] - RETURN dst[127:0] -} - -tmp_dst[127:0] := INTERLEAVE_HIGH_QWORDS(a[127:0], b[127:0]) -tmp_dst[255:128] := INTERLEAVE_HIGH_QWORDS(a[255:128], b[255:128]) -tmp_dst[383:256] := INTERLEAVE_HIGH_QWORDS(a[383:256], b[383:256]) -tmp_dst[511:384] := INTERLEAVE_HIGH_QWORDS(a[511:384], b[511:384]) - -FOR j := 0 to 7 - i := j*64 +FOR j := 0 to 1 + i := 64*j + k := 8*j + dst[k+7:k] := Truncate8(a[i+63:i]) +ENDFOR +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Integer + AVX512VL + AVX512F + Convert + + + + + Convert packed 64-bit integers in "a" to packed 8-bit integers with truncation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +FOR j := 0 to 1 + i := 64*j + l := 8*j IF k[j] - dst[i+63:i] := tmp_dst[i+63:i] + dst[l+7:l] := Truncate8(a[i+63:i]) ELSE - dst[i+63:i] := src[i+63:i] + dst[l+7:l] := src[l+7:l] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer + AVX512VL AVX512F - Swizzle - - - - Unpack and interleave 64-bit integers from the high half of each 128-bit lane in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Convert + Store + + + + + Convert packed 64-bit integers in "a" to packed 8-bit integers with truncation, and store the active results (those with their respective bit set in writemask "k") to unaligned memory at "base_addr". -INTERLEAVE_HIGH_QWORDS(src1[127:0], src2[127:0]){ - dst[63:0] := src1[127:64] - dst[127:64] := src2[127:64] - RETURN dst[127:0] -} - -tmp_dst[127:0] := INTERLEAVE_HIGH_QWORDS(a[127:0], b[127:0]) -tmp_dst[255:128] := INTERLEAVE_HIGH_QWORDS(a[255:128], b[255:128]) -tmp_dst[383:256] := INTERLEAVE_HIGH_QWORDS(a[383:256], b[383:256]) -tmp_dst[511:384] := INTERLEAVE_HIGH_QWORDS(a[511:384], b[511:384]) - -FOR j := 0 to 7 - i := j*64 +FOR j := 0 to 1 + i := 64*j + l := 8*j IF k[j] - dst[i+63:i] := tmp_dst[i+63:i] + MEM[base_addr+l+7:base_addr+l] := Truncate8(a[i+63:i]) + FI +ENDFOR + + +
immintrin.h
+ + + Integer + AVX512VL + AVX512F + Convert + + + + Convert packed 64-bit integers in "a" to packed 8-bit integers with truncation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + +FOR j := 0 to 1 + i := 64*j + l := 8*j + IF k[j] + dst[l+7:l] := Truncate8(a[i+63:i]) ELSE - dst[i+63:i] := 0 + dst[l+7:l] := 0 FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer + AVX512VL AVX512F - Swizzle - - - Unpack and interleave 64-bit integers from the high half of each 128-bit lane in "a" and "b", and store the results in "dst". + Convert + + + Convert packed 64-bit integers in "a" to packed 32-bit integers with truncation, and store the results in "dst". -INTERLEAVE_HIGH_QWORDS(src1[127:0], src2[127:0]){ - dst[63:0] := src1[127:64] - dst[127:64] := src2[127:64] - RETURN dst[127:0] -} - -dst[127:0] := INTERLEAVE_HIGH_QWORDS(a[127:0], b[127:0]) -dst[255:128] := INTERLEAVE_HIGH_QWORDS(a[255:128], b[255:128]) -dst[383:256] := INTERLEAVE_HIGH_QWORDS(a[383:256], b[383:256]) -dst[511:384] := INTERLEAVE_HIGH_QWORDS(a[511:384], b[511:384]) -dst[MAX:512] := 0 +FOR j := 0 to 3 + i := 64*j + k := 32*j + dst[k+31:k] := Truncate32(a[i+63:i]) +ENDFOR +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer + AVX512VL AVX512F - Swizzle - - - - - Unpack and interleave 32-bit integers from the low half of each 128-bit lane in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Convert + + + + + Convert packed 64-bit integers in "a" to packed 32-bit integers with truncation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -INTERLEAVE_DWORDS(src1[127:0], src2[127:0]){ - dst[31:0] := src1[31:0] - dst[63:32] := src2[31:0] - dst[95:64] := src1[63:32] - dst[127:96] := src2[63:32] - RETURN dst[127:0] -} - -tmp_dst[127:0] := INTERLEAVE_DWORDS(a[127:0], b[127:0]) -tmp_dst[255:128] := INTERLEAVE_DWORDS(a[255:128], b[255:128]) -tmp_dst[383:256] := INTERLEAVE_DWORDS(a[383:256], b[383:256]) -tmp_dst[511:384] := INTERLEAVE_DWORDS(a[511:384], b[511:384]) - -FOR j := 0 to 15 - i := j*32 +FOR j := 0 to 3 + i := 64*j + l := 32*j IF k[j] - dst[i+31:i] := tmp_dst[i+31:i] + dst[l+31:l] := Truncate32(a[i+63:i]) ELSE - dst[i+31:i] := src[i+31:i] + dst[l+31:l] := src[l+31:l] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer + AVX512VL AVX512F - Swizzle - - - - Unpack and interleave 32-bit integers from the low half of each 128-bit lane in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Convert + Store + + + + + Convert packed 64-bit integers in "a" to packed 32-bit integers with truncation, and store the active results (those with their respective bit set in writemask "k") to unaligned memory at "base_addr". -INTERLEAVE_DWORDS(src1[127:0], src2[127:0]){ - dst[31:0] := src1[31:0] - dst[63:32] := src2[31:0] - dst[95:64] := src1[63:32] - dst[127:96] := src2[63:32] - RETURN dst[127:0] -} - -tmp_dst[127:0] := INTERLEAVE_DWORDS(a[127:0], b[127:0]) -tmp_dst[255:128] := INTERLEAVE_DWORDS(a[255:128], b[255:128]) -tmp_dst[383:256] := INTERLEAVE_DWORDS(a[383:256], b[383:256]) -tmp_dst[511:384] := INTERLEAVE_DWORDS(a[511:384], b[511:384]) - -FOR j := 0 to 15 - i := j*32 +FOR j := 0 to 3 + i := 64*j + l := 32*j IF k[j] - dst[i+31:i] := tmp_dst[i+31:i] + MEM[base_addr+l+31:base_addr+l] := Truncate32(a[i+63:i]) + FI +ENDFOR + + +
immintrin.h
+ + + Integer + AVX512VL + AVX512F + Convert + + + + Convert packed 64-bit integers in "a" to packed 32-bit integers with truncation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + +FOR j := 0 to 3 + i := 64*j + l := 32*j + IF k[j] + dst[l+31:l] := Truncate32(a[i+63:i]) ELSE - dst[i+31:i] := 0 + dst[l+31:l] := 0 FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer + AVX512VL AVX512F - Swizzle - - - Unpack and interleave 32-bit integers from the low half of each 128-bit lane in "a" and "b", and store the results in "dst". + Convert + + + Convert packed 64-bit integers in "a" to packed 32-bit integers with truncation, and store the results in "dst". -INTERLEAVE_DWORDS(src1[127:0], src2[127:0]){ - dst[31:0] := src1[31:0] - dst[63:32] := src2[31:0] - dst[95:64] := src1[63:32] - dst[127:96] := src2[63:32] - RETURN dst[127:0] -} - -dst[127:0] := INTERLEAVE_DWORDS(a[127:0], b[127:0]) -dst[255:128] := INTERLEAVE_DWORDS(a[255:128], b[255:128]) -dst[383:256] := INTERLEAVE_DWORDS(a[383:256], b[383:256]) -dst[511:384] := INTERLEAVE_DWORDS(a[511:384], b[511:384]) -dst[MAX:512] := 0 +FOR j := 0 to 1 + i := 64*j + k := 32*j + dst[k+31:k] := Truncate32(a[i+63:i]) +ENDFOR +dst[MAX:64] := 0 - +
immintrin.h
 - + Integer + AVX512VL AVX512F - Swizzle - - - - - Unpack and interleave 64-bit integers from the low half of each 128-bit lane in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Convert + + + + + Convert packed 64-bit integers in "a" to packed 32-bit integers with truncation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -INTERLEAVE_QWORDS(src1[127:0], src2[127:0]){ - dst[63:0] := src1[63:0] - dst[127:64] := src2[63:0] - RETURN dst[127:0] -} - -tmp_dst[127:0] := INTERLEAVE_QWORDS(a[127:0], b[127:0]) -tmp_dst[255:128] := INTERLEAVE_QWORDS(a[255:128], b[255:128]) -tmp_dst[383:256] := INTERLEAVE_QWORDS(a[383:256], b[383:256]) -tmp_dst[511:384] := INTERLEAVE_QWORDS(a[511:384], b[511:384]) - -FOR j := 0 to 7 - i := j*64 +FOR j := 0 to 1 + i := 64*j + l := 32*j IF k[j] - dst[i+63:i] := tmp_dst[i+63:i] + dst[l+31:l] := Truncate32(a[i+63:i]) ELSE - dst[i+63:i] := src[i+63:i] + dst[l+31:l] := src[l+31:l] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:64] := 0 - +
immintrin.h
 - + Integer + AVX512VL AVX512F - Swizzle - - - - Unpack and interleave 64-bit integers from the low half of each 128-bit lane in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Convert + Store + + + + + Convert packed 64-bit integers in "a" to packed 32-bit integers with truncation, and store the active results (those with their respective bit set in writemask "k") to unaligned memory at "base_addr". -INTERLEAVE_QWORDS(src1[127:0], src2[127:0]){ - dst[63:0] := src1[63:0] - dst[127:64] := src2[63:0] - RETURN dst[127:0] -} - -tmp_dst[127:0] := INTERLEAVE_QWORDS(a[127:0], b[127:0]) -tmp_dst[255:128] := INTERLEAVE_QWORDS(a[255:128], b[255:128]) -tmp_dst[383:256] := INTERLEAVE_QWORDS(a[383:256], b[383:256]) -tmp_dst[511:384] := INTERLEAVE_QWORDS(a[511:384], b[511:384]) - -FOR j := 0 to 7 - i := j*64 +FOR j := 0 to 1 + i := 64*j + l := 32*j IF k[j] - dst[i+63:i] := tmp_dst[i+63:i] + MEM[base_addr+l+31:base_addr+l] := Truncate32(a[i+63:i]) + FI +ENDFOR + + +
immintrin.h
+ + + Integer + AVX512VL + AVX512F + Convert + + + + Convert packed 64-bit integers in "a" to packed 32-bit integers with truncation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + +FOR j := 0 to 1 + i := 64*j + l := 32*j + IF k[j] + dst[l+31:l] := Truncate32(a[i+63:i]) ELSE - dst[i+63:i] := 0 + dst[l+31:l] := 0 FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:64] := 0 - +
immintrin.h
 - + Integer + AVX512VL AVX512F - Swizzle - - - Unpack and interleave 64-bit integers from the low half of each 128-bit lane in "a" and "b", and store the results in "dst". + Convert + + + Convert packed 64-bit integers in "a" to packed 16-bit integers with truncation, and store the results in "dst". -INTERLEAVE_QWORDS(src1[127:0], src2[127:0]){ - dst[63:0] := src1[63:0] - dst[127:64] := src2[63:0] - RETURN dst[127:0] -} - -dst[127:0] := INTERLEAVE_QWORDS(a[127:0], b[127:0]) -dst[255:128] := INTERLEAVE_QWORDS(a[255:128], b[255:128]) -dst[383:256] := INTERLEAVE_QWORDS(a[383:256], b[383:256]) -dst[511:384] := INTERLEAVE_QWORDS(a[511:384], b[511:384]) -dst[MAX:512] := 0 +FOR j := 0 to 3 + i := 64*j + k := 16*j + dst[k+15:k] := Truncate16(a[i+63:i]) +ENDFOR +dst[MAX:64] := 0 - +
immintrin.h
 - + Integer - AVX512F/KNCNI - Logical - - - - - Compute the bitwise XOR of packed 32-bit integers in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + AVX512VL + AVX512F + Convert + + + + + Convert packed 64-bit integers in "a" to packed 16-bit integers with truncation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*32 +FOR j := 0 to 3 + i := 64*j + l := 16*j IF k[j] - dst[i+31:i] := a[i+31:i] XOR b[i+31:i] + dst[l+15:l] := Truncate16(a[i+63:i]) ELSE - dst[i+31:i] := src[i+31:i] + dst[l+15:l] := src[l+15:l] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:64] := 0 - +
immintrin.h
 - + Integer + AVX512VL AVX512F - Logical - - - - Compute the bitwise XOR of packed 32-bit integers in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Convert + Store + + + + + Convert packed 64-bit integers in "a" to packed 16-bit integers with truncation, and store the active results (those with their respective bit set in writemask "k") to unaligned memory at "base_addr". -FOR j := 0 to 15 - i := j*32 +FOR j := 0 to 3 + i := 64*j + l := 16*j IF k[j] - dst[i+31:i] := a[i+31:i] XOR b[i+31:i] - ELSE - dst[i+31:i] := 0 + MEM[base_addr+l+15:base_addr+l] := Truncate16(a[i+63:i]) FI ENDFOR -dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX512F/KNCNI - Logical - - - Compute the bitwise XOR of packed 32-bit integers in "a" and "b", and store the results in "dst". + AVX512VL + AVX512F + Convert + + + + Convert packed 64-bit integers in "a" to packed 16-bit integers with truncation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*32 - dst[i+31:i] := a[i+31:i] XOR b[i+31:i] +FOR j := 0 to 3 + i := 64*j + l := 16*j + IF k[j] + dst[l+15:l] := Truncate16(a[i+63:i]) + ELSE + dst[l+15:l] := 0 + FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:64] := 0 - +
immintrin.h
 - + Integer - AVX512F/KNCNI - Logical - - - Compute the bitwise XOR of 512 bits (representing integer data) in "a" and "b", and store the result in "dst". + AVX512VL + AVX512F + Convert + + + Convert packed 64-bit integers in "a" to packed 16-bit integers with truncation, and store the results in "dst". -dst[511:0] := (a[511:0] XOR b[511:0]) -dst[MAX:512] := 0 +FOR j := 0 to 1 + i := 64*j + k := 16*j + dst[k+15:k] := Truncate16(a[i+63:i]) +ENDFOR +dst[MAX:32] := 0 - +
immintrin.h
 - + Integer - AVX512F/KNCNI - Logical - - - - - Compute the bitwise XOR of packed 64-bit integers in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + AVX512VL + AVX512F + Convert + + + + + Convert packed 64-bit integers in "a" to packed 16-bit integers with truncation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := j*64 +FOR j := 0 to 1 + i := 64*j + l := 16*j IF k[j] - dst[i+63:i] := a[i+63:i] XOR b[i+63:i] + dst[l+15:l] := Truncate16(a[i+63:i]) ELSE - dst[i+63:i] := src[i+63:i] + dst[l+15:l] := src[l+15:l] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:32] := 0 - +
immintrin.h
 - + Integer + AVX512VL AVX512F - Logical - - - - Compute the bitwise XOR of packed 64-bit integers in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Convert + Store + + + + + Convert packed 64-bit integers in "a" to packed 16-bit integers with truncation, and store the active results (those with their respective bit set in writemask "k") to unaligned memory at "base_addr". -FOR j := 0 to 7 - i := j*64 +FOR j := 0 to 1 + i := 64*j + l := 16*j IF k[j] - dst[i+63:i] := a[i+63:i] XOR b[i+63:i] + MEM[base_addr+l+15:base_addr+l] := Truncate16(a[i+63:i]) + FI +ENDFOR + + +
immintrin.h
+ + + Integer + AVX512VL + AVX512F + Convert + + + + Convert packed 64-bit integers in "a" to packed 16-bit integers with truncation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + +FOR j := 0 to 1 + i := 64*j + l := 16*j + IF k[j] + dst[l+15:l] := Truncate16(a[i+63:i]) ELSE - dst[i+63:i] := 0 + dst[l+15:l] := 0 FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:32] := 0 - +
immintrin.h
 - + Integer - AVX512F/KNCNI - Logical - - - Compute the bitwise XOR of packed 64-bit integers in "a" and "b", and store the results in "dst". - + AVX512VL + AVX512F + Convert + + + Convert packed signed 32-bit integers in "a" to packed 8-bit integers with signed saturation, and store the results in "dst". FOR j := 0 to 7 - i := j*64 - dst[i+63:i] := a[i+63:i] XOR b[i+63:i] + i := 32*j + k := 8*j + dst[k+7:k] := Saturate8(a[i+31:i]) ENDFOR -dst[MAX:512] := 0 +dst[MAX:64] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Elementary Math Functions - - - - Compute the approximate reciprocal of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). The maximum relative error for this approximation is less than 2^-14. + Convert + + + + + Convert packed signed 32-bit integers in "a" to packed 8-bit integers with signed saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). FOR j := 0 to 7 - i := j*64 + i := 32*j + l := 8*j IF k[j] - dst[i+63:i] := APPROXIMATE(1.0/a[i+63:i]) + dst[l+7:l] := Saturate8(a[i+31:i]) ELSE - dst[i+63:i] := src[i+63:i] + dst[l+7:l] := src[l+7:l] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:64] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Elementary Math Functions - - - Compute the approximate reciprocal of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). The maximum relative error for this approximation is less than 2^-14. + Convert + Store + + + + + Convert packed signed 32-bit integers in "a" to packed 8-bit integers with signed saturation, and store the active results (those with their respective bit set in writemask "k") to unaligned memory at "base_addr". FOR j := 0 to 7 - i := j*64 + i := 32*j + l := 8*j IF k[j] - dst[i+63:i] := APPROXIMATE(1.0/a[i+63:i]) - ELSE - dst[i+63:i] := 0 + MEM[base_addr+l+7:base_addr+l] := Saturate8(a[i+31:i]) FI ENDFOR -dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Elementary Math Functions - - Compute the approximate reciprocal of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst". The maximum relative error for this approximation is less than 2^-14. + Convert + + + + Convert packed signed 32-bit integers in "a" to packed 8-bit integers with signed saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). FOR j := 0 to 7 - i := j*64 - dst[i+63:i] := APPROXIMATE(1.0/a[i+63:i]) + i := 32*j + l := 8*j + IF k[j] + dst[l+7:l] := Saturate8(a[i+31:i]) + ELSE + dst[l+7:l] := 0 + FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:64] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Elementary Math Functions - - - - Compute the approximate reciprocal of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). The maximum relative error for this approximation is less than 2^-14. + Convert + + + Convert packed signed 32-bit integers in "a" to packed 8-bit integers with signed saturation, and store the results in "dst". -FOR j := 0 to 15 - i := j*32 - IF k[j] - dst[i+31:i] := APPROXIMATE(1.0/a[i+31:i]) - ELSE - dst[i+31:i] := src[i+31:i] - FI +FOR j := 0 to 3 + i := 32*j + k := 8*j + dst[k+7:k] := Saturate8(a[i+31:i]) ENDFOR -dst[MAX:512] := 0 +dst[MAX:32] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Elementary Math Functions - - - Compute the approximate reciprocal of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). The maximum relative error for this approximation is less than 2^-14. + Convert + + + + + Convert packed signed 32-bit integers in "a" to packed 8-bit integers with signed saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*32 +FOR j := 0 to 3 + i := 32*j + l := 8*j IF k[j] - dst[i+31:i] := APPROXIMATE(1.0/a[i+31:i]) + dst[l+7:l] := Saturate8(a[i+31:i]) ELSE - dst[i+31:i] := 0 + dst[l+7:l] := src[l+7:l] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:32] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Elementary Math Functions - - Compute the approximate reciprocal of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst". The maximum relative error for this approximation is less than 2^-14. + Convert + Store + + + + + Convert packed signed 32-bit integers in "a" to packed 8-bit integers with signed saturation, and store the active results (those with their respective bit set in writemask "k") to unaligned memory at "base_addr". -FOR j := 0 to 15 - i := j*32 - dst[i+31:i] := APPROXIMATE(1.0/a[i+31:i]) +FOR j := 0 to 3 + i := 32*j + l := 8*j + IF k[j] + MEM[base_addr+l+7:base_addr+l] := Saturate8(a[i+31:i]) + FI ENDFOR -dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Elementary Math Functions - - - - - Compute the approximate reciprocal of the lower double-precision (64-bit) floating-point element in "b", store the result in the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". The maximum relative error for this approximation is less than 2^-14. + Convert + + + + Convert packed signed 32-bit integers in "a" to packed 8-bit integers with signed saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -IF k[0] - dst[63:0] := APPROXIMATE(1.0/b[63:0]) -ELSE - dst[63:0] := src[63:0] -FI -dst[127:64] := a[127:64] -dst[MAX:128] := 0 +FOR j := 0 to 3 + i := 32*j + l := 8*j + IF k[j] + dst[l+7:l] := Saturate8(a[i+31:i]) + ELSE + dst[l+7:l] := 0 + FI +ENDFOR +dst[MAX:32] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Elementary Math Functions - - - - Compute the approximate reciprocal of the lower double-precision (64-bit) floating-point element in "b", store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". The maximum relative error for this approximation is less than 2^-14. + Convert + + + Convert packed signed 32-bit integers in "a" to packed 16-bit integers with signed saturation, and store the results in "dst". -IF k[0] - dst[63:0] := APPROXIMATE(1.0/b[63:0]) -ELSE - dst[63:0] := 0 -FI -dst[127:64] := a[127:64] +FOR j := 0 to 7 + i := 32*j + k := 16*j + dst[k+15:k] := Saturate16(a[i+31:i]) +ENDFOR dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Elementary Math Functions - - - Compute the approximate reciprocal of the lower double-precision (64-bit) floating-point element in "b", store the result in the lower element of "dst", and copy the upper element from "a" to the upper element of "dst". The maximum relative error for this approximation is less than 2^-14. + Convert + + + + + Convert packed signed 32-bit integers in "a" to packed 16-bit integers with signed saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -dst[63:0] := APPROXIMATE(1.0/b[63:0]) -dst[127:64] := a[127:64] +FOR j := 0 to 7 + i := 32*j + l := 16*j + IF k[j] + dst[l+15:l] := Saturate16(a[i+31:i]) + ELSE + dst[l+15:l] := src[l+15:l] + FI +ENDFOR dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point - AVX512F - Elementary Math Functions - - - - - Compute the approximate reciprocal of the lower single-precision (32-bit) floating-point element in "b", store the result in the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". The maximum relative error for this approximation is less than 2^-14. + + Integer + AVX512VL + AVX512F + Convert + Store + + + + + Convert packed signed 32-bit integers in "a" to packed 16-bit integers with signed saturation, and store the active results (those with their respective bit set in writemask "k") to unaligned memory at "base_addr". -IF k[0] - dst[31:0] := APPROXIMATE(1.0/b[31:0]) -ELSE - dst[31:0] := src[31:0] -FI -dst[127:32] := a[127:32] -dst[MAX:128] := 0 +FOR j := 0 to 7 + i := 32*j + l := 16*j + IF k[j] + MEM[base_addr+l+15:base_addr+l] := Saturate16(a[i+31:i]) + FI +ENDFOR - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Elementary Math Functions - - - - Compute the approximate reciprocal of the lower single-precision (32-bit) floating-point element in "b", store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". The maximum relative error for this approximation is less than 2^-14. + Convert + + + + Convert packed signed 32-bit integers in "a" to packed 16-bit integers with signed saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -IF k[0] - dst[31:0] := APPROXIMATE(1.0/b[31:0]) -ELSE - dst[31:0] := 0 -FI -dst[127:32] := a[127:32] +FOR j := 0 to 7 + i := 32*j + l := 16*j + IF k[j] + dst[l+15:l] := Saturate16(a[i+31:i]) + ELSE + dst[l+15:l] := 0 + FI +ENDFOR dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Elementary Math Functions - - - Compute the approximate reciprocal of the lower single-precision (32-bit) floating-point element in "b", store the result in the lower element of "dst", and copy the upper 3 packed elements from "a" to the upper elements of "dst". The maximum relative error for this approximation is less than 2^-14. + Convert + + + Convert packed signed 32-bit integers in "a" to packed 16-bit integers with signed saturation, and store the results in "dst". -dst[31:0] := APPROXIMATE(1.0/b[31:0]) -dst[127:32] := a[127:32] -dst[MAX:128] := 0 +FOR j := 0 to 3 + i := 32*j + k := 16*j + dst[k+15:k] := Saturate16(a[i+31:i]) +ENDFOR +dst[MAX:64] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Miscellaneous - - - - - Round packed double-precision (64-bit) floating-point elements in "a" to the number of fraction bits specified by "imm8", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Convert + + + + + Convert packed signed 32-bit integers in "a" to packed 16-bit integers with signed saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -RoundTo_IntegerPD(src[63:0], imm8[7:0]){ - IF(imm8[2] == 1) - rounding_direction := MXCSR.RC //Use the rounding mode specified by MXCSR.RC - ELSE - rounding_direction := imm8[1:0] //Use the rounding mode specified by imm8[1:0] - FI - - M := imm8[7:4] // The scaling factor (number of fraction bits to round to) - - CASE(rounding_direction) - 0: tmp[63:0] := round_to_nearest_even_integer(2^M * src[63:0]) - 1: tmp[63:0] := round_to_equal_or_smaller_integer(2^M * src[63:0]) - 2: tmp[63:0] := round_to_equal_or_larger_integer(2^M * src[63:0]) - 3: tmp[63:0] := round_to_nearest_smallest_magnitude_integer(2^M * src[63:0]) - ESAC - - dst[63:0] := 2^-M * tmp[63:0] // scale back down - - IF imm8[3] == 0 //check SPE - IF src[63:0] != dst[63:0] //check if precision has been lost - set_precision() //set #PE - FI - FI - RETURN dst[63:0] -} - -FOR j := 0 to 7 - i := j*64 +FOR j := 0 to 3 + i := 32*j + l := 16*j IF k[j] - dst[i+63:i] := RoundTo_IntegerPD(a[i+63:i], imm8[7:0]) + dst[l+15:l] := Saturate16(a[i+31:i]) ELSE - dst[i+63:i] := src[i+63:i] + dst[l+15:l] := src[l+15:l] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:64] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Miscellaneous - - - - - - Round packed double-precision (64-bit) floating-point elements in "a" to the number of fraction bits specified by "imm8", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - [round_note] + Convert + Store + + + + + Convert packed signed 32-bit integers in "a" to packed 16-bit integers with signed saturation, and store the active results (those with their respective bit set in writemask "k") to unaligned memory at "base_addr". -RoundTo_IntegerPD(src[63:0], imm8[7:0]){ - IF(imm8[2] == 1) - rounding_direction := MXCSR.RC //Use the rounding mode specified by MXCSR.RC - ELSE - rounding_direction := imm8[1:0] //Use the rounding mode specified by imm8[1:0] - FI - - M := imm8[7:4] // The scaling factor (number of fraction bits to round to) - - CASE(rounding_direction) - 0: tmp[63:0] := round_to_nearest_even_integer(2^M * src[63:0]) - 1: tmp[63:0] := round_to_equal_or_smaller_integer(2^M * src[63:0]) - 2: tmp[63:0] := round_to_equal_or_larger_integer(2^M * src[63:0]) - 3: tmp[63:0] := round_to_nearest_smallest_magnitude_integer(2^M * src[63:0]) - ESAC - - dst[63:0] := 2^-M * tmp[63:0] // scale back down - - IF imm8[3] == 0 //check SPE - IF src[63:0] != dst[63:0] //check if precision has been lost - set_precision() //set #PE - FI - FI - RETURN dst[63:0] -} - -FOR j := 0 to 7 - i := j*64 +FOR j := 0 to 3 + i := 32*j + l := 16*j IF k[j] - dst[i+63:i] := RoundTo_IntegerPD(a[i+63:i], imm8[7:0]) - ELSE - dst[i+63:i] := src[i+63:i] + MEM[base_addr+l+15:base_addr+l] := Saturate16(a[i+31:i]) FI ENDFOR -dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Miscellaneous - - - - Round packed double-precision (64-bit) floating-point elements in "a" to the number of fraction bits specified by "imm8", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Convert + + + + Convert packed signed 32-bit integers in "a" to packed 16-bit integers with signed saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -RoundTo_IntegerPD(src[63:0], imm8[7:0]){ - IF(imm8[2] == 1) - rounding_direction := MXCSR.RC //Use the rounding mode specified by MXCSR.RC - ELSE - rounding_direction := imm8[1:0] //Use the rounding mode specified by imm8[1:0] - FI - - M := imm8[7:4] // The scaling factor (number of fraction bits to round to) - - CASE(rounding_direction) - 0: tmp[63:0] := round_to_nearest_even_integer(2^M * src[63:0]) - 1: tmp[63:0] := round_to_equal_or_smaller_integer(2^M * src[63:0]) - 2: tmp[63:0] := round_to_equal_or_larger_integer(2^M * src[63:0]) - 3: tmp[63:0] := round_to_nearest_smallest_magnitude_integer(2^M * src[63:0]) - ESAC - - dst[63:0] := 2^-M * tmp[63:0] // scale back down - - IF imm8[3] == 0 //check SPE - IF src[63:0] != dst[63:0] //check if precision has been lost - set_precision() //set #PE - FI - FI - RETURN dst[63:0] -} - -FOR j := 0 to 7 - i := j*64 +FOR j := 0 to 3 + i := 32*j + l := 16*j IF k[j] - dst[i+63:i] := RoundTo_IntegerPD(a[i+63:i], imm8[7:0]) + dst[l+15:l] := Saturate16(a[i+31:i]) ELSE - dst[i+63:i] := 0 + dst[l+15:l] := 0 FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:64] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Miscellaneous - - - - - Round packed double-precision (64-bit) floating-point elements in "a" to the number of fraction bits specified by "imm8", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - [round_note] + Convert + + + Convert packed signed 64-bit integers in "a" to packed 8-bit integers with signed saturation, and store the results in "dst". -RoundTo_IntegerPD(src[63:0], imm8[7:0]){ - IF(imm8[2] == 1) - rounding_direction := MXCSR.RC //Use the rounding mode specified by MXCSR.RC - ELSE - rounding_direction := imm8[1:0] //Use the rounding mode specified by imm8[1:0] - FI - - M := imm8[7:4] // The scaling factor (number of fraction bits to round to) - - CASE(rounding_direction) - 0: tmp[63:0] := round_to_nearest_even_integer(2^M * src[63:0]) - 1: tmp[63:0] := round_to_equal_or_smaller_integer(2^M * src[63:0]) - 2: tmp[63:0] := round_to_equal_or_larger_integer(2^M * src[63:0]) - 3: tmp[63:0] := round_to_nearest_smallest_magnitude_integer(2^M * src[63:0]) - ESAC - - dst[63:0] := 2^-M * tmp[63:0] // scale back down - - IF imm8[3] == 0 //check SPE - IF src[63:0] != dst[63:0] //check if precision has been lost - set_precision() //set #PE - FI - FI - RETURN dst[63:0] -} - -FOR j := 0 to 7 - i := j*64 - IF k[j] - dst[i+63:i] := RoundTo_IntegerPD(a[i+63:i], imm8[7:0]) - ELSE - dst[i+63:i] := 0 - FI +FOR j := 0 to 3 + i := 64*j + k := 8*j + dst[k+7:k] := Saturate8(a[i+63:i]) ENDFOR -dst[MAX:512] := 0 +dst[MAX:32] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Miscellaneous - - - Round packed double-precision (64-bit) floating-point elements in "a" to the number of fraction bits specified by "imm8", and store the results in "dst". - + Convert + + + + + Convert packed signed 64-bit integers in "a" to packed 8-bit integers with signed saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -RoundTo_IntegerPD(src[63:0], imm8[7:0]){ - IF(imm8[2] == 1) - rounding_direction := MXCSR.RC //Use the rounding mode specified by MXCSR.RC +FOR j := 0 to 3 + i := 64*j + l := 8*j + IF k[j] + dst[l+7:l] := Saturate8(a[i+63:i]) ELSE - rounding_direction := imm8[1:0] //Use the rounding mode specified by imm8[1:0] - FI - - M := imm8[7:4] // The scaling factor (number of fraction bits to round to) - - CASE(rounding_direction) - 0: tmp[63:0] := round_to_nearest_even_integer(2^M * src[63:0]) - 1: tmp[63:0] := round_to_equal_or_smaller_integer(2^M * src[63:0]) - 2: tmp[63:0] := round_to_equal_or_larger_integer(2^M * src[63:0]) - 3: tmp[63:0] := round_to_nearest_smallest_magnitude_integer(2^M * src[63:0]) - ESAC - - dst[63:0] := 2^-M * tmp[63:0] // scale back down - - IF imm8[3] == 0 //check SPE - IF src[63:0] != dst[63:0] //check if precision has been lost - set_precision() //set #PE - FI + dst[l+7:l] := src[l+7:l] FI - RETURN dst[63:0] -} - -FOR j := 0 to 7 - i := j*64 - dst[i+63:i] := RoundTo_IntegerPD(a[i+63:i], imm8[7:0]) ENDFOR -dst[MAX:512] := 0 +dst[MAX:32] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Miscellaneous - - - - Round packed double-precision (64-bit) floating-point elements in "a" to the number of fraction bits specified by "imm8", and store the results in "dst". - [round_note] + Convert + Store + + + + + Convert packed signed 64-bit integers in "a" to packed 8-bit integers with signed saturation, and store the active results (those with their respective bit set in writemask "k") to unaligned memory at "base_addr". -RoundTo_IntegerPD(src[63:0], imm8[7:0]){ - IF(imm8[2] == 1) - rounding_direction := MXCSR.RC //Use the rounding mode specified by MXCSR.RC - ELSE - rounding_direction := imm8[1:0] //Use the rounding mode specified by imm8[1:0] - FI - - M := imm8[7:4] // The scaling factor (number of fraction bits to round to) - - CASE(rounding_direction) - 0: tmp[63:0] := round_to_nearest_even_integer(2^M * src[63:0]) - 1: tmp[63:0] := round_to_equal_or_smaller_integer(2^M * src[63:0]) - 2: tmp[63:0] := round_to_equal_or_larger_integer(2^M * src[63:0]) - 3: tmp[63:0] := round_to_nearest_smallest_magnitude_integer(2^M * src[63:0]) - ESAC - - dst[63:0] := 2^-M * tmp[63:0] // scale back down - - IF imm8[3] == 0 //check SPE - IF src[63:0] != dst[63:0] //check if precision has been lost - set_precision() //set #PE - FI +FOR j := 0 to 3 + i := 64*j + l := 8*j + IF k[j] + MEM[base_addr+l+7:base_addr+l] := Saturate8(a[i+63:i]) FI - RETURN dst[63:0] -} - -FOR j := 0 to 7 - i := j*64 - dst[i+63:i] := RoundTo_IntegerPD(a[i+63:i], imm8[7:0]) ENDFOR -dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Miscellaneous - - - - - Round packed single-precision (32-bit) floating-point elements in "a" to the number of fraction bits specified by "imm8", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Convert + + + + Convert packed signed 64-bit integers in "a" to packed 8-bit integers with signed saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -RoundTo_IntegerPS(src[31:0], imm8[7:0]){ - IF(imm8[2] == 1) - rounding_direction := MXCSR.RC //Use the rounding mode specified by MXCSR.RC - ELSE - rounding_direction := imm8[1:0] //Use the rounding mode specified by imm8[1:0] - FI - - M := imm8[7:4] // The scaling factor (number of fraction bits to round to) - - CASE(rounding_direction) - 0: tmp[31:0] := round_to_nearest_even_integer(2^M * src[31:0]) - 1: tmp[31:0] := round_to_equal_or_smaller_integer(2^M * src[31:0]) - 2: tmp[31:0] := round_to_equal_or_larger_integer(2^M * src[31:0]) - 3: tmp[31:0] := round_to_nearest_smallest_magnitude_integer(2^M * src[31:0]) - ESAC - - dst[31:0] := 2^-M * tmp[31:0] // scale back down - - IF imm8[3] == 0 //check SPE - IF src[31:0] != dst[31:0] //check if precision has been lost - set_precision() //set #PE - FI - FI - RETURN dst[31:0] -} - -FOR j := 0 to 15 - i := j*32 +FOR j := 0 to 3 + i := 64*j + l := 8*j IF k[j] - dst[i+31:i] := RoundTo_IntegerPS(a[i+31:i], imm8[7:0]) + dst[l+7:l] := Saturate8(a[i+63:i]) ELSE - dst[i+31:i] := src[i+31:i] + dst[l+7:l] := 0 FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:32] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Miscellaneous - - - - - - Round packed single-precision (32-bit) floating-point elements in "a" to the number of fraction bits specified by "imm8", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - [round_note] + Convert + + + Convert packed signed 64-bit integers in "a" to packed 8-bit integers with signed saturation, and store the results in "dst". -RoundTo_IntegerPS(src[31:0], imm8[7:0]){ - IF(imm8[2] == 1) - rounding_direction := MXCSR.RC //Use the rounding mode specified by MXCSR.RC - ELSE - rounding_direction := imm8[1:0] //Use the rounding mode specified by imm8[1:0] - FI - - M := imm8[7:4] // The scaling factor (number of fraction bits to round to) - - CASE(rounding_direction) - 0: tmp[31:0] := round_to_nearest_even_integer(2^M * src[31:0]) - 1: tmp[31:0] := round_to_equal_or_smaller_integer(2^M * src[31:0]) - 2: tmp[31:0] := round_to_equal_or_larger_integer(2^M * src[31:0]) - 3: tmp[31:0] := round_to_nearest_smallest_magnitude_integer(2^M * src[31:0]) - ESAC - - dst[31:0] := 2^-M * tmp[31:0] // scale back down - - IF imm8[3] == 0 //check SPE - IF src[31:0] != dst[31:0] //check if precision has been lost - set_precision() //set #PE - FI - FI - RETURN dst[31:0] -} - -FOR j := 0 to 15 - i := j*32 - IF k[j] - dst[i+31:i] := RoundTo_IntegerPS(a[i+31:i], imm8[7:0]) - ELSE - dst[i+31:i] := src[i+31:i] - FI +FOR j := 0 to 1 + i := 64*j + k := 8*j + dst[k+7:k] := Saturate8(a[i+63:i]) ENDFOR -dst[MAX:512] := 0 +dst[MAX:16] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Miscellaneous - - - - Round packed single-precision (32-bit) floating-point elements in "a" to the number of fraction bits specified by "imm8", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Convert + + + + + Convert packed signed 64-bit integers in "a" to packed 8-bit integers with signed saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -RoundTo_IntegerPS(src[31:0], imm8[7:0]){ - IF(imm8[2] == 1) - rounding_direction := MXCSR.RC //Use the rounding mode specified by MXCSR.RC - ELSE - rounding_direction := imm8[1:0] //Use the rounding mode specified by imm8[1:0] - FI - - M := imm8[7:4] // The scaling factor (number of fraction bits to round to) - - CASE(rounding_direction) - 0: tmp[31:0] := round_to_nearest_even_integer(2^M * src[31:0]) - 1: tmp[31:0] := round_to_equal_or_smaller_integer(2^M * src[31:0]) - 2: tmp[31:0] := round_to_equal_or_larger_integer(2^M * src[31:0]) - 3: tmp[31:0] := round_to_nearest_smallest_magnitude_integer(2^M * src[31:0]) - ESAC - - dst[31:0] := 2^-M * tmp[31:0] // scale back down - - IF imm8[3] == 0 //check SPE - IF src[31:0] != dst[31:0] //check if precision has been lost - set_precision() //set #PE - FI - FI - RETURN dst[31:0] -} - -FOR j := 0 to 15 - i := j*32 +FOR j := 0 to 1 + i := 64*j + l := 8*j IF k[j] - dst[i+31:i] := RoundTo_IntegerPS(a[i+31:i], imm8[7:0]) + dst[l+7:l] := Saturate8(a[i+63:i]) ELSE - dst[i+31:i] := 0 + dst[l+7:l] := src[l+7:l] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:16] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Miscellaneous - - - - - Round packed single-precision (32-bit) floating-point elements in "a" to the number of fraction bits specified by "imm8", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - [round_note] + Convert + Store + + + + + Convert packed signed 64-bit integers in "a" to packed 8-bit integers with signed saturation, and store the active results (those with their respective bit set in writemask "k") to unaligned memory at "base_addr". -RoundTo_IntegerPS(src[31:0], imm8[7:0]){ - IF(imm8[2] == 1) - rounding_direction := MXCSR.RC //Use the rounding mode specified by MXCSR.RC - ELSE - rounding_direction := imm8[1:0] //Use the rounding mode specified by imm8[1:0] - FI - - M := imm8[7:4] // The scaling factor (number of fraction bits to round to) - - CASE(rounding_direction) - 0: tmp[31:0] := round_to_nearest_even_integer(2^M * src[31:0]) - 1: tmp[31:0] := round_to_equal_or_smaller_integer(2^M * src[31:0]) - 2: tmp[31:0] := round_to_equal_or_larger_integer(2^M * src[31:0]) - 3: tmp[31:0] := round_to_nearest_smallest_magnitude_integer(2^M * src[31:0]) - ESAC - - dst[31:0] := 2^-M * tmp[31:0] // scale back down - - IF imm8[3] == 0 //check SPE - IF src[31:0] != dst[31:0] //check if precision has been lost - set_precision() //set #PE - FI - FI - RETURN dst[31:0] -} - -FOR j := 0 to 15 - i := j*32 +FOR j := 0 to 1 + i := 64*j + l := 8*j IF k[j] - dst[i+31:i] := RoundTo_IntegerPS(a[i+31:i], imm8[7:0]) - ELSE - dst[i+31:i] := 0 + MEM[base_addr+l+7:base_addr+l] := Saturate8(a[i+63:i]) FI ENDFOR -dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Miscellaneous - - - Round packed single-precision (32-bit) floating-point elements in "a" to the number of fraction bits specified by "imm8", and store the results in "dst". - + Convert + + + + Convert packed signed 64-bit integers in "a" to packed 8-bit integers with signed saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -RoundTo_IntegerPS(src[31:0], imm8[7:0]){ - IF(imm8[2] == 1) - rounding_direction := MXCSR.RC //Use the rounding mode specified by MXCSR.RC +FOR j := 0 to 1 + i := 64*j + l := 8*j + IF k[j] + dst[l+7:l] := Saturate8(a[i+63:i]) ELSE - rounding_direction := imm8[1:0] //Use the rounding mode specified by imm8[1:0] - FI - - M := imm8[7:4] // The scaling factor (number of fraction bits to round to) - - CASE(rounding_direction) - 0: tmp[31:0] := round_to_nearest_even_integer(2^M * src[31:0]) - 1: tmp[31:0] := round_to_equal_or_smaller_integer(2^M * src[31:0]) - 2: tmp[31:0] := round_to_equal_or_larger_integer(2^M * src[31:0]) - 3: tmp[31:0] := round_to_nearest_smallest_magnitude_integer(2^M * src[31:0]) - ESAC - - dst[31:0] := 2^-M * tmp[31:0] // scale back down - - IF imm8[3] == 0 //check SPE - IF src[31:0] != dst[31:0] //check if precision has been lost - set_precision() //set #PE - FI + dst[l+7:l] := 0 FI - RETURN dst[31:0] -} - -FOR j := 0 to 15 - i := j*32 - dst[i+31:i] := RoundTo_IntegerPS(a[i+31:i], imm8[7:0]) ENDFOR -dst[MAX:512] := 0 +dst[MAX:16] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Miscellaneous - - - - Round packed single-precision (32-bit) floating-point elements in "a" to the number of fraction bits specified by "imm8", and store the results in "dst". - [round_note] + Convert + + + Convert packed signed 64-bit integers in "a" to packed 32-bit integers with signed saturation, and store the results in "dst". -RoundTo_IntegerPS(src[31:0], imm8[7:0]){ - IF(imm8[2] == 1) - rounding_direction := MXCSR.RC //Use the rounding mode specified by MXCSR.RC - ELSE - rounding_direction := imm8[1:0] //Use the rounding mode specified by imm8[1:0] - FI - - M := imm8[7:4] // The scaling factor (number of fraction bits to round to) - - CASE(rounding_direction) - 0: tmp[31:0] := round_to_nearest_even_integer(2^M * src[31:0]) - 1: tmp[31:0] := round_to_equal_or_smaller_integer(2^M * src[31:0]) - 2: tmp[31:0] := round_to_equal_or_larger_integer(2^M * src[31:0]) - 3: tmp[31:0] := round_to_nearest_smallest_magnitude_integer(2^M * src[31:0]) - ESAC - - dst[31:0] := 2^-M * tmp[31:0] // scale back down - - IF imm8[3] == 0 //check SPE - IF src[31:0] != dst[31:0] //check if precision has been lost - set_precision() //set #PE - FI - FI - RETURN dst[31:0] -} - -FOR j := 0 to 15 - i := j*32 - dst[i+31:i] := RoundTo_IntegerPS(a[i+31:i], imm8[7:0]) +FOR j := 0 to 3 + i := 64*j + k := 32*j + dst[k+31:k] := Saturate32(a[i+63:i]) ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Miscellaneous - - - - - - - Round the lower double-precision (64-bit) floating-point element in "a" to the number of fraction bits specified by "imm8", store the result in the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and copy the upper element from "b" to the upper element of "dst". - [round_note] + Convert + + + + + Convert packed signed 64-bit integers in "a" to packed 32-bit integers with signed saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -RoundTo_IntegerPD(src[63:0], imm8[7:0]){ - IF(imm8[2] == 1) - rounding_direction := MXCSR.RC //Use the rounding mode specified by MXCSR.RC +FOR j := 0 to 3 + i := 64*j + l := 32*j + IF k[j] + dst[l+31:l] := Saturate32(a[i+63:i]) ELSE - rounding_direction := imm8[1:0] //Use the rounding mode specified by imm8[1:0] - FI - - M := imm8[7:4] // The scaling factor (number of fraction bits to round to) - - CASE(rounding_direction) - 0: tmp[63:0] := round_to_nearest_even_integer(2^M * src[63:0]) - 1: tmp[63:0] := round_to_equal_or_smaller_integer(2^M * src[63:0]) - 2: tmp[63:0] := round_to_equal_or_larger_integer(2^M * src[63:0]) - 3: tmp[63:0] := round_to_nearest_smallest_magnitude_integer(2^M * src[63:0]) - ESAC - - dst[63:0] := 2^-M * tmp[63:0] // scale back down - - IF imm8[3] == 0 //check SPE - IF src[63:0] != dst[63:0] //check if precision has been lost - set_precision() //set #PE - FI + dst[l+31:l] := src[l+31:l] FI - RETURN dst[63:0] -} - -IF k[0] - dst[63:0] := RoundTo_IntegerPD(a[63:0], imm8[7:0]) -ELSE - dst[63:0] := src[63:0] -FI -dst[127:64] := b[127:64] +ENDFOR dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Miscellaneous - - - - - - Round the lower double-precision (64-bit) floating-point element in "a" to the number of fraction bits specified by "imm8", store the result in the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and copy the upper element from "b" to the upper element of "dst". + Convert + Store + + + + + Convert packed signed 64-bit integers in "a" to packed 32-bit integers with signed saturation, and store the active results (those with their respective bit set in writemask "k") to unaligned memory at "base_addr". -RoundTo_IntegerPD(src[63:0], imm8[7:0]){ - IF(imm8[2] == 1) - rounding_direction := MXCSR.RC //Use the rounding mode specified by MXCSR.RC - ELSE - rounding_direction := imm8[1:0] //Use the rounding mode specified by imm8[1:0] - FI - - M := imm8[7:4] // The scaling factor (number of fraction bits to round to) - - CASE(rounding_direction) - 0: tmp[63:0] := round_to_nearest_even_integer(2^M * src[63:0]) - 1: tmp[63:0] := round_to_equal_or_smaller_integer(2^M * src[63:0]) - 2: tmp[63:0] := round_to_equal_or_larger_integer(2^M * src[63:0]) - 3: tmp[63:0] := round_to_nearest_smallest_magnitude_integer(2^M * src[63:0]) - ESAC - - dst[63:0] := 2^-M * tmp[63:0] // scale back down - - IF imm8[3] == 0 //check SPE - IF src[63:0] != dst[63:0] //check if precision has been lost - set_precision() //set #PE - FI +FOR j := 0 to 3 + i := 64*j + l := 32*j + IF k[j] + MEM[base_addr+l+31:base_addr+l] := Saturate32(a[i+63:i]) FI - RETURN dst[63:0] -} - -IF k[0] - dst[63:0] := RoundTo_IntegerPD(a[63:0], imm8[7:0]) -ELSE - dst[63:0] := src[63:0] -FI -dst[127:64] := b[127:64] -dst[MAX:128] := 0 +ENDFOR - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Miscellaneous - - - - - - Round the lower double-precision (64-bit) floating-point element in "a" to the number of fraction bits specified by "imm8", store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper element from "b" to the upper element of "dst". - [round_note] + Convert + + + + Convert packed signed 64-bit integers in "a" to packed 32-bit integers with signed saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -RoundTo_IntegerPD(src[63:0], imm8[7:0]){ - IF(imm8[2] == 1) - rounding_direction := MXCSR.RC //Use the rounding mode specified by MXCSR.RC +FOR j := 0 to 3 + i := 64*j + l := 32*j + IF k[j] + dst[l+31:l] := Saturate32(a[i+63:i]) ELSE - rounding_direction := imm8[1:0] //Use the rounding mode specified by imm8[1:0] - FI - - M := imm8[7:4] // The scaling factor (number of fraction bits to round to) - - CASE(rounding_direction) - 0: tmp[63:0] := round_to_nearest_even_integer(2^M * src[63:0]) - 1: tmp[63:0] := round_to_equal_or_smaller_integer(2^M * src[63:0]) - 2: tmp[63:0] := round_to_equal_or_larger_integer(2^M * src[63:0]) - 3: tmp[63:0] := round_to_nearest_smallest_magnitude_integer(2^M * src[63:0]) - ESAC - - dst[63:0] := 2^-M * tmp[63:0] // scale back down - - IF imm8[3] == 0 //check SPE - IF src[63:0] != dst[63:0] //check if precision has been lost - set_precision() //set #PE - FI + dst[l+31:l] := 0 FI - RETURN dst[63:0] -} - -IF k[0] - dst[63:0] := RoundTo_IntegerPD(a[63:0], imm8[7:0]) -ELSE - dst[63:0] := 0 -FI -dst[127:64] := b[127:64] +ENDFOR dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Miscellaneous - - - - - Round the lower double-precision (64-bit) floating-point element in "a" to the number of fraction bits specified by "imm8", store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper element from "b" to the upper element of "dst". + Convert + + + Convert packed signed 64-bit integers in "a" to packed 32-bit integers with signed saturation, and store the results in "dst". -RoundTo_IntegerPD(src[63:0], imm8[7:0]){ - IF(imm8[2] == 1) - rounding_direction := MXCSR.RC //Use the rounding mode specified by MXCSR.RC - ELSE - rounding_direction := imm8[1:0] //Use the rounding mode specified by imm8[1:0] - FI - - M := imm8[7:4] // The scaling factor (number of fraction bits to round to) - - CASE(rounding_direction) - 0: tmp[63:0] := round_to_nearest_even_integer(2^M * src[63:0]) - 1: tmp[63:0] := round_to_equal_or_smaller_integer(2^M * src[63:0]) - 2: tmp[63:0] := round_to_equal_or_larger_integer(2^M * src[63:0]) - 3: tmp[63:0] := round_to_nearest_smallest_magnitude_integer(2^M * src[63:0]) - ESAC - - dst[63:0] := 2^-M * tmp[63:0] // scale back down - - IF imm8[3] == 0 //check SPE - IF src[63:0] != dst[63:0] //check if precision has been lost - set_precision() //set #PE - FI - FI - RETURN dst[63:0] -} - -IF k[0] - dst[63:0] := RoundTo_IntegerPD(a[63:0], imm8[7:0]) -ELSE - dst[63:0] := 0 -FI -dst[127:64] := b[127:64] -dst[MAX:128] := 0 +FOR j := 0 to 1 + i := 64*j + k := 32*j + dst[k+31:k] := Saturate32(a[i+63:i]) +ENDFOR +dst[MAX:64] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Miscellaneous - - - - - Round the lower double-precision (64-bit) floating-point element in "a" to the number of fraction bits specified by "imm8", store the result in the lower element of "dst", and copy the upper element from "b" to the upper element of "dst". - [round_note] + Convert + + + + + Convert packed signed 64-bit integers in "a" to packed 32-bit integers with signed saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -RoundTo_IntegerPD(src[63:0], imm8[7:0]){ - IF(imm8[2] == 1) - rounding_direction := MXCSR.RC //Use the rounding mode specified by MXCSR.RC +FOR j := 0 to 1 + i := 64*j + l := 32*j + IF k[j] + dst[l+31:l] := Saturate32(a[i+63:i]) ELSE - rounding_direction := imm8[1:0] //Use the rounding mode specified by imm8[1:0] - FI - - M := imm8[7:4] // The scaling factor (number of fraction bits to round to) - - CASE(rounding_direction) - 0: tmp[63:0] := round_to_nearest_even_integer(2^M * src[63:0]) - 1: tmp[63:0] := round_to_equal_or_smaller_integer(2^M * src[63:0]) - 2: tmp[63:0] := round_to_equal_or_larger_integer(2^M * src[63:0]) - 3: tmp[63:0] := round_to_nearest_smallest_magnitude_integer(2^M * src[63:0]) - ESAC - - dst[63:0] := 2^-M * tmp[63:0] // scale back down - - IF imm8[3] == 0 //check SPE - IF src[63:0] != dst[63:0] //check if precision has been lost - set_precision() //set #PE - FI + dst[l+31:l] := src[l+31:l] FI - RETURN dst[63:0] -} - -dst[63:0] := RoundTo_IntegerPD(a[63:0], imm8[7:0]) -dst[127:64] := b[127:64] -dst[MAX:128] := 0 +ENDFOR +dst[MAX:64] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Miscellaneous - - - - Round the lower double-precision (64-bit) floating-point element in "a" to the number of fraction bits specified by "imm8", store the result in the lower element of "dst", and copy the upper element from "b" to the upper element of "dst". + Convert + Store + + + + + Convert packed signed 64-bit integers in "a" to packed 32-bit integers with signed saturation, and store the active results (those with their respective bit set in writemask "k") to unaligned memory at "base_addr". -RoundTo_IntegerPD(src[63:0], imm8[7:0]){ - IF(imm8[2] == 1) - rounding_direction := MXCSR.RC //Use the rounding mode specified by MXCSR.RC - ELSE - rounding_direction := imm8[1:0] //Use the rounding mode specified by imm8[1:0] - FI - - M := imm8[7:4] // The scaling factor (number of fraction bits to round to) - - CASE(rounding_direction) - 0: tmp[63:0] := round_to_nearest_even_integer(2^M * src[63:0]) - 1: tmp[63:0] := round_to_equal_or_smaller_integer(2^M * src[63:0]) - 2: tmp[63:0] := round_to_equal_or_larger_integer(2^M * src[63:0]) - 3: tmp[63:0] := round_to_nearest_smallest_magnitude_integer(2^M * src[63:0]) - ESAC - - dst[63:0] := 2^-M * tmp[63:0] // scale back down - - IF imm8[3] == 0 //check SPE - IF src[63:0] != dst[63:0] //check if precision has been lost - set_precision() //set #PE - FI +FOR j := 0 to 1 + i := 64*j + l := 32*j + IF k[j] + MEM[base_addr+l+31:base_addr+l] := Saturate32(a[i+63:i]) FI - RETURN dst[63:0] -} - -dst[63:0] := RoundTo_IntegerPD(a[63:0], imm8[7:0]) -dst[127:64] := b[127:64] -dst[MAX:128] := 0 +ENDFOR - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Miscellaneous - - - - - - - Round the lower single-precision (32-bit) floating-point element in "a" to the number of fraction bits specified by "imm8", store the result in the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and copy the upper 3 packed elements from "b" to the upper elements of "dst". - [round_note] + Convert + + + + Convert packed signed 64-bit integers in "a" to packed 32-bit integers with signed saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -RoundTo_IntegerPS(src[31:0], imm8[7:0]){ - IF(imm8[2] == 1) - rounding_direction := MXCSR.RC //Use the rounding mode specified by MXCSR.RC +FOR j := 0 to 1 + i := 64*j + l := 32*j + IF k[j] + dst[l+31:l] := Saturate32(a[i+63:i]) ELSE - rounding_direction := imm8[1:0] //Use the rounding mode specified by imm8[1:0] - FI - - M := imm8[7:4] // The scaling factor (number of fraction bits to round to) - - CASE(rounding_direction) - 0: tmp[31:0] := round_to_nearest_even_integer(2^M * src[31:0]) - 1: tmp[31:0] := round_to_equal_or_smaller_integer(2^M * src[31:0]) - 2: tmp[31:0] := round_to_equal_or_larger_integer(2^M * src[31:0]) - 3: tmp[31:0] := round_to_nearest_smallest_magnitude_integer(2^M * src[31:0]) - ESAC - - dst[31:0] := 2^-M * tmp[31:0] // scale back down - - IF imm8[3] == 0 //check SPE - IF src[31:0] != dst[31:0] //check if precision has been lost - set_precision() //set #PE - FI + dst[l+31:l] := 0 FI - RETURN dst[31:0] -} - -IF k[0] - dst[31:0] := RoundTo_IntegerPS(a[31:0], imm8[7:0]) -ELSE - dst[31:0] := src[31:0] -FI -dst[127:32] := b[127:32] -dst[MAX:128] := 0 +ENDFOR +dst[MAX:64] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Miscellaneous - - - - - - Round the lower single-precision (32-bit) floating-point element in "a" to the number of fraction bits specified by "imm8", store the result in the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and copy the upper 3 packed elements from "b" to the upper elements of "dst". + Convert + + + Convert packed signed 64-bit integers in "a" to packed 16-bit integers with signed saturation, and store the results in "dst". -RoundTo_IntegerPS(src[31:0], imm8[7:0]){ - IF(imm8[2] == 1) - rounding_direction := MXCSR.RC //Use the rounding mode specified by MXCSR.RC - ELSE - rounding_direction := imm8[1:0] //Use the rounding mode specified by imm8[1:0] - FI - - M := imm8[7:4] // The scaling factor (number of fraction bits to round to) - - CASE(rounding_direction) - 0: tmp[31:0] := round_to_nearest_even_integer(2^M * src[31:0]) - 1: tmp[31:0] := round_to_equal_or_smaller_integer(2^M * src[31:0]) - 2: tmp[31:0] := round_to_equal_or_larger_integer(2^M * src[31:0]) - 3: tmp[31:0] := round_to_nearest_smallest_magnitude_integer(2^M * src[31:0]) - ESAC - - dst[31:0] := 2^-M * tmp[31:0] // scale back down - - IF imm8[3] == 0 //check SPE - IF src[31:0] != dst[31:0] //check if precision has been lost - set_precision() //set #PE - FI - FI - RETURN dst[31:0] -} - -IF k[0] - dst[31:0] := RoundTo_IntegerPS(a[31:0], imm8[7:0]) -ELSE - dst[31:0] := src[31:0] -FI -dst[127:32] := b[127:32] -dst[MAX:128] := 0 +FOR j := 0 to 3 + i := 64*j + k := 16*j + dst[k+15:k] := Saturate16(a[i+63:i]) +ENDFOR +dst[MAX:64] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Miscellaneous - - - - - - Round the lower single-precision (32-bit) floating-point element in "a" to the number of fraction bits specified by "imm8", store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper 3 packed elements from "b" to the upper elements of "dst". - [round_note] + Convert + + + + + Convert packed signed 64-bit integers in "a" to packed 16-bit integers with signed saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -RoundTo_IntegerPS(src[31:0], imm8[7:0]){ - IF(imm8[2] == 1) - rounding_direction := MXCSR.RC //Use the rounding mode specified by MXCSR.RC +FOR j := 0 to 3 + i := 64*j + l := 16*j + IF k[j] + dst[l+15:l] := Saturate16(a[i+63:i]) ELSE - rounding_direction := imm8[1:0] //Use the rounding mode specified by imm8[1:0] - FI - - M := imm8[7:4] // The scaling factor (number of fraction bits to round to) - - CASE(rounding_direction) - 0: tmp[31:0] := round_to_nearest_even_integer(2^M * src[31:0]) - 1: tmp[31:0] := round_to_equal_or_smaller_integer(2^M * src[31:0]) - 2: tmp[31:0] := round_to_equal_or_larger_integer(2^M * src[31:0]) - 3: tmp[31:0] := round_to_nearest_smallest_magnitude_integer(2^M * src[31:0]) - ESAC - - dst[31:0] := 2^-M * tmp[31:0] // scale back down - - IF imm8[3] == 0 //check SPE - IF src[31:0] != dst[31:0] //check if precision has been lost - set_precision() //set #PE - FI + dst[l+15:l] := src[l+15:l] FI - RETURN dst[31:0] -} - -IF k[0] - dst[31:0] := RoundTo_IntegerPS(a[31:0], imm8[7:0]) -ELSE - dst[31:0] := 0 -FI -dst[127:32] := b[127:32] -dst[MAX:128] := 0 +ENDFOR +dst[MAX:64] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Miscellaneous - - - - - Round the lower single-precision (32-bit) floating-point element in "a" to the number of fraction bits specified by "imm8", store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper 3 packed elements from "b" to the upper elements of "dst". + Convert + Store + + + + + Convert packed signed 64-bit integers in "a" to packed 16-bit integers with signed saturation, and store the active results (those with their respective bit set in writemask "k") to unaligned memory at "base_addr". -RoundTo_IntegerPS(src[31:0], imm8[7:0]){ - IF(imm8[2] == 1) - rounding_direction := MXCSR.RC //Use the rounding mode specified by MXCSR.RC - ELSE - rounding_direction := imm8[1:0] //Use the rounding mode specified by imm8[1:0] - FI - - M := imm8[7:4] // The scaling factor (number of fraction bits to round to) - - CASE(rounding_direction) - 0: tmp[31:0] := round_to_nearest_even_integer(2^M * src[31:0]) - 1: tmp[31:0] := round_to_equal_or_smaller_integer(2^M * src[31:0]) - 2: tmp[31:0] := round_to_equal_or_larger_integer(2^M * src[31:0]) - 3: tmp[31:0] := round_to_nearest_smallest_magnitude_integer(2^M * src[31:0]) - ESAC - - dst[31:0] := 2^-M * tmp[31:0] // scale back down - - IF imm8[3] == 0 //check SPE - IF src[31:0] != dst[31:0] //check if precision has been lost - set_precision() //set #PE - FI +FOR j := 0 to 3 + i := 64*j + l := 16*j + IF k[j] + MEM[base_addr+l+15:base_addr+l] := Saturate16(a[i+63:i]) FI - RETURN dst[31:0] -} - -IF k[0] - dst[31:0] := RoundTo_IntegerPS(a[31:0], imm8[7:0]) -ELSE - dst[31:0] := 0 -FI -dst[127:32] := b[127:32] -dst[MAX:128] := 0 +ENDFOR - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Miscellaneous - - - - - Round the lower single-precision (32-bit) floating-point element in "a" to the number of fraction bits specified by "imm8", store the result in the lower element of "dst", and copy the upper 3 packed elements from "b" to the upper elements of "dst". - [round_note] + Convert + + + + Convert packed signed 64-bit integers in "a" to packed 16-bit integers with signed saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -RoundTo_IntegerPS(src[31:0], imm8[7:0]){ - IF(imm8[2] == 1) - rounding_direction := MXCSR.RC //Use the rounding mode specified by MXCSR.RC +FOR j := 0 to 3 + i := 64*j + l := 16*j + IF k[j] + dst[l+15:l] := Saturate16(a[i+63:i]) ELSE - rounding_direction := imm8[1:0] //Use the rounding mode specified by imm8[1:0] - FI - - M := imm8[7:4] // The scaling factor (number of fraction bits to round to) - - CASE(rounding_direction) - 0: tmp[31:0] := round_to_nearest_even_integer(2^M * src[31:0]) - 1: tmp[31:0] := round_to_equal_or_smaller_integer(2^M * src[31:0]) - 2: tmp[31:0] := round_to_equal_or_larger_integer(2^M * src[31:0]) - 3: tmp[31:0] := round_to_nearest_smallest_magnitude_integer(2^M * src[31:0]) - ESAC - - dst[31:0] := 2^-M * tmp[31:0] // scale back down - - IF imm8[3] == 0 //check SPE - IF src[31:0] != dst[31:0] //check if precision has been lost - set_precision() //set #PE - FI + dst[l+15:l] := 0 FI - RETURN dst[31:0] -} - -dst[31:0] := RoundTo_IntegerPS(a[31:0], imm8[7:0]) -dst[127:32] := b[127:32] -dst[MAX:128] := 0 +ENDFOR +dst[MAX:64] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Miscellaneous - - - - Round the lower single-precision (32-bit) floating-point element in "a" to the number of fraction bits specified by "imm8", store the result in the lower element of "dst", and copy the upper 3 packed elements from "b" to the upper elements of "dst". + Convert + + + Convert packed signed 64-bit integers in "a" to packed 16-bit integers with signed saturation, and store the results in "dst". -RoundTo_IntegerPS(src[31:0], imm8[7:0]){ - IF(imm8[2] == 1) - rounding_direction := MXCSR.RC //Use the rounding mode specified by MXCSR.RC - ELSE - rounding_direction := imm8[1:0] //Use the rounding mode specified by imm8[1:0] - FI - - M := imm8[7:4] // The scaling factor (number of fraction bits to round to) - - CASE(rounding_direction) - 0: tmp[31:0] := round_to_nearest_even_integer(2^M * src[31:0]) - 1: tmp[31:0] := round_to_equal_or_smaller_integer(2^M * src[31:0]) - 2: tmp[31:0] := round_to_equal_or_larger_integer(2^M * src[31:0]) - 3: tmp[31:0] := round_to_nearest_smallest_magnitude_integer(2^M * src[31:0]) - ESAC - - dst[31:0] := 2^-M * tmp[31:0] // scale back down - - IF imm8[3] == 0 //check SPE - IF src[31:0] != dst[31:0] //check if precision has been lost - set_precision() //set #PE - FI - FI - RETURN dst[31:0] -} - -dst[31:0] := RoundTo_IntegerPS(a[31:0], imm8[7:0]) -dst[127:32] := b[127:32] -dst[MAX:128] := 0 +FOR j := 0 to 1 + i := 64*j + k := 16*j + dst[k+15:k] := Saturate16(a[i+63:i]) +ENDFOR +dst[MAX:32] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Elementary Math Functions - - - - Compute the approximate reciprocal square root of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). The maximum relative error for this approximation is less than 2^-14. + Convert + + + + + Convert packed signed 64-bit integers in "a" to packed 16-bit integers with signed saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := j*64 +FOR j := 0 to 1 + i := 64*j + l := 16*j IF k[j] - dst[i+63:i] := APPROXIMATE(1.0 / SQRT(a[i+63:i])) + dst[l+15:l] := Saturate16(a[i+63:i]) ELSE - dst[i+63:i] := src[i+63:i] + dst[l+15:l] := src[l+15:l] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:32] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Elementary Math Functions - - - Compute the approximate reciprocal square root of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). The maximum relative error for this approximation is less than 2^-14. + Convert + Store + + + + + Convert packed signed 64-bit integers in "a" to packed 16-bit integers with signed saturation, and store the active results (those with their respective bit set in writemask "k") to unaligned memory at "base_addr". -FOR j := 0 to 7 - i := j*64 +FOR j := 0 to 1 + i := 64*j + l := 16*j IF k[j] - dst[i+63:i] := APPROXIMATE(1.0 / SQRT(a[i+63:i])) - ELSE - dst[i+63:i] := 0 + MEM[base_addr+l+15:base_addr+l] := Saturate16(a[i+63:i]) FI ENDFOR -dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Elementary Math Functions - - Compute the approximate reciprocal square root of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst". The maximum relative error for this approximation is less than 2^-14. + Convert + + + + Convert packed signed 64-bit integers in "a" to packed 16-bit integers with signed saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := j*64 - dst[i+63:i] := APPROXIMATE(1.0 / SQRT(a[i+63:i])) +FOR j := 0 to 1 + i := 64*j + l := 16*j + IF k[j] + dst[l+15:l] := Saturate16(a[i+63:i]) + ELSE + dst[l+15:l] := 0 + FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:32] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Elementary Math Functions - - - - Compute the approximate reciprocal square root of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). The maximum relative error for this approximation is less than 2^-14. + Convert + + + + + Sign extend packed 8-bit integers in the low 8 bytes of "a" to packed 32-bit integers, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*32 +FOR j := 0 to 7 + i := 32*j + l := 8*j IF k[j] - dst[i+31:i] := APPROXIMATE(1.0 / SQRT(a[i+31:i])) + dst[i+31:i] := SignExtend32(a[l+7:l]) ELSE dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Elementary Math Functions - - - Compute the approximate reciprocal square root of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). The maximum relative error for this approximation is less than 2^-14. + Convert + + + + Sign extend packed 8-bit integers in the low 8 bytes of "a" to packed 32-bit integers, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*32 +FOR j := 0 to 7 + i := 32*j + l := 8*j IF k[j] - dst[i+31:i] := APPROXIMATE(1.0 / SQRT(a[i+31:i])) + dst[i+31:i] := SignExtend32(a[l+7:l]) ELSE dst[i+31:i] := 0 FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Elementary Math Functions - - Compute the approximate reciprocal square root of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst". The maximum relative error for this approximation is less than 2^-14. + Convert + + + + + Sign extend packed 8-bit integers in the low 4 bytes of "a" to packed 32-bit integers, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*32 - dst[i+31:i] := APPROXIMATE(1.0 / SQRT(a[i+31:i])) +FOR j := 0 to 3 + i := 32*j + l := 8*j + IF k[j] + dst[i+31:i] := SignExtend32(a[l+7:l]) + ELSE + dst[i+31:i] := src[i+31:i] + FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Elementary Math Functions - - - - - Compute the approximate reciprocal square root of the lower double-precision (64-bit) floating-point element in "b", store the result in the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". The maximum relative error for this approximation is less than 2^-14. + Convert + + + + Sign extend packed 8-bit integers in the low 4 bytes of "a" to packed 32-bit integers, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -IF k[0] - dst[63:0] := APPROXIMATE(1.0 / SQRT(b[63:0])) -ELSE - dst[63:0] := src[63:0] -FI -dst[127:64] := a[127:64] +FOR j := 0 to 3 + i := 32*j + l := 8*j + IF k[j] + dst[i+31:i] := SignExtend32(a[l+7:l]) + ELSE + dst[i+31:i] := 0 + FI +ENDFOR dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Elementary Math Functions - - - - Compute the approximate reciprocal square root of the lower double-precision (64-bit) floating-point element in "b", store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". The maximum relative error for this approximation is less than 2^-14. + Convert + + + + + Sign extend packed 8-bit integers in the low 4 bytes of "a" to packed 64-bit integers, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -IF k[0] - dst[63:0] := APPROXIMATE(1.0 / SQRT(b[63:0])) -ELSE - dst[63:0] := 0 -FI -dst[127:64] := a[127:64] -dst[MAX:128] := 0 +FOR j := 0 to 3 + i := 64*j + l := 8*j + IF k[j] + dst[i+63:i] := SignExtend64(a[l+7:l]) + ELSE + dst[i+63:i] := src[i+63:i] + FI +ENDFOR +dst[MAX:256] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Elementary Math Functions - - - Compute the approximate reciprocal square root of the lower double-precision (64-bit) floating-point element in "b", store the result in the lower element of "dst", and copy the upper element from "a" to the upper element of "dst". The maximum relative error for this approximation is less than 2^-14. + Convert + + + + Sign extend packed 8-bit integers in the low 4 bytes of "a" to packed 64-bit integers, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -dst[63:0] := APPROXIMATE(1.0 / SQRT(b[63:0])) -dst[127:64] := a[127:64] -dst[MAX:128] := 0 +FOR j := 0 to 3 + i := 64*j + l := 8*j + IF k[j] + dst[i+63:i] := SignExtend64(a[l+7:l]) + ELSE + dst[i+63:i] := 0 + FI +ENDFOR +dst[MAX:256] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Elementary Math Functions - - - - - Compute the approximate reciprocal square root of the lower single-precision (32-bit) floating-point element in "b", store the result in the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". The maximum relative error for this approximation is less than 2^-14. + Convert + + + + + Sign extend packed 8-bit integers in the low 2 bytes of "a" to packed 64-bit integers, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -IF k[0] - dst[31:0] := APPROXIMATE(1.0 / SQRT(b[31:0])) -ELSE - dst[31:0] := src[31:0] -FI -dst[127:32] := a[127:32] +FOR j := 0 to 1 + i := 64*j + l := 8*j + IF k[j] + dst[i+63:i] := SignExtend64(a[l+7:l]) + ELSE + dst[i+63:i] := src[i+63:i] + FI +ENDFOR dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Elementary Math Functions - - - - Compute the approximate reciprocal square root of the lower single-precision (32-bit) floating-point element in "b", store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". The maximum relative error for this approximation is less than 2^-14. + Convert + + + + Sign extend packed 8-bit integers in the low 2 bytes of "a" to packed 64-bit integers, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -IF k[0] - dst[31:0] := APPROXIMATE(1.0 / SQRT(b[31:0])) -ELSE - dst[31:0] := 0 -FI -dst[127:32] := a[127:32] +FOR j := 0 to 1 + i := 64*j + l := 8*j + IF k[j] + dst[i+63:i] := SignExtend64(a[l+7:l]) + ELSE + dst[i+63:i] := 0 + FI +ENDFOR dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Elementary Math Functions - - - Compute the approximate reciprocal square root of the lower single-precision (32-bit) floating-point element in "b", store the result in the lower element of "dst", and copy the upper 3 packed elements from "a" to the upper elements of "dst". The maximum relative error for this approximation is less than 2^-14. + Convert + + + + + Sign extend packed 32-bit integers in "a" to packed 64-bit integers, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -dst[31:0] := APPROXIMATE(1.0 / SQRT(b[31:0])) -dst[127:32] := a[127:32] -dst[MAX:128] := 0 +FOR j := 0 to 3 + i := 64*j + l := 32*j + IF k[j] + dst[i+63:i] := SignExtend64(a[l+31:l]) + ELSE + dst[i+63:i] := src[i+63:i] + FI +ENDFOR +dst[MAX:256] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Miscellaneous - - - - - Scale the packed double-precision (64-bit) floating-point elements in "a" using values from "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Convert + + + + Sign extend packed 32-bit integers in "a" to packed 64-bit integers, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -SCALE(src1, src2){ - IF (src2 == NaN) - IF (src2 == SNaN) - RETURN QNAN(src2) - FI - ELSE IF (src1 == NaN) - IF (src1 == SNaN) - RETURN QNAN(src1) - FI - IF (src2 != INF) - RETURN QNAN(src1) - FI - ELSE - tmp_src2 := src2 - tmp_src1 := src1 - IF (src2 is denormal AND MXCSR.DAZ) - tmp_src2 := 0 - FI - IF (src1 is denormal AND MXCSR.DAZ) - tmp_src1 := 0 - FI - FI - dst[63:0] := tmp_src1[63:0] * POW(2, FLOOR(tmp_src2[63:0])) - RETURN dst[63:0] -} - -FOR j := 0 to 7 - i := j*64 +FOR j := 0 to 3 + i := 64*j + l := 32*j IF k[j] - dst[i+63:i] := SCALE(a[i+63:0], b[i+63:i]) + dst[i+63:i] := SignExtend64(a[l+31:l]) ELSE - dst[i+63:i] := src[i+63:i] + dst[i+63:i] := 0 FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Miscellaneous - - - - - - Scale the packed double-precision (64-bit) floating-point elements in "a" using values from "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - [round_note] + Convert + + + + + Sign extend packed 32-bit integers in "a" to packed 64-bit integers, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -SCALE(src1, src2){ - IF (src2 == NaN) - IF (src2 == SNaN) - RETURN QNAN(src2) - FI - ELSE IF (src1 == NaN) - IF (src1 == SNaN) - RETURN QNAN(src1) - FI - IF (src2 != INF) - RETURN QNAN(src1) - FI - ELSE - tmp_src2 := src2 - tmp_src1 := src1 - IF (src2 is denormal AND MXCSR.DAZ) - tmp_src2 := 0 - FI - IF (src1 is denormal AND MXCSR.DAZ) - tmp_src1 := 0 - FI - FI - dst[63:0] := tmp_src1[63:0] * POW(2, FLOOR(tmp_src2[63:0])) - RETURN dst[63:0] -} - -FOR j := 0 to 7 - i := j*64 +FOR j := 0 to 1 + i := 64*j + l := 32*j IF k[j] - dst[i+63:i] := SCALE(a[i+63:0], b[i+63:i]) + dst[i+63:i] := SignExtend64(a[l+31:l]) ELSE dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Miscellaneous - - - - Scale the packed double-precision (64-bit) floating-point elements in "a" using values from "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Convert + + + + Sign extend packed 32-bit integers in "a" to packed 64-bit integers, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -SCALE(src1, src2){ - IF (src2 == NaN) - IF (src2 == SNaN) - RETURN QNAN(src2) - FI - ELSE IF (src1 == NaN) - IF (src1 == SNaN) - RETURN QNAN(src1) - FI - IF (src2 != INF) - RETURN QNAN(src1) - FI - ELSE - tmp_src2 := src2 - tmp_src1 := src1 - IF (src2 is denormal AND MXCSR.DAZ) - tmp_src2 := 0 - FI - IF (src1 is denormal AND MXCSR.DAZ) - tmp_src1 := 0 - FI - FI - dst[63:0] := tmp_src1[63:0] * POW(2, FLOOR(tmp_src2[63:0])) - RETURN dst[63:0] -} - -FOR j := 0 to 7 - i := j*64 +FOR j := 0 to 1 + i := 64*j + l := 32*j IF k[j] - dst[i+63:i] := SCALE(a[i+63:0], b[i+63:i]) + dst[i+63:i] := SignExtend64(a[l+31:l]) ELSE dst[i+63:i] := 0 FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Miscellaneous - - - - - Scale the packed double-precision (64-bit) floating-point elements in "a" using values from "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - [round_note] + Convert + + + + + Sign extend packed 16-bit integers in "a" to packed 32-bit integers, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -SCALE(src1, src2){ - IF (src2 == NaN) - IF (src2 == SNaN) - RETURN QNAN(src2) - FI - ELSE IF (src1 == NaN) - IF (src1 == SNaN) - RETURN QNAN(src1) - FI - IF (src2 != INF) - RETURN QNAN(src1) - FI - ELSE - tmp_src2 := src2 - tmp_src1 := src1 - IF (src2 is denormal AND MXCSR.DAZ) - tmp_src2 := 0 - FI - IF (src1 is denormal AND MXCSR.DAZ) - tmp_src1 := 0 - FI - FI - dst[63:0] := tmp_src1[63:0] * POW(2, FLOOR(tmp_src2[63:0])) - RETURN dst[63:0] -} - FOR j := 0 to 7 - i := j*64 + i := j*32 + l := j*16 IF k[j] - dst[i+63:i] := SCALE(a[i+63:0], b[i+63:i]) + dst[i+31:i] := SignExtend32(a[l+15:l]) ELSE - dst[i+63:i] := 0 + dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Miscellaneous - - - Scale the packed double-precision (64-bit) floating-point elements in "a" using values from "b", and store the results in "dst". + Convert + + + + Sign extend packed 16-bit integers in "a" to packed 32-bit integers, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -SCALE(src1, src2){ - IF (src2 == NaN) - IF (src2 == SNaN) - RETURN QNAN(src2) - FI - ELSE IF (src1 == NaN) - IF (src1 == SNaN) - RETURN QNAN(src1) - FI - IF (src2 != INF) - RETURN QNAN(src1) - FI +FOR j := 0 to 7 + i := 32*j + l := 16*j + IF k[j] + dst[i+31:i] := SignExtend32(a[l+15:l]) ELSE - tmp_src2 := src2 - tmp_src1 := src1 - IF (src2 is denormal AND MXCSR.DAZ) - tmp_src2 := 0 - FI - IF (src1 is denormal AND MXCSR.DAZ) - tmp_src1 := 0 - FI + dst[i+31:i] := 0 FI - dst[63:0] := tmp_src1[63:0] * POW(2, FLOOR(tmp_src2[63:0])) - RETURN dst[63:0] -} - -FOR j := 0 to 7 - i := j*64 - dst[i+63:i] := SCALE(a[i+63:0], b[i+63:i]) ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Miscellaneous - - - - Scale the packed double-precision (64-bit) floating-point elements in "a" using values from "b", and store the results in "dst". - [round_note] + Convert + + + + + Sign extend packed 16-bit integers in "a" to packed 32-bit integers, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -SCALE(src1, src2){ - IF (src2 == NaN) - IF (src2 == SNaN) - RETURN QNAN(src2) - FI - ELSE IF (src1 == NaN) - IF (src1 == SNaN) - RETURN QNAN(src1) - FI - IF (src2 != INF) - RETURN QNAN(src1) - FI +FOR j := 0 to 3 + i := j*32 + l := j*16 + IF k[j] + dst[i+31:i] := SignExtend32(a[l+15:l]) ELSE - tmp_src2 := src2 - tmp_src1 := src1 - IF (src2 is denormal AND MXCSR.DAZ) - tmp_src2 := 0 - FI - IF (src1 is denormal AND MXCSR.DAZ) - tmp_src1 := 0 - FI + dst[i+31:i] := src[i+31:i] FI - dst[63:0] := tmp_src1[63:0] * POW(2, FLOOR(tmp_src2[63:0])) - RETURN dst[63:0] -} - -FOR j := 0 to 7 - i := j*64 - dst[i+63:i] := SCALE(a[i+63:0], b[i+63:i]) ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Miscellaneous - - - - - Scale the packed single-precision (32-bit) floating-point elements in "a" using values from "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Convert + + + + Sign extend packed 16-bit integers in "a" to packed 32-bit integers, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -SCALE(src1, src2){ - IF (src2 == NaN) - IF (src2 == SNaN) - RETURN QNAN(src2) - FI - ELSE IF (src1 == NaN) - IF (src1 == SNaN) - RETURN QNAN(src1) - FI - IF (src2 != INF) - RETURN QNAN(src1) - FI - ELSE - tmp_src2 := src2 - tmp_src1 := src1 - IF (src2 is denormal AND MXCSR.DAZ) - tmp_src2 := 0 - FI - IF (src1 is denormal AND MXCSR.DAZ) - tmp_src1 := 0 - FI - FI - dst[31:0] := tmp_src1[31:0] * POW(2, FLOOR(tmp_src2[31:0])) - RETURN dst[31:0] -} - -FOR j := 0 to 15 - i := j*32 +FOR j := 0 to 3 + i := 32*j + l := 16*j IF k[j] - dst[i+31:i] := SCALE(a[i+31:0], b[i+31:i]) + dst[i+31:i] := SignExtend32(a[l+15:l]) ELSE - dst[i+31:i] := src[i+31:i] + dst[i+31:i] := 0 FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Miscellaneous - - - - - - Scale the packed single-precision (32-bit) floating-point elements in "a" using values from "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - [round_note] + Convert + + + + + Sign extend packed 16-bit integers in the low 8 bytes of "a" to packed 64-bit integers, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -SCALE(src1, src2){ - IF (src2 == NaN) - IF (src2 == SNaN) - RETURN QNAN(src2) - FI - ELSE IF (src1 == NaN) - IF (src1 == SNaN) - RETURN QNAN(src1) - FI - IF (src2 != INF) - RETURN QNAN(src1) - FI - ELSE - tmp_src2 := src2 - tmp_src1 := src1 - IF (src2 is denormal AND MXCSR.DAZ) - tmp_src2 := 0 - FI - IF (src1 is denormal AND MXCSR.DAZ) - tmp_src1 := 0 - FI - FI - dst[31:0] := tmp_src1[31:0] * POW(2, FLOOR(tmp_src2[31:0])) - RETURN dst[31:0] -} - -FOR j := 0 to 15 - i := j*32 +FOR j := 0 to 3 + i := 64*j + l := 16*j IF k[j] - dst[i+31:i] := SCALE(a[i+31:0], b[i+31:i]) + dst[i+63:i] := SignExtend64(a[l+15:l]) ELSE - dst[i+31:i] := src[i+31:i] + dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Miscellaneous - - - - Scale the packed single-precision (32-bit) floating-point elements in "a" using values from "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Convert + + + + Sign extend packed 16-bit integers in the low 8 bytes of "a" to packed 64-bit integers, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -SCALE(src1, src2){ - IF (src2 == NaN) - IF (src2 == SNaN) - RETURN QNAN(src2) - FI - ELSE IF (src1 == NaN) - IF (src1 == SNaN) - RETURN QNAN(src1) - FI - IF (src2 != INF) - RETURN QNAN(src1) - FI - ELSE - tmp_src2 := src2 - tmp_src1 := src1 - IF (src2 is denormal AND MXCSR.DAZ) - tmp_src2 := 0 - FI - IF (src1 is denormal AND MXCSR.DAZ) - tmp_src1 := 0 - FI - FI - dst[31:0] := tmp_src1[31:0] * POW(2, FLOOR(tmp_src2[31:0])) - RETURN dst[31:0] -} - -FOR j := 0 to 15 - i := j*32 +FOR j := 0 to 3 + i := 64*j + l := 16*j IF k[j] - dst[i+31:i] := SCALE(a[i+31:0], b[i+31:i]) + dst[i+63:i] := SignExtend64(a[l+15:l]) ELSE - dst[i+31:i] := 0 + dst[i+63:i] := 0 FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Miscellaneous - - - - - Scale the packed single-precision (32-bit) floating-point elements in "a" using values from "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - [round_note] + Convert + + + + + Sign extend packed 16-bit integers in the low 4 bytes of "a" to packed 64-bit integers, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -SCALE(src1, src2){ - IF (src2 == NaN) - IF (src2 == SNaN) - RETURN QNAN(src2) - FI - ELSE IF (src1 == NaN) - IF (src1 == SNaN) - RETURN QNAN(src1) - FI - IF (src2 != INF) - RETURN QNAN(src1) - FI - ELSE - tmp_src2 := src2 - tmp_src1 := src1 - IF (src2 is denormal AND MXCSR.DAZ) - tmp_src2 := 0 - FI - IF (src1 is denormal AND MXCSR.DAZ) - tmp_src1 := 0 - FI - FI - dst[31:0] := tmp_src1[31:0] * POW(2, FLOOR(tmp_src2[31:0])) - RETURN dst[31:0] -} - -FOR j := 0 to 15 - i := j*32 +FOR j := 0 to 1 + i := 64*j + l := 16*j IF k[j] - dst[i+31:i] := SCALE(a[i+31:0], b[i+31:i]) + dst[i+63:i] := SignExtend64(a[l+15:l]) ELSE - dst[i+31:i] := 0 + dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Miscellaneous - - - Scale the packed single-precision (32-bit) floating-point elements in "a" using values from "b", and store the results in "dst". + Convert + + + + Sign extend packed 16-bit integers in the low 4 bytes of "a" to packed 64-bit integers, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -SCALE(src1, src2){ - IF (src2 == NaN) - IF (src2 == SNaN) - RETURN QNAN(src2) - FI - ELSE IF (src1 == NaN) - IF (src1 == SNaN) - RETURN QNAN(src1) - FI - IF (src2 != INF) - RETURN QNAN(src1) - FI +FOR j := 0 to 1 + i := 64*j + l := 16*j + IF k[j] + dst[i+63:i] := SignExtend64(a[l+15:l]) ELSE - tmp_src2 := src2 - tmp_src1 := src1 - IF (src2 is denormal AND MXCSR.DAZ) - tmp_src2 := 0 - FI - IF (src1 is denormal AND MXCSR.DAZ) - tmp_src1 := 0 - FI + dst[i+63:i] := 0 FI - dst[31:0] := tmp_src1[31:0] * POW(2, FLOOR(tmp_src2[31:0])) - RETURN dst[31:0] -} - -FOR j := 0 to 15 - i := j*32 - dst[i+31:i] := SCALE(a[i+31:0], b[i+31:i]) ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Miscellaneous - - - - Scale the packed single-precision (32-bit) floating-point elements in "a" using values from "b", and store the results in "dst". - [round_note] + Convert + + + Convert packed unsigned 32-bit integers in "a" to packed unsigned 8-bit integers with unsigned saturation, and store the results in "dst". -SCALE(src1, src2){ - IF (src2 == NaN) - IF (src2 == SNaN) - RETURN QNAN(src2) - FI - ELSE IF (src1 == NaN) - IF (src1 == SNaN) - RETURN QNAN(src1) - FI - IF (src2 != INF) - RETURN QNAN(src1) - FI - ELSE - tmp_src2 := src2 - tmp_src1 := src1 - IF (src2 is denormal AND MXCSR.DAZ) - tmp_src2 := 0 - FI - IF (src1 is denormal AND MXCSR.DAZ) - tmp_src1 := 0 - FI - FI - dst[31:0] := tmp_src1[31:0] * POW(2, FLOOR(tmp_src2[31:0])) - RETURN dst[31:0] -} - -FOR j := 0 to 15 - i := j*32 - dst[i+31:i] := SCALE(a[i+31:0], b[i+31:i]) +FOR j := 0 to 7 + i := 32*j + k := 8*j + dst[k+7:k] := SaturateU8(a[i+31:i]) ENDFOR -dst[MAX:512] := 0 +dst[MAX:64] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Miscellaneous - - - - - - Scale the packed double-precision (64-bit) floating-point elements in "a" using values from "b", store the result in the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and copy the upper element from "b" to the upper element of "dst". - [round_note] + Convert + + + + + Convert packed unsigned 32-bit integers in "a" to packed unsigned 8-bit integers with unsigned saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -SCALE(src1, src2){ - IF (src2 == NaN) - IF (src2 == SNaN) - RETURN QNAN(src2) - FI - ELSE IF (src1 == NaN) - IF (src1 == SNaN) - RETURN QNAN(src1) - FI - IF (src2 != INF) - RETURN QNAN(src1) - FI +FOR j := 0 to 7 + i := 32*j + l := 8*j + IF k[j] + dst[l+7:l] := SaturateU8(a[i+31:i]) ELSE - tmp_src2 := src2 - tmp_src1 := src1 - IF (src2 is denormal AND MXCSR.DAZ) - tmp_src2 := 0 - FI - IF (src1 is denormal AND MXCSR.DAZ) - tmp_src1 := 0 - FI + dst[l+7:l] := src[l+7:l] FI - dst[63:0] := tmp_src1[63:0] * POW(2, FLOOR(tmp_src2[63:0])) - RETURN dst[63:0] -} - -IF k[0] - dst[63:0] := SCALE(a[63:0], b[63:0]) -ELSE - dst[63:0] := src[63:0] -FI -dst[127:64] := b[127:64] -dst[MAX:128] := 0 +ENDFOR +dst[MAX:64] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Miscellaneous - - - - - Scale the packed double-precision (64-bit) floating-point elements in "a" using values from "b", store the result in the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and copy the upper element from "b" to the upper element of "dst". + Convert + Store + + + + + Convert packed unsigned 32-bit integers in "a" to packed unsigned 8-bit integers with unsigned saturation, and store the active results (those with their respective bit set in writemask "k") to unaligned memory at "base_addr". -SCALE(src1, src2){ - IF (src2 == NaN) - IF (src2 == SNaN) - RETURN QNAN(src2) - FI - ELSE IF (src1 == NaN) - IF (src1 == SNaN) - RETURN QNAN(src1) - FI - IF (src2 != INF) - RETURN QNAN(src1) - FI - ELSE - tmp_src2 := src2 - tmp_src1 := src1 - IF (src2 is denormal AND MXCSR.DAZ) - tmp_src2 := 0 - FI - IF (src1 is denormal AND MXCSR.DAZ) - tmp_src1 := 0 - FI +FOR j := 0 to 7 + i := 32*j + l := 8*j + IF k[j] + MEM[base_addr+l+7:base_addr+l] := SaturateU8(a[i+31:i]) FI - dst[63:0] := tmp_src1[63:0] * POW(2, FLOOR(tmp_src2[63:0])) - RETURN dst[63:0] -} - -IF k[0] - dst[63:0] := SCALE(a[63:0], b[63:0]) -ELSE - dst[63:0] := src[63:0] -FI -dst[127:64] := b[127:64] -dst[MAX:128] := 0 +ENDFOR - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Miscellaneous - - - - - Scale the packed double-precision (64-bit) floating-point elements in "a" using values from "b", store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper element from "b" to the upper element of "dst". - [round_note] + Convert + + + + Convert packed unsigned 32-bit integers in "a" to packed unsigned 8-bit integers with unsigned saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -SCALE(src1, src2){ - IF (src2 == NaN) - IF (src2 == SNaN) - RETURN QNAN(src2) - FI - ELSE IF (src1 == NaN) - IF (src1 == SNaN) - RETURN QNAN(src1) - FI - IF (src2 != INF) - RETURN QNAN(src1) - FI +FOR j := 0 to 7 + i := 32*j + l := 8*j + IF k[j] + dst[l+7:l] := SaturateU8(a[i+31:i]) ELSE - tmp_src2 := src2 - tmp_src1 := src1 - IF (src2 is denormal AND MXCSR.DAZ) - tmp_src2 := 0 - FI - IF (src1 is denormal AND MXCSR.DAZ) - tmp_src1 := 0 - FI + dst[l+7:l] := 0 FI - dst[63:0] := tmp_src1[63:0] * POW(2, FLOOR(tmp_src2[63:0])) - RETURN dst[63:0] -} - -IF k[0] - dst[63:0] := SCALE(a[63:0], b[63:0]) -ELSE - dst[63:0] := 0 -FI -dst[127:64] := b[127:64] -dst[MAX:128] := 0 +ENDFOR +dst[MAX:64] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Miscellaneous - - - - Scale the packed double-precision (64-bit) floating-point elements in "a" using values from "b", store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper element from "b" to the upper element of "dst". + Convert + + + Convert packed unsigned 32-bit integers in "a" to packed unsigned 8-bit integers with unsigned saturation, and store the results in "dst". -SCALE(src1, src2){ - IF (src2 == NaN) - IF (src2 == SNaN) - RETURN QNAN(src2) - FI - ELSE IF (src1 == NaN) - IF (src1 == SNaN) - RETURN QNAN(src1) - FI - IF (src2 != INF) - RETURN QNAN(src1) - FI - ELSE - tmp_src2 := src2 - tmp_src1 := src1 - IF (src2 is denormal AND MXCSR.DAZ) - tmp_src2 := 0 - FI - IF (src1 is denormal AND MXCSR.DAZ) - tmp_src1 := 0 - FI - FI - dst[63:0] := tmp_src1[63:0] * POW(2, FLOOR(tmp_src2[63:0])) - RETURN dst[63:0] -} - -IF k[0] - dst[63:0] := SCALE(a[63:0], b[63:0]) -ELSE - dst[63:0] := 0 -FI -dst[127:64] := b[127:64] -dst[MAX:128] := 0 +FOR j := 0 to 3 + i := 32*j + k := 8*j + dst[k+7:k] := SaturateU8(a[i+31:i]) +ENDFOR +dst[MAX:32] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Miscellaneous - - - - Scale the packed double-precision (64-bit) floating-point elements in "a" using values from "b", store the result in the lower element of "dst", and copy the upper element from "b" to the upper element of "dst". - [round_note] + Convert + + + + + Convert packed unsigned 32-bit integers in "a" to packed unsigned 8-bit integers with unsigned saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -SCALE(src1, src2){ - IF (src2 == NaN) - IF (src2 == SNaN) - RETURN QNAN(src2) - FI - ELSE IF (src1 == NaN) - IF (src1 == SNaN) - RETURN QNAN(src1) - FI - IF (src2 != INF) - RETURN QNAN(src1) - FI +FOR j := 0 to 3 + i := 32*j + l := 8*j + IF k[j] + dst[l+7:l] := SaturateU8(a[i+31:i]) ELSE - tmp_src2 := src2 - tmp_src1 := src1 - IF (src2 is denormal AND MXCSR.DAZ) - tmp_src2 := 0 - FI - IF (src1 is denormal AND MXCSR.DAZ) - tmp_src1 := 0 - FI + dst[l+7:l] := src[l+7:l] FI - dst[63:0] := tmp_src1[63:0] * POW(2, FLOOR(tmp_src2[63:0])) - RETURN dst[63:0] -} - -dst[63:0] := SCALE(a[63:0], b[63:0]) -dst[127:64] := b[127:64] -dst[MAX:128] := 0 +ENDFOR +dst[MAX:32] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Miscellaneous - - - Scale the packed double-precision (64-bit) floating-point elements in "a" using values from "b", store the result in the lower element of "dst", and copy the upper element from "b" to the upper element of "dst". + Convert + Store + + + + + Convert packed unsigned 32-bit integers in "a" to packed unsigned 8-bit integers with unsigned saturation, and store the active results (those with their respective bit set in writemask "k") to unaligned memory at "base_addr". -SCALE(src1, src2){ - IF (src2 == NaN) - IF (src2 == SNaN) - RETURN QNAN(src2) - FI - ELSE IF (src1 == NaN) - IF (src1 == SNaN) - RETURN QNAN(src1) - FI - IF (src2 != INF) - RETURN QNAN(src1) - FI - ELSE - tmp_src2 := src2 - tmp_src1 := src1 - IF (src2 is denormal AND MXCSR.DAZ) - tmp_src2 := 0 - FI - IF (src1 is denormal AND MXCSR.DAZ) - tmp_src1 := 0 - FI +FOR j := 0 to 3 + i := 32*j + l := 8*j + IF k[j] + MEM[base_addr+l+7:base_addr+l] := SaturateU8(a[i+31:i]) FI - dst[63:0] := tmp_src1[63:0] * POW(2, FLOOR(tmp_src2[63:0])) - RETURN dst[63:0] -} - -dst[63:0] := SCALE(a[63:0], b[63:0]) -dst[127:64] := b[127:64] -dst[MAX:128] := 0 +ENDFOR - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Miscellaneous - - - - - - Scale the packed single-precision (32-bit) floating-point elements in "a" using values from "b", store the result in the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and copy the upper 3 packed elements from "b" to the upper elements of "dst". - [round_note] + Convert + + + + Convert packed unsigned 32-bit integers in "a" to packed unsigned 8-bit integers with unsigned saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -SCALE(src1, src2){ - IF (src2 == NaN) - IF (src2 == SNaN) - RETURN QNAN(src2) - FI - ELSE IF (src1 == NaN) - IF (src1 == SNaN) - RETURN QNAN(src1) - FI - IF (src2 != INF) - RETURN QNAN(src1) - FI +FOR j := 0 to 3 + i := 32*j + l := 8*j + IF k[j] + dst[l+7:l] := SaturateU8(a[i+31:i]) ELSE - tmp_src2 := src2 - tmp_src1 := src1 - IF (src2 is denormal AND MXCSR.DAZ) - tmp_src2 := 0 - FI - IF (src1 is denormal AND MXCSR.DAZ) - tmp_src1 := 0 - FI + dst[l+7:l] := 0 FI - dst[31:0] := tmp_src1[31:0] * POW(2, FLOOR(tmp_src2[31:0])) - RETURN dst[63:0] -} - -IF k[0] - dst[31:0] := SCALE(a[31:0], b[31:0]) -ELSE - dst[31:0] := src[31:0] -FI -dst[127:32] := b[127:32] -dst[MAX:128] := 0 +ENDFOR +dst[MAX:32] := 0 - +
immintrin.h
 - - Floating Point - AVX512F - Miscellaneous - - - - - Scale the packed single-precision (32-bit) floating-point elements in "a" using values from "b", store the result in the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and copy the upper 3 packed elements from "b" to the upper elements of "dst". + + Integer + AVX512VL + AVX512F + Convert + + + Convert packed unsigned 32-bit integers in "a" to packed unsigned 16-bit integers with unsigned saturation, and store the results in "dst". -SCALE(src1, src2){ - IF (src2 == NaN) - IF (src2 == SNaN) - RETURN QNAN(src2) - FI - ELSE IF (src1 == NaN) - IF (src1 == SNaN) - RETURN QNAN(src1) - FI - IF (src2 != INF) - RETURN QNAN(src1) - FI - ELSE - tmp_src2 := src2 - tmp_src1 := src1 - IF (src2 is denormal AND MXCSR.DAZ) - tmp_src2 := 0 - FI - IF (src1 is denormal AND MXCSR.DAZ) - tmp_src1 := 0 - FI - FI - dst[31:0] := tmp_src1[31:0] * POW(2, FLOOR(tmp_src2[31:0])) - RETURN dst[63:0] -} - -IF k[0] - dst[31:0] := SCALE(a[31:0], b[31:0]) -ELSE - dst[31:0] := src[31:0] -FI -dst[127:32] := b[127:32] +FOR j := 0 to 7 + i := 32*j + k := 16*j + dst[k+15:k] := SaturateU16(a[i+31:i]) +ENDFOR dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Miscellaneous - - - - - Scale the packed single-precision (32-bit) floating-point elements in "a" using values from "b", store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper 3 packed elements from "b" to the upper elements of "dst". - [round_note] + Convert + + + + + Convert packed unsigned 32-bit integers in "a" to packed unsigned 16-bit integers with unsigned saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -SCALE(src1, src2){ - IF (src2 == NaN) - IF (src2 == SNaN) - RETURN QNAN(src2) - FI - ELSE IF (src1 == NaN) - IF (src1 == SNaN) - RETURN QNAN(src1) - FI - IF (src2 != INF) - RETURN QNAN(src1) - FI +FOR j := 0 to 7 + i := 32*j + l := 16*j + IF k[j] + dst[l+15:l] := SaturateU16(a[i+31:i]) ELSE - tmp_src2 := src2 - tmp_src1 := src1 - IF (src2 is denormal AND MXCSR.DAZ) - tmp_src2 := 0 - FI - IF (src1 is denormal AND MXCSR.DAZ) - tmp_src1 := 0 - FI + dst[l+15:l] := src[l+15:l] FI - dst[31:0] := tmp_src1[31:0] * POW(2, FLOOR(tmp_src2[31:0])) - RETURN dst[63:0] -} - -IF k[0] - dst[31:0] := SCALE(a[31:0], b[31:0]) -ELSE - dst[31:0] := 0 -FI -dst[127:32] := b[127:32] +ENDFOR dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Miscellaneous - - - - Scale the packed single-precision (32-bit) floating-point elements in "a" using values from "b", store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper 3 packed elements from "b" to the upper elements of "dst". + Convert + Store + + + + + Convert packed unsigned 32-bit integers in "a" to packed unsigned 16-bit integers with unsigned saturation, and store the active results (those with their respective bit set in writemask "k") to unaligned memory at "base_addr". -SCALE(src1, src2){ - IF (src2 == NaN) - IF (src2 == SNaN) - RETURN QNAN(src2) - FI - ELSE IF (src1 == NaN) - IF (src1 == SNaN) - RETURN QNAN(src1) - FI - IF (src2 != INF) - RETURN QNAN(src1) - FI - ELSE - tmp_src2 := src2 - tmp_src1 := src1 - IF (src2 is denormal AND MXCSR.DAZ) - tmp_src2 := 0 - FI - IF (src1 is denormal AND MXCSR.DAZ) - tmp_src1 := 0 - FI +FOR j := 0 to 7 + i := 32*j + l := 16*j + IF k[j] + MEM[base_addr+l+15:base_addr+l] := SaturateU16(a[i+31:i]) FI - dst[31:0] := tmp_src1[31:0] * POW(2, FLOOR(tmp_src2[31:0])) - RETURN dst[63:0] -} - -IF k[0] - dst[31:0] := SCALE(a[31:0], b[31:0]) -ELSE - dst[31:0] := 0 -FI -dst[127:32] := b[127:32] -dst[MAX:128] := 0 +ENDFOR - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Miscellaneous - - - - Scale the packed single-precision (32-bit) floating-point elements in "a" using values from "b", store the result in the lower element of "dst", and copy the upper 3 packed elements from "b" to the upper elements of "dst". - [round_note] + Convert + + + + Convert packed unsigned 32-bit integers in "a" to packed unsigned 16-bit integers with unsigned saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -SCALE(src1, src2){ - IF (src2 == NaN) - IF (src2 == SNaN) - RETURN QNAN(src2) - FI - ELSE IF (src1 == NaN) - IF (src1 == SNaN) - RETURN QNAN(src1) - FI - IF (src2 != INF) - RETURN QNAN(src1) - FI +FOR j := 0 to 7 + i := 32*j + l := 16*j + IF k[j] + dst[l+15:l] := SaturateU16(a[i+31:i]) ELSE - tmp_src2 := src2 - tmp_src1 := src1 - IF (src2 is denormal AND MXCSR.DAZ) - tmp_src2 := 0 - FI - IF (src1 is denormal AND MXCSR.DAZ) - tmp_src1 := 0 - FI + dst[l+15:l] := 0 FI - dst[31:0] := tmp_src1[31:0] * POW(2, FLOOR(tmp_src2[31:0])) - RETURN dst[63:0] -} - -dst[31:0] := SCALE(a[31:0], b[31:0]) -dst[127:32] := b[127:32] +ENDFOR dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Miscellaneous - - - Scale the packed single-precision (32-bit) floating-point elements in "a" using values from "b", store the result in the lower element of "dst", and copy the upper 3 packed elements from "b" to the upper elements of "dst". + Convert + + + Convert packed unsigned 32-bit integers in "a" to packed unsigned 16-bit integers with unsigned saturation, and store the results in "dst". -SCALE(src1, src2){ - IF (src2 == NaN) - IF (src2 == SNaN) - RETURN QNAN(src2) - FI - ELSE IF (src1 == NaN) - IF (src1 == SNaN) - RETURN QNAN(src1) - FI - IF (src2 != INF) - RETURN QNAN(src1) - FI +FOR j := 0 to 3 + i := 32*j + k := 16*j + dst[k+15:k] := SaturateU16(a[i+31:i]) +ENDFOR +dst[MAX:64] := 0 + + +
immintrin.h
+ + + Integer + AVX512VL + AVX512F + Convert + + + + + Convert packed unsigned 32-bit integers in "a" to packed unsigned 16-bit integers with unsigned saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +FOR j := 0 to 3 + i := 32*j + l := 16*j + IF k[j] + dst[l+15:l] := SaturateU16(a[i+31:i]) ELSE - tmp_src2 := src2 - tmp_src1 := src1 - IF (src2 is denormal AND MXCSR.DAZ) - tmp_src2 := 0 - FI - IF (src1 is denormal AND MXCSR.DAZ) - tmp_src1 := 0 - FI + dst[l+15:l] := src[l+15:l] FI - dst[31:0] := tmp_src1[31:0] * POW(2, FLOOR(tmp_src2[31:0])) - RETURN dst[63:0] -} - -dst[31:0] := SCALE(a[31:0], b[31:0]) -dst[127:32] := b[127:32] -dst[MAX:128] := 0 +ENDFOR +dst[MAX:64] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F + Convert Store - - - - - Scatter double-precision (64-bit) floating-point elements from "a" into memory using 32-bit indices. 64-bit elements are stored at addresses starting at "base_addr" and offset by each 32-bit element in "vindex" (each index is scaled by the factor in "scale"). "scale" should be 1, 2, 4 or 8. - + + + + + Convert packed unsigned 32-bit integers in "a" to packed unsigned 16-bit integers with unsigned saturation, and store the active results (those with their respective bit set in writemask "k") to unaligned memory at "base_addr". -FOR j := 0 to 7 - i := j*64 - l := j*32 - MEM[base_addr + SignExtend(vindex[l+31:l])*scale] := a[i+63:i] +FOR j := 0 to 3 + i := 32*j + l := 16*j + IF k[j] + MEM[base_addr+l+15:base_addr+l] := SaturateU16(a[i+31:i]) + FI ENDFOR - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Store - - - - - - Scatter double-precision (64-bit) floating-point elements from "a" into memory using 32-bit indices. 64-bit elements are stored at addresses starting at "base_addr" and offset by each 32-bit element in "vindex" (each index is scaled by the factor in "scale") subject to mask "k" (elements are not stored when the corresponding mask bit is not set). "scale" should be 1, 2, 4 or 8. - + Convert + + + + Convert packed unsigned 32-bit integers in "a" to packed unsigned 16-bit integers with unsigned saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := j*64 - l := j*32 +FOR j := 0 to 3 + i := 32*j + l := 16*j IF k[j] - MEM[base_addr + SignExtend(vindex[l+31:l])*scale] := a[i+63:i] - k[j] := 0 + dst[l+15:l] := SaturateU16(a[i+31:i]) + ELSE + dst[l+15:l] := 0 FI ENDFOR -k[MAX:8] := 0 +dst[MAX:64] := 0 - +
immintrin.h
 - - Floating Point - AVX512F/KNCNI - Store - - - - - Scatter single-precision (32-bit) floating-point elements from "a" into memory using 32-bit indices. 32-bit elements are stored at addresses starting at "base_addr" and offset by each 32-bit element in "vindex" (each index is scaled by the factor in "scale"). "scale" should be 1, 2, 4 or 8. + + Integer + AVX512VL + AVX512F + Convert + + + Convert packed unsigned 64-bit integers in "a" to packed unsigned 8-bit integers with unsigned saturation, and store the results in "dst". -FOR j := 0 to 15 - i := j*32 - MEM[base_addr + SignExtend(vindex[i+31:i])*scale] := a[i+31:i] +FOR j := 0 to 3 + i := 64*j + k := 8*j + dst[k+7:k] := SaturateU8(a[i+63:i]) ENDFOR +dst[MAX:32] := 0 - +
immintrin.h
 - - Floating Point - AVX512F/KNCNI - Store - - - - - - Scatter single-precision (32-bit) floating-point elements from "a" into memory using 32-bit indices. 32-bit elements are stored at addresses starting at "base_addr" and offset by each 32-bit element in "vindex" (each index is scaled by the factor in "scale") subject to mask "k" (elements are not stored when the corresponding mask bit is not set). "scale" should be 1, 2, 4 or 8. + + Integer + AVX512VL + AVX512F + Convert + + + + + Convert packed unsigned 64-bit integers in "a" to packed unsigned 8-bit integers with unsigned saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*32 +FOR j := 0 to 3 + i := 64*j + l := 8*j IF k[j] - MEM[base_addr + SignExtend(vindex[i+31:i])*scale] := a[i+31:i] - k[j] := 0 + dst[l+7:l] := SaturateU8(a[i+63:i]) + ELSE + dst[l+7:l] := src[l+7:l] FI ENDFOR -k[MAX:16] := 0 +dst[MAX:32] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F + Convert Store - - - - - Scatter double-precision (64-bit) floating-point elements from "a" into memory using 64-bit indices. 64-bit elements are stored at addresses starting at "base_addr" and offset by each 64-bit element in "vindex" (each index is scaled by the factor in "scale"). "scale" should be 1, 2, 4 or 8. - + + + + + Convert packed unsigned 64-bit integers in "a" to packed unsigned 8-bit integers with unsigned saturation, and store the active results (those with their respective bit set in writemask "k") to unaligned memory at "base_addr". -FOR j := 0 to 7 - i := j*64 - MEM[base_addr + SignExtend(vindex[i+63:i])*scale] := a[i+63:i] +FOR j := 0 to 3 + i := 64*j + l := 8*j + IF k[j] + MEM[base_addr+l+7:base_addr+l] := SaturateU8(a[i+63:i]) + FI ENDFOR - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Store - - - - - - Scatter double-precision (64-bit) floating-point elements from "a" into memory using 64-bit indices. 64-bit elements are stored at addresses starting at "base_addr" and offset by each 64-bit element in "vindex" (each index is scaled by the factor in "scale") subject to mask "k" (elements are not stored when the corresponding mask bit is not set). "scale" should be 1, 2, 4 or 8. - + Convert + + + + Convert packed unsigned 64-bit integers in "a" to packed unsigned 8-bit integers with unsigned saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := j*64 +FOR j := 0 to 3 + i := 64*j + l := 8*j IF k[j] - MEM[base_addr + SignExtend(vindex[i+63:i])*scale] := a[i+63:i] - k[j] := 0 + dst[l+7:l] := SaturateU8(a[i+63:i]) + ELSE + dst[l+7:l] := 0 FI ENDFOR -k[MAX:8] := 0 +dst[MAX:32] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Store - - - - - Scatter single-precision (32-bit) floating-point elements from "a" into memory using 64-bit indices. 32-bit elements are stored at addresses starting at "base_addr" and offset by each 64-bit element in "vindex" (each index is scaled by the factor in "scale") subject to mask "k" (elements are not stored when the corresponding mask bit is not set). "scale" should be 1, 2, 4 or 8. - + Convert + + + Convert packed unsigned 64-bit integers in "a" to packed unsigned 8-bit integers with unsigned saturation, and store the results in "dst". -FOR j := 0 to 7 - i := j*32 - l := j*64 - MEM[base_addr + SignExtend(vindex[l+63:l])*scale] := a[i+31:i] +FOR j := 0 to 1 + i := 64*j + k := 8*j + dst[k+7:k] := SaturateU8(a[i+63:i]) ENDFOR +dst[MAX:16] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Store - - - - - - Scatter single-precision (32-bit) floating-point elements from "a" into memory using 64-bit indices. 32-bit elements are stored at addresses starting at "base_addr" and offset by each 64-bit element in "vindex" (each index is scaled by the factor in "scale") subject to mask "k" (elements are not stored when the corresponding mask bit is not set). "scale" should be 1, 2, 4 or 8. - + Convert + + + + + Convert packed unsigned 64-bit integers in "a" to packed unsigned 8-bit integers with unsigned saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := j*32 - l := j*64 +FOR j := 0 to 1 + i := 64*j + l := 8*j IF k[j] - MEM[base_addr + SignExtend(vindex[l+63:l])*scale] := a[i+31:i] - k[j] := 0 + dst[l+7:l] := SaturateU8(a[i+63:i]) + ELSE + dst[l+7:l] := src[l+7:l] FI ENDFOR -k[MAX:8] := 0 +dst[MAX:16] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Swizzle - - - - - - Shuffle 128-bits (composed of 4 single-precision (32-bit) floating-point elements) selected by "imm8" from "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -SELECT4(src, control){ - CASE(control[1:0]) - 0: tmp[127:0] := src[127:0] - 1: tmp[127:0] := src[255:128] - 2: tmp[127:0] := src[383:256] - 3: tmp[127:0] := src[511:384] - ESAC - RETURN tmp[127:0] -} - -tmp_dst[127:0] := SELECT4(a[511:0], imm8[1:0]) -tmp_dst[255:128] := SELECT4(a[511:0], imm8[3:2]) -tmp_dst[383:256] := SELECT4(b[511:0], imm8[5:4]) -tmp_dst[511:384] := SELECT4(b[511:0], imm8[7:6]) - -FOR j := 0 to 15 - i := j*32 + Convert + Store + + + + + Convert packed unsigned 64-bit integers in "a" to packed unsigned 8-bit integers with unsigned saturation, and store the active results (those with their respective bit set in writemask "k") to unaligned memory at "base_addr". + +FOR j := 0 to 1 + i := 64*j + l := 8*j IF k[j] - dst[i+31:i] := tmp_dst[i+31:i] - ELSE - dst[i+31:i] := src[i+31:i] + MEM[base_addr+l+7:base_addr+l] := SaturateU8(a[i+63:i]) FI ENDFOR -dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Swizzle - - - - - Shuffle 128-bits (composed of 4 single-precision (32-bit) floating-point elements) selected by "imm8" from "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Convert + + + + Convert packed unsigned 64-bit integers in "a" to packed unsigned 8-bit integers with unsigned saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -SELECT4(src, control){ - CASE(control[1:0]) - 0: tmp[127:0] := src[127:0] - 1: tmp[127:0] := src[255:128] - 2: tmp[127:0] := src[383:256] - 3: tmp[127:0] := src[511:384] - ESAC - RETURN tmp[127:0] -} - -tmp_dst[127:0] := SELECT4(a[511:0], imm8[1:0]) -tmp_dst[255:128] := SELECT4(a[511:0], imm8[3:2]) -tmp_dst[383:256] := SELECT4(b[511:0], imm8[5:4]) -tmp_dst[511:384] := SELECT4(b[511:0], imm8[7:6]) - -FOR j := 0 to 15 - i := j*32 +FOR j := 0 to 1 + i := 64*j + l := 8*j IF k[j] - dst[i+31:i] := tmp_dst[i+31:i] + dst[l+7:l] := SaturateU8(a[i+63:i]) ELSE - dst[i+31:i] := 0 + dst[l+7:l] := 0 FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:16] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Swizzle - - - - Shuffle 128-bits (composed of 4 single-precision (32-bit) floating-point elements) selected by "imm8" from "a" and "b", and store the results in "dst". - - -SELECT4(src, control){ - CASE(control[1:0]) - 0: tmp[127:0] := src[127:0] - 1: tmp[127:0] := src[255:128] - 2: tmp[127:0] := src[383:256] - 3: tmp[127:0] := src[511:384] - ESAC - RETURN tmp[127:0] -} - -dst[127:0] := SELECT4(a[511:0], imm8[1:0]) -dst[255:128] := SELECT4(a[511:0], imm8[3:2]) -dst[383:256] := SELECT4(b[511:0], imm8[5:4]) -dst[511:384] := SELECT4(b[511:0], imm8[7:6]) -dst[MAX:512] := 0 + Convert + + + Convert packed unsigned 64-bit integers in "a" to packed unsigned 32-bit integers with unsigned saturation, and store the results in "dst". + +FOR j := 0 to 3 + i := 64*j + k := 32*j + dst[k+31:k] := SaturateU32(a[i+63:i]) +ENDFOR +dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Swizzle - - - - - - Shuffle 128-bits (composed of 2 double-precision (64-bit) floating-point elements) selected by "imm8" from "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -SELECT4(src, control){ - CASE(control[1:0]) - 0: tmp[127:0] := src[127:0] - 1: tmp[127:0] := src[255:128] - 2: tmp[127:0] := src[383:256] - 3: tmp[127:0] := src[511:384] - ESAC - RETURN tmp[127:0] -} - -tmp_dst[127:0] := SELECT4(a[511:0], imm8[1:0]) -tmp_dst[255:128] := SELECT4(a[511:0], imm8[3:2]) -tmp_dst[383:256] := SELECT4(b[511:0], imm8[5:4]) -tmp_dst[511:384] := SELECT4(b[511:0], imm8[7:6]) - -FOR j := 0 to 7 - i := j*64 + Convert + + + + + Convert packed unsigned 64-bit integers in "a" to packed unsigned 32-bit integers with unsigned saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +FOR j := 0 to 3 + i := 64*j + l := 32*j IF k[j] - dst[i+63:i] := tmp_dst[i+63:i] + dst[l+31:l] := SaturateU32(a[i+63:i]) ELSE - dst[i+63:i] := src[i+63:i] + dst[l+31:l] := src[l+31:l] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Swizzle - - - - - Shuffle 128-bits (composed of 2 double-precision (64-bit) floating-point elements) selected by "imm8" from "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Convert + Store + + + + + Convert packed unsigned 64-bit integers in "a" to packed unsigned 32-bit integers with unsigned saturation, and store the active results (those with their respective bit set in writemask "k") to unaligned memory at "base_addr". -SELECT4(src, control){ - CASE(control[1:0]) - 0: tmp[127:0] := src[127:0] - 1: tmp[127:0] := src[255:128] - 2: tmp[127:0] := src[383:256] - 3: tmp[127:0] := src[511:384] - ESAC - RETURN tmp[127:0] -} - -tmp_dst[127:0] := SELECT4(a[511:0], imm8[1:0]) -tmp_dst[255:128] := SELECT4(a[511:0], imm8[3:2]) -tmp_dst[383:256] := SELECT4(b[511:0], imm8[5:4]) -tmp_dst[511:384] := SELECT4(b[511:0], imm8[7:6]) - -FOR j := 0 to 7 - i := j*64 +FOR j := 0 to 3 + i := 64*j + l := 32*j IF k[j] - dst[i+63:i] := tmp_dst[i+63:i] + MEM[base_addr+l+31:base_addr+l] := SaturateU32(a[i+63:i]) + FI +ENDFOR + + +
immintrin.h
+ + + Integer + AVX512VL + AVX512F + Convert + + + + Convert packed unsigned 64-bit integers in "a" to packed unsigned 32-bit integers with unsigned saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + +FOR j := 0 to 3 + i := 64*j + l := 32*j + IF k[j] + dst[l+31:l] := SaturateU32(a[i+63:i]) ELSE - dst[i+63:i] := 0 + dst[l+31:l] := 0 FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Swizzle - - - - Shuffle 128-bits (composed of 2 double-precision (64-bit) floating-point elements) selected by "imm8" from "a" and "b", and store the results in "dst". + Convert + + + Convert packed unsigned 64-bit integers in "a" to packed unsigned 32-bit integers with unsigned saturation, and store the results in "dst". -SELECT4(src, control){ - CASE(control[1:0]) - 0: tmp[127:0] := src[127:0] - 1: tmp[127:0] := src[255:128] - 2: tmp[127:0] := src[383:256] - 3: tmp[127:0] := src[511:384] - ESAC - RETURN tmp[127:0] -} - -dst[127:0] := SELECT4(a[511:0], imm8[1:0]) -dst[255:128] := SELECT4(a[511:0], imm8[3:2]) -dst[383:256] := SELECT4(b[511:0], imm8[5:4]) -dst[511:384] := SELECT4(b[511:0], imm8[7:6]) -dst[MAX:512] := 0 +FOR j := 0 to 1 + i := 64*j + k := 32*j + dst[k+31:k] := SaturateU32(a[i+63:i]) +ENDFOR +dst[MAX:64] := 0 - +
immintrin.h
 - + Integer + AVX512VL AVX512F - Swizzle - - - - - - Shuffle 128-bits (composed of 4 32-bit integers) selected by "imm8" from "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -SELECT4(src, control){ - CASE(control[1:0]) - 0: tmp[127:0] := src[127:0] - 1: tmp[127:0] := src[255:128] - 2: tmp[127:0] := src[383:256] - 3: tmp[127:0] := src[511:384] - ESAC - RETURN tmp[127:0] -} - -tmp_dst[127:0] := SELECT4(a[511:0], imm8[1:0]) -tmp_dst[255:128] := SELECT4(a[511:0], imm8[3:2]) -tmp_dst[383:256] := SELECT4(b[511:0], imm8[5:4]) -tmp_dst[511:384] := SELECT4(b[511:0], imm8[7:6]) - -FOR j := 0 to 15 - i := j*32 + Convert + + + + + Convert packed unsigned 64-bit integers in "a" to packed unsigned 32-bit integers with unsigned saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +FOR j := 0 to 1 + i := 64*j + l := 32*j IF k[j] - dst[i+31:i] := tmp_dst[i+31:i] + dst[l+31:l] := SaturateU32(a[i+63:i]) ELSE - dst[i+31:i] := src[i+31:i] + dst[l+31:l] := src[l+31:l] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:64] := 0 - +
immintrin.h
 - + Integer + AVX512VL AVX512F - Swizzle - - - - - Shuffle 128-bits (composed of 4 32-bit integers) selected by "imm8" from "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Convert + Store + + + + + Convert packed unsigned 64-bit integers in "a" to packed unsigned 32-bit integers with unsigned saturation, and store the active results (those with their respective bit set in writemask "k") to unaligned memory at "base_addr". -SELECT4(src, control){ - CASE(control[1:0]) - 0: tmp[127:0] := src[127:0] - 1: tmp[127:0] := src[255:128] - 2: tmp[127:0] := src[383:256] - 3: tmp[127:0] := src[511:384] - ESAC - RETURN tmp[127:0] -} - -tmp_dst[127:0] := SELECT4(a[511:0], imm8[1:0]) -tmp_dst[255:128] := SELECT4(a[511:0], imm8[3:2]) -tmp_dst[383:256] := SELECT4(b[511:0], imm8[5:4]) -tmp_dst[511:384] := SELECT4(b[511:0], imm8[7:6]) - -FOR j := 0 to 15 - i := j*32 +FOR j := 0 to 1 + i := 64*j + l := 32*j IF k[j] - dst[i+31:i] := tmp_dst[i+31:i] - ELSE - dst[i+31:i] := 0 + MEM[base_addr+l+31:base_addr+l] := SaturateU32(a[i+63:i]) FI ENDFOR -dst[MAX:512] := 0 - +
immintrin.h
 - + Integer + AVX512VL AVX512F - Swizzle - - - - Shuffle 128-bits (composed of 4 32-bit integers) selected by "imm8" from "a" and "b", and store the results in "dst". + Convert + + + + Convert packed unsigned 64-bit integers in "a" to packed unsigned 32-bit integers with unsigned saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -SELECT4(src, control){ - CASE(control[1:0]) - 0: tmp[127:0] := src[127:0] - 1: tmp[127:0] := src[255:128] - 2: tmp[127:0] := src[383:256] - 3: tmp[127:0] := src[511:384] - ESAC - RETURN tmp[127:0] -} - -dst[127:0] := SELECT4(a[511:0], imm8[1:0]) -dst[255:128] := SELECT4(a[511:0], imm8[3:2]) -dst[383:256] := SELECT4(b[511:0], imm8[5:4]) -dst[511:384] := SELECT4(b[511:0], imm8[7:6]) -dst[MAX:512] := 0 - - -
immintrin.h
- - - Integer - AVX512F - Swizzle - - - - - - Shuffle 128-bits (composed of 2 64-bit integers) selected by "imm8" from "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -SELECT4(src, control){ - CASE(control[1:0]) - 0: tmp[127:0] := src[127:0] - 1: tmp[127:0] := src[255:128] - 2: tmp[127:0] := src[383:256] - 3: tmp[127:0] := src[511:384] - ESAC - RETURN tmp[127:0] -} - -tmp_dst[127:0] := SELECT4(a[511:0], imm8[1:0]) -tmp_dst[255:128] := SELECT4(a[511:0], imm8[3:2]) -tmp_dst[383:256] := SELECT4(b[511:0], imm8[5:4]) -tmp_dst[511:384] := SELECT4(b[511:0], imm8[7:6]) - -FOR j := 0 to 7 - i := j*64 +FOR j := 0 to 1 + i := 64*j + l := 32*j IF k[j] - dst[i+63:i] := tmp_dst[i+63:i] + dst[l+31:l] := SaturateU32(a[i+63:i]) ELSE - dst[i+63:i] := src[i+63:i] + dst[l+31:l] := 0 FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:64] := 0 - +
immintrin.h
 - + Integer + AVX512VL AVX512F - Swizzle - - - - - Shuffle 128-bits (composed of 2 64-bit integers) selected by "imm8" from "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Convert + + + Convert packed unsigned 64-bit integers in "a" to packed unsigned 16-bit integers with unsigned saturation, and store the results in "dst". -SELECT4(src, control){ - CASE(control[1:0]) - 0: tmp[127:0] := src[127:0] - 1: tmp[127:0] := src[255:128] - 2: tmp[127:0] := src[383:256] - 3: tmp[127:0] := src[511:384] - ESAC - RETURN tmp[127:0] -} - -tmp_dst[127:0] := SELECT4(a[511:0], imm8[1:0]) -tmp_dst[255:128] := SELECT4(a[511:0], imm8[3:2]) -tmp_dst[383:256] := SELECT4(b[511:0], imm8[5:4]) -tmp_dst[511:384] := SELECT4(b[511:0], imm8[7:6]) - -FOR j := 0 to 7 - i := j*64 - IF k[j] - dst[i+63:i] := tmp_dst[i+63:i] - ELSE - dst[i+63:i] := 0 - FI +FOR j := 0 to 3 + i := 64*j + k := 16*j + dst[k+15:k] := SaturateU16(a[i+63:i]) ENDFOR -dst[MAX:512] := 0 +dst[MAX:64] := 0 - +
immintrin.h
 - + Integer + AVX512VL AVX512F - Swizzle - - - - Shuffle 128-bits (composed of 2 64-bit integers) selected by "imm8" from "a" and "b", and store the results in "dst". + Convert + + + + + Convert packed unsigned 64-bit integers in "a" to packed unsigned 16-bit integers with unsigned saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -SELECT4(src, control){ - CASE(control[1:0]) - 0: tmp[127:0] := src[127:0] - 1: tmp[127:0] := src[255:128] - 2: tmp[127:0] := src[383:256] - 3: tmp[127:0] := src[511:384] - ESAC - RETURN tmp[127:0] -} - -dst[127:0] := SELECT4(a[511:0], imm8[1:0]) -dst[255:128] := SELECT4(a[511:0], imm8[3:2]) -dst[383:256] := SELECT4(b[511:0], imm8[5:4]) -dst[511:384] := SELECT4(b[511:0], imm8[7:6]) -dst[MAX:512] := 0 +FOR j := 0 to 3 + i := 64*j + l := 16*j + IF k[j] + dst[l+15:l] := SaturateU16(a[i+63:i]) + ELSE + dst[l+15:l] := src[l+15:l] + FI +ENDFOR +dst[MAX:64] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Swizzle - - - - - - Shuffle double-precision (64-bit) floating-point elements within 128-bit lanes using the control in "imm8", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Convert + Store + + + + + Convert packed unsigned 64-bit integers in "a" to packed unsigned 16-bit integers with unsigned saturation, and store the active results (those with their respective bit set in writemask "k") to unaligned memory at "base_addr". -tmp_dst[63:0] := (imm8[0] == 0) ? a[63:0] : a[127:64] -tmp_dst[127:64] := (imm8[1] == 0) ? b[63:0] : b[127:64] -tmp_dst[191:128] := (imm8[2] == 0) ? a[191:128] : a[255:192] -tmp_dst[255:192] := (imm8[3] == 0) ? b[191:128] : b[255:192] -tmp_dst[319:256] := (imm8[4] == 0) ? a[319:256] : a[383:320] -tmp_dst[383:320] := (imm8[5] == 0) ? b[319:256] : b[383:320] -tmp_dst[447:384] := (imm8[6] == 0) ? a[447:384] : a[511:448] -tmp_dst[511:448] := (imm8[7] == 0) ? b[447:384] : b[511:448] - -FOR j := 0 to 7 - i := j*64 +FOR j := 0 to 3 + i := 64*j + l := 16*j IF k[j] - dst[i+63:i] := tmp_dst[i+63:i] - ELSE - dst[i+63:i] := src[i+63:i] + MEM[base_addr+l+15:base_addr+l] := SaturateU16(a[i+63:i]) FI ENDFOR -dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Swizzle - - - - - Shuffle double-precision (64-bit) floating-point elements within 128-bit lanes using the control in "imm8", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Convert + + + + Convert packed unsigned 64-bit integers in "a" to packed unsigned 16-bit integers with unsigned saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -tmp_dst[63:0] := (imm8[0] == 0) ? a[63:0] : a[127:64] -tmp_dst[127:64] := (imm8[1] == 0) ? b[63:0] : b[127:64] -tmp_dst[191:128] := (imm8[2] == 0) ? a[191:128] : a[255:192] -tmp_dst[255:192] := (imm8[3] == 0) ? b[191:128] : b[255:192] -tmp_dst[319:256] := (imm8[4] == 0) ? a[319:256] : a[383:320] -tmp_dst[383:320] := (imm8[5] == 0) ? b[319:256] : b[383:320] -tmp_dst[447:384] := (imm8[6] == 0) ? a[447:384] : a[511:448] -tmp_dst[511:448] := (imm8[7] == 0) ? b[447:384] : b[511:448] - -FOR j := 0 to 7 - i := j*64 +FOR j := 0 to 3 + i := 64*j + l := 16*j IF k[j] - dst[i+63:i] := tmp_dst[i+63:i] + dst[l+15:l] := SaturateU16(a[i+63:i]) ELSE - dst[i+63:i] := 0 + dst[l+15:l] := 0 FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:64] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Swizzle - - - - Shuffle double-precision (64-bit) floating-point elements within 128-bit lanes using the control in "imm8", and store the results in "dst". + Convert + + + Convert packed unsigned 64-bit integers in "a" to packed unsigned 16-bit integers with unsigned saturation, and store the results in "dst". -dst[63:0] := (imm8[0] == 0) ? a[63:0] : a[127:64] -dst[127:64] := (imm8[1] == 0) ? b[63:0] : b[127:64] -dst[191:128] := (imm8[2] == 0) ? a[191:128] : a[255:192] -dst[255:192] := (imm8[3] == 0) ? b[191:128] : b[255:192] -dst[319:256] := (imm8[4] == 0) ? a[319:256] : a[383:320] -dst[383:320] := (imm8[5] == 0) ? b[319:256] : b[383:320] -dst[447:384] := (imm8[6] == 0) ? a[447:384] : a[511:448] -dst[511:448] := (imm8[7] == 0) ? b[447:384] : b[511:448] -dst[MAX:512] := 0 +FOR j := 0 to 1 + i := 64*j + k := 16*j + dst[k+15:k] := SaturateU16(a[i+63:i]) +ENDFOR +dst[MAX:32] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Swizzle - - - - - - Shuffle single-precision (32-bit) floating-point elements in "a" within 128-bit lanes using the control in "imm8", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Convert + + + + + Convert packed unsigned 64-bit integers in "a" to packed unsigned 16-bit integers with unsigned saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -SELECT4(src, control){ - CASE(control[1:0]) - 0: tmp[31:0] := src[31:0] - 1: tmp[31:0] := src[63:32] - 2: tmp[31:0] := src[95:64] - 3: tmp[31:0] := src[127:96] - ESAC - RETURN tmp[31:0] -} - -tmp_dst[31:0] := SELECT4(a[127:0], imm8[1:0]) -tmp_dst[63:32] := SELECT4(a[127:0], imm8[3:2]) -tmp_dst[95:64] := SELECT4(b[127:0], imm8[5:4]) -tmp_dst[127:96] := SELECT4(b[127:0], imm8[7:6]) -tmp_dst[159:128] := SELECT4(a[255:128], imm8[1:0]) -tmp_dst[191:160] := SELECT4(a[255:128], imm8[3:2]) -tmp_dst[223:192] := SELECT4(b[255:128], imm8[5:4]) -tmp_dst[255:224] := SELECT4(b[255:128], imm8[7:6]) -tmp_dst[287:256] := SELECT4(a[383:256], imm8[1:0]) -tmp_dst[319:288] := SELECT4(a[383:256], imm8[3:2]) -tmp_dst[351:320] := SELECT4(b[383:256], imm8[5:4]) -tmp_dst[383:352] := SELECT4(b[383:256], imm8[7:6]) -tmp_dst[415:384] := SELECT4(a[511:384], imm8[1:0]) -tmp_dst[447:416] := SELECT4(a[511:384], imm8[3:2]) -tmp_dst[479:448] := SELECT4(b[511:384], imm8[5:4]) -tmp_dst[511:480] := SELECT4(b[511:384], imm8[7:6]) - -FOR j := 0 to 15 - i := j*32 +FOR j := 0 to 1 + i := 64*j + l := 16*j IF k[j] - dst[i+31:i] := tmp_dst[i+31:i] + dst[l+15:l] := SaturateU16(a[i+63:i]) ELSE - dst[i+31:i] := src[i+31:i] + dst[l+15:l] := src[l+15:l] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:32] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Swizzle - - - - - Shuffle single-precision (32-bit) floating-point elements in "a" within 128-bit lanes using the control in "imm8", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Convert + Store + + + + + Convert packed unsigned 64-bit integers in "a" to packed unsigned 16-bit integers with unsigned saturation, and store the active results (those with their respective bit set in writemask "k") to unaligned memory at "base_addr". -SELECT4(src, control){ - CASE(control[1:0]) - 0: tmp[31:0] := src[31:0] - 1: tmp[31:0] := src[63:32] - 2: tmp[31:0] := src[95:64] - 3: tmp[31:0] := src[127:96] - ESAC - RETURN tmp[31:0] -} - -tmp_dst[31:0] := SELECT4(a[127:0], imm8[1:0]) -tmp_dst[63:32] := SELECT4(a[127:0], imm8[3:2]) -tmp_dst[95:64] := SELECT4(b[127:0], imm8[5:4]) -tmp_dst[127:96] := SELECT4(b[127:0], imm8[7:6]) -tmp_dst[159:128] := SELECT4(a[255:128], imm8[1:0]) -tmp_dst[191:160] := SELECT4(a[255:128], imm8[3:2]) -tmp_dst[223:192] := SELECT4(b[255:128], imm8[5:4]) -tmp_dst[255:224] := SELECT4(b[255:128], imm8[7:6]) -tmp_dst[287:256] := SELECT4(a[383:256], imm8[1:0]) -tmp_dst[319:288] := SELECT4(a[383:256], imm8[3:2]) -tmp_dst[351:320] := SELECT4(b[383:256], imm8[5:4]) -tmp_dst[383:352] := SELECT4(b[383:256], imm8[7:6]) -tmp_dst[415:384] := SELECT4(a[511:384], imm8[1:0]) -tmp_dst[447:416] := SELECT4(a[511:384], imm8[3:2]) -tmp_dst[479:448] := SELECT4(b[511:384], imm8[5:4]) -tmp_dst[511:480] := SELECT4(b[511:384], imm8[7:6]) - -FOR j := 0 to 15 - i := j*32 +FOR j := 0 to 1 + i := 64*j + l := 16*j IF k[j] - dst[i+31:i] := tmp_dst[i+31:i] - ELSE - dst[i+31:i] := 0 + MEM[base_addr+l+15:base_addr+l] := SaturateU16(a[i+63:i]) FI ENDFOR -dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Swizzle - - - - Shuffle single-precision (32-bit) floating-point elements in "a" within 128-bit lanes using the control in "imm8", and store the results in "dst". + Convert + + + + Convert packed unsigned 64-bit integers in "a" to packed unsigned 16-bit integers with unsigned saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -SELECT4(src, control){ - CASE(control[1:0]) - 0: tmp[31:0] := src[31:0] - 1: tmp[31:0] := src[63:32] - 2: tmp[31:0] := src[95:64] - 3: tmp[31:0] := src[127:96] - ESAC - RETURN tmp[31:0] -} - -dst[31:0] := SELECT4(a[127:0], imm8[1:0]) -dst[63:32] := SELECT4(a[127:0], imm8[3:2]) -dst[95:64] := SELECT4(b[127:0], imm8[5:4]) -dst[127:96] := SELECT4(b[127:0], imm8[7:6]) -dst[159:128] := SELECT4(a[255:128], imm8[1:0]) -dst[191:160] := SELECT4(a[255:128], imm8[3:2]) -dst[223:192] := SELECT4(b[255:128], imm8[5:4]) -dst[255:224] := SELECT4(b[255:128], imm8[7:6]) -dst[287:256] := SELECT4(a[383:256], imm8[1:0]) -dst[319:288] := SELECT4(a[383:256], imm8[3:2]) -dst[351:320] := SELECT4(b[383:256], imm8[5:4]) -dst[383:352] := SELECT4(b[383:256], imm8[7:6]) -dst[415:384] := SELECT4(a[511:384], imm8[1:0]) -dst[447:416] := SELECT4(a[511:384], imm8[3:2]) -dst[479:448] := SELECT4(b[511:384], imm8[5:4]) -dst[511:480] := SELECT4(b[511:384], imm8[7:6]) -dst[MAX:512] := 0 +FOR j := 0 to 1 + i := 64*j + l := 16*j + IF k[j] + dst[l+15:l] := SaturateU16(a[i+63:i]) + ELSE + dst[l+15:l] := 0 + FI +ENDFOR +dst[MAX:32] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Elementary Math Functions - - - - Compute the square root of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Convert + + + + + Zero extend packed unsigned 8-bit integers in the low 8 bytes of "a" to packed 32-bit integers, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). FOR j := 0 to 7 - i := j*64 + i := 32*j + l := 8*j IF k[j] - dst[i+63:i] := SQRT(a[i+63:i]) + dst[i+31:i] := ZeroExtend32(a[l+7:l]) ELSE - dst[i+63:i] := src[i+63:i] + dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Elementary Math Functions - - - - - Compute the square root of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - [round_note] + Convert + + + + Zero extend packed unsigned 8-bit integers in the low 8 bytes of "a" to packed 32-bit integers, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). FOR j := 0 to 7 - i := j*64 + i := 32*j + l := 8*j IF k[j] - dst[i+63:i] := SQRT(a[i+63:i]) + dst[i+31:i] := ZeroExtend32(a[l+7:l]) ELSE - dst[i+63:i] := src[i+63:i] + dst[i+31:i] := 0 FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Elementary Math Functions - - - Compute the square root of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Convert + + + + + Zero extend packed unsigned 8-bit integers in the low 4 bytes of "a" to packed 32-bit integers, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := j*64 +FOR j := 0 to 3 + i := 32*j + l := 8*j IF k[j] - dst[i+63:i] := SQRT(a[i+63:i]) + dst[i+31:i] := ZeroExtend32(a[l+7:l]) ELSE - dst[i+63:i] := 0 + dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Elementary Math Functions - - - - Compute the square root of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - [round_note]. + Convert + + + + Zero extend packed unsigned 8-bit integers in th elow 4 bytes of "a" to packed 32-bit integers, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := j*64 +FOR j := 0 to 3 + i := 32*j + l := 8*j IF k[j] - dst[i+63:i] := SQRT(a[i+63:i]) + dst[i+31:i] := ZeroExtend32(a[l+7:l]) ELSE - dst[i+63:i] := 0 + dst[i+31:i] := 0 FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Elementary Math Functions - - Compute the square root of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst". + Convert + + + + + Zero extend packed unsigned 8-bit integers in the low 4 bytes of "a" to packed 64-bit integers, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := j*64 - dst[i+63:i] := SQRT(a[i+63:i]) +FOR j := 0 to 3 + i := 64*j + l := 8*j + IF k[j] + dst[i+63:i] := ZeroExtend64(a[l+7:l]) + ELSE + dst[i+63:i] := src[i+63:i] + FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Elementary Math Functions - - - Compute the square root of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst". - [round_note]. + Convert + + + + Zero extend packed unsigned 8-bit integers in the low 4 bytes of "a" to packed 64-bit integers, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := j*64 - dst[i+63:i] := SQRT(a[i+63:i]) +FOR j := 0 to 3 + i := 64*j + l := 8*j + IF k[j] + dst[i+63:i] := ZeroExtend64(a[l+7:l]) + ELSE + dst[i+63:i] := 0 + FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Elementary Math Functions - - - - Compute the square root of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Convert + + + + + Zero extend packed unsigned 8-bit integers in the low 2 bytes of "a" to packed 64-bit integers, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*32 +FOR j := 0 to 1 + i := 64*j + l := 8*j IF k[j] - dst[i+31:i] := SQRT(a[i+31:i]) + dst[i+63:i] := ZeroExtend64(a[l+7:l]) ELSE - dst[i+31:i] := src[i+31:i] + dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Elementary Math Functions - - - - - Compute the square root of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - [round_note] + Convert + + + + Zero extend packed unsigned 8-bit integers in the low 2 bytes of "a" to packed 64-bit integers, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*32 +FOR j := 0 to 1 + i := 64*j + l := 8*j IF k[j] - dst[i+31:i] := SQRT(a[i+31:i]) + dst[i+63:i] := ZeroExtend64(a[l+7:l]) ELSE - dst[i+31:i] := src[i+31:i] + dst[i+63:i] := 0 FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Elementary Math Functions - - - Compute the square root of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Convert + + + + + Zero extend packed unsigned 32-bit integers in "a" to packed 64-bit integers, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*32 +FOR j := 0 to 3 + i := 64*j + l := 32*j IF k[j] - dst[i+31:i] := SQRT(a[i+31:i]) + dst[i+63:i] := ZeroExtend64(a[l+31:l]) ELSE - dst[i+31:i] := 0 + dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Elementary Math Functions - - - - Compute the square root of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - [round_note] + Convert + + + + Zero extend packed unsigned 32-bit integers in "a" to packed 64-bit integers, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*32 +FOR j := 0 to 3 + i := 64*j + l := 32*j IF k[j] - dst[i+31:i] := SQRT(a[i+31:i]) - ELSE - dst[i+31:i] := 0 + dst[i+63:i] := ZeroExtend64(a[l+31:l]) + ELSE + dst[i+63:i] := 0 FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Elementary Math Functions - - Compute the square root of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst". + Convert + + + + + Zero extend packed unsigned 32-bit integers in "a" to packed 64-bit integers, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*32 - dst[i+31:i] := SQRT(a[i+31:i]) +FOR j := 0 to 1 + i := 64*j + l := 32*j + IF k[j] + dst[i+63:i] := ZeroExtend64(a[l+31:l]) + ELSE + dst[i+63:i] := src[i+63:i] + FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Elementary Math Functions - - - Compute the square root of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst". - [round_note]. + Convert + + + + Zero extend packed unsigned 32-bit integers in "a" to packed 64-bit integers, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*32 - dst[i+31:i] := SQRT(a[i+31:i]) +FOR j := 0 to 1 + i := 64*j + l := 32*j + IF k[j] + dst[i+63:i] := ZeroExtend64(a[l+31:l]) + ELSE + dst[i+63:i] := 0 + FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Elementary Math Functions - - - - - - Compute the square root of the lower double-precision (64-bit) floating-point element in "b", store the result in the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". - [round_note] - + Convert + + + + + Zero extend packed unsigned 16-bit integers in "a" to packed 32-bit integers, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -IF k[0] - dst[63:0] := SQRT(b[63:0]) -ELSE - dst[63:0] := src[63:0] -FI -dst[127:64] := a[127:64] -dst[MAX:128] := 0 - - -
immintrin.h
- - - Floating Point - AVX512F - Elementary Math Functions - - - - - Compute the square root of the lower double-precision (64-bit) floating-point element in "b", store the result in the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". - -IF k[0] - dst[63:0] := SQRT(b[63:0]) -ELSE - dst[63:0] := src[63:0] -FI -dst[127:64] := a[127:64] -dst[MAX:128] := 0 +FOR j := 0 to 7 + i := 32*j + l := 16*j + IF k[j] + dst[i+31:i] := ZeroExtend32(a[l+15:l]) + ELSE + dst[i+31:i] := src[i+31:i] + FI +ENDFOR +dst[MAX:256] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Elementary Math Functions - - - - - Compute the square root of the lower double-precision (64-bit) floating-point element in "b", store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". - [round_note] - + Convert + + + + Zero extend packed unsigned 16-bit integers in "a" to packed 32-bit integers, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -IF k[0] - dst[63:0] := SQRT(b[63:0]) -ELSE - dst[63:0] := 0 -FI -dst[127:64] := a[127:64] -dst[MAX:128] := 0 +FOR j := 0 to 7 + i := 32*j + l := 16*j + IF k[j] + dst[i+31:i] := ZeroExtend32(a[l+15:l]) + ELSE + dst[i+31:i] := 0 + FI +ENDFOR +dst[MAX:256] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Elementary Math Functions - - - - Compute the square root of the lower double-precision (64-bit) floating-point element in "b", store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". + Convert + + + + + Zero extend packed unsigned 16-bit integers in "a" to packed 32-bit integers, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -IF k[0] - dst[63:0] := SQRT(b[63:0]) -ELSE - dst[63:0] := 0 -FI -dst[127:64] := a[127:64] +FOR j := 0 to 3 + i := 32*j + l := 16*j + IF k[j] + dst[i+31:i] := ZeroExtend32(a[l+15:l]) + ELSE + dst[i+31:i] := src[i+31:i] + FI +ENDFOR dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Elementary Math Functions - - - - Compute the square root of the lower double-precision (64-bit) floating-point element in "b", store the result in the lower element of "dst", and copy the upper element from "a" to the upper element of "dst". - [round_note] - + Convert + + + + Zero extend packed unsigned 16-bit integers in "a" to packed 32-bit integers, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -dst[63:0] := SQRT(b[63:0]) -dst[127:64] := a[127:64] +FOR j := 0 to 3 + i := 32*j + l := 16*j + IF k[j] + dst[i+31:i] := ZeroExtend32(a[l+15:l]) + ELSE + dst[i+31:i] := 0 + FI +ENDFOR dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Elementary Math Functions - - - - - - Compute the square root of the lower single-precision (32-bit) floating-point element in "a", store the result in the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and copy the upper 3 packed elements from "b" to the upper elements of "dst". - [round_note] - + Convert + + + + + Zero extend packed unsigned 16-bit integers in the low 8 bytes of "a" to packed 64-bit integers, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -IF k[0] - dst[31:0] := SQRT(b[31:0]) -ELSE - dst[31:0] := src[31:0] -FI -dst[127:32] := a[127:32] -dst[MAX:128] := 0 +FOR j := 0 to 3 + i := 64*j + l := 16*j + IF k[j] + dst[i+63:i] := ZeroExtend64(a[l+15:l]) + ELSE + dst[i+63:i] := src[i+63:i] + FI +ENDFOR +dst[MAX:256] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Elementary Math Functions - - - - - Compute the square root of the lower single-precision (32-bit) floating-point element in "a", store the result in the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and copy the upper 3 packed elements from "b" to the upper elements of "dst". + Convert + + + + Zero extend packed unsigned 16-bit integers in the low 8 bytes of "a" to packed 64-bit integers, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -IF k[0] - dst[31:0] := SQRT(b[31:0]) -ELSE - dst[31:0] := src[31:0] -FI -dst[127:32] := a[127:32] -dst[MAX:128] := 0 +FOR j := 0 to 3 + i := 64*j + l := 16*j + IF k[j] + dst[i+63:i] := ZeroExtend64(a[l+15:l]) + ELSE + dst[i+63:i] := 0 + FI +ENDFOR +dst[MAX:256] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Elementary Math Functions - - - - - Compute the square root of the lower single-precision (32-bit) floating-point element in "a", store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper 3 packed elements from "b" to the upper elements of "dst". - [round_note] - + Convert + + + + + Zero extend packed unsigned 16-bit integers in the low 4 bytes of "a" to packed 64-bit integers, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -IF k[0] - dst[31:0] := SQRT(b[31:0]) -ELSE - dst[31:0] := 0 -FI -dst[127:32] := a[127:32] +FOR j := 0 to 1 + i := 64*j + l := 16*j + IF k[j] + dst[i+63:i] := ZeroExtend64(a[l+15:l]) + ELSE + dst[i+63:i] := src[i+63:i] + FI +ENDFOR dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Elementary Math Functions - - - - Compute the square root of the lower single-precision (32-bit) floating-point element in "a", store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper 3 packed elements from "b" to the upper elements of "dst". + Convert + + + + Zero extend packed unsigned 16-bit integers in the low 4 bytes of "a" to packed 64-bit integers, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -IF k[0] - dst[31:0] := SQRT(b[31:0]) -ELSE - dst[31:0] := 0 -FI -dst[127:32] := a[127:32] +FOR j := 0 to 1 + i := 64*j + l := 16*j + IF k[j] + dst[i+63:i] := ZeroExtend64(a[l+15:l]) + ELSE + dst[i+63:i] := 0 + FI +ENDFOR dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Elementary Math Functions - - - - Compute the square root of the lower single-precision (32-bit) floating-point element in "a", store the result in the lower element of "dst", and copy the upper 3 packed elements from "b" to the upper elements of "dst". - [round_note] - - -dst[31:0] := SQRT(b[31:0]) -dst[127:32] := a[127:32] -dst[MAX:128] := 0 - - -
immintrin.h
- - - Floating Point - AVX512F/KNCNI Arithmetic - - - - - Subtract packed double-precision (64-bit) floating-point elements in "b" from packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + + + + + Multiply the low signed 32-bit integers from each packed 64-bit element in "a" and "b", and store the signed 64-bit results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 7 +FOR j := 0 to 3 i := j*64 IF k[j] - dst[i+63:i] := a[i+63:i] - b[i+63:i] + dst[i+63:i] := SignExtend64(a[i+31:i]) * SignExtend64(b[i+31:i]) ELSE dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - - Floating Point - AVX512F/KNCNI + + Integer + AVX512VL + AVX512F Arithmetic - - - - - - Subtract packed double-precision (64-bit) floating-point elements in "b" from packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - [round_note] + + + + + Multiply the low signed 32-bit integers from each packed 64-bit element in "a" and "b", and store the signed 64-bit results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 +FOR j := 0 to 3 i := j*64 IF k[j] - dst[i+63:i] := a[i+63:i] - b[i+63:i] + dst[i+63:i] := SignExtend64(a[i+31:i]) * SignExtend64(b[i+31:i]) ELSE - dst[i+63:i] := src[i+63:i] + dst[i+63:i] := 0 FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F Arithmetic - - - - Subtract packed double-precision (64-bit) floating-point elements in "b" from packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + + + + + Multiply the low signed 32-bit integers from each packed 64-bit element in "a" and "b", and store the signed 64-bit results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 7 +FOR j := 0 to 1 i := j*64 IF k[j] - dst[i+63:i] := a[i+63:i] - b[i+63:i] + dst[i+63:i] := SignExtend64(a[i+31:i]) * SignExtend64(b[i+31:i]) ELSE - dst[i+63:i] := 0 + dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F Arithmetic - - - - - Subtract packed double-precision (64-bit) floating-point elements in "b" from packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - [round_note] = + + + + + Multiply the low signed 32-bit integers from each packed 64-bit element in "a" and "b", and store the signed 64-bit results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 +FOR j := 0 to 1 i := j*64 IF k[j] - dst[i+63:i] := a[i+63:i] - b[i+63:i] + dst[i+63:i] := SignExtend64(a[i+31:i]) * SignExtend64(b[i+31:i]) ELSE dst[i+63:i] := 0 FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point - AVX512F/KNCNI + + Integer + AVX512VL + AVX512F Arithmetic - - - Subtract packed double-precision (64-bit) floating-point elements in "b" from packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst". + + + + + + Multiply the packed 32-bit integers in "a" and "b", producing intermediate 64-bit integers, and store the low 32 bits of the intermediate integers in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). FOR j := 0 to 7 - i := j*64 - dst[i+63:i] := a[i+63:i] - b[i+63:i] + i := j*32 + IF k[j] + tmp[63:0] := a[i+31:i] * b[i+31:i] + dst[i+31:i] := tmp[31:0] + ELSE + dst[i+31:i] := src[i+31:i] + FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - - Floating Point - AVX512F/KNCNI + + Integer + AVX512VL + AVX512F Arithmetic - - - - Subtract packed double-precision (64-bit) floating-point elements in "b" from packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst". - [round_note] + + + + + Multiply the packed 32-bit integers in "a" and "b", producing intermediate 64-bit integers, and store the low 32 bits of the intermediate integers in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). FOR j := 0 to 7 - i := j*64 - dst[i+63:i] := a[i+63:i] - b[i+63:i] + i := j*32 + IF k[j] + tmp[63:0] := a[i+31:i] * b[i+31:i] + dst[i+31:i] := tmp[31:0] + ELSE + dst[i+31:i] := 0 + FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - - Floating Point - AVX512F/KNCNI + + Integer + AVX512F Arithmetic - - - - - Subtract packed single-precision (32-bit) floating-point elements in "b" from packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + + + + Multiply the packed 32-bit integers in "a" and "b", producing intermediate 64-bit integers, and store the low 32 bits of the intermediate integers in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). FOR j := 0 to 15 i := j*32 IF k[j] - dst[i+31:i] := a[i+31:i] - b[i+31:i] + tmp[63:0] := a[i+31:i] * b[i+31:i] + dst[i+31:i] := tmp[31:0] ELSE - dst[i+31:i] := src[i+31:i] - FI + dst[i+31:i] := 0 + FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point - AVX512F/KNCNI + + Integer + AVX512VL + AVX512F Arithmetic - - - - - - Subtract packed single-precision (32-bit) floating-point elements in "b" from packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - [round_note] + + + + + + Multiply the packed 32-bit integers in "a" and "b", producing intermediate 64-bit integers, and store the low 32 bits of the intermediate integers in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 15 +FOR j := 0 to 3 i := j*32 IF k[j] - dst[i+31:i] := a[i+31:i] - b[i+31:i] + tmp[63:0] := a[i+31:i] * b[i+31:i] + dst[i+31:i] := tmp[31:0] ELSE dst[i+31:i] := src[i+31:i] - FI + FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F Arithmetic - - - - Subtract packed single-precision (32-bit) floating-point elements in "b" from packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + + + + Multiply the packed 32-bit integers in "a" and "b", producing intermediate 64-bit integers, and store the low 32 bits of the intermediate integers in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 15 +FOR j := 0 to 3 i := j*32 IF k[j] - dst[i+31:i] := a[i+31:i] - b[i+31:i] + tmp[63:0] := a[i+31:i] * b[i+31:i] + dst[i+31:i] := tmp[31:0] ELSE dst[i+31:i] := 0 - FI + FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F Arithmetic - - - - - Subtract packed single-precision (32-bit) floating-point elements in "b" from packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - [round_note] + + + + + + Multiply the low unsigned 32-bit integers from each packed 64-bit element in "a" and "b", and store the unsigned 64-bit results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*32 +FOR j := 0 to 3 + i := j*64 IF k[j] - dst[i+31:i] := a[i+31:i] - b[i+31:i] + dst[i+63:i] := a[i+31:i] * b[i+31:i] ELSE - dst[i+31:i] := 0 - FI + dst[i+63:i] := src[i+63:i] + FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - - Floating Point - AVX512F/KNCNI + + Integer + AVX512VL + AVX512F Arithmetic - - - Subtract packed single-precision (32-bit) floating-point elements in "b" from packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst". + + + + + Multiply the low unsigned 32-bit integers from each packed 64-bit element in "a" and "b", and store the unsigned 64-bit results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*32 - dst[i+31:i] := a[i+31:i] - b[i+31:i] +FOR j := 0 to 3 + i := j*64 + IF k[j] + dst[i+63:i] := a[i+31:i] * b[i+31:i] + ELSE + dst[i+63:i] := 0 + FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - - Floating Point - AVX512F/KNCNI + + Integer + AVX512VL + AVX512F Arithmetic - - - - Subtract packed single-precision (32-bit) floating-point elements in "b" from packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst". - [round_note] + + + + + + Multiply the low unsigned 32-bit integers from each packed 64-bit element in "a" and "b", and store the unsigned 64-bit results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*32 - dst[i+31:i] := a[i+31:i] - b[i+31:i] +FOR j := 0 to 1 + i := j*64 + IF k[j] + dst[i+63:i] := a[i+31:i] * b[i+31:i] + ELSE + dst[i+63:i] := src[i+63:i] + FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F Arithmetic - - - - - - Subtract the lower double-precision (64-bit) floating-point element in "b" from the lower double-precision (64-bit) floating-point element in "a", store the result in the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". - [round_note] - + + + + + Multiply the low unsigned 32-bit integers from each packed 64-bit element in "a" and "b", and store the unsigned 64-bit results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -IF k[0] - dst[63:0] := a[63:0] - b[63:0] -ELSE - dst[63:0] := src[63:0] -FI -dst[127:64] := a[127:64] +FOR j := 0 to 1 + i := j*64 + IF k[j] + dst[i+63:i] := a[i+31:i] * b[i+31:i] + ELSE + dst[i+63:i] := 0 + FI +ENDFOR dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Arithmetic - - - - - Subtract the lower double-precision (64-bit) floating-point element in "b" from the lower double-precision (64-bit) floating-point element in "a", store the result in the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". + Logical + + + + + + Compute the bitwise OR of packed 32-bit integers in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -IF k[0] - dst[63:0] := a[63:0] - b[63:0] -ELSE - dst[63:0] := src[63:0] -FI -dst[127:64] := a[127:64] -dst[MAX:128] := 0 +FOR j := 0 to 7 + i := j*32 + IF k[j] + dst[i+31:i] := a[i+31:i] OR b[i+31:i] + ELSE + dst[i+31:i] := src[i+31:i] + FI +ENDFOR +dst[MAX:256] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Arithmetic - - - - - Subtract the lower double-precision (64-bit) floating-point element in "b" from the lower double-precision (64-bit) floating-point element in "a", store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". - [round_note] - + Logical + + + + + Compute the bitwise OR of packed 32-bit integers in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -IF k[0] - dst[63:0] := a[63:0] - b[63:0] -ELSE - dst[63:0] := 0 -FI -dst[127:64] := a[127:64] -dst[MAX:128] := 0 +FOR j := 0 to 7 + i := j*32 + IF k[j] + dst[i+31:i] := a[i+31:i] OR b[i+31:i] + ELSE + dst[i+31:i] := 0 + FI +ENDFOR +dst[MAX:256] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Arithmetic - - - - Subtract the lower double-precision (64-bit) floating-point element in "b" from the lower double-precision (64-bit) floating-point element in "a", store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". + Logical + + + + + + Compute the bitwise OR of packed 32-bit integers in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -IF k[0] - dst[63:0] := a[63:0] - b[63:0] -ELSE - dst[63:0] := 0 -FI -dst[127:64] := a[127:64] +FOR j := 0 to 3 + i := j*32 + IF k[j] + dst[i+31:i] := a[i+31:i] OR b[i+31:i] + ELSE + dst[i+31:i] := src[i+31:i] + FI +ENDFOR dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Arithmetic - - - - Subtract the lower double-precision (64-bit) floating-point element in "b" from the lower double-precision (64-bit) floating-point element in "a", store the result in the lower element of "dst", and copy the upper element from "a" to the upper element of "dst". - [round_note] - + Logical + + + + + Compute the bitwise OR of packed 32-bit integers in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -dst[63:0] := a[63:0] - b[63:0] -dst[127:64] := a[127:64] +FOR j := 0 to 3 + i := j*32 + IF k[j] + dst[i+31:i] := a[i+31:i] OR b[i+31:i] + ELSE + dst[i+31:i] := 0 + FI +ENDFOR dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Arithmetic - - - - - - Subtract the lower single-precision (32-bit) floating-point element in "b" from the lower single-precision (32-bit) floating-point element in "a", store the result in the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". - [round_note] - + Logical + + + + + + Compute the bitwise OR of packed 64-bit integers in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -IF k[0] - dst[31:0] := a[31:0] - b[31:0] -ELSE - dst[31:0] := src[31:0] -FI -dst[127:32] := a[127:32] -dst[MAX:128] := 0 +FOR j := 0 to 3 + i := j*64 + IF k[j] + dst[i+63:i] := a[i+63:i] OR b[i+63:i] + ELSE + dst[i+63:i] := src[i+63:i] + FI +ENDFOR +dst[MAX:256] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Arithmetic - - - - - Subtract the lower single-precision (32-bit) floating-point element in "b" from the lower single-precision (32-bit) floating-point element in "a", store the result in the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". + Logical + + + + + Compute the bitwise OR of packed 64-bit integers in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -IF k[0] - dst[31:0] := a[31:0] - b[31:0] -ELSE - dst[31:0] := src[31:0] -FI -dst[127:32] := a[127:32] -dst[MAX:128] := 0 +FOR j := 0 to 3 + i := j*64 + IF k[j] + dst[i+63:i] := a[i+63:i] OR b[i+63:i] + ELSE + dst[i+63:i] := 0 + FI +ENDFOR +dst[MAX:256] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Arithmetic - - - - - Subtract the lower single-precision (32-bit) floating-point element in "b" from the lower single-precision (32-bit) floating-point element in "a", store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". - [round_note] - + Logical + + + + + + Compute the bitwise OR of packed 64-bit integers in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -IF k[0] - dst[31:0] := a[31:0] - b[31:0] -ELSE - dst[31:0] := 0 -FI -dst[127:32] := a[127:32] -dst[MAX:128] := 0 - - -
immintrin.h
- - - Floating Point - AVX512F - Arithmetic - - - - Subtract the lower single-precision (32-bit) floating-point element in "b" from the lower single-precision (32-bit) floating-point element in "a", store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". - -IF k[0] - dst[31:0] := a[31:0] - b[31:0] -ELSE - dst[31:0] := 0 -FI -dst[127:32] := a[127:32] +FOR j := 0 to 1 + i := j*64 + IF k[j] + dst[i+63:i] := a[i+63:i] OR b[i+63:i] + ELSE + dst[i+63:i] := src[i+63:i] + FI +ENDFOR dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Arithmetic - - - - Subtract the lower single-precision (32-bit) floating-point element in "b" from the lower single-precision (32-bit) floating-point element in "a", store the result in the lower element of "dst", and copy the upper 3 packed elements from "a" to the upper elements of "dst". - [round_note] - + Logical + + + + + Compute the bitwise OR of packed 64-bit integers in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -dst[31:0] := a[31:0] - b[31:0] -dst[127:32] := a[127:32] +FOR j := 0 to 1 + i := j*64 + IF k[j] + dst[i+63:i] := a[i+63:i] OR b[i+63:i] + ELSE + dst[i+63:i] := 0 + FI +ENDFOR dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Swizzle - - - - - Unpack and interleave double-precision (64-bit) floating-point elements from the high half of each 128-bit lane in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -INTERLEAVE_HIGH_QWORDS(src1[127:0], src2[127:0]){ - dst[63:0] := src1[127:64] - dst[127:64] := src2[127:64] - RETURN dst[127:0] -} - -tmp_dst[127:0] := INTERLEAVE_HIGH_QWORDS(a[127:0], b[127:0]) -tmp_dst[255:128] := INTERLEAVE_HIGH_QWORDS(a[255:128], b[255:128]) -tmp_dst[383:256] := INTERLEAVE_HIGH_QWORDS(a[383:256], b[383:256]) -tmp_dst[511:384] := INTERLEAVE_HIGH_QWORDS(a[511:384], b[511:384]) - + Shift + + + + + + Rotate the bits in each packed 32-bit integer in "a" to the left by the number of bits specified in "imm8", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +DEFINE LEFT_ROTATE_DWORDS(src, count_src) { + count := count_src % 32 + RETURN (src << count) OR (src >> (32 - count)) +} FOR j := 0 to 7 - i := j*64 + i := j*32 IF k[j] - dst[i+63:i] := tmp_dst[i+63:i] + dst[i+31:i] := LEFT_ROTATE_DWORDS(a[i+31:i], imm8[7:0]) ELSE - dst[i+63:i] := src[i+63:i] + dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Swizzle - - - - Unpack and interleave double-precision (64-bit) floating-point elements from the high half of each 128-bit lane in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - -INTERLEAVE_HIGH_QWORDS(src1[127:0], src2[127:0]){ - dst[63:0] := src1[127:64] - dst[127:64] := src2[127:64] - RETURN dst[127:0] -} - -tmp_dst[127:0] := INTERLEAVE_HIGH_QWORDS(a[127:0], b[127:0]) -tmp_dst[255:128] := INTERLEAVE_HIGH_QWORDS(a[255:128], b[255:128]) -tmp_dst[383:256] := INTERLEAVE_HIGH_QWORDS(a[383:256], b[383:256]) -tmp_dst[511:384] := INTERLEAVE_HIGH_QWORDS(a[511:384], b[511:384]) - + Shift + + + + + Rotate the bits in each packed 32-bit integer in "a" to the left by the number of bits specified in "imm8", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + +DEFINE LEFT_ROTATE_DWORDS(src, count_src) { + count := count_src % 32 + RETURN (src << count) OR (src >> (32 - count)) +} FOR j := 0 to 7 - i := j*64 + i := j*32 IF k[j] - dst[i+63:i] := tmp_dst[i+63:i] + dst[i+31:i] := LEFT_ROTATE_DWORDS(a[i+31:i], imm8[7:0]) ELSE - dst[i+63:i] := 0 + dst[i+31:i] := 0 FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Swizzle - - - Unpack and interleave double-precision (64-bit) floating-point elements from the high half of each 128-bit lane in "a" and "b", and store the results in "dst". + Shift + + + + Rotate the bits in each packed 32-bit integer in "a" to the left by the number of bits specified in "imm8", and store the results in "dst". -INTERLEAVE_HIGH_QWORDS(src1[127:0], src2[127:0]){ - dst[63:0] := src1[127:64] - dst[127:64] := src2[127:64] - RETURN dst[127:0] -} - -dst[127:0] := INTERLEAVE_HIGH_QWORDS(a[127:0], b[127:0]) -dst[255:128] := INTERLEAVE_HIGH_QWORDS(a[255:128], b[255:128]) -dst[383:256] := INTERLEAVE_HIGH_QWORDS(a[383:256], b[383:256]) -dst[511:384] := INTERLEAVE_HIGH_QWORDS(a[511:384], b[511:384]) -dst[MAX:512] := 0 +DEFINE LEFT_ROTATE_DWORDS(src, count_src) { + count := count_src % 32 + RETURN (src << count) OR (src >> (32 - count)) +} +FOR j := 0 to 7 + i := j*32 + dst[i+31:i] := LEFT_ROTATE_DWORDS(a[i+31:i], imm8[7:0]) +ENDFOR +dst[MAX:256] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Swizzle - - - - - Unpack and interleave single-precision (32-bit) floating-point elements from the high half of each 128-bit lane in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -INTERLEAVE_HIGH_DWORDS(src1[127:0], src2[127:0]){ - dst[31:0] := src1[95:64] - dst[63:32] := src2[95:64] - dst[95:64] := src1[127:96] - dst[127:96] := src2[127:96] - RETURN dst[127:0] -} - -tmp_dst[127:0] := INTERLEAVE_HIGH_DWORDS(a[127:0], b[127:0]) -tmp_dst[255:128] := INTERLEAVE_HIGH_DWORDS(a[255:128], b[255:128]) -tmp_dst[383:256] := INTERLEAVE_HIGH_DWORDS(a[383:256], b[383:256]) -tmp_dst[511:384] := INTERLEAVE_HIGH_DWORDS(a[511:384], b[511:384]) - -FOR j := 0 to 15 + Shift + + + + + + Rotate the bits in each packed 32-bit integer in "a" to the left by the number of bits specified in "imm8", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +DEFINE LEFT_ROTATE_DWORDS(src, count_src) { + count := count_src % 32 + RETURN (src << count) OR (src >> (32 - count)) +} +FOR j := 0 to 3 i := j*32 IF k[j] - dst[i+31:i] := tmp_dst[i+31:i] + dst[i+31:i] := LEFT_ROTATE_DWORDS(a[i+31:i], imm8[7:0]) ELSE dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Swizzle - - - - Unpack and interleave single-precision (32-bit) floating-point elements from the high half of each 128-bit lane in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - -INTERLEAVE_HIGH_DWORDS(src1[127:0], src2[127:0]){ - dst[31:0] := src1[95:64] - dst[63:32] := src2[95:64] - dst[95:64] := src1[127:96] - dst[127:96] := src2[127:96] - RETURN dst[127:0] -} - -tmp_dst[127:0] := INTERLEAVE_HIGH_DWORDS(a[127:0], b[127:0]) -tmp_dst[255:128] := INTERLEAVE_HIGH_DWORDS(a[255:128], b[255:128]) -tmp_dst[383:256] := INTERLEAVE_HIGH_DWORDS(a[383:256], b[383:256]) -tmp_dst[511:384] := INTERLEAVE_HIGH_DWORDS(a[511:384], b[511:384]) - -FOR j := 0 to 15 + Shift + + + + + Rotate the bits in each packed 32-bit integer in "a" to the left by the number of bits specified in "imm8", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + +DEFINE LEFT_ROTATE_DWORDS(src, count_src) { + count := count_src % 32 + RETURN (src << count) OR (src >> (32 - count)) +} +FOR j := 0 to 3 i := j*32 IF k[j] - dst[i+31:i] := tmp_dst[i+31:i] + dst[i+31:i] := LEFT_ROTATE_DWORDS(a[i+31:i], imm8[7:0]) ELSE dst[i+31:i] := 0 FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Swizzle - - - Unpack and interleave single-precision (32-bit) floating-point elements from the high half of each 128-bit lane in "a" and "b", and store the results in "dst". + Shift + + + + Rotate the bits in each packed 32-bit integer in "a" to the left by the number of bits specified in "imm8", and store the results in "dst". -INTERLEAVE_HIGH_DWORDS(src1[127:0], src2[127:0]){ - dst[31:0] := src1[95:64] - dst[63:32] := src2[95:64] - dst[95:64] := src1[127:96] - dst[127:96] := src2[127:96] - RETURN dst[127:0] -} - -dst[127:0] := INTERLEAVE_HIGH_DWORDS(a[127:0], b[127:0]) -dst[255:128] := INTERLEAVE_HIGH_DWORDS(a[255:128], b[255:128]) -dst[383:256] := INTERLEAVE_HIGH_DWORDS(a[383:256], b[383:256]) -dst[511:384] := INTERLEAVE_HIGH_DWORDS(a[511:384], b[511:384]) -dst[MAX:512] := 0 +DEFINE LEFT_ROTATE_DWORDS(src, count_src) { + count := count_src % 32 + RETURN (src << count) OR (src >> (32 - count)) +} +FOR j := 0 to 3 + i := j*32 + dst[i+31:i] := LEFT_ROTATE_DWORDS(a[i+31:i], imm8[7:0]) +ENDFOR +dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Swizzle - - - - - Unpack and interleave double-precision (64-bit) floating-point elements from the low half of each 128-bit lane in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -INTERLEAVE_QWORDS(src1[127:0], src2[127:0]){ - dst[63:0] := src1[63:0] - dst[127:64] := src2[63:0] - RETURN dst[127:0] + Shift + + + + + + Rotate the bits in each packed 64-bit integer in "a" to the left by the number of bits specified in "imm8", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +DEFINE LEFT_ROTATE_QWORDS(src, count_src) { + count := count_src % 64 + RETURN (src << count) OR (src >> (64 - count)) } - -tmp_dst[127:0] := INTERLEAVE_QWORDS(a[127:0], b[127:0]) -tmp_dst[255:128] := INTERLEAVE_QWORDS(a[255:128], b[255:128]) -tmp_dst[383:256] := INTERLEAVE_QWORDS(a[383:256], b[383:256]) -tmp_dst[511:384] := INTERLEAVE_QWORDS(a[511:384], b[511:384]) - -FOR j := 0 to 7 +FOR j := 0 to 3 i := j*64 IF k[j] - dst[i+63:i] := tmp_dst[i+63:i] + dst[i+63:i] := LEFT_ROTATE_QWORDS(a[i+63:i], imm8[7:0]) ELSE dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Swizzle - - - - Unpack and interleave double-precision (64-bit) floating-point elements from the low half of each 128-bit lane in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - -INTERLEAVE_QWORDS(src1[127:0], src2[127:0]){ - dst[63:0] := src1[63:0] - dst[127:64] := src2[63:0] - RETURN dst[127:0] + Shift + + + + + Rotate the bits in each packed 64-bit integer in "a" to the left by the number of bits specified in "imm8", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + +DEFINE LEFT_ROTATE_QWORDS(src, count_src) { + count := count_src % 64 + RETURN (src << count) OR (src >> (64 - count)) } - -tmp_dst[127:0] := INTERLEAVE_QWORDS(a[127:0], b[127:0]) -tmp_dst[255:128] := INTERLEAVE_QWORDS(a[255:128], b[255:128]) -tmp_dst[383:256] := INTERLEAVE_QWORDS(a[383:256], b[383:256]) -tmp_dst[511:384] := INTERLEAVE_QWORDS(a[511:384], b[511:384]) - -FOR j := 0 to 7 +FOR j := 0 to 3 i := j*64 IF k[j] - dst[i+63:i] := tmp_dst[i+63:i] + dst[i+63:i] := LEFT_ROTATE_QWORDS(a[i+63:i], imm8[7:0]) ELSE dst[i+63:i] := 0 FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Swizzle - - - Unpack and interleave double-precision (64-bit) floating-point elements from the low half of each 128-bit lane in "a" and "b", and store the results in "dst". + Shift + + + + Rotate the bits in each packed 64-bit integer in "a" to the left by the number of bits specified in "imm8", and store the results in "dst". -INTERLEAVE_QWORDS(src1[127:0], src2[127:0]){ - dst[63:0] := src1[63:0] - dst[127:64] := src2[63:0] - RETURN dst[127:0] +DEFINE LEFT_ROTATE_QWORDS(src, count_src) { + count := count_src % 64 + RETURN (src << count) OR (src >> (64 - count)) } - -dst[127:0] := INTERLEAVE_QWORDS(a[127:0], b[127:0]) -dst[255:128] := INTERLEAVE_QWORDS(a[255:128], b[255:128]) -dst[383:256] := INTERLEAVE_QWORDS(a[383:256], b[383:256]) -dst[511:384] := INTERLEAVE_QWORDS(a[511:384], b[511:384]) -dst[MAX:512] := 0 +FOR j := 0 to 3 + i := j*64 + dst[i+63:i] := LEFT_ROTATE_QWORDS(a[i+63:i], imm8[7:0]) +ENDFOR +dst[MAX:256] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Swizzle - - - - - Unpack and interleave single-precision (32-bit) floating-point elements from the low half of each 128-bit lane in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -INTERLEAVE_DWORDS(src1[127:0], src2[127:0]){ - dst[31:0] := src1[31:0] - dst[63:32] := src2[31:0] - dst[95:64] := src1[63:32] - dst[127:96] := src2[63:32] - RETURN dst[127:0] -} - -tmp_dst[127:0] := INTERLEAVE_DWORDS(a[127:0], b[127:0]) -tmp_dst[255:128] := INTERLEAVE_DWORDS(a[255:128], b[255:128]) -tmp_dst[383:256] := INTERLEAVE_DWORDS(a[383:256], b[383:256]) -tmp_dst[511:384] := INTERLEAVE_DWORDS(a[511:384], b[511:384]) - -FOR j := 0 to 15 - i := j*32 + Shift + + + + + + Rotate the bits in each packed 64-bit integer in "a" to the left by the number of bits specified in "imm8", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +DEFINE LEFT_ROTATE_QWORDS(src, count_src) { + count := count_src % 64 + RETURN (src << count) OR (src >> (64 - count)) +} +FOR j := 0 to 1 + i := j*64 IF k[j] - dst[i+31:i] := tmp_dst[i+31:i] + dst[i+63:i] := LEFT_ROTATE_QWORDS(a[i+63:i], imm8[7:0]) ELSE - dst[i+31:i] := src[i+31:i] + dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Swizzle - - - - Unpack and interleave single-precision (32-bit) floating-point elements from the low half of each 128-bit lane in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - -INTERLEAVE_DWORDS(src1[127:0], src2[127:0]){ - dst[31:0] := src1[31:0] - dst[63:32] := src2[31:0] - dst[95:64] := src1[63:32] - dst[127:96] := src2[63:32] - RETURN dst[127:0] -} - -tmp_dst[127:0] := INTERLEAVE_DWORDS(a[127:0], b[127:0]) -tmp_dst[255:128] := INTERLEAVE_DWORDS(a[255:128], b[255:128]) -tmp_dst[383:256] := INTERLEAVE_DWORDS(a[383:256], b[383:256]) -tmp_dst[511:384] := INTERLEAVE_DWORDS(a[511:384], b[511:384]) - -FOR j := 0 to 15 - i := j*32 + Shift + + + + + Rotate the bits in each packed 64-bit integer in "a" to the left by the number of bits specified in "imm8", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + +DEFINE LEFT_ROTATE_QWORDS(src, count_src) { + count := count_src % 64 + RETURN (src << count) OR (src >> (64 - count)) +} +FOR j := 0 to 1 + i := j*64 IF k[j] - dst[i+31:i] := tmp_dst[i+31:i] + dst[i+63:i] := LEFT_ROTATE_QWORDS(a[i+63:i], imm8[7:0]) ELSE - dst[i+31:i] := 0 + dst[i+63:i] := 0 FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Swizzle - - - Unpack and interleave single-precision (32-bit) floating-point elements from the low half of each 128-bit lane in "a" and "b", and store the results in "dst". + Shift + + + + Rotate the bits in each packed 64-bit integer in "a" to the left by the number of bits specified in "imm8", and store the results in "dst". -INTERLEAVE_DWORDS(src1[127:0], src2[127:0]){ - dst[31:0] := src1[31:0] - dst[63:32] := src2[31:0] - dst[95:64] := src1[63:32] - dst[127:96] := src2[63:32] - RETURN dst[127:0] -} - -dst[127:0] := INTERLEAVE_DWORDS(a[127:0], b[127:0]) -dst[255:128] := INTERLEAVE_DWORDS(a[255:128], b[255:128]) -dst[383:256] := INTERLEAVE_DWORDS(a[383:256], b[383:256]) -dst[511:384] := INTERLEAVE_DWORDS(a[511:384], b[511:384]) -dst[MAX:512] := 0 +DEFINE LEFT_ROTATE_QWORDS(src, count_src) { + count := count_src % 64 + RETURN (src << count) OR (src >> (64 - count)) +} +FOR j := 0 to 1 + i := j*64 + dst[i+63:i] := LEFT_ROTATE_QWORDS(a[i+63:i], imm8[7:0]) +ENDFOR +dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Cast - - Cast vector of type __m128d to type __m512d; the upper 384 bits of the result are undefined. - This intrinsic is only used for compilation and does not generate any instructions, thus it has zero latency. + Shift + + + + + + Rotate the bits in each packed 32-bit integer in "a" to the left by the number of bits specified in the corresponding element of "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +DEFINE LEFT_ROTATE_DWORDS(src, count_src) { + count := count_src % 32 + RETURN (src << count) OR (src >> (32 - count)) +} +FOR j := 0 to 7 + i := j*32 + IF k[j] + dst[i+31:i] := LEFT_ROTATE_DWORDS(a[i+31:i], b[i+31:i]) + ELSE + dst[i+31:i] := src[i+31:i] + FI +ENDFOR +dst[MAX:256] := 0 + +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Cast - - Cast vector of type __m256d to type __m512d; the upper 256 bits of the result are undefined. - This intrinsic is only used for compilation and does not generate any instructions, thus it has zero latency. + Shift + + + + + Rotate the bits in each packed 32-bit integer in "a" to the left by the number of bits specified in the corresponding element of "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + +DEFINE LEFT_ROTATE_DWORDS(src, count_src) { + count := count_src % 32 + RETURN (src << count) OR (src >> (32 - count)) +} +FOR j := 0 to 7 + i := j*32 + IF k[j] + dst[i+31:i] := LEFT_ROTATE_DWORDS(a[i+31:i], b[i+31:i]) + ELSE + dst[i+31:i] := 0 + FI +ENDFOR +dst[MAX:256] := 0 + +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Cast - - Cast vector of type __m512d to type __m128d. - This intrinsic is only used for compilation and does not generate any instructions, thus it has zero latency. + Shift + + + + Rotate the bits in each packed 32-bit integer in "a" to the left by the number of bits specified in the corresponding element of "b", and store the results in "dst". + +DEFINE LEFT_ROTATE_DWORDS(src, count_src) { + count := count_src % 32 + RETURN (src << count) OR (src >> (32 - count)) +} +FOR j := 0 to 7 + i := j*32 + dst[i+31:i] := LEFT_ROTATE_DWORDS(a[i+31:i], b[i+31:i]) +ENDFOR +dst[MAX:256] := 0 + +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Cast - - Cast vector of type __m512 to type __m128. - This intrinsic is only used for compilation and does not generate any instructions, thus it has zero latency. + Shift + + + + + + Rotate the bits in each packed 32-bit integer in "a" to the left by the number of bits specified in the corresponding element of "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +DEFINE LEFT_ROTATE_DWORDS(src, count_src) { + count := count_src % 32 + RETURN (src << count) OR (src >> (32 - count)) +} +FOR j := 0 to 3 + i := j*32 + IF k[j] + dst[i+31:i] := LEFT_ROTATE_DWORDS(a[i+31:i], b[i+31:i]) + ELSE + dst[i+31:i] := src[i+31:i] + FI +ENDFOR +dst[MAX:128] := 0 + +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Cast - - Cast vector of type __m512d to type __m256d. - This intrinsic is only used for compilation and does not generate any instructions, thus it has zero latency. -
immintrin.h
- - - Floating Point - AVX512F/KNCNI - Cast - - Cast vector of type __m512d to type __m512. - This intrinsic is only used for compilation and does not generate any instructions, thus it has zero latency. + Shift + + + + + Rotate the bits in each packed 32-bit integer in "a" to the left by the number of bits specified in the corresponding element of "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + +DEFINE LEFT_ROTATE_DWORDS(src, count_src) { + count := count_src % 32 + RETURN (src << count) OR (src >> (32 - count)) +} +FOR j := 0 to 3 + i := j*32 + IF k[j] + dst[i+31:i] := LEFT_ROTATE_DWORDS(a[i+31:i], b[i+31:i]) + ELSE + dst[i+31:i] := 0 + FI +ENDFOR +dst[MAX:128] := 0 + +
immintrin.h
 - - Floating Point + Integer - AVX512F/KNCNI - Cast - - Cast vector of type __m512d to type __m512i. - This intrinsic is only used for compilation and does not generate any instructions, thus it has zero latency. -
immintrin.h
- - - Floating Point + AVX512VL AVX512F - Cast - - Cast vector of type __m128 to type __m512; the upper 384 bits of the result are undefined. - This intrinsic is only used for compilation and does not generate any instructions, thus it has zero latency. + Shift + + + + Rotate the bits in each packed 32-bit integer in "a" to the left by the number of bits specified in the corresponding element of "b", and store the results in "dst". + +DEFINE LEFT_ROTATE_DWORDS(src, count_src) { + count := count_src % 32 + RETURN (src << count) OR (src >> (32 - count)) +} +FOR j := 0 to 3 + i := j*32 + dst[i+31:i] := LEFT_ROTATE_DWORDS(a[i+31:i], b[i+31:i]) +ENDFOR +dst[MAX:128] := 0 + +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Cast - - Cast vector of type __m256 to type __m512; the upper 256 bits of the result are undefined. - This intrinsic is only used for compilation and does not generate any instructions, thus it has zero latency. + Shift + + + + + + Rotate the bits in each packed 64-bit integer in "a" to the left by the number of bits specified in the corresponding element of "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +DEFINE LEFT_ROTATE_QWORDS(src, count_src) { + count := count_src % 64 + RETURN (src << count) OR (src >> (64 - count)) +} +FOR j := 0 to 3 + i := j*64 + IF k[j] + dst[i+63:i] := LEFT_ROTATE_QWORDS(a[i+63:i], b[i+63:i]) + ELSE + dst[i+63:i] := src[i+63:i] + FI +ENDFOR +dst[MAX:256] := 0 + +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Cast - - Cast vector of type __m512 to type __m256. - This intrinsic is only used for compilation and does not generate any instructions, thus it has zero latency. -
immintrin.h
- - - Floating Point - AVX512F/KNCNI - Cast - - Cast vector of type __m512 to type __m512d. - This intrinsic is only used for compilation and does not generate any instructions, thus it has zero latency. + Shift + + + + + Rotate the bits in each packed 64-bit integer in "a" to the left by the number of bits specified in the corresponding element of "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + +DEFINE LEFT_ROTATE_QWORDS(src, count_src) { + count := count_src % 64 + RETURN (src << count) OR (src >> (64 - count)) +} +FOR j := 0 to 3 + i := j*64 + IF k[j] + dst[i+63:i] := LEFT_ROTATE_QWORDS(a[i+63:i], b[i+63:i]) + ELSE + dst[i+63:i] := 0 + FI +ENDFOR +dst[MAX:256] := 0 + +
immintrin.h
 - - Floating Point + Integer - AVX512F/KNCNI - Cast - - Cast vector of type __m512 to type __m512i. - This intrinsic is only used for compilation and does not generate any instructions, thus it has zero latency. + AVX512VL + AVX512F + Shift + + + + Rotate the bits in each packed 64-bit integer in "a" to the left by the number of bits specified in the corresponding element of "b", and store the results in "dst". + +DEFINE LEFT_ROTATE_QWORDS(src, count_src) { + count := count_src % 64 + RETURN (src << count) OR (src >> (64 - count)) +} +FOR j := 0 to 3 + i := j*64 + dst[i+63:i] := LEFT_ROTATE_QWORDS(a[i+63:i], b[i+63:i]) +ENDFOR +dst[MAX:256] := 0 + +
immintrin.h
 - + Integer + AVX512VL AVX512F - Cast - - Cast vector of type __m128i to type __m512i; the upper 384 bits of the result are undefined. - This intrinsic is only used for compilation and does not generate any instructions, thus it has zero latency. + Shift + + + + + + Rotate the bits in each packed 64-bit integer in "a" to the left by the number of bits specified in the corresponding element of "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +DEFINE LEFT_ROTATE_QWORDS(src, count_src) { + count := count_src % 64 + RETURN (src << count) OR (src >> (64 - count)) +} +FOR j := 0 to 1 + i := j*64 + IF k[j] + dst[i+63:i] := LEFT_ROTATE_QWORDS(a[i+63:i], b[i+63:i]) + ELSE + dst[i+63:i] := src[i+63:i] + FI +ENDFOR +dst[MAX:128] := 0 + +
immintrin.h
 - + Integer + AVX512VL AVX512F - Cast - - Cast vector of type __m256i to type __m512i; the upper 256 bits of the result are undefined. - This intrinsic is only used for compilation and does not generate any instructions, thus it has zero latency. -
immintrin.h
- - - Floating Point - AVX512F/KNCNI - Cast - - Cast vector of type __m512i to type __m512d. - This intrinsic is only used for compilation and does not generate any instructions, thus it has zero latency. -
immintrin.h
- - - Floating Point - AVX512F/KNCNI - Cast - - Cast vector of type __m512i to type __m512. - This intrinsic is only used for compilation and does not generate any instructions, thus it has zero latency. + Shift + + + + + Rotate the bits in each packed 64-bit integer in "a" to the left by the number of bits specified in the corresponding element of "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + +DEFINE LEFT_ROTATE_QWORDS(src, count_src) { + count := count_src % 64 + RETURN (src << count) OR (src >> (64 - count)) +} +FOR j := 0 to 1 + i := j*64 + IF k[j] + dst[i+63:i] := LEFT_ROTATE_QWORDS(a[i+63:i], b[i+63:i]) + ELSE + dst[i+63:i] := 0 + FI +ENDFOR +dst[MAX:128] := 0 + +
immintrin.h
 - + Integer + AVX512VL AVX512F - Cast - - Cast vector of type __m512i to type __m128i. - This intrinsic is only used for compilation and does not generate any instructions, thus it has zero latency. + Shift + + + + Rotate the bits in each packed 64-bit integer in "a" to the left by the number of bits specified in the corresponding element of "b", and store the results in "dst". + +DEFINE LEFT_ROTATE_QWORDS(src, count_src) { + count := count_src % 64 + RETURN (src << count) OR (src >> (64 - count)) +} +FOR j := 0 to 1 + i := j*64 + dst[i+63:i] := LEFT_ROTATE_QWORDS(a[i+63:i], b[i+63:i]) +ENDFOR +dst[MAX:128] := 0 + +
immintrin.h
 - + Integer + AVX512VL AVX512F - Cast - - Cast vector of type __m512i to type __m256i. - This intrinsic is only used for compilation and does not generate any instructions, thus it has zero latency. -
immintrin.h
- - - Floating Point - AVX512F - Cast - - Casts vector of type __m128d to type __m512d; the upper 384 bits of the result are zeroed. This intrinsic is only used for compilation and does not generate any instructions, thus it has zero latency. - -
immintrin.h
- - - Floating Point - AVX512F - Cast - - Casts vector of type __m128 to type __m512; the upper 384 bits of the result are zeroed. This intrinsic is only used for compilation and does not generate any instructions, thus it has zero latency. - + Shift + + + + + + Rotate the bits in each packed 32-bit integer in "a" to the right by the number of bits specified in "imm8", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +DEFINE RIGHT_ROTATE_DWORDS(src, count_src) { + count := count_src % 32 + RETURN (src >>count) OR (src << (32 - count)) +} +FOR j := 0 to 7 + i := j*32 + IF k[j] + dst[i+31:i] := RIGHT_ROTATE_DWORDS(a[i+31:i], imm8[7:0]) + ELSE + dst[i+31:i] := src[i+31:i] + FI +ENDFOR +dst[MAX:256] := 0 + +
immintrin.h
 - + Integer + AVX512VL AVX512F - Cast - - Casts vector of type __m128i to type __m512i; the upper 384 bits of the result are zeroed. This intrinsic is only used for compilation and does not generate any instructions, thus it has zero latency. - + Shift + + + + + Rotate the bits in each packed 32-bit integer in "a" to the right by the number of bits specified in "imm8", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + +DEFINE RIGHT_ROTATE_DWORDS(src, count_src) { + count := count_src % 32 + RETURN (src >>count) OR (src << (32 - count)) +} +FOR j := 0 to 7 + i := j*32 + IF k[j] + dst[i+31:i] := RIGHT_ROTATE_DWORDS(a[i+31:i], imm8[7:0]) + ELSE + dst[i+31:i] := 0 + FI +ENDFOR +dst[MAX:256] := 0 + +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Cast - - Casts vector of type __m256d to type __m512d; the upper 256 bits of the result are zeroed. This intrinsic is only used for compilation and does not generate any instructions, thus it has zero latency. - + Shift + + + + Rotate the bits in each packed 32-bit integer in "a" to the right by the number of bits specified in "imm8", and store the results in "dst". + +DEFINE RIGHT_ROTATE_DWORDS(src, count_src) { + count := count_src % 32 + RETURN (src >>count) OR (src << (32 - count)) +} +FOR j := 0 to 7 + i := j*32 + dst[i+31:i] := RIGHT_ROTATE_DWORDS(a[i+31:i], imm8[7:0]) +ENDFOR +dst[MAX:256] := 0 + +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Cast - - Casts vector of type __m256 to type __m512; the upper 256 bits of the result are zeroed. This intrinsic is only used for compilation and does not generate any instructions, thus it has zero latency. - + Shift + + + + + + Rotate the bits in each packed 32-bit integer in "a" to the right by the number of bits specified in "imm8", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +DEFINE RIGHT_ROTATE_DWORDS(src, count_src) { + count := count_src % 32 + RETURN (src >>count) OR (src << (32 - count)) +} +FOR j := 0 to 3 + i := j*32 + IF k[j] + dst[i+31:i] := RIGHT_ROTATE_DWORDS(a[i+31:i], imm8[7:0]) + ELSE + dst[i+31:i] := src[i+31:i] + FI +ENDFOR +dst[MAX:128] := 0 + +
immintrin.h
 - + Integer + AVX512VL AVX512F - Cast - - Casts vector of type __m256i to type __m512i; the upper 256 bits of the result are zeroed. This intrinsic is only used for compilation and does not generate any instructions, thus it has zero latency. - + Shift + + + + + Rotate the bits in each packed 32-bit integer in "a" to the right by the number of bits specified in "imm8", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + +DEFINE RIGHT_ROTATE_DWORDS(src, count_src) { + count := count_src % 32 + RETURN (src >>count) OR (src << (32 - count)) +} +FOR j := 0 to 3 + i := j*32 + IF k[j] + dst[i+31:i] := RIGHT_ROTATE_DWORDS(a[i+31:i], imm8[7:0]) + ELSE + dst[i+31:i] := 0 + FI +ENDFOR +dst[MAX:128] := 0 + +
immintrin.h
 - + Integer - AVX512F/KNCNI - Arithmetic - - - Reduce the packed 32-bit integers in "a" by addition using mask "k". Returns the sum of all active elements in "a". + AVX512VL + AVX512F + Shift + + + + Rotate the bits in each packed 32-bit integer in "a" to the right by the number of bits specified in "imm8", and store the results in "dst". -sum[31:0] := 0 -FOR j := 0 to 15 +DEFINE RIGHT_ROTATE_DWORDS(src, count_src) { + count := count_src % 32 + RETURN (src >>count) OR (src << (32 - count)) +} +FOR j := 0 to 3 i := j*32 - IF k[j] - sum[31:0] := sum[31:0] + a[i+31:i] - FI + dst[i+31:i] := RIGHT_ROTATE_DWORDS(a[i+31:i], imm8[7:0]) ENDFOR -RETURN sum[31:0] +dst[MAX:128] := 0 +
immintrin.h
 - + Integer - AVX512F/KNCNI - Arithmetic - - - Reduce the packed 64-bit integers in "a" by addition using mask "k". Returns the sum of all active elements in "a". - -sum[63:0] := 0 -FOR j := 0 to 7 + AVX512VL + AVX512F + Shift + + + + + + Rotate the bits in each packed 64-bit integer in "a" to the right by the number of bits specified in "imm8", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +DEFINE RIGHT_ROTATE_QWORDS(src, count_src) { + count := count_src % 64 + RETURN (src >> count) OR (src << (64 - count)) +} +FOR j := 0 to 3 i := j*64 IF k[j] - sum[63:0] := sum[63:0] + a[i+63:i] + dst[i+63:i] := RIGHT_ROTATE_QWORDS(a[i+63:i], imm8[7:0]) + ELSE + dst[i+63:i] := src[i+63:i] FI ENDFOR -RETURN sum[63:0] +dst[MAX:256] := 0 +
immintrin.h
 - - Floating Point - AVX512F/KNCNI - Arithmetic - - - Reduce the packed double-precision (64-bit) floating-point elements in "a" by addition using mask "k". Returns the sum of all active elements in "a". - -sum[63:0] := 0 -FOR j := 0 to 7 + + Integer + AVX512VL + AVX512F + Shift + + + + + Rotate the bits in each packed 64-bit integer in "a" to the right by the number of bits specified in "imm8", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + +DEFINE RIGHT_ROTATE_QWORDS(src, count_src) { + count := count_src % 64 + RETURN (src >> count) OR (src << (64 - count)) +} +FOR j := 0 to 3 i := j*64 IF k[j] - sum[63:0] := sum[63:0] + a[i+63:i] + dst[i+63:i] := RIGHT_ROTATE_QWORDS(a[i+63:i], imm8[7:0]) + ELSE + dst[i+63:i] := 0 FI ENDFOR -RETURN sum[63:0] +dst[MAX:256] := 0 +
immintrin.h
 - - Floating Point - AVX512F/KNCNI - Arithmetic - - - Reduce the packed single-precision (32-bit) floating-point elements in "a" by addition using mask "k". Returns the sum of all active elements in "a". + + Integer + AVX512VL + AVX512F + Shift + + + + Rotate the bits in each packed 64-bit integer in "a" to the right by the number of bits specified in "imm8", and store the results in "dst". -sum[31:0] := 0 -FOR j := 0 to 15 - i := j*32 - IF k[j] - sum[31:0] := sum[31:0] + a[i+31:i] - FI +DEFINE RIGHT_ROTATE_QWORDS(src, count_src) { + count := count_src % 64 + RETURN (src >> count) OR (src << (64 - count)) +} +FOR j := 0 to 3 + i := j*64 + dst[i+63:i] := RIGHT_ROTATE_QWORDS(a[i+63:i], imm8[7:0]) ENDFOR -RETURN sum[31:0] +dst[MAX:256] := 0 +
immintrin.h
 - + Integer - AVX512F/KNCNI - Logical - - - Reduce the packed 32-bit integers in "a" by bitwise AND using mask "k". Returns the bitwise AND of all active elements in "a". - -reduced[31:0] := 0xFFFFFFFF -FOR j := 0 to 15 - i := j*32 + AVX512VL + AVX512F + Shift + + + + + + Rotate the bits in each packed 64-bit integer in "a" to the right by the number of bits specified in "imm8", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +DEFINE RIGHT_ROTATE_QWORDS(src, count_src) { + count := count_src % 64 + RETURN (src >> count) OR (src << (64 - count)) +} +FOR j := 0 to 1 + i := j*64 IF k[j] - reduced[31:0] := reduced[31:0] AND a[i+31:i] + dst[i+63:i] := RIGHT_ROTATE_QWORDS(a[i+63:i], imm8[7:0]) + ELSE + dst[i+63:i] := src[i+63:i] FI ENDFOR -RETURN reduced[31:0] +dst[MAX:128] := 0 +
immintrin.h
 - + Integer - AVX512F/KNCNI - Logical - - - Reduce the packed 64-bit integers in "a" by bitwise AND using mask "k". Returns the bitwise AND of all active elements in "a". - -reduced[63:0] := 0xFFFFFFFFFFFFFFFF -FOR j := 0 to 7 + AVX512VL + AVX512F + Shift + + + + + Rotate the bits in each packed 64-bit integer in "a" to the right by the number of bits specified in "imm8", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + +DEFINE RIGHT_ROTATE_QWORDS(src, count_src) { + count := count_src % 64 + RETURN (src >> count) OR (src << (64 - count)) +} +FOR j := 0 to 1 i := j*64 IF k[j] - reduced[63:0] := reduced[63:0] AND a[i+63:i] + dst[i+63:i] := RIGHT_ROTATE_QWORDS(a[i+63:i], imm8[7:0]) + ELSE + dst[i+63:i] := 0 FI ENDFOR -RETURN reduced[63:0] +dst[MAX:128] := 0 +
immintrin.h
 - + Integer - AVX512F/KNCNI - Special Math Functions - - - Reduce the packed 32-bit integers in "a" by maximum using mask "k". Returns the maximum of all active elements in "a". + AVX512VL + AVX512F + Shift + + + + Rotate the bits in each packed 64-bit integer in "a" to the right by the number of bits specified in "imm8", and store the results in "dst". -max[31:0] := MIN_INT -FOR j := 0 to 15 - i := j*32 - IF k[j] - max[31:0] := MAXIMUM(max[31:0], a[i+31:i]) - FI +DEFINE RIGHT_ROTATE_QWORDS(src, count_src) { + count := count_src % 64 + RETURN (src >> count) OR (src << (64 - count)) +} +FOR j := 0 to 1 + i := j*64 + dst[i+63:i] := RIGHT_ROTATE_QWORDS(a[i+63:i], imm8[7:0]) ENDFOR -RETURN max[31:0] +dst[MAX:128] := 0 +
immintrin.h
 - + Integer - AVX512F/KNCNI - Special Math Functions - - - Reduce the packed 64-bit integers in "a" by maximum using mask "k". Returns the maximum of all active elements in "a". - -max[63:0] := MIN_INT + AVX512VL + AVX512F + Shift + + + + + + Rotate the bits in each packed 32-bit integer in "a" to the right by the number of bits specified in the corresponding element of "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +DEFINE RIGHT_ROTATE_DWORDS(src, count_src) { + count := count_src % 32 + RETURN (src >>count) OR (src << (32 - count)) +} FOR j := 0 to 7 - i := j*64 + i := j*32 IF k[j] - max[63:0] := MAXIMUM(max[63:0], a[i+63:i]) + dst[i+31:i] := RIGHT_ROTATE_DWORDS(a[i+31:i], b[i+31:i]) + ELSE + dst[i+31:i] := src[i+31:i] FI ENDFOR -RETURN max[63:0] +dst[MAX:256] := 0 +
immintrin.h
 - + Integer - AVX512F/KNCNI - Special Math Functions - - - Reduce the packed unsigned 32-bit integers in "a" by maximum using mask "k". Returns the maximum of all active elements in "a". - -max[31:0] := 0 -FOR j := 0 to 15 + AVX512VL + AVX512F + Shift + + + + + Rotate the bits in each packed 32-bit integer in "a" to the right by the number of bits specified in the corresponding element of "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + +DEFINE RIGHT_ROTATE_DWORDS(src, count_src) { + count := count_src % 32 + RETURN (src >>count) OR (src << (32 - count)) +} +FOR j := 0 to 7 i := j*32 IF k[j] - max[31:0] := MAXIMUM(max[31:0], a[i+31:i]) + dst[i+31:i] := RIGHT_ROTATE_DWORDS(a[i+31:i], b[i+31:i]) + ELSE + dst[i+31:i] := 0 FI ENDFOR -RETURN max[31:0] +dst[MAX:256] := 0 +
immintrin.h
 - + Integer - AVX512F/KNCNI - Special Math Functions - - - Reduce the packed unsigned 64-bit integers in "a" by maximum using mask "k". Returns the maximum of all active elements in "a". + AVX512VL + AVX512F + Shift + + + + Rotate the bits in each packed 32-bit integer in "a" to the right by the number of bits specified in the corresponding element of "b", and store the results in "dst". -max[63:0] := 0 +DEFINE RIGHT_ROTATE_DWORDS(src, count_src) { + count := count_src % 32 + RETURN (src >>count) OR (src << (32 - count)) +} FOR j := 0 to 7 - i := j*64 - IF k[j] - max[63:0] := MAXIMUM(max[63:0], a[i+63:i]) - FI + i := j*32 + dst[i+31:i] := RIGHT_ROTATE_DWORDS(a[i+31:i], b[i+31:i]) ENDFOR -RETURN max[63:0] +dst[MAX:256] := 0 +
immintrin.h
 - - Floating Point - AVX512F/KNCNI - Special Math Functions - - - Reduce the packed double-precision (64-bit) floating-point elements in "a" by maximum using mask "k". Returns the maximum of all active elements in "a". - -max[63:0] := MIN_DOUBLE -FOR j := 0 to 7 - i := j*64 + + Integer + AVX512VL + AVX512F + Shift + + + + + + Rotate the bits in each packed 32-bit integer in "a" to the right by the number of bits specified in the corresponding element of "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +DEFINE RIGHT_ROTATE_DWORDS(src, count_src) { + count := count_src % 32 + RETURN (src >>count) OR (src << (32 - count)) +} +FOR j := 0 to 3 + i := j*32 IF k[j] - max[63:0] := MAXIMUM(max[63:0], a[i+63:i]) + dst[i+31:i] := RIGHT_ROTATE_DWORDS(a[i+31:i], b[i+31:i]) + ELSE + dst[i+31:i] := src[i+31:i] FI ENDFOR -RETURN max[63:0] +dst[MAX:128] := 0 +
immintrin.h
 - - Floating Point - AVX512F/KNCNI - Special Math Functions - - - Reduce the packed single-precision (32-bit) floating-point elements in "a" by maximum using mask "k". Returns the maximum of all active elements in "a". - -max[31:0] := MIN_FLOAT -FOR j := 0 to 15 + + Integer + AVX512VL + AVX512F + Shift + + + + + Rotate the bits in each packed 32-bit integer in "a" to the right by the number of bits specified in the corresponding element of "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + +DEFINE RIGHT_ROTATE_DWORDS(src, count_src) { + count := count_src % 32 + RETURN (src >>count) OR (src << (32 - count)) +} +FOR j := 0 to 3 i := j*32 IF k[j] - max[31:0] := MAXIMUM(max[31:0], a[i+31:i]) + dst[i+31:i] := RIGHT_ROTATE_DWORDS(a[i+31:i], b[i+31:i]) + ELSE + dst[i+31:i] := 0 FI ENDFOR -RETURN max[31:0] +dst[MAX:128] := 0 +
immintrin.h
 - + Integer - AVX512F/KNCNI - Special Math Functions - - - Reduce the packed 32-bit integers in "a" by maximum using mask "k". Returns the minimum of all active elements in "a". + AVX512VL + AVX512F + Shift + + + + Rotate the bits in each packed 32-bit integer in "a" to the right by the number of bits specified in the corresponding element of "b", and store the results in "dst". -min[31:0] := MAX_INT -FOR j := 0 to 15 +DEFINE RIGHT_ROTATE_DWORDS(src, count_src) { + count := count_src % 32 + RETURN (src >>count) OR (src << (32 - count)) +} +FOR j := 0 to 3 i := j*32 - IF k[j] - min[31:0] := MINIMUM(min[31:0], a[i+31:i]) - FI + dst[i+31:i] := RIGHT_ROTATE_DWORDS(a[i+31:i], b[i+31:i]) ENDFOR -RETURN min[31:0] +dst[MAX:128] := 0 +
immintrin.h
 - + Integer - AVX512F/KNCNI - Special Math Functions - - - Reduce the packed 64-bit integers in "a" by maximum using mask "k". Returns the minimum of all active elements in "a". - -min[63:0] := MAX_INT -FOR j := 0 to 7 + AVX512VL + AVX512F + Shift + + + + + + Rotate the bits in each packed 64-bit integer in "a" to the right by the number of bits specified in the corresponding element of "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +DEFINE RIGHT_ROTATE_QWORDS(src, count_src) { + count := count_src % 64 + RETURN (src >> count) OR (src << (64 - count)) +} +FOR j := 0 to 3 i := j*64 IF k[j] - min[63:0] := MINIMUM(min[63:0], a[i+63:i]) + dst[i+63:i] := RIGHT_ROTATE_QWORDS(a[i+63:i], b[i+63:i]) + ELSE + dst[i+63:i] := src[i+63:i] FI ENDFOR -RETURN min[63:0] +dst[MAX:256] := 0 +
immintrin.h
 - + Integer - AVX512F/KNCNI - Special Math Functions - - - Reduce the packed unsigned 32-bit integers in "a" by maximum using mask "k". Returns the minimum of all active elements in "a". - -min[31:0] := MAX_UINT -FOR j := 0 to 15 - i := j*32 + AVX512VL + AVX512F + Shift + + + + + Rotate the bits in each packed 64-bit integer in "a" to the right by the number of bits specified in the corresponding element of "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + +DEFINE RIGHT_ROTATE_QWORDS(src, count_src) { + count := count_src % 64 + RETURN (src >> count) OR (src << (64 - count)) +} +FOR j := 0 to 3 + i := j*64 IF k[j] - min[31:0] := MINIMUM(min[31:0], a[i+31:i]) + dst[i+63:i] := RIGHT_ROTATE_QWORDS(a[i+63:i], b[i+63:i]) + ELSE + dst[i+63:i] := 0 FI ENDFOR -RETURN min[31:0] +dst[MAX:256] := 0 +
immintrin.h
 - + Integer - AVX512F/KNCNI - Special Math Functions - - - Reduce the packed unsigned 64-bit integers in "a" by minimum using mask "k". Returns the minimum of all active elements in "a". + AVX512VL + AVX512F + Shift + + + + Rotate the bits in each packed 64-bit integer in "a" to the right by the number of bits specified in the corresponding element of "b", and store the results in "dst". -min[63:0] := MAX_UINT -FOR j := 0 to 7 +DEFINE RIGHT_ROTATE_QWORDS(src, count_src) { + count := count_src % 64 + RETURN (src >> count) OR (src << (64 - count)) +} +FOR j := 0 to 3 i := j*64 - IF k[j] - min[63:0] := MINIMUM(min[63:0], a[i+63:i]) - FI + dst[i+63:i] := RIGHT_ROTATE_QWORDS(a[i+63:i], b[i+63:i]) ENDFOR -RETURN min[63:0] +dst[MAX:256] := 0 +
immintrin.h
 - - Floating Point - AVX512F/KNCNI - Special Math Functions - - - Reduce the packed double-precision (64-bit) floating-point elements in "a" by maximum using mask "k". Returns the minimum of all active elements in "a". - -min[63:0] := MAX_DOUBLE -FOR j := 0 to 7 + + Integer + AVX512VL + AVX512F + Shift + + + + + + Rotate the bits in each packed 64-bit integer in "a" to the right by the number of bits specified in the corresponding element of "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +DEFINE RIGHT_ROTATE_QWORDS(src, count_src) { + count := count_src % 64 + RETURN (src >> count) OR (src << (64 - count)) +} +FOR j := 0 to 1 i := j*64 IF k[j] - min[63:0] := MINIMUM(min[63:0], a[i+63:i]) + dst[i+63:i] := RIGHT_ROTATE_QWORDS(a[i+63:i], b[i+63:i]) + ELSE + dst[i+63:i] := src[i+63:i] FI ENDFOR -RETURN min[63:0] +dst[MAX:128] := 0 +
immintrin.h
 - - Floating Point - AVX512F/KNCNI - Special Math Functions - - - Reduce the packed single-precision (32-bit) floating-point elements in "a" by maximum using mask "k". Returns the minimum of all active elements in "a". - -min[31:0] := MAX_FLOAT -FOR j := 0 to 15 - i := j*32 + + Integer + AVX512VL + AVX512F + Shift + + + + + Rotate the bits in each packed 64-bit integer in "a" to the right by the number of bits specified in the corresponding element of "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + +DEFINE RIGHT_ROTATE_QWORDS(src, count_src) { + count := count_src % 64 + RETURN (src >> count) OR (src << (64 - count)) +} +FOR j := 0 to 1 + i := j*64 IF k[j] - min[31:0] := MINIMUM(min[31:0], a[i+31:i]) + dst[i+63:i] := RIGHT_ROTATE_QWORDS(a[i+63:i], b[i+63:i]) + ELSE + dst[i+63:i] := 0 FI ENDFOR -RETURN min[31:0] +dst[MAX:128] := 0 +
immintrin.h
 - + Integer - AVX512F/KNCNI - Arithmetic - - - Reduce the packed 32-bit integers in "a" by multiplication using mask "k". Returns the product of all active elements in "a". + AVX512VL + AVX512F + Shift + + + + Rotate the bits in each packed 64-bit integer in "a" to the right by the number of bits specified in the corresponding element of "b", and store the results in "dst". -prod[31:0] := 1 -FOR j := 0 to 15 - i := j*32 - IF k[j] - prod[31:0] := prod[31:0] * a[i+31:i] - FI +DEFINE RIGHT_ROTATE_QWORDS(src, count_src) { + count := count_src % 64 + RETURN (src >> count) OR (src << (64 - count)) +} +FOR j := 0 to 1 + i := j*64 + dst[i+63:i] := RIGHT_ROTATE_QWORDS(a[i+63:i], b[i+63:i]) ENDFOR -RETURN prod[31:0] +dst[MAX:128] := 0 +
immintrin.h
 - + Integer - AVX512F/KNCNI - Arithmetic - - - Reduce the packed 64-bit integers in "a" by multiplication using mask "k". Returns the product of all active elements in "a". + AVX512VL + AVX512F + Store + + + + + + Scatter 32-bit integers from "a" into memory using 32-bit indices. 32-bit elements are stored at addresses starting at "base_addr" and offset by each 32-bit element in "vindex" (each index is scaled by the factor in "scale"). "scale" should be 1, 2, 4 or 8. -prod[63:0] := 1 FOR j := 0 to 7 - i := j*64 - IF k[j] - prod[63:0] := prod[63:0] * a[i+63:i] - FI + i := j*32 + m := j*32 + addr := base_addr + SignExtend64(vindex[m+31:m]) * ZeroExtend64(scale) * 8 + MEM[addr+31:addr] := a[i+31:i] ENDFOR -RETURN prod[63:0] +
immintrin.h
 - - Floating Point - AVX512F/KNCNI - Arithmetic - - - Reduce the packed double-precision (64-bit) floating-point elements in "a" by multiplication using mask "k". Returns the product of all active elements in "a". + + Integer + AVX512VL + AVX512F + Store + + + + + + + Scatter 32-bit integers from "a" into memory using 32-bit indices. 32-bit elements are stored at addresses starting at "base_addr" and offset by each 32-bit element in "vindex" (each index is scaled by the factor in "scale") subject to mask "k" (elements are not stored when the corresponding mask bit is not set). "scale" should be 1, 2, 4 or 8. -prod[63:0] := 1 FOR j := 0 to 7 - i := j*64 + i := j*32 + m := j*32 IF k[j] - prod[63:0] := prod[63:0] * a[i+63:i] + addr := base_addr + SignExtend64(vindex[m+31:m]) * ZeroExtend64(scale) * 8 + MEM[addr+31:addr] := a[i+31:i] FI ENDFOR -RETURN prod[63:0] +
immintrin.h
 - - Floating Point - AVX512F/KNCNI - Arithmetic - - - Reduce the packed single-precision (32-bit) floating-point elements in "a" by multiplication using mask "k". Returns the product of all active elements in "a". + + Integer + AVX512VL + AVX512F + Store + + + + + + Scatter 32-bit integers from "a" into memory using 32-bit indices. 32-bit elements are stored at addresses starting at "base_addr" and offset by each 32-bit element in "vindex" (each index is scaled by the factor in "scale"). "scale" should be 1, 2, 4 or 8. -prod[31:0] := 1 -FOR j := 0 to 15 +FOR j := 0 to 3 i := j*32 - IF k[j] - prod[31:0] := prod[31:0] * a[i+31:i] - FI + m := j*32 + addr := base_addr + SignExtend64(vindex[m+31:m]) * ZeroExtend64(scale) * 8 + MEM[addr+31:addr] := a[i+31:i] ENDFOR -RETURN prod[31:0] +
immintrin.h
 - + Integer - AVX512F/KNCNI - Logical - - - Reduce the packed 32-bit integers in "a" by bitwise OR using mask "k". Returns the bitwise OR of all active elements in "a". + AVX512VL + AVX512F + Store + + + + + + + Scatter 32-bit integers from "a" into memory using 32-bit indices. 32-bit elements are stored at addresses starting at "base_addr" and offset by each 32-bit element in "vindex" (each index is scaled by the factor in "scale") subject to mask "k" (elements are not stored when the corresponding mask bit is not set). "scale" should be 1, 2, 4 or 8. -reduced[31:0] := 0 -FOR j := 0 to 15 +FOR j := 0 to 3 i := j*32 + m := j*32 IF k[j] - reduced[31:0] := reduced[31:0] OR a[i+31:i] + addr := base_addr + SignExtend64(vindex[m+31:m]) * ZeroExtend64(scale) * 8 + MEM[addr+31:addr] := a[i+31:i] FI ENDFOR -RETURN reduced[31:0] +
immintrin.h
 - + Integer - AVX512F/KNCNI - Logical - - - Reduce the packed 64-bit integers in "a" by bitwise OR using mask "k". Returns the bitwise OR of all active elements in "a". + AVX512VL + AVX512F + Store + + + + + + Scatter 64-bit integers from "a" into memory using 32-bit indices. 64-bit elements are stored at addresses starting at "base_addr" and offset by each 32-bit element in "vindex" (each index is scaled by the factor in "scale"). "scale" should be 1, 2, 4 or 8. -reduced[63:0] := 0 -FOR j := 0 to 7 +FOR j := 0 to 3 i := j*64 - IF k[j] - reduced[63:0] := reduced[63:0] OR a[i+63:i] - FI + m := j*32 + addr := base_addr + SignExtend64(vindex[m+31:m]) * ZeroExtend64(scale) * 8 + MEM[addr+63:addr] := a[i+63:i] ENDFOR -RETURN reduced[63:0] +
immintrin.h
 - + Integer - AVX512F/KNCNI - Arithmetic - - Reduce the packed 32-bit integers in "a" by addition. Returns the sum of all elements in "a". + AVX512VL + AVX512F + Store + + + + + + + Scatter 64-bit integers from "a" into memory using 32-bit indices. 64-bit elements are stored at addresses starting at "base_addr" and offset by each 32-bit element in "vindex" (each index is scaled by the factor in "scale") subject to mask "k" (elements are not stored when the corresponding mask bit is not set). "scale" should be 1, 2, 4 or 8. -sum[31:0] := 0 -FOR j := 0 to 15 - i := j*32 - sum[31:0] := sum[31:0] + a[i+31:i] +FOR j := 0 to 3 + i := j*64 + m := j*32 + IF k[j] + addr := base_addr + SignExtend64(vindex[m+31:m]) * ZeroExtend64(scale) * 8 + MEM[addr+63:addr] := a[i+63:i] + FI ENDFOR -RETURN sum[63:0] +
immintrin.h
 - + Integer - AVX512F/KNCNI - Arithmetic - - Reduce the packed 64-bit integers in "a" by addition. Returns the sum of all elements in "a". + AVX512VL + AVX512F + Store + + + + + + Scatter 64-bit integers from "a" into memory using 32-bit indices. 64-bit elements are stored at addresses starting at "base_addr" and offset by each 32-bit element in "vindex" (each index is scaled by the factor in "scale"). "scale" should be 1, 2, 4 or 8. -sum[63:0] := 0 -FOR j := 0 to 7 +FOR j := 0 to 1 i := j*64 - sum[63:0] := sum[63:0] + a[i+63:i] + m := j*32 + addr := base_addr + SignExtend64(vindex[m+31:m]) * ZeroExtend64(scale) * 8 + MEM[addr+63:addr] := a[i+63:i] ENDFOR -RETURN sum[63:0] +
immintrin.h
 - - Floating Point - AVX512F/KNCNI - Arithmetic - - Reduce the packed double-precision (64-bit) floating-point elements in "a" by addition. Returns the sum of all elements in "a". + + Integer + AVX512VL + AVX512F + Store + + + + + + + Scatter 64-bit integers from "a" into memory using 32-bit indices. 64-bit elements are stored at addresses starting at "base_addr" and offset by each 32-bit element in "vindex" (each index is scaled by the factor in "scale") subject to mask "k" (elements are not stored when the corresponding mask bit is not set). "scale" should be 1, 2, 4 or 8. -sum[63:0] := 0 -FOR j := 0 to 7 +FOR j := 0 to 1 i := j*64 - sum[63:0] := sum[63:0] + a[i+63:i] + m := j*32 + IF k[j] + addr := base_addr + SignExtend64(vindex[m+31:m]) * ZeroExtend64(scale) * 8 + MEM[addr+63:addr] := a[i+63:i] + FI ENDFOR -RETURN sum[63:0] +
immintrin.h
 - - Floating Point - AVX512F/KNCNI - Arithmetic - - Reduce the packed single-precision (32-bit) floating-point elements in "a" by addition. Returns the sum of all elements in "a". + + Integer + AVX512VL + AVX512F + Store + + + + + + Scatter 32-bit integers from "a" into memory using 64-bit indices. 32-bit elements are stored at addresses starting at "base_addr" and offset by each 64-bit element in "vindex" (each index is scaled by the factor in "scale"). "scale" should be 1, 2, 4 or 8. -sum[31:0] := 0 -FOR j := 0 to 15 +FOR j := 0 to 3 i := j*32 - sum[31:0] := sum[31:0] + a[i+31:i] + m := j*64 + addr := base_addr + vindex[m+63:m] * ZeroExtend64(scale) * 8 + MEM[addr+31:addr] := a[i+31:i] ENDFOR -RETURN sum[63:0] +
immintrin.h
 - + Integer - AVX512F/KNCNI - Logical - - Reduce the packed 32-bit integers in "a" by bitwise AND. Returns the bitwise AND of all elements in "a". + AVX512VL + AVX512F + Store + + + + + + + Scatter 32-bit integers from "a" into memory using 64-bit indices. 32-bit elements are stored at addresses starting at "base_addr" and offset by each 64-bit element in "vindex" (each index is scaled by the factor in "scale") subject to mask "k" (elements are not stored when the corresponding mask bit is not set). "scale" should be 1, 2, 4 or 8. -reduced[31:0] := 0xFFFFFFFF -FOR j := 0 to 15 +FOR j := 0 to 3 i := j*32 - reduced[31:0] := reduced[31:0] AND a[i+31:i] + m := j*64 + IF k[j] + addr := base_addr + vindex[m+63:m] * ZeroExtend64(scale) * 8 + MEM[addr+31:addr] := a[i+31:i] + FI ENDFOR -RETURN reduced[31:0] +
immintrin.h
 - + Integer - AVX512F/KNCNI - Logical - - Reduce the packed 64-bit integers in "a" by bitwise AND. Returns the bitwise AND of all elements in "a". + AVX512VL + AVX512F + Store + + + + + + Scatter 32-bit integers from "a" into memory using 64-bit indices. 32-bit elements are stored at addresses starting at "base_addr" and offset by each 64-bit element in "vindex" (each index is scaled by the factor in "scale"). "scale" should be 1, 2, 4 or 8. -reduced[63:0] := 0xFFFFFFFFFFFFFFFF -FOR j := 0 to 7 - i := j*64 - reduced[63:0] := reduced[63:0] AND a[i+63:i] +FOR j := 0 to 1 + i := j*32 + m := j*64 + addr := base_addr + vindex[m+63:m] * ZeroExtend64(scale) * 8 + MEM[addr+31:addr] := a[i+31:i] ENDFOR -RETURN reduced[63:0] +
immintrin.h
 - + Integer - AVX512F/KNCNI - Special Math Functions - - Reduce the packed 32-bit integers in "a" by maximum. Returns the maximum of all elements in "a". + AVX512VL + AVX512F + Store + + + + + + + Scatter 32-bit integers from "a" into memory using 64-bit indices. 32-bit elements are stored at addresses starting at "base_addr" and offset by each 64-bit element in "vindex" (each index is scaled by the factor in "scale") subject to mask "k" (elements are not stored when the corresponding mask bit is not set). "scale" should be 1, 2, 4 or 8. -max[31:0] := MIN_INT -FOR j := 0 to 15 +FOR j := 0 to 1 i := j*32 - max[31:0] := MAXIMUM(max[31:0], a[i+31:i]) + m := j*64 + IF k[j] + addr := base_addr + vindex[m+63:m] * ZeroExtend64(scale) * 8 + MEM[addr+31:addr] := a[i+31:i] + FI ENDFOR -RETURN max[31:0] +
immintrin.h
 - + Integer - AVX512F/KNCNI - Special Math Functions - - Reduce the packed 64-bit integers in "a" by maximum. Returns the maximum of all elements in "a". + AVX512VL + AVX512F + Store + + + + + + Scatter 64-bit integers from "a" into memory using 64-bit indices. 64-bit elements are stored at addresses starting at "base_addr" and offset by each 64-bit element in "vindex" (each index is scaled by the factor in "scale"). "scale" should be 1, 2, 4 or 8. -max[63:0] := MIN_INT -FOR j := 0 to 7 +FOR j := 0 to 3 i := j*64 - max[63:0] := MAXIMUM(max[63:0], a[i+63:i]) + m := j*64 + addr := base_addr + vindex[m+63:m] * ZeroExtend64(scale) * 8 + MEM[addr+63:addr] := a[i+63:i] ENDFOR -RETURN max[63:0] +
immintrin.h
 - + Integer - AVX512F/KNCNI - Special Math Functions - - Reduce the packed unsigned 32-bit integers in "a" by maximum. Returns the maximum of all elements in "a". + AVX512VL + AVX512F + Store + + + + + + + Scatter 64-bit integers from "a" into memory using 64-bit indices. 64-bit elements are stored at addresses starting at "base_addr" and offset by each 64-bit element in "vindex" (each index is scaled by the factor in "scale") subject to mask "k" (elements are not stored when the corresponding mask bit is not set). "scale" should be 1, 2, 4 or 8. -max[31:0] := 0 -FOR j := 0 to 15 - i := j*32 - max[31:0] := MAXIMUM(max[31:0], a[i+31:i]) +FOR j := 0 to 3 + i := j*64 + m := j*64 + IF k[j] + addr := base_addr + vindex[m+63:m] * ZeroExtend64(scale) * 8 + MEM[addr+63:addr] := a[i+63:i] + FI ENDFOR -RETURN max[31:0] +
immintrin.h
 - + Integer - AVX512F/KNCNI - Special Math Functions - - Reduce the packed unsigned 64-bit integers in "a" by maximum. Returns the maximum of all elements in "a". + AVX512VL + AVX512F + Store + + + + + + Scatter 64-bit integers from "a" into memory using 64-bit indices. 64-bit elements are stored at addresses starting at "base_addr" and offset by each 64-bit element in "vindex" (each index is scaled by the factor in "scale"). "scale" should be 1, 2, 4 or 8. -max[63:0] := 0 -FOR j := 0 to 7 +FOR j := 0 to 1 i := j*64 - max[63:0] := MAXIMUM(max[63:0], a[i+63:i]) + m := j*64 + addr := base_addr + vindex[m+63:m] * ZeroExtend64(scale) * 8 + MEM[addr+63:addr] := a[i+63:i] ENDFOR -RETURN max[63:0] +
immintrin.h
 - - Floating Point - AVX512F/KNCNI - Special Math Functions - - Reduce the packed double-precision (64-bit) floating-point elements in "a" by maximum. Returns the maximum of all elements in "a". + + Integer + AVX512VL + AVX512F + Store + + + + + + + Scatter 64-bit integers from "a" into memory using 64-bit indices. 64-bit elements are stored at addresses starting at "base_addr" and offset by each 64-bit element in "vindex" (each index is scaled by the factor in "scale") subject to mask "k" (elements are not stored when the corresponding mask bit is not set). "scale" should be 1, 2, 4 or 8. -max[63:0] := MIN_DOUBLE -FOR j := 0 to 7 +FOR j := 0 to 1 i := j*64 - max[63:0] := MAXIMUM(max[63:0], a[i+63:i]) + m := j*64 + IF k[j] + addr := base_addr + vindex[m+63:m] * ZeroExtend64(scale) * 8 + MEM[addr+63:addr] := a[i+63:i] + FI ENDFOR -RETURN max[63:0] +
immintrin.h
 - - Floating Point - AVX512F/KNCNI - Special Math Functions - - Reduce the packed single-precision (32-bit) floating-point elements in "a" by maximum. Returns the maximum of all elements in "a". + + Integer + AVX512VL + AVX512F + Miscellaneous + + + + + + Shuffle 32-bit integers in "a" within 128-bit lanes using the control in "imm8", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -max[31:0] := MIN_FLOAT -FOR j := 0 to 15 +DEFINE SELECT4(src, control) { + CASE(control[1:0]) OF + 0: tmp[31:0] := src[31:0] + 1: tmp[31:0] := src[63:32] + 2: tmp[31:0] := src[95:64] + 3: tmp[31:0] := src[127:96] + ESAC + RETURN tmp[31:0] +} +tmp_dst[31:0] := SELECT4(a[127:0], imm8[1:0]) +tmp_dst[63:32] := SELECT4(a[127:0], imm8[3:2]) +tmp_dst[95:64] := SELECT4(a[127:0], imm8[5:4]) +tmp_dst[127:96] := SELECT4(a[127:0], imm8[7:6]) +tmp_dst[159:128] := SELECT4(a[255:128], imm8[1:0]) +tmp_dst[191:160] := SELECT4(a[255:128], imm8[3:2]) +tmp_dst[223:192] := SELECT4(a[255:128], imm8[5:4]) +tmp_dst[255:224] := SELECT4(a[255:128], imm8[7:6]) +FOR j := 0 to 7 i := j*32 - max[31:0] := MAXIMUM(max[31:0], a[i+31:i]) + IF k[j] + dst[i+31:i] := tmp_dst[i+31:i] + ELSE + dst[i+31:i] := src[i+31:i] + FI ENDFOR -RETURN max[31:0] +dst[MAX:256] := 0 +
immintrin.h
 - + Integer - AVX512F/KNCNI - Special Math Functions - - Reduce the packed 32-bit integers in "a" by minimum. Returns the minimum of all elements in "a". + AVX512VL + AVX512F + Miscellaneous + + + + + Shuffle 32-bit integers in "a" within 128-bit lanes using the control in "imm8", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -min[31:0] := MAX_INT -FOR j := 0 to 15 +DEFINE SELECT4(src, control) { + CASE(control[1:0]) OF + 0: tmp[31:0] := src[31:0] + 1: tmp[31:0] := src[63:32] + 2: tmp[31:0] := src[95:64] + 3: tmp[31:0] := src[127:96] + ESAC + RETURN tmp[31:0] +} +tmp_dst[31:0] := SELECT4(a[127:0], imm8[1:0]) +tmp_dst[63:32] := SELECT4(a[127:0], imm8[3:2]) +tmp_dst[95:64] := SELECT4(a[127:0], imm8[5:4]) +tmp_dst[127:96] := SELECT4(a[127:0], imm8[7:6]) +tmp_dst[159:128] := SELECT4(a[255:128], imm8[1:0]) +tmp_dst[191:160] := SELECT4(a[255:128], imm8[3:2]) +tmp_dst[223:192] := SELECT4(a[255:128], imm8[5:4]) +tmp_dst[255:224] := SELECT4(a[255:128], imm8[7:6]) +FOR j := 0 to 7 i := j*32 - min[31:0] := MINIMUM(min[31:0], a[i+31:i]) + IF k[j] + dst[i+31:i] := tmp_dst[i+31:i] + ELSE + dst[i+31:i] := 0 + FI ENDFOR -RETURN min[31:0] +dst[MAX:256] := 0 +
immintrin.h
 - + Integer - AVX512F/KNCNI - Special Math Functions - - Reduce the packed 64-bit integers in "a" by minimum. Returns the minimum of all elements in "a". + AVX512VL + AVX512F + Miscellaneous + + + + + + Shuffle 32-bit integers in "a" using the control in "imm8", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -min[63:0] := MAX_INT -FOR j := 0 to 7 - i := j*64 - min[63:0] := MINIMUM(min[63:0], a[i+63:i]) +DEFINE SELECT4(src, control) { + CASE(control[1:0]) OF + 0: tmp[31:0] := src[31:0] + 1: tmp[31:0] := src[63:32] + 2: tmp[31:0] := src[95:64] + 3: tmp[31:0] := src[127:96] + ESAC + RETURN tmp[31:0] +} +tmp_dst[31:0] := SELECT4(a[127:0], imm8[1:0]) +tmp_dst[63:32] := SELECT4(a[127:0], imm8[3:2]) +tmp_dst[95:64] := SELECT4(a[127:0], imm8[5:4]) +tmp_dst[127:96] := SELECT4(a[127:0], imm8[7:6]) +FOR j := 0 to 3 + i := j*32 + IF k[j] + dst[i+31:i] := tmp_dst[i+31:i] + ELSE + dst[i+31:i] := src[i+31:i] + FI ENDFOR -RETURN min[63:0] +dst[MAX:128] := 0 +
immintrin.h
 - + Integer - AVX512F/KNCNI - Special Math Functions - - Reduce the packed unsigned 32-bit integers in "a" by minimum. Returns the minimum of all elements in "a". + AVX512VL + AVX512F + Miscellaneous + + + + + Shuffle 32-bit integers in "a" using the control in "imm8", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -min[31:0] := MAX_UINT -FOR j := 0 to 15 +DEFINE SELECT4(src, control) { + CASE(control[1:0]) OF + 0: tmp[31:0] := src[31:0] + 1: tmp[31:0] := src[63:32] + 2: tmp[31:0] := src[95:64] + 3: tmp[31:0] := src[127:96] + ESAC + RETURN tmp[31:0] +} +tmp_dst[31:0] := SELECT4(a[127:0], imm8[1:0]) +tmp_dst[63:32] := SELECT4(a[127:0], imm8[3:2]) +tmp_dst[95:64] := SELECT4(a[127:0], imm8[5:4]) +tmp_dst[127:96] := SELECT4(a[127:0], imm8[7:6]) +FOR j := 0 to 3 i := j*32 - min[31:0] := MINIMUM(min[31:0], a[i+31:i]) + IF k[j] + dst[i+31:i] := tmp_dst[i+31:i] + ELSE + dst[i+31:i] := 0 + FI ENDFOR -RETURN min[31:0] +dst[MAX:128] := 0 +
immintrin.h
 - + Integer - AVX512F/KNCNI - Special Math Functions - - Reduce the packed unsigned 64-bit integers in "a" by minimum. Returns the minimum of all elements in "a". + AVX512VL + AVX512F + Shift + + + + + + Shift packed 32-bit integers in "a" left by "count" while shifting in zeros, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -min[63:0] := MAX_UINT FOR j := 0 to 7 - i := j*64 - min[63:0] := MINIMUM(min[63:0], a[i+63:i]) + i := j*32 + IF k[j] + IF count[63:0] > 31 + dst[i+31:i] := 0 + ELSE + dst[i+31:i] := ZeroExtend32(a[i+31:i] << count[63:0]) + FI + ELSE + dst[i+31:i] := src[i+31:i] + FI ENDFOR -RETURN min[63:0] +dst[MAX:256] := 0 +
immintrin.h
 - - Floating Point - AVX512F/KNCNI - Special Math Functions - - Reduce the packed double-precision (64-bit) floating-point elements in "a" by minimum. Returns the minimum of all elements in "a". + + Integer + AVX512VL + AVX512F + Shift + + + + + + Shift packed 32-bit integers in "a" left by "imm8" while shifting in zeros, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -min[63:0] := MAX_DOUBLE FOR j := 0 to 7 - i := j*64 - min[63:0] := MINIMUM(min[63:0], a[i+63:i]) + i := j*32 + IF k[j] + IF imm8[7:0] > 31 + dst[i+31:i] := 0 + ELSE + dst[i+31:i] := ZeroExtend32(a[i+31:i] << imm8[7:0]) + FI + ELSE + dst[i+31:i] := src[i+31:i] + FI ENDFOR -RETURN min[63:0] +dst[MAX:256] := 0 +
immintrin.h
 - - Floating Point - AVX512F/KNCNI - Special Math Functions - - Reduce the packed single-precision (32-bit) floating-point elements in "a" by minimum. Returns the minimum of all elements in "a". + + Integer + AVX512VL + AVX512F + Shift + + + + + Shift packed 32-bit integers in "a" left by "count" while shifting in zeros, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -min[31:0] := MAX_INT -FOR j := 0 to 15 +FOR j := 0 to 7 i := j*32 - min[31:0] := MINIMUM(min[31:0], a[i+31:i]) + IF k[j] + IF count[63:0] > 31 + dst[i+31:i] := 0 + ELSE + dst[i+31:i] := ZeroExtend32(a[i+31:i] << count[63:0]) + FI + ELSE + dst[i+31:i] := 0 + FI ENDFOR -RETURN min[31:0] +dst[MAX:256] := 0 +
immintrin.h
 - + Integer - AVX512F/KNCNI - Arithmetic - - Reduce the packed 32-bit integers in "a" by multiplication. Returns the product of all elements in "a". + AVX512VL + AVX512F + Shift + + + + + Shift packed 32-bit integers in "a" left by "imm8" while shifting in zeros, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -prod[31:0] := 1 -FOR j := 0 to 15 +FOR j := 0 to 7 i := j*32 - prod[31:0] := prod[31:0] * a[i+31:i] + IF k[j] + IF imm8[7:0] > 31 + dst[i+31:i] := 0 + ELSE + dst[i+31:i] := ZeroExtend32(a[i+31:i] << imm8[7:0]) + FI + ELSE + dst[i+31:i] := 0 + FI ENDFOR -RETURN prod[31:0] +dst[MAX:256] := 0 +
immintrin.h
 - + Integer - AVX512F/KNCNI - Arithmetic - - Reduce the packed 64-bit integers in "a" by multiplication. Returns the product of all elements in "a". + AVX512VL + AVX512F + Shift + + + + + + Shift packed 32-bit integers in "a" left by "count" while shifting in zeros, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -prod[63:0] := 1 -FOR j := 0 to 7 - i := j*64 - prod[63:0] := prod[63:0] * a[i+63:i] +FOR j := 0 to 3 + i := j*32 + IF k[j] + IF count[63:0] > 31 + dst[i+31:i] := 0 + ELSE + dst[i+31:i] := ZeroExtend32(a[i+31:i] << count[63:0]) + FI + ELSE + dst[i+31:i] := src[i+31:i] + FI ENDFOR -RETURN prod[63:0] +dst[MAX:128] := 0 +
immintrin.h
 - - Floating Point - AVX512F/KNCNI - Arithmetic - - Reduce the packed double-precision (64-bit) floating-point elements in "a" by multiplication. Returns the product of all elements in "a". + + Integer + AVX512VL + AVX512F + Shift + + + + + + Shift packed 32-bit integers in "a" left by "imm8" while shifting in zeros, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -prod[63:0] := 1 -FOR j := 0 to 7 - i := j*64 - prod[63:0] := prod[63:0] * a[i+63:i] +FOR j := 0 to 3 + i := j*32 + IF k[j] + IF imm8[7:0] > 31 + dst[i+31:i] := 0 + ELSE + dst[i+31:i] := ZeroExtend32(a[i+31:i] << imm8[7:0]) + FI + ELSE + dst[i+31:i] := src[i+31:i] + FI ENDFOR -RETURN prod[63:0] +dst[MAX:128] := 0 +
immintrin.h
 - - Floating Point - AVX512F/KNCNI - Arithmetic - - Reduce the packed single-precision (32-bit) floating-point elements in "a" by multiplication. Returns the product of all elements in "a". + + Integer + AVX512VL + AVX512F + Shift + + + + + Shift packed 32-bit integers in "a" left by "count" while shifting in zeros, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -prod[31:0] := 1 -FOR j := 0 to 15 +FOR j := 0 to 3 i := j*32 - prod[31:0] := prod[31:0] * a[i+31:i] + IF k[j] + IF count[63:0] > 31 + dst[i+31:i] := 0 + ELSE + dst[i+31:i] := ZeroExtend32(a[i+31:i] << count[63:0]) + FI + ELSE + dst[i+31:i] := 0 + FI ENDFOR -RETURN prod[31:0] +dst[MAX:128] := 0 +
immintrin.h
 - + Integer - AVX512F/KNCNI - Logical - - Reduce the packed 32-bit integers in "a" by bitwise OR. Returns the bitwise OR of all elements in "a". + AVX512VL + AVX512F + Shift + + + + + Shift packed 32-bit integers in "a" left by "imm8" while shifting in zeros, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -reduced[31:0] := 0 -FOR j := 0 to 15 +FOR j := 0 to 3 i := j*32 - reduced[31:0] := reduced[31:0] OR a[i+31:i] + IF k[j] + IF imm8[7:0] > 31 + dst[i+31:i] := 0 + ELSE + dst[i+31:i] := ZeroExtend32(a[i+31:i] << imm8[7:0]) + FI + ELSE + dst[i+31:i] := 0 + FI ENDFOR -RETURN reduced[31:0] +dst[MAX:128] := 0 +
immintrin.h
 - + Integer - AVX512F/KNCNI - Logical - - Reduce the packed 64-bit integers in "a" by bitwise OR. Returns the bitwise OR of all elements in "a". + AVX512VL + AVX512F + Shift + + + + + + Shift packed 64-bit integers in "a" left by "count" while shifting in zeros, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -reduced[63:0] := 0 -FOR j := 0 to 7 +FOR j := 0 to 3 i := j*64 - reduced[63:0] := reduced[63:0] OR a[i+63:i] + IF k[j] + IF count[63:0] > 63 + dst[i+63:i] := 0 + ELSE + dst[i+63:i] := ZeroExtend64(a[i+63:i] << count[63:0]) + FI + ELSE + dst[i+63:i] := src[i+63:i] + FI ENDFOR -RETURN reduced[63:0] +dst[MAX:256] := 0 +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Set - - Broadcast double-precision (64-bit) floating-point value "a" to all elements of "dst". + Shift + + + + + + Shift packed 64-bit integers in "a" left by "imm8" while shifting in zeros, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 7 +FOR j := 0 to 3 i := j*64 - dst[i+63:i] := a[63:0] + IF k[j] + IF imm8[7:0] > 63 + dst[i+63:i] := 0 + ELSE + dst[i+63:i] := ZeroExtend64(a[i+63:i] << imm8[7:0]) + FI + ELSE + dst[i+63:i] := src[i+63:i] + FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Set - - Broadcast single-precision (32-bit) floating-point value "a" to all elements of "dst". + Shift + + + + + Shift packed 64-bit integers in "a" left by "count" while shifting in zeros, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*32 - dst[i+31:i] := a[31:0] +FOR j := 0 to 3 + i := j*64 + IF k[j] + IF count[63:0] > 63 + dst[i+63:i] := 0 + ELSE + dst[i+63:i] := ZeroExtend64(a[i+63:i] << count[63:0]) + FI + ELSE + dst[i+63:i] := 0 + FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 +
immintrin.h
 - + Integer + AVX512VL AVX512F - Set - - - - - Set packed 32-bit integers in "dst" with the repeated 4 element sequence. + Shift + + + + + Shift packed 64-bit integers in "a" left by "imm8" while shifting in zeros, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -dst[31:0] := d -dst[63:32] := c -dst[95:64] := b -dst[127:96] := a -dst[159:128] := d -dst[191:160] := c -dst[223:192] := b -dst[255:224] := a -dst[287:256] := d -dst[319:288] := c -dst[351:320] := b -dst[383:352] := a -dst[415:384] := d -dst[447:416] := c -dst[479:448] := b -dst[511:480] := a -dst[MAX:512] := 0 +FOR j := 0 to 3 + i := j*64 + IF k[j] + IF imm8[7:0] > 63 + dst[i+63:i] := 0 + ELSE + dst[i+63:i] := ZeroExtend64(a[i+63:i] << imm8[7:0]) + FI + ELSE + dst[i+63:i] := 0 + FI +ENDFOR +dst[MAX:256] := 0 +
immintrin.h
 - + Integer + AVX512VL AVX512F - Set - - - - - Set packed 64-bit integers in "dst" with the repeated 4 element sequence. + Shift + + + + + + Shift packed 64-bit integers in "a" left by "count" while shifting in zeros, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -dst[63:0] := d -dst[127:64] := c -dst[191:128] := b -dst[255:192] := a -dst[319:256] := d -dst[383:320] := c -dst[447:384] := b -dst[511:448] := a -dst[MAX:512] := 0 +FOR j := 0 to 1 + i := j*64 + IF k[j] + IF count[63:0] > 63 + dst[i+63:i] := 0 + ELSE + dst[i+63:i] := ZeroExtend64(a[i+63:i] << count[63:0]) + FI + ELSE + dst[i+63:i] := src[i+63:i] + FI +ENDFOR +dst[MAX:128] := 0 +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Set - - - - - Set packed double-precision (64-bit) floating-point elements in "dst" with the repeated 4 element sequence. + Shift + + + + + + Shift packed 64-bit integers in "a" left by "imm8" while shifting in zeros, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -dst[63:0] := d -dst[127:64] := c -dst[191:128] := b -dst[255:192] := a -dst[319:256] := d -dst[383:320] := c -dst[447:384] := b -dst[511:448] := a -dst[MAX:512] := 0 +FOR j := 0 to 1 + i := j*64 + IF k[j] + IF imm8[7:0] > 63 + dst[i+63:i] := 0 + ELSE + dst[i+63:i] := ZeroExtend64(a[i+63:i] << imm8[7:0]) + FI + ELSE + dst[i+63:i] := src[i+63:i] + FI +ENDFOR +dst[MAX:128] := 0 +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Set - - - - - Set packed single-precision (32-bit) floating-point elements in "dst" with the repeated 4 element sequence. + Shift + + + + + Shift packed 64-bit integers in "a" left by "count" while shifting in zeros, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -dst[31:0] := d -dst[63:32] := c -dst[95:64] := b -dst[127:96] := a -dst[159:128] := d -dst[191:160] := c -dst[223:192] := b -dst[255:224] := a -dst[287:256] := d -dst[319:288] := c -dst[351:320] := b -dst[383:352] := a -dst[415:384] := d -dst[447:416] := c -dst[479:448] := b -dst[511:480] := a -dst[MAX:512] := 0 +FOR j := 0 to 1 + i := j*64 + IF k[j] + IF count[63:0] > 63 + dst[i+63:i] := 0 + ELSE + dst[i+63:i] := ZeroExtend64(a[i+63:i] << count[63:0]) + FI + ELSE + dst[i+63:i] := 0 + FI +ENDFOR +dst[MAX:128] := 0 +
immintrin.h
 - + Integer + AVX512VL AVX512F - Set - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - Set packed 8-bit integers in "dst" with the supplied values in reverse order. + Shift + + + + + Shift packed 64-bit integers in "a" left by "imm8" while shifting in zeros, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -dst[7:0] := e0 -dst[15:8] := e1 -dst[23:16] := e2 -dst[31:24] := e3 -dst[39:32] := e4 -dst[47:40] := e5 -dst[55:48] := e6 -dst[63:56] := e7 -dst[71:64] := e8 -dst[79:72] := e9 -dst[87:80] := e10 -dst[95:88] := e11 -dst[103:96] := e12 -dst[111:104] := e13 -dst[119:112] := e14 -dst[127:120] := e15 -dst[135:128] := e16 -dst[143:136] := e17 -dst[151:144] := e18 -dst[159:152] := e19 -dst[167:160] := e20 -dst[175:168] := e21 -dst[183:176] := e22 -dst[191:184] := e23 -dst[199:192] := e24 -dst[207:200] := e25 -dst[215:208] := e26 -dst[223:216] := e27 -dst[231:224] := e28 -dst[239:232] := e29 -dst[247:240] := e30 -dst[255:248] := e31 -dst[263:256] := e32 -dst[271:264] := e33 -dst[279:272] := e34 -dst[287:280] := e35 -dst[295:288] := e36 -dst[303:296] := e37 -dst[311:304] := e38 -dst[319:312] := e39 -dst[327:320] := e40 -dst[335:328] := e41 -dst[343:336] := e42 -dst[351:344] := e43 -dst[359:352] := e44 -dst[367:360] := e45 -dst[375:368] := e46 -dst[383:376] := e47 -dst[391:384] := e48 -dst[399:392] := e49 -dst[407:400] := e50 -dst[415:408] := e51 -dst[423:416] := e52 -dst[431:424] := e53 -dst[439:432] := e54 -dst[447:440] := e55 -dst[455:448] := e56 -dst[463:456] := e57 -dst[471:464] := e58 -dst[479:472] := e59 -dst[487:480] := e60 -dst[495:488] := e61 -dst[503:496] := e62 -dst[511:503] := e63 -dst[MAX:512] := 0 +FOR j := 0 to 1 + i := j*64 + IF k[j] + IF imm8[7:0] > 63 + dst[i+63:i] := 0 + ELSE + dst[i+63:i] := ZeroExtend64(a[i+63:i] << imm8[7:0]) + FI + ELSE + dst[i+63:i] := 0 + FI +ENDFOR +dst[MAX:128] := 0 +
immintrin.h
 - + Integer + AVX512VL AVX512F - Set - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - Set packed 16-bit integers in "dst" with the supplied values in reverse order. + Shift + + + + + + Shift packed 32-bit integers in "a" left by the amount specified by the corresponding element in "count" while shifting in zeros, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -dst[15:0] := e0 -dst[31:16] := e1 -dst[47:32] := e2 -dst[63:48] := e3 -dst[79:64] := e4 -dst[95:80] := e5 -dst[111:96] := e6 -dst[127:112] := e7 -dst[145:128] := e8 -dst[159:144] := e9 -dst[175:160] := e10 -dst[191:176] := e11 -dst[207:192] := e12 -dst[223:208] := e13 -dst[239:224] := e14 -dst[255:240] := e15 -dst[271:256] := e16 -dst[287:272] := e17 -dst[303:288] := e18 -dst[319:304] := e19 -dst[335:320] := e20 -dst[351:336] := e21 -dst[367:352] := e22 -dst[383:368] := e23 -dst[399:384] := e24 -dst[415:400] := e25 -dst[431:416] := e26 -dst[447:432] := e27 -dst[463:448] := e28 -dst[479:464] := e29 -dst[495:480] := e30 -dst[511:496] := e31 -dst[MAX:512] := 0 +FOR j := 0 to 7 + i := j*32 + IF k[j] + IF count[i+31:i] < 32 + dst[i+31:i] := ZeroExtend32(a[i+31:i] << count[i+31:i]) + ELSE + dst[i+31:i] := 0 + FI + ELSE + dst[i+31:i] := src[i+31:i] + FI +ENDFOR +dst[MAX:256] := 0 +
immintrin.h
 - + Integer + AVX512VL AVX512F - Set - - - - - - - - - - - - - - - - - Set packed 32-bit integers in "dst" with the supplied values. + Shift + + + + + Shift packed 32-bit integers in "a" left by the amount specified by the corresponding element in "count" while shifting in zeros, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -dst[31:0] := e0 -dst[63:32] := e1 -dst[95:64] := e2 -dst[127:96] := e3 -dst[159:128] := e4 -dst[191:160] := e5 -dst[223:192] := e6 -dst[255:224] := e7 -dst[287:256] := e8 -dst[319:288] := e9 -dst[351:320] := e10 -dst[383:352] := e11 -dst[415:384] := e12 -dst[447:416] := e13 -dst[479:448] := e14 -dst[511:480] := e15 -dst[MAX:512] := 0 +FOR j := 0 to 7 + i := j*32 + IF k[j] + IF count[i+31:i] < 32 + dst[i+31:i] := ZeroExtend32(a[i+31:i] << count[i+31:i]) + ELSE + dst[i+31:i] := 0 + FI + ELSE + dst[i+31:i] := 0 + FI +ENDFOR +dst[MAX:256] := 0 +
immintrin.h
 - + Integer + AVX512VL AVX512F - Set - - - - - - - - - Set packed 64-bit integers in "dst" with the supplied values. - -dst[63:0] := e0 -dst[127:64] := e1 -dst[191:128] := e2 -dst[255:192] := e3 -dst[319:256] := e4 -dst[383:320] := e5 -dst[447:384] := e6 -dst[511:448] := e7 -dst[MAX:512] := 0 - -
immintrin.h
- - - Floating Point - AVX512F - Set - - - - - - - - - Set packed double-precision (64-bit) floating-point elements in "dst" with the supplied values. + Shift + + + + + + Shift packed 32-bit integers in "a" left by the amount specified by the corresponding element in "count" while shifting in zeros, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -dst[63:0] := e0 -dst[127:64] := e1 -dst[191:128] := e2 -dst[255:192] := e3 -dst[319:256] := e4 -dst[383:320] := e5 -dst[447:384] := e6 -dst[511:448] := e7 -dst[MAX:512] := 0 +FOR j := 0 to 3 + i := j*32 + IF k[j] + IF count[i+31:i] < 32 + dst[i+31:i] := ZeroExtend32(a[i+31:i] << count[i+31:i]) + ELSE + dst[i+31:i] := 0 + FI + ELSE + dst[i+31:i] := src[i+31:i] + FI +ENDFOR +dst[MAX:128] := 0 +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Set - - - - - - - - - - - - - - - - - Set packed single-precision (32-bit) floating-point elements in "dst" with the supplied values. + Shift + + + + + Shift packed 32-bit integers in "a" left by the amount specified by the corresponding element in "count" while shifting in zeros, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -dst[31:0] := e0 -dst[63:32] := e1 -dst[95:64] := e2 -dst[127:96] := e3 -dst[159:128] := e4 -dst[191:160] := e5 -dst[223:192] := e6 -dst[255:224] := e7 -dst[287:256] := e8 -dst[319:288] := e9 -dst[351:320] := e10 -dst[383:352] := e11 -dst[415:384] := e12 -dst[447:416] := e13 -dst[479:448] := e14 -dst[511:480] := e15 -dst[MAX:512] := 0 +FOR j := 0 to 3 + i := j*32 + IF k[j] + IF count[i+31:i] < 32 + dst[i+31:i] := ZeroExtend32(a[i+31:i] << count[i+31:i]) + ELSE + dst[i+31:i] := 0 + FI + ELSE + dst[i+31:i] := 0 + FI +ENDFOR +dst[MAX:128] := 0 +
immintrin.h
 - + Integer + AVX512VL AVX512F - Set - - - - - Set packed 32-bit integers in "dst" with the repeated 4 element sequence in reverse order. + Shift + + + + + + Shift packed 64-bit integers in "a" left by the amount specified by the corresponding element in "count" while shifting in zeros, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -dst[31:0] := a -dst[63:32] := b -dst[95:64] := c -dst[127:96] := d -dst[159:128] := a -dst[191:160] := b -dst[223:192] := c -dst[255:224] := d -dst[287:256] := a -dst[319:288] := b -dst[351:320] := c -dst[383:352] := d -dst[415:384] := a -dst[447:416] := b -dst[479:448] := c -dst[511:480] := d -dst[MAX:512] := 0 +FOR j := 0 to 3 + i := j*64 + IF k[j] + IF count[i+63:i] < 64 + dst[i+63:i] := ZeroExtend64(a[i+63:i] << count[i+63:i]) + ELSE + dst[i+63:i] := 0 + FI + ELSE + dst[i+63:i] := src[i+63:i] + FI +ENDFOR +dst[MAX:256] := 0 +
immintrin.h
 - + Integer + AVX512VL AVX512F - Set - - - - - Set packed 64-bit integers in "dst" with the repeated 4 element sequence in reverse order. + Shift + + + + + Shift packed 64-bit integers in "a" left by the amount specified by the corresponding element in "count" while shifting in zeros, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -dst[63:0] := a -dst[127:64] := b -dst[191:128] := c -dst[255:192] := d -dst[319:256] := a -dst[383:320] := b -dst[447:384] := c -dst[511:448] := d -dst[MAX:512] := 0 +FOR j := 0 to 3 + i := j*64 + IF k[j] + IF count[i+63:i] < 64 + dst[i+63:i] := ZeroExtend64(a[i+63:i] << count[i+63:i]) + ELSE + dst[i+63:i] := 0 + FI + ELSE + dst[i+63:i] := 0 + FI +ENDFOR +dst[MAX:256] := 0 +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Set - - - - - Set packed double-precision (64-bit) floating-point elements in "dst" with the repeated 4 element sequence in reverse order. + Shift + + + + + + Shift packed 64-bit integers in "a" left by the amount specified by the corresponding element in "count" while shifting in zeros, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -dst[63:0] := a -dst[127:64] := b -dst[191:128] := c -dst[255:192] := d -dst[319:256] := a -dst[383:320] := b -dst[447:384] := c -dst[511:448] := d -dst[MAX:512] := 0 +FOR j := 0 to 1 + i := j*64 + IF k[j] + IF count[i+63:i] < 64 + dst[i+63:i] := ZeroExtend64(a[i+63:i] << count[i+63:i]) + ELSE + dst[i+63:i] := 0 + FI + ELSE + dst[i+63:i] := src[i+63:i] + FI +ENDFOR +dst[MAX:128] := 0 +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Set - - - - - Set packed single-precision (32-bit) floating-point elements in "dst" with the repeated 4 element sequence in reverse order. + Shift + + + + + Shift packed 64-bit integers in "a" left by the amount specified by the corresponding element in "count" while shifting in zeros, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -dst[31:0] := a -dst[63:32] := b -dst[95:64] := c -dst[127:96] := d -dst[159:128] := a -dst[191:160] := b -dst[223:192] := c -dst[255:224] := d -dst[287:256] := a -dst[319:288] := b -dst[351:320] := c -dst[383:352] := d -dst[415:384] := a -dst[447:416] := b -dst[479:448] := c -dst[511:480] := d -dst[MAX:512] := 0 +FOR j := 0 to 1 + i := j*64 + IF k[j] + IF count[i+63:i] < 64 + dst[i+63:i] := ZeroExtend64(a[i+63:i] << count[i+63:i]) + ELSE + dst[i+63:i] := 0 + FI + ELSE + dst[i+63:i] := 0 + FI +ENDFOR +dst[MAX:128] := 0 +
immintrin.h
 - + Integer + AVX512VL AVX512F - Set - - - - - - - - - - - - - - - - - Set packed 32-bit integers in "dst" with the supplied values in reverse order. + Shift + + + + + + Shift packed 32-bit integers in "a" right by "count" while shifting in sign bits, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -dst[31:0] := e15 -dst[63:32] := e14 -dst[95:64] := e13 -dst[127:96] := e12 -dst[159:128] := e11 -dst[191:160] := e10 -dst[223:192] := e9 -dst[255:224] := e8 -dst[287:256] := e7 -dst[319:288] := e6 -dst[351:320] := e5 -dst[383:352] := e4 -dst[415:384] := e3 -dst[447:416] := e2 -dst[479:448] := e1 -dst[511:480] := e0 -dst[MAX:512] := 0 +FOR j := 0 to 7 + i := j*32 + IF k[j] + IF count[63:0] > 31 + dst[i+31:i] := (a[i+31] ? 0xFFFFFFFF : 0x0) + ELSE + dst[i+31:i] := SignExtend32(a[i+31:i] >> count[63:0]) + FI + ELSE + dst[i+31:i] := src[i+31:i] + FI +ENDFOR +dst[MAX:256] := 0 +
immintrin.h
 - + Integer + AVX512VL AVX512F - Set - - - - - - - - - Set packed 64-bit integers in "dst" with the supplied values in reverse order. + Shift + + + + + + Shift packed 32-bit integers in "a" right by "imm8" while shifting in sign bits, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -dst[63:0] := e7 -dst[127:64] := e6 -dst[191:128] := e5 -dst[255:192] := e4 -dst[319:256] := e3 -dst[383:320] := e2 -dst[447:384] := e1 -dst[511:448] := e0 -dst[MAX:512] := 0 +FOR j := 0 to 7 + i := j*32 + IF k[j] + IF imm8[7:0] > 31 + dst[i+31:i] := (a[i+31] ? 0xFFFFFFFF : 0x0) + ELSE + dst[i+31:i] := SignExtend32(a[i+31:i] >> imm8[7:0]) + FI + ELSE + dst[i+31:i] := src[i+31:i] + FI +ENDFOR +dst[MAX:256] := 0 +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Set - - - - - - - - - Set packed double-precision (64-bit) floating-point elements in "dst" with the supplied values in reverse order. + Shift + + + + + Shift packed 32-bit integers in "a" right by "count" while shifting in sign bits, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -dst[63:0] := e7 -dst[127:64] := e6 -dst[191:128] := e5 -dst[255:192] := e4 -dst[319:256] := e3 -dst[383:320] := e2 -dst[447:384] := e1 -dst[511:448] := e0 -dst[MAX:512] := 0 +FOR j := 0 to 7 + i := j*32 + IF k[j] + IF count[63:0] > 31 + dst[i+31:i] := (a[i+31] ? 0xFFFFFFFF : 0x0) + ELSE + dst[i+31:i] := SignExtend32(a[i+31:i] >> count[63:0]) + FI + ELSE + dst[i+31:i] := 0 + FI +ENDFOR +dst[MAX:256] := 0 +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Set - - - - - - - - - - - - - - - - - Set packed single-precision (32-bit) floating-point elements in "dst" with the supplied values in reverse order. + Shift + + + + + Shift packed 32-bit integers in "a" right by "imm8" while shifting in sign bits, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -dst[31:0] := e15 -dst[63:32] := e14 -dst[95:64] := e13 -dst[127:96] := e12 -dst[159:128] := e11 -dst[191:160] := e10 -dst[223:192] := e9 -dst[255:224] := e8 -dst[287:256] := e7 -dst[319:288] := e6 -dst[351:320] := e5 -dst[383:352] := e4 -dst[415:384] := e3 -dst[447:416] := e2 -dst[479:448] := e1 -dst[511:480] := e0 -dst[MAX:512] := 0 +FOR j := 0 to 7 + i := j*32 + IF k[j] + IF imm8[7:0] > 31 + dst[i+31:i] := (a[i+31] ? 0xFFFFFFFF : 0x0) + ELSE + dst[i+31:i] := SignExtend32(a[i+31:i] >> imm8[7:0]) + FI + ELSE + dst[i+31:i] := 0 + FI +ENDFOR +dst[MAX:256] := 0 +
immintrin.h
 - + + Integer + AVX512VL AVX512F - Set - - Return vector of type __m512 with all elements set to zero. + Shift + + + + + + Shift packed 32-bit integers in "a" right by "count" while shifting in sign bits, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -dst[MAX:0] := 0 +FOR j := 0 to 3 + i := j*32 + IF k[j] + IF count[63:0] > 31 + dst[i+31:i] := (a[i+31] ? 0xFFFFFFFF : 0x0) + ELSE + dst[i+31:i] := SignExtend32(a[i+31:i] >> count[63:0]) + FI + ELSE + dst[i+31:i] := src[i+31:i] + FI +ENDFOR +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer + AVX512VL AVX512F - Set - Return vector of type __m512i with all elements set to zero. + Shift + + + + + + Shift packed 32-bit integers in "a" right by "imm8" while shifting in sign bits, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -dst[MAX:0] := 0 +FOR j := 0 to 3 + i := j*32 + IF k[j] + IF imm8[7:0] > 31 + dst[i+31:i] := (a[i+31] ? 0xFFFFFFFF : 0x0) + ELSE + dst[i+31:i] := SignExtend32(a[i+31:i] >> imm8[7:0]) + FI + ELSE + dst[i+31:i] := src[i+31:i] + FI +ENDFOR +dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Set - Return vector of type __m512d with all elements set to zero. + Shift + + + + + Shift packed 32-bit integers in "a" right by "count" while shifting in sign bits, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -dst[MAX:0] := 0 +FOR j := 0 to 3 + i := j*32 + IF k[j] + IF count[63:0] > 31 + dst[i+31:i] := (a[i+31] ? 0xFFFFFFFF : 0x0) + ELSE + dst[i+31:i] := SignExtend32(a[i+31:i] >> count[63:0]) + FI + ELSE + dst[i+31:i] := 0 + FI +ENDFOR +dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512VL AVX512F - Set - Return vector of type __m512 with all elements set to zero. + Shift + + + + + Shift packed 32-bit integers in "a" right by "imm8" while shifting in sign bits, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -dst[MAX:0] := 0 +FOR j := 0 to 3 + i := j*32 + IF k[j] + IF imm8[7:0] > 31 + dst[i+31:i] := (a[i+31] ? 0xFFFFFFFF : 0x0) + ELSE + dst[i+31:i] := SignExtend32(a[i+31:i] >> imm8[7:0]) + FI + ELSE + dst[i+31:i] := 0 + FI +ENDFOR +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer + AVX512VL AVX512F - Set - Return vector of type __m512i with all elements set to zero. + Shift + + + + + + Shift packed 64-bit integers in "a" right by "count" while shifting in sign bits, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -dst[MAX:0] := 0 +FOR j := 0 to 3 + i := j*64 + IF k[j] + IF count[63:0] > 63 + dst[i+63:i] := (a[i+63] ? 0xFFFFFFFFFFFFFFFF : 0x0) + ELSE + dst[i+63:i] := SignExtend64(a[i+63:i] >> count[63:0]) + FI + ELSE + dst[i+63:i] := src[i+63:i] + FI +ENDFOR +dst[MAX:256] := 0 - -
immintrin.h
- - - AVX512F - General Support - - Return vector of type __m512 with undefined elements. +
immintrin.h
 - + Integer + AVX512VL AVX512F - General Support - Return vector of type __m512i with undefined elements. -
immintrin.h
- - - Floating Point - AVX512F - General Support - Return vector of type __m512d with undefined elements. -
immintrin.h
- - - Floating Point - AVX512F - General Support - Return vector of type __m512 with undefined elements. -
immintrin.h
- - - Floating Point - AVX512PF - Load - - - - - Prefetch single-precision (32-bit) floating-point elements from memory using 64-bit indices. 32-bit elements are loaded from addresses starting at "base_addr" and offset by each 64-bit element in "vindex" (each index is scaled by the factor in "scale"). Gathered elements are merged in cache. "scale" should be 1, 2, 4 or 8. The "hint" parameter may be 1 (_MM_HINT_T0) for prefetching to L1 cache, or 2 (_MM_HINT_T1) for prefetching to L2 cache. + Shift + + + + + + Shift packed 64-bit integers in "a" right by "imm8" while shifting in sign bits, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j:= 0 to 7 - i := j*64; - Prefetch([base_addr + SignExtend(vindex[i+63:i]) * scale], hint, RFO=0); -ENDFOR; +FOR j := 0 to 3 + i := j*64 + IF k[j] + IF imm8[7:0] > 63 + dst[i+63:i] := (a[i+63] ? 0xFFFFFFFFFFFFFFFF : 0x0) + ELSE + dst[i+63:i] := SignExtend64(a[i+63:i] >> imm8[7:0]) + FI + ELSE + dst[i+63:i] := src[i+63:i] + FI +ENDFOR +dst[MAX:256] := 0 - - +
immintrin.h
 - - Floating Point - AVX512PF - Load - - - - - - Prefetch single-precision (32-bit) floating-point elements from memory using 64-bit indices. 32-bit elements are loaded from addresses starting at "base_addr" and offset by each 64-bit element in "vindex" (each index is scaled by the factor in "scale"). Gathered elements are merged in cache using writemask "k" (elements are only brought into cache when their corresponding mask bit is set). "scale" should be 1, 2, 4 or 8.. The "hint" parameter may be 1 (_MM_HINT_T0) for prefetching to L1 cache, or 2 (_MM_HINT_T1) for prefetching to L2 cache. + + Integer + AVX512VL + AVX512F + Shift + + + + + Shift packed 64-bit integers in "a" right by "count" while shifting in sign bits, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j:= 0 to 7 - i := j*64; - IF mask[j] THEN - Prefetch([base_addr + SignExtend(vindex[i+63:i]) * scale], hint, RFO=0); +FOR j := 0 to 3 + i := j*64 + IF k[j] + IF count[63:0] > 63 + dst[i+63:i] := (a[i+63] ? 0xFFFFFFFFFFFFFFFF : 0x0) + ELSE + dst[i+63:i] := SignExtend64(a[i+63:i] >> count[63:0]) + FI + ELSE + dst[i+63:i] := 0 FI -ENDFOR; +ENDFOR +dst[MAX:256] := 0 - - +
immintrin.h
 - - Floating Point - AVX512PF - Store - - - - - Prefetch single-precision (32-bit) floating-point elements with intent to write into memory using 64-bit indices. Elements are prefetched into cache level "hint", where "hint" is 0 or 1. 32-bit elements are stored at addresses starting at "base_addr" and offset by each 64-bit element in "vindex" (each index is scaled by the factor in "scale"). "scale" should be 1, 2, 4 or 8. + + Integer + AVX512VL + AVX512F + Shift + + + + + Shift packed 64-bit integers in "a" right by "imm8" while shifting in sign bits, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := j*64; - Prefetch([base_addr + SignExtend(vindex[i+63:i]) * scale], Level=hint, RFO=1); -ENDFOR; +FOR j := 0 to 3 + i := j*64 + IF k[j] + IF imm8[7:0] > 63 + dst[i+63:i] := (a[i+63] ? 0xFFFFFFFFFFFFFFFF : 0x0) + ELSE + dst[i+63:i] := SignExtend64(a[i+63:i] >> imm8[7:0]) + FI + ELSE + dst[i+63:i] := 0 + FI +ENDFOR +dst[MAX:256] := 0 - - +
immintrin.h
 - - Floating Point - AVX512PF - Store - - - - - - Prefetch single-precision (32-bit) floating-point elements with intent to write into memory using 64-bit indices. The "hint" parameter may be 1 (_MM_HINT_T0) for prefetching to L1 cache, or 2 (_MM_HINT_T1) for prefetching to L2 cache. 32-bit elements are stored at addresses starting at "base_addr" and offset by each 64-bit element in "vindex" (each index is scaled by the factor in "scale") subject to mask "k" (elements are not brought into cache when the corresponding mask bit is not set). "scale" should be 1, 2, 4 or 8. + + Integer + AVX512VL + AVX512F + Shift + + + + Shift packed 64-bit integers in "a" right by "count" while shifting in sign bits, and store the results in "dst". -FOR j := 0 to 7 - i := j*64; - IF mask[j] THEN - Prefetch([base_addr + SignExtend(vindex[i+63:i]) * scale], Level=hint, RFO=1); +FOR j := 0 to 3 + i := j*64 + IF count[63:0] > 63 + dst[i+63:i] := (a[i+63] ? 0xFFFFFFFFFFFFFFFF : 0x0) + ELSE + dst[i+63:i] := SignExtend64(a[i+63:i] >> count[63:0]) FI -ENDFOR; +ENDFOR +dst[MAX:256] := 0 - - +
immintrin.h
 - - Floating Point - AVX512PF - Load - - - - - Prefetch double-precision (64-bit) floating-point elements from memory using 32-bit indices. 64-bit elements are loaded from addresses starting at "base_addr" and offset by each 32-bit element in "vindex" (each index is scaled by the factor in "scale"). Gathered elements are merged in cache. "scale" should be 1, 2, 4 or 8. The "hint" parameter may be 1 (_MM_HINT_T0) for prefetching to L1 cache, or 2 (_MM_HINT_T1) for prefetching to L2 cache. + + Integer + AVX512VL + AVX512F + Shift + + + + Shift packed 64-bit integers in "a" right by "imm8" while shifting in sign bits, and store the results in "dst". -FOR j := 0 to 7 - i := j*32; - Prefetch([base_addr + SignExtend(vindex[i*31:i]) * scale], hint, RFO=0); -ENDFOR; +FOR j := 0 to 3 + i := j*64 + IF imm8[7:0] > 63 + dst[i+63:i] := (a[i+63] ? 0xFFFFFFFFFFFFFFFF : 0x0) + ELSE + dst[i+63:i] := SignExtend64(a[i+63:i] >> imm8[7:0]) + FI +ENDFOR +dst[MAX:256] := 0 - - +
immintrin.h
 - - Floating Point - AVX512PF - Load - - - - - - Prefetch double-precision (64-bit) floating-point elements from memory using 32-bit indices. 64-bit elements are loaded from addresses starting at "base_addr" and offset by each 32-bit element in "vindex" (each index is scaled by the factor in "scale"). Gathered elements are merged in cache using writemask "k" (elements are brought into cache only when their corresponding mask bits are set). "scale" should be 1, 2, 4 or 8. The "hint" parameter may be 1 (_MM_HINT_T0) for prefetching to L1 cache, or 2 (_MM_HINT_T1) for prefetching to L2 cache. + + Integer + AVX512VL + AVX512F + Shift + + + + + + Shift packed 64-bit integers in "a" right by "count" while shifting in sign bits, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := j*32; - IF mask[j] THEN - Prefetch([base_addr + SignExtend(vindex[i*31:i]) * scale], hint, RFO=0); +FOR j := 0 to 1 + i := j*64 + IF k[j] + IF count[63:0] > 63 + dst[i+63:i] := (a[i+63] ? 0xFFFFFFFFFFFFFFFF : 0x0) + ELSE + dst[i+63:i] := SignExtend64(a[i+63:i] >> count[63:0]) + FI + ELSE + dst[i+63:i] := src[i+63:i] FI -ENDFOR; +ENDFOR +dst[MAX:128] := 0 - - +
immintrin.h
 - - Floating Point - AVX512PF/KNCNI - Load - - - - - - Prefetch single-precision (32-bit) floating-point elements from memory using 32-bit indices. 32-bit elements are loaded from addresses starting at "base_addr" and offset by each 32-bit element in "vindex" (each index is scaled by the factor in "scale"). Gathered elements are merged in cache using writemask "k" (elements are brought into cache only when their corresponding mask bits are set). "scale" should be 1, 2, 4 or 8. The "hint" parameter may be 1 (_MM_HINT_T0) for prefetching to L1 cache, or 2 (_MM_HINT_T1) for prefetching to L2 cache. + + Integer + AVX512VL + AVX512F + Shift + + + + + + Shift packed 64-bit integers in "a" right by "imm8" while shifting in sign bits, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*16; - IF mask[j] THEN - Prefetch([base_addr + SignExtend(vindex[i*31:i]) * scale], hint, RFO=0); +FOR j := 0 to 1 + i := j*64 + IF k[j] + IF imm8[7:0] > 63 + dst[i+63:i] := (a[i+63] ? 0xFFFFFFFFFFFFFFFF : 0x0) + ELSE + dst[i+63:i] := SignExtend64(a[i+63:i] >> imm8[7:0]) + FI + ELSE + dst[i+63:i] := src[i+63:i] FI -ENDFOR; +ENDFOR +dst[MAX:128] := 0 - - +
immintrin.h
 - - Floating Point - AVX512PF - Store - - - - - Prefetch double-precision (64-bit) floating-point elements with intent to write using 32-bit indices. The "hint" parameter may be 1 (_MM_HINT_T0) for prefetching to L1 cache, or 2 (_MM_HINT_T1) for prefetching to L2 cache. 64-bit elements are brought into cache from addresses starting at "base_addr" and offset by each 32-bit element in "vindex" (each index is scaled by the factor in "scale"). "scale" should be 1, 2, 4 or 8. + + Integer + AVX512VL + AVX512F + Shift + + + + + Shift packed 64-bit integers in "a" right by "count" while shifting in sign bits, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 TO 7 - i := j*32; - Prefetch(base_addr + SignExtend(vindex[i+31:i]) * scale], Level=hint, RFO=1); -ENDFOR; +FOR j := 0 to 1 + i := j*64 + IF k[j] + IF count[63:0] > 63 + dst[i+63:i] := (a[i+63] ? 0xFFFFFFFFFFFFFFFF : 0x0) + ELSE + dst[i+63:i] := SignExtend64(a[i+63:i] >> count[63:0]) + FI + ELSE + dst[i+63:i] := 0 + FI +ENDFOR +dst[MAX:128] := 0 - - +
immintrin.h
 - - Floating Point - AVX512PF - Store - - - - - - Prefetch double-precision (64-bit) floating-point elements with intent to write using 32-bit indices. The "hint" parameter may be 1 (_MM_HINT_T0) for prefetching to L1 cache, or 2 (_MM_HINT_T1) for prefetching to L2 cache. 64-bit elements are brought into cache from addresses starting at "base_addr" and offset by each 32-bit element in "vindex" (each index is scaled by the factor in "scale") subject to mask "k" (elements are not brought into cache when the corresponding mask bit is not set). "scale" should be 1, 2, 4 or 8. + + Integer + AVX512VL + AVX512F + Shift + + + + + Shift packed 64-bit integers in "a" right by "imm8" while shifting in sign bits, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 TO 7 - i := j*32; - IF mask[j] THEN - Prefetch(base_addr + SignExtend(vindex[i+31:i]) * scale], Level=hint, RFO=1); +FOR j := 0 to 1 + i := j*64 + IF k[j] + IF imm8[7:0] > 63 + dst[i+63:i] := (a[i+63] ? 0xFFFFFFFFFFFFFFFF : 0x0) + ELSE + dst[i+63:i] := SignExtend64(a[i+63:i] >> imm8[7:0]) + FI + ELSE + dst[i+63:i] := 0 FI -ENDFOR; +ENDFOR +dst[MAX:128] := 0 - - +
immintrin.h
 - - Floating Point - AVX512PF - Load - - - - - Prefetch double-precision (64-bit) floating-point elements from memory into cache level specified by "hint" using 64-bit indices. 64-bit elements are loaded from addresses starting at "base_addr" and offset by each 64-bit element in "vindex" (each index is scaled by the factor in "scale"). "scale" should be 1, 2, 4 or 8. The "hint" parameter may be 1 (_MM_HINT_T0) for prefetching to L1 cache, or 2 (_MM_HINT_T1) for prefetching to L2 cache. + + Integer + AVX512VL + AVX512F + Shift + + + + Shift packed 64-bit integers in "a" right by "count" while shifting in sign bits, and store the results in "dst". -FOR j := 0 to 7 - i := j*64; - Prefetch([base_addr + SignExtend(vindex[i*63:i] * scale]), Level=hint, RFO=0); -ENDFOR; +FOR j := 0 to 1 + i := j*64 + IF count[63:0] > 63 + dst[i+63:i] := (a[i+63] ? 0xFFFFFFFFFFFFFFFF : 0x0) + ELSE + dst[i+63:i] := SignExtend64(a[i+63:i] >> count[63:0]) + FI +ENDFOR +dst[MAX:128] := 0 - - +
immintrin.h
 - - Floating Point - AVX512PF - Load - - - - - - Prefetch double-precision (64-bit) floating-point elements from memory into cache level specified by "hint" using 64-bit indices. 64-bit elements are loaded from addresses starting at "base_addr" and offset by each 64-bit element in "vindex" (each index is scaled by the factor in "scale"). Prefetched elements are merged in cache using writemask "k" (elements are copied from memory when the corresponding mask bit is set). "scale" should be 1, 2, 4 or 8. The "hint" parameter may be 1 (_MM_HINT_T0) for prefetching to L1 cache, or 2 (_MM_HINT_T1) for prefetching to L2 cache. + + Integer + AVX512VL + AVX512F + Shift + + + + Shift packed 64-bit integers in "a" right by "imm8" while shifting in sign bits, and store the results in "dst". -FOR j := 0 to 7 - i := j*64; - IF mask[j] THEN - Prefetch([base_addr + SignExtend(vindex[i*63:i] * scale]), Level=hint, RFO=0); +FOR j := 0 to 1 + i := j*64 + IF imm8[7:0] > 63 + dst[i+63:i] := (a[i+63] ? 0xFFFFFFFFFFFFFFFF : 0x0) + ELSE + dst[i+63:i] := SignExtend64(a[i+63:i] >> imm8[7:0]) FI -ENDFOR; +ENDFOR +dst[MAX:128] := 0 - - +
immintrin.h
 - - Floating Point - AVX512PF - Store - - - - - Prefetch double-precision (64-bit) floating-point elements with intent to write into memory using 64-bit indices. The "hint" parameter may be 1 (_MM_HINT_T0) for prefetching to L1 cache, or 2 (_MM_HINT_T1) for prefetching to L2 cache. 64-bit elements are brought into cache from addresses starting at "base_addr" and offset by each 64-bit element in "vindex" (each index is scaled by the factor in "scale"). "scale" should be 1, 2, 4 or 8. + + Integer + AVX512VL + AVX512F + Shift + + + + + + Shift packed 32-bit integers in "a" right by the amount specified by the corresponding element in "count" while shifting in sign bits, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). FOR j := 0 to 7 - i := j*64; - Prefetch([base_addr + SignExtend(vindex[i+63:i]) * scale], Level=hint, RFO=1); -ENDFOR; + i := j*32 + IF k[j] + IF count[i+31:i] < 32 + dst[i+31:i] := SignExtend32(a[i+31:i] >> count[i+31:i]) + ELSE + dst[i+31:i] := (a[i+31] ? 0xFFFFFFFF : 0) + FI + ELSE + dst[i+31:i] := src[i+31:i] + FI +ENDFOR +dst[MAX:256] := 0 - - +
immintrin.h
 - - Floating Point - AVX512PF - Store - - - - - - Prefetch double-precision (64-bit) floating-point elements with intent to write into memory using 64-bit indices. The "hint" parameter may be 1 (_MM_HINT_T0) for prefetching to L1 cache, or 2 (_MM_HINT_T1) for prefetching to L2 cache. 64-bit elements are brought into cache from addresses starting at "base_addr" and offset by each 64-bit element in "vindex" (each index is scaled by the factor in "scale") subject to mask "k" (elements are not brought into cache when the corresponding mask bit is not set). "scale" should be 1, 2, 4 or 8. + + Integer + AVX512VL + AVX512F + Shift + + + + + Shift packed 32-bit integers in "a" right by the amount specified by the corresponding element in "count" while shifting in sign bits, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). FOR j := 0 to 7 - i := j*64; - IF mask[j] THEN - Prefetch([base_addr + SignExtend(vindex[i+63:i]) * scale], Level=hint, RFO=1); - FI -ENDFOR; + i := j*32 + IF k[j] + IF count[i+31:i] < 32 + dst[i+31:i] := SignExtend32(a[i+31:i] >> count[i+31:i]) + ELSE + dst[i+31:i] := (a[i+31] ? 0xFFFFFFFF : 0) + FI + ELSE + dst[i+31:i] := 0 + FI +ENDFOR +dst[MAX:256] := 0 - - +
immintrin.h
 - - Floating Point - AVX512ER - Elementary Math Functions - - - Compute the approximate exponential value of 2 raised to the power of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst". The maximum relative error for this approximation is less than 2^-23. [round_note] + + Integer + AVX512VL + AVX512F + Shift + + + + + + Shift packed 32-bit integers in "a" right by the amount specified by the corresponding element in "count" while shifting in sign bits, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*32; - dst[i+31:i] := EXP_2_23_SP(a[i+31:i]); -ENDFOR; -dst[MAX:512] := 0 +FOR j := 0 to 3 + i := j*32 + IF k[j] + IF count[i+31:i] < 32 + dst[i+31:i] := SignExtend32(a[i+31:i] >> count[i+31:i]) + ELSE + dst[i+31:i] := (a[i+31] ? 0xFFFFFFFF : 0) + FI + ELSE + dst[i+31:i] := src[i+31:i] + FI +ENDFOR +dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point - AVX512ER - Elementary Math Functions - - Compute the approximate exponential value of 2 raised to the power of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst". The maximum relative error for this approximation is less than 2^-23. + + Integer + AVX512VL + AVX512F + Shift + + + + + Shift packed 32-bit integers in "a" right by the amount specified by the corresponding element in "count" while shifting in sign bits, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*32; - dst[i+31:i] := EXP_2_23_SP(a[i+31:i]); -ENDFOR; -dst[MAX:512] := 0 +FOR j := 0 to 3 + i := j*32 + IF k[j] + IF count[i+31:i] < 32 + dst[i+31:i] := SignExtend32(a[i+31:i] >> count[i+31:i]) + ELSE + dst[i+31:i] := (a[i+31] ? 0xFFFFFFFF : 0) + FI + ELSE + dst[i+31:i] := 0 + FI +ENDFOR +dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point - AVX512ER - Elementary Math Functions - - - - - Compute the approximate exponential value of 2 raised to the power of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). The maximum relative error for this approximation is less than 2^-23. [round_note] + + Integer + AVX512VL + AVX512F + Shift + + + + + + Shift packed 64-bit integers in "a" right by the amount specified by the corresponding element in "count" while shifting in sign bits, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*32; - IF k[j] THEN - dst[i+31:i] := EXP_2_23_SP(a[i+31:i]); +FOR j := 0 to 3 + i := j*64 + IF k[j] + IF count[i+63:i] < 64 + dst[i+63:i] := SignExtend64(a[i+63:i] >> count[i+63:i]) + ELSE + dst[i+63:i] := (a[i+63] ? 0xFFFFFFFFFFFFFFFF : 0) + FI ELSE - dst[i*31:i] := src[i*31:i]; - FI -ENDFOR; -dst[MAX:512] := 0 + dst[i+63:i] := src[i+63:i] + FI +ENDFOR +dst[MAX:256] := 0 - +
immintrin.h
 - - Floating Point - AVX512ER - Elementary Math Functions - - - - Compute the approximate exponential value of 2 raised to the power of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). The maximum relative error for this approximation is less than 2^-23. + + Integer + AVX512VL + AVX512F + Shift + + + + + Shift packed 64-bit integers in "a" right by the amount specified by the corresponding element in "count" while shifting in sign bits, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*32; - IF k[j] THEN - dst[i+31:i] := EXP_2_23_SP(a[i+31:i]); +FOR j := 0 to 3 + i := j*64 + IF k[j] + IF count[i+63:i] < 64 + dst[i+63:i] := SignExtend64(a[i+63:i] >> count[i+63:i]) + ELSE + dst[i+63:i] := (a[i+63] ? 0xFFFFFFFFFFFFFFFF : 0) + FI ELSE - dst[i*31:i] := src[i*31:i]; - FI -ENDFOR; -dst[MAX:512] := 0 + dst[i+63:i] := 0 + FI +ENDFOR +dst[MAX:256] := 0 - +
immintrin.h
 - - Floating Point -AVX512ER - Elementary Math Functions - - - - Compute the approximate exponential value of 2 raised to the power of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). The maximum relative error for this approximation is less than 2^-23. [round_note] + + Integer + AVX512VL + AVX512F + Shift + + + + Shift packed 64-bit integers in "a" right by the amount specified by the corresponding element in "count" while shifting in sign bits, and store the results in "dst". -FOR j := 0 to 15 - i := j*32; - IF k[j] THEN - dst[i+31:i] := EXP_2_23_SP(a[i+31:i]); +FOR j := 0 to 3 + i := j*64 + IF count[i+63:i] < 64 + dst[i+63:i] := SignExtend64(a[i+63:i] >> count[i+63:i]) ELSE - dst[i*31:i] := 0; + dst[i+63:i] := (a[i+63] ? 0xFFFFFFFFFFFFFFFF : 0) FI -ENDFOR; -dst[MAX:512] := 0 +ENDFOR +dst[MAX:256] := 0 - +
immintrin.h
 - - Floating Point - AVX512ER - Elementary Math Functions - - - Compute the approximate exponential value of 2 raised to the power of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). The maximum relative error for this approximation is less than 2^-23. + + Integer + AVX512VL + AVX512F + Shift + + + + + + Shift packed 64-bit integers in "a" right by the amount specified by the corresponding element in "count" while shifting in sign bits, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*32; - IF k[j] THEN - dst[i+31:i] := EXP_2_23_SP(a[i+31:i]); +FOR j := 0 to 1 + i := j*64 + IF k[j] + IF count[i+63:i] < 64 + dst[i+63:i] := SignExtend64(a[i+63:i] >> count[i+63:i]) + ELSE + dst[i+63:i] := (a[i+63] ? 0xFFFFFFFFFFFFFFFF : 0) + FI ELSE - dst[i*31:i] := 0; - FI -ENDFOR; -dst[MAX:512] := 0 + dst[i+63:i] := src[i+63:i] + FI +ENDFOR +dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point - AVX512ER - Elementary Math Functions - - - Compute the approximate exponential value of 2 raised to the power of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst". The maximum relative error for this approximation is less than 2^-23. [round_note] + + Integer + AVX512VL + AVX512F + Shift + + + + + Shift packed 64-bit integers in "a" right by the amount specified by the corresponding element in "count" while shifting in sign bits, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := j*64; - dst[i+63:i] := EXP_2_23_DP(a[i+63:i]); -ENDFOR; -dst[MAX:512] := 0 +FOR j := 0 to 1 + i := j*64 + IF k[j] + IF count[i+63:i] < 64 + dst[i+63:i] := SignExtend64(a[i+63:i] >> count[i+63:i]) + ELSE + dst[i+63:i] := (a[i+63] ? 0xFFFFFFFFFFFFFFFF : 0) + FI + ELSE + dst[i+63:i] := 0 + FI +ENDFOR +dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point - AVX512ER - Elementary Math Functions - - Compute the approximate exponential value of 2 raised to the power of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst". The maximum relative error for this approximation is less than 2^-23. + + Integer + AVX512VL + AVX512F + Shift + + + + Shift packed 64-bit integers in "a" right by the amount specified by the corresponding element in "count" while shifting in sign bits, and store the results in "dst". -FOR j := 0 to 7 - i := j*64; - dst[i+63:i] := EXP_2_23_DP(a[i+63:i]); -ENDFOR; -dst[MAX:512] := 0 +FOR j := 0 to 1 + i := j*64 + IF count[i+63:i] < 64 + dst[i+63:i] := SignExtend64(a[i+63:i] >> count[i+63:i]) + ELSE + dst[i+63:i] := (a[i+63] ? 0xFFFFFFFFFFFFFFFF : 0) + FI +ENDFOR +dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point - AVX512ER - Elementary Math Functions - - - - - Compute the approximate exponential value of 2 raised to the power of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). The maximum relative error for this approximation is less than 2^-23. [round_note] + + Integer + AVX512VL + AVX512F + Shift + + + + + + Shift packed 32-bit integers in "a" right by "count" while shifting in zeros, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). FOR j := 0 to 7 - i := j*64; - IF k[j] THEN - dst[i+63:i] := EXP_2_23_DP(a[i+63:i]); + i := j*32 + IF k[j] + IF count[63:0] > 31 + dst[i+31:i] := 0 + ELSE + dst[i+31:i] := ZeroExtend32(a[i+31:i] >> count[63:0]) + FI ELSE - dst[i+63:i] := src[i+63:i]; + dst[i+31:i] := src[i+31:i] FI -ENDFOR; -dst[MAX:512] := 0 +ENDFOR +dst[MAX:256] := 0 - +
immintrin.h
 - - Floating Point - AVX512ER - Elementary Math Functions - - - - Compute the approximate exponential value of 2 raised to the power of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). The maximum relative error for this approximation is less than 2^-23. + + Integer + AVX512VL + AVX512F + Shift + + + + + + Shift packed 32-bit integers in "a" right by "imm8" while shifting in zeros, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). FOR j := 0 to 7 - i := j*64; - IF k[j] THEN - dst[i+63:i] := EXP_2_23_DP(a[i+63:i]); + i := j*32 + IF k[j] + IF imm8[7:0] > 31 + dst[i+31:i] := 0 + ELSE + dst[i+31:i] := ZeroExtend32(a[i+31:i] >> imm8[7:0]) + FI ELSE - dst[i+63:i] := src[i+63:i]; + dst[i+31:i] := src[i+31:i] FI -ENDFOR; -dst[MAX:512] := 0 +ENDFOR +dst[MAX:256] := 0 - +
immintrin.h
 - - Floating Point - AVX512ER - Elementary Math Functions - - - - Compute the approximate exponential value of 2 raised to the power of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). The maximum relative error for this approximation is less than 2^-23. [round_note] + + Integer + AVX512VL + AVX512F + Shift + + + + + Shift packed 32-bit integers in "a" right by "count" while shifting in zeros, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). FOR j := 0 to 7 - i := j*64; - IF k[j] THEN - dst[i+63:i] := EXP_2_23_DP(a[i+63:i]); + i := j*32 + IF k[j] + IF count[63:0] > 31 + dst[i+31:i] := 0 + ELSE + dst[i+31:i] := ZeroExtend32(a[i+31:i] >> count[63:0]) + FI ELSE - dst[i+63:i] := 0; + dst[i+31:i] := 0 FI -ENDFOR; -dst[MAX:512] := 0 +ENDFOR +dst[MAX:256] := 0 - +
immintrin.h
 - - Floating Point - AVX512ER - Elementary Math Functions - - - Compute the approximate exponential value of 2 raised to the power of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). The maximum relative error for this approximation is less than 2^-23. + + Integer + AVX512VL + AVX512F + Shift + + + + + Shift packed 32-bit integers in "a" right by "imm8" while shifting in zeros, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). FOR j := 0 to 7 - i := j*64; - IF k[j] THEN - dst[i+63:i] := EXP_2_23_DP(a[i+63:i]); + i := j*32 + IF k[j] + IF imm8[7:0] > 31 + dst[i+31:i] := 0 + ELSE + dst[i+31:i] := ZeroExtend32(a[i+31:i] >> imm8[7:0]) + FI ELSE - dst[i+63:i] := 0; + dst[i+31:i] := 0 FI -ENDFOR; -dst[MAX:512] := 0 +ENDFOR +dst[MAX:256] := 0 - +
immintrin.h
 - - Floating Point - AVX512ER - Elementary Math Functions - - - - Compute the approximate reciprocal of the lower double-precision (64-bit) floating-point element in "b", store the result in the lower element of "dst", and copy the upper element from "a" to the upper element of "dst". The maximum relative error for this approximation is less than 2^-28. [round_note] + + Integer + AVX512VL + AVX512F + Shift + + + + + + Shift packed 32-bit integers in "a" right by "count" while shifting in zeros, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -dst[63:0] := RCP_28_DP(1.0/b[63:0]; -dst[127:64] := a[127:64]; -dst[MAX:128] := 0; +FOR j := 0 to 3 + i := j*32 + IF k[j] + IF count[63:0] > 31 + dst[i+31:i] := 0 + ELSE + dst[i+31:i] := ZeroExtend32(a[i+31:i] >> count[63:0]) + FI + ELSE + dst[i+31:i] := src[i+31:i] + FI +ENDFOR +dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point - AVX512ER - Elementary Math Functions - - - Compute the approximate reciprocal of the lower double-precision (64-bit) floating-point element in "b", store the result in the lower element of "dst", and copy the upper element from "a" to the upper element of "dst". The maximum relative error for this approximation is less than 2^-28. + + Integer + AVX512VL + AVX512F + Shift + + + + + + Shift packed 32-bit integers in "a" right by "imm8" while shifting in zeros, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -dst[63:0] := RCP_28_DP(1.0/b[63:0]; -dst[127:64] := a[127:64]; -dst[MAX:128] := 0; - - -
immintrin.h
- - - Floating Point - AVX512ER - Elementary Math Functions - - - - - - Compute the approximate reciprocal of the lower double-precision (64-bit) floating-point element in "b", store the result in the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". The maximum relative error for this approximation is less than 2^-28. [round_note] - -IF k[0] THEN - dst[63:0] := RCP_28_DP(1.0/b[63:0]; -ELSE - dst[63:0] := src[63:0]; -FI -dst[127:64] := a[127:64]; -dst[MAX:128] := 0; +FOR j := 0 to 3 + i := j*32 + IF k[j] + IF imm8[7:0] > 31 + dst[i+31:i] := 0 + ELSE + dst[i+31:i] := ZeroExtend32(a[i+31:i] >> imm8[7:0]) + FI + ELSE + dst[i+31:i] := src[i+31:i] + FI +ENDFOR +dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point - AVX512ER - Elementary Math Functions - - - - - Compute the approximate reciprocal of the lower double-precision (64-bit) floating-point element in "b", store the result in the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". The maximum relative error for this approximation is less than 2^-28. + + Integer + AVX512VL + AVX512F + Shift + + + + + Shift packed 32-bit integers in "a" right by "count" while shifting in zeros, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -IF k[0] THEN - dst[63:0] := RCP_28_DP(1.0/b[63:0]; -ELSE - dst[63:0] := src[63:0]; -FI -dst[127:64] := a[127:64]; -dst[MAX:128] := 0; - - -
immintrin.h
- - - Floating Point - AVX512ER - Elementary Math Functions - - - - - Compute the approximate reciprocal of the lower double-precision (64-bit) floating-point element in "b", store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". The maximum relative error for this approximation is less than 2^-28. [round_note] - -IF k[0] THEN - dst[63:0] := RCP_28_DP(1.0/b[63:0]; -ELSE - dst[63:0] := 0; -FI -dst[127:64] := a[127:64]; -dst[MAX:128] := 0; +FOR j := 0 to 3 + i := j*32 + IF k[j] + IF count[63:0] > 31 + dst[i+31:i] := 0 + ELSE + dst[i+31:i] := ZeroExtend32(a[i+31:i] >> count[63:0]) + FI + ELSE + dst[i+31:i] := 0 + FI +ENDFOR +dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point - AVX512ER - Elementary Math Functions - - - - Compute the approximate reciprocal of the lower double-precision (64-bit) floating-point element in "b", store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". The maximum relative error for this approximation is less than 2^-28. + + Integer + AVX512VL + AVX512F + Shift + + + + + Shift packed 32-bit integers in "a" right by "imm8" while shifting in zeros, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -IF k[0] THEN - dst[63:0] := RCP_28_DP(1.0/b[63:0]; -ELSE - dst[63:0] := 0; -FI -dst[127:64] := a[127:64]; -dst[MAX:128] := 0; +FOR j := 0 to 3 + i := j*32 + IF k[j] + IF imm8[7:0] > 31 + dst[i+31:i] := 0 + ELSE + dst[i+31:i] := ZeroExtend32(a[i+31:i] >> imm8[7:0]) + FI + ELSE + dst[i+31:i] := 0 + FI +ENDFOR +dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point - AVX512ER - Elementary Math Functions - - - - Compute the approximate reciprocal of the lower single-precision (32-bit) floating-point element in "b", store the result in the lower element of "dst". The maximum relative error for this approximation is less than 2^-28, and copy the upper 3 packed elements from "a" to the upper elements of "dst". [round_note] + + Integer + AVX512VL + AVX512F + Shift + + + + + + Shift packed 64-bit integers in "a" right by "count" while shifting in zeros, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -dst[31:0] := RCP_28_DP(1.0/b[31:0]; -dst[127:32] := a[127:32]; -dst[MAX:128] := 0; +FOR j := 0 to 3 + i := j*64 + IF k[j] + IF count[63:0] > 63 + dst[i+63:i] := 0 + ELSE + dst[i+63:i] := ZeroExtend64(a[i+63:i] >> count[63:0]) + FI + ELSE + dst[i+63:i] := src[i+63:i] + FI +ENDFOR +dst[MAX:256] := 0 - +
immintrin.h
 - - Floating Point - AVX512ER - Elementary Math Functions - - - Compute the approximate reciprocal of the lower single-precision (32-bit) floating-point element in "b", store the result in the lower element of "dst", and copy the upper 3 packed elements from "a" to the upper elements of "dst". The maximum relative error for this approximation is less than 2^-28. + + Integer + AVX512VL + AVX512F + Shift + + + + + + Shift packed 64-bit integers in "a" right by "imm8" while shifting in zeros, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -dst[31:0] := RCP_28_DP(1.0/b[31:0]; -dst[127:32] := a[127:32]; -dst[MAX:128] := 0; +FOR j := 0 to 3 + i := j*64 + IF k[j] + IF imm8[7:0] > 63 + dst[i+63:i] := 0 + ELSE + dst[i+63:i] := ZeroExtend64(a[i+63:i] >> imm8[7:0]) + FI + ELSE + dst[i+63:i] := src[i+63:i] + FI +ENDFOR +dst[MAX:256] := 0 - +
immintrin.h
 - - Floating Point - AVX512ER - Elementary Math Functions - - - - - - Compute the approximate reciprocal of the lower single-precision (32-bit) floating-point element in "b", store the result in the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". The maximum relative error for this approximation is less than 2^-28. [round_note] - -IF k[0] THEN - dst[31:0] := RCP_28_DP(1.0/b[31:0]; -ELSE - dst[31:0] := src[31:0]; -FI -dst[127:32] := a[127:32]; -dst[MAX:128] := 0; + + Integer + AVX512VL + AVX512F + Shift + + + + + Shift packed 64-bit integers in "a" right by "count" while shifting in zeros, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + +FOR j := 0 to 3 + i := j*64 + IF k[j] + IF count[63:0] > 63 + dst[i+63:i] := 0 + ELSE + dst[i+63:i] := ZeroExtend64(a[i+63:i] >> count[63:0]) + FI + ELSE + dst[i+63:i] := 0 + FI +ENDFOR +dst[MAX:256] := 0 - +
immintrin.h
 - - Floating Point - AVX512ER - Elementary Math Functions - - - - - Compute the approximate reciprocal of the lower single-precision (32-bit) floating-point element in "b", store the result in the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". The maximum relative error for this approximation is less than 2^-28. + + Integer + AVX512VL + AVX512F + Shift + + + + + Shift packed 64-bit integers in "a" right by "imm8" while shifting in zeros, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -IF k[0] THEN - dst[31:0] := RCP_28_DP(1.0/b[31:0]; -ELSE - dst[31:0] := src[31:0]; -FI -dst[127:32] := a[127:32]; -dst[MAX:128] := 0; +FOR j := 0 to 3 + i := j*64 + IF k[j] + IF imm8[7:0] > 63 + dst[i+63:i] := 0 + ELSE + dst[i+63:i] := ZeroExtend64(a[i+63:i] >> imm8[7:0]) + FI + ELSE + dst[i+63:i] := 0 + FI +ENDFOR +dst[MAX:256] := 0 - +
immintrin.h
 - - Floating Point - AVX512ER - Elementary Math Functions - - - - - Compute the approximate reciprocal of the lower single-precision (32-bit) floating-point element in "b", store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". The maximum relative error for this approximation is less than 2^-28. [round_note] - -IF k[0] THEN - dst[31:0] := RCP_28_DP(1.0/b[31:0]; -ELSE - dst[31:0] := 0; -FI -dst[127:32] := a[127:32]; -dst[MAX:128] := 0; + + Integer + AVX512VL + AVX512F + Shift + + + + + + Shift packed 64-bit integers in "a" right by "count" while shifting in zeros, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +FOR j := 0 to 1 + i := j*64 + IF k[j] + IF count[63:0] > 63 + dst[i+63:i] := 0 + ELSE + dst[i+63:i] := ZeroExtend64(a[i+63:i] >> count[63:0]) + FI + ELSE + dst[i+63:i] := src[i+63:i] + FI +ENDFOR +dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point - AVX512ER - Elementary Math Functions - - - - Compute the approximate reciprocal of the lower single-precision (32-bit) floating-point element in "b", store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". The maximum relative error for this approximation is less than 2^-28. + + Integer + AVX512VL + AVX512F + Shift + + + + + + Shift packed 64-bit integers in "a" right by "imm8" while shifting in zeros, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -IF k[0] THEN - dst[31:0] := RCP_28_DP(1.0/b[31:0]; -ELSE - dst[31:0] := 0; -FI -dst[127:32] := a[127:32]; -dst[MAX:128] := 0; +FOR j := 0 to 1 + i := j*64 + IF k[j] + IF imm8[7:0] > 63 + dst[i+63:i] := 0 + ELSE + dst[i+63:i] := ZeroExtend64(a[i+63:i] >> imm8[7:0]) + FI + ELSE + dst[i+63:i] := src[i+63:i] + FI +ENDFOR +dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point - AVX512ER - Elementary Math Functions - - - Compute the approximate reciprocal of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst". The maximum relative error for this approximation is less than 2^-28. [round_note] + + Integer + AVX512VL + AVX512F + Shift + + + + + Shift packed 64-bit integers in "a" right by "count" while shifting in zeros, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*32; - dst[i+31:i] := RCP_28_SP(1.0/a[i+31:i]; -ENDFOR; +FOR j := 0 to 1 + i := j*64 + IF k[j] + IF count[63:0] > 63 + dst[i+63:i] := 0 + ELSE + dst[i+63:i] := ZeroExtend64(a[i+63:i] >> count[63:0]) + FI + ELSE + dst[i+63:i] := 0 + FI +ENDFOR +dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point - AVX512ER - Elementary Math Functions - - Compute the approximate reciprocal of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst". The maximum relative error for this approximation is less than 2^-28. + + Integer + AVX512VL + AVX512F + Shift + + + + + Shift packed 64-bit integers in "a" right by "imm8" while shifting in zeros, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*32; - dst[i+31:i] := RCP_28_SP(1.0/a[i+31:i]; -ENDFOR; +FOR j := 0 to 1 + i := j*64 + IF k[j] + IF imm8[7:0] > 63 + dst[i+63:i] := 0 + ELSE + dst[i+63:i] := ZeroExtend64(a[i+63:i] >> imm8[7:0]) + FI + ELSE + dst[i+63:i] := 0 + FI +ENDFOR +dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point - AVX512ER - Elementary Math Functions - - - - - Compute the approximate reciprocal of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). The maximum relative error for this approximation is less than 2^-28. [round_note] + + Integer + AVX512VL + AVX512F + Shift + + + + + + Shift packed 32-bit integers in "a" right by the amount specified by the corresponding element in "count" while shifting in zeros, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*32; - IF k[j] THEN - dst[i+31:i] := RCP_28_SP(1.0/a[i+31:i]; +FOR j := 0 to 7 + i := j*32 + IF k[j] + IF count[i+31:i] < 32 + dst[i+31:i] := ZeroExtend32(a[i+31:i] >> count[i+31:i]) + ELSE + dst[i+31:i] := 0 + FI ELSE - dst[i+31:i] := src[i+31:i]; - FI -ENDFOR; + dst[i+31:i] := src[i+31:i] + FI +ENDFOR +dst[MAX:256] := 0 - +
immintrin.h
 - - Floating Point - AVX512ER - Elementary Math Functions - - - - Compute the approximate reciprocal of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). The maximum relative error for this approximation is less than 2^-28. + + Integer + AVX512VL + AVX512F + Shift + + + + + Shift packed 32-bit integers in "a" right by the amount specified by the corresponding element in "count" while shifting in zeros, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*32; - IF k[j] THEN - dst[i+31:i] := RCP_28_SP(1.0/a[i+31:i]; +FOR j := 0 to 7 + i := j*32 + IF k[j] + IF count[i+31:i] < 32 + dst[i+31:i] := ZeroExtend32(a[i+31:i] >> count[i+31:i]) + ELSE + dst[i+31:i] := 0 + FI ELSE - dst[i+31:i] := src[i+31:i]; - FI -ENDFOR; + dst[i+31:i] := 0 + FI +ENDFOR +dst[MAX:256] := 0 - +
immintrin.h
 - - Floating Point - AVX512ER - Elementary Math Functions - - - - Compute the approximate reciprocal of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). The maximum relative error for this approximation is less than 2^-28. [round_note] + + Integer + AVX512VL + AVX512F + Shift + + + + + + Shift packed 32-bit integers in "a" right by the amount specified by the corresponding element in "count" while shifting in zeros, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*32; - IF k[j] THEN - dst[i+31:i] := RCP_28_SP(1.0/a[i+31:i]; +FOR j := 0 to 3 + i := j*32 + IF k[j] + IF count[i+31:i] < 32 + dst[i+31:i] := ZeroExtend32(a[i+31:i] >> count[i+31:i]) + ELSE + dst[i+31:i] := 0 + FI ELSE - dst[i+31:i] := 0; - FI -ENDFOR; + dst[i+31:i] := src[i+31:i] + FI +ENDFOR +dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point - AVX512ER - Elementary Math Functions - - - Compute the approximate reciprocal of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). The maximum relative error for this approximation is less than 2^-28. + + Integer + AVX512VL + AVX512F + Shift + + + + + Shift packed 32-bit integers in "a" right by the amount specified by the corresponding element in "count" while shifting in zeros, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*32; - IF k[j] THEN - dst[i+31:i] := RCP_28_SP(1.0/a[i+31:i]; +FOR j := 0 to 3 + i := j*32 + IF k[j] + IF count[i+31:i] < 32 + dst[i+31:i] := ZeroExtend32(a[i+31:i] >> count[i+31:i]) + ELSE + dst[i+31:i] := 0 + FI ELSE - dst[i+31:i] := 0; - FI -ENDFOR; + dst[i+31:i] := 0 + FI +ENDFOR +dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point - AVX512ER - Elementary Math Functions - - - Compute the approximate reciprocal of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst". The maximum relative error for this approximation is less than 2^-28. [round_note] + + Integer + AVX512VL + AVX512F + Shift + + + + + + Shift packed 64-bit integers in "a" right by the amount specified by the corresponding element in "count" while shifting in zeros, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := j*64; - dst[i+63:i] := RCP_28_SP(1.0/a[i+63:i]; -ENDFOR; +FOR j := 0 to 3 + i := j*64 + IF k[j] + IF count[i+63:i] < 64 + dst[i+63:i] := ZeroExtend64(a[i+63:i] >> count[i+63:i]) + ELSE + dst[i+63:i] := 0 + FI + ELSE + dst[i+63:i] := src[i+63:i] + FI +ENDFOR +dst[MAX:256] := 0 - +
immintrin.h
 - - Floating Point - AVX512ER - Elementary Math Functions - - Compute the approximate reciprocal of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst". The maximum relative error for this approximation is less than 2^-28. + + Integer + AVX512VL + AVX512F + Shift + + + + + Shift packed 64-bit integers in "a" right by the amount specified by the corresponding element in "count" while shifting in zeros, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := j*64; - dst[i+63:i] := RCP_28_SP(1.0/a[i+63:i]; -ENDFOR; +FOR j := 0 to 3 + i := j*64 + IF k[j] + IF count[i+63:i] < 64 + dst[i+63:i] := ZeroExtend64(a[i+63:i] >> count[i+63:i]) + ELSE + dst[i+63:i] := 0 + FI + ELSE + dst[i+63:i] := 0 + FI +ENDFOR +dst[MAX:256] := 0 - +
immintrin.h
 - - Floating Point - AVX512ER - Elementary Math Functions - - - - - Compute the approximate reciprocal of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). The maximum relative error for this approximation is less than 2^-28. [round_note] + + Integer + AVX512VL + AVX512F + Shift + + + + + + Shift packed 64-bit integers in "a" right by the amount specified by the corresponding element in "count" while shifting in zeros, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := j*64; - IF k[j] THEN - dst[i+63:i] := RCP_28_SP(1.0/a[i+63:i]; +FOR j := 0 to 1 + i := j*64 + IF k[j] + IF count[i+63:i] < 64 + dst[i+63:i] := ZeroExtend64(a[i+63:i] >> count[i+63:i]) + ELSE + dst[i+63:i] := 0 + FI ELSE - dst[i+63:i] := src[i+63:i]; - FI -ENDFOR; + dst[i+63:i] := src[i+63:i] + FI +ENDFOR +dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point - AVX512ER - Elementary Math Functions - - - - Compute the approximate reciprocal of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). The maximum relative error for this approximation is less than 2^-28. + + Integer + AVX512VL + AVX512F + Shift + + + + + Shift packed 64-bit integers in "a" right by the amount specified by the corresponding element in "count" while shifting in zeros, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := j*64; - IF k[j] THEN - dst[i+63:i] := RCP_28_SP(1.0/a[i+63:i]; +FOR j := 0 to 1 + i := j*64 + IF k[j] + IF count[i+63:i] < 64 + dst[i+63:i] := ZeroExtend64(a[i+63:i] >> count[i+63:i]) + ELSE + dst[i+63:i] := 0 + FI ELSE - dst[i+63:i] := src[i+63:i]; - FI -ENDFOR; + dst[i+63:i] := 0 + FI +ENDFOR +dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point - AVX512ER - Elementary Math Functions - - - - Compute the approximate reciprocal of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). The maximum relative error for this approximation is less than 2^-28. [round_note] + + Integer + AVX512VL + AVX512F + Arithmetic + + + + + + Subtract packed 32-bit integers in "b" from packed 32-bit integers in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). FOR j := 0 to 7 - i := j*64; - IF k[j] THEN - dst[i+63:i] := RCP_28_SP(1.0/a[i+63:i]; + i := j*32 + IF k[j] + dst[i+31:i] := a[i+31:i] - b[i+31:i] ELSE - dst[i+63:i] := 0; - FI -ENDFOR; + dst[i+31:i] := src[i+31:i] + FI +ENDFOR +dst[MAX:256] := 0 - +
immintrin.h
 - - Floating Point - AVX512ER - Elementary Math Functions - - - Compute the approximate reciprocal of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). The maximum relative error for this approximation is less than 2^-28. + + Integer + AVX512VL + AVX512F + Arithmetic + + + + + Subtract packed 32-bit integers in "b" from packed 32-bit integers in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). FOR j := 0 to 7 - i := j*64; - IF k[j] THEN - dst[i+63:i] := RCP_28_SP(1.0/a[i+63:i]; + i := j*32 + IF k[j] + dst[i+31:i] := a[i+31:i] - b[i+31:i] ELSE - dst[i+63:i] := 0; - FI -ENDFOR; + dst[i+31:i] := 0 + FI +ENDFOR +dst[MAX:256] := 0 - +
immintrin.h
 - - Floating Point - AVX512ER - Elementary Math Functions - - - - Compute the approximate reciprocal square root of the lower double-precision (64-bit) floating-point element in "b", store the result in the lower element of "dst", and copy the upper element from "a" to the upper element of "dst". The maximum relative error for this approximation is less than 2^-28. [round_note] + + Integer + AVX512VL + AVX512F + Arithmetic + + + + + + Subtract packed 32-bit integers in "b" from packed 32-bit integers in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -dst[63:0] := (1.0/SQRT(b[63:0])); -dst[127:64] := a[127:64]; -dst[MAX:128] := 0; +FOR j := 0 to 3 + i := j*32 + IF k[j] + dst[i+31:i] := a[i+31:i] - b[i+31:i] + ELSE + dst[i+31:i] := src[i+31:i] + FI +ENDFOR +dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point - AVX512ER - Elementary Math Functions - - - Compute the approximate reciprocal square root of the lower double-precision (64-bit) floating-point element in "b", store the result in the lower element of "dst", and copy the upper element from "a" to the upper element of "dst". The maximum relative error for this approximation is less than 2^-28. + + Integer + AVX512VL + AVX512F + Arithmetic + + + + + Subtract packed 32-bit integers in "b" from packed 32-bit integers in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -dst[63:0] := (1.0/SQRT(b[63:0])); -dst[127:64] := a[127:64]; -dst[MAX:128] := 0; +FOR j := 0 to 3 + i := j*32 + IF k[j] + dst[i+31:i] := a[i+31:i] - b[i+31:i] + ELSE + dst[i+31:i] := 0 + FI +ENDFOR +dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point - AVX512ER - Elementary Math Functions - - - - - - Compute the approximate reciprocal square root of the lower double-precision (64-bit) floating-point element in "b", store the result in the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". The maximum relative error for this approximation is less than 2^-28. [round_note] - -IF k[0] THEN - dst[63:0] := (1.0/SQRT(b[63:0])); -ELSE - dst[63:0] := src[63:0]; -FI -dst[127:64] := a[127:64]; -dst[MAX:128] := 0; + + Integer + AVX512VL + AVX512F + Arithmetic + + + + + + Subtract packed 64-bit integers in "b" from packed 64-bit integers in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +FOR j := 0 to 3 + i := j*64 + IF k[j] + dst[i+63:i] := a[i+63:i] - b[i+63:i] + ELSE + dst[i+63:i] := src[i+63:i] + FI +ENDFOR +dst[MAX:256] := 0 - +
immintrin.h
 - - Floating Point - AVX512ER - Elementary Math Functions - - - - - Compute the approximate reciprocal square root of the lower double-precision (64-bit) floating-point element in "b", store the result in the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". The maximum relative error for this approximation is less than 2^-28. + + Integer + AVX512VL + AVX512F + Arithmetic + + + + + Subtract packed 64-bit integers in "b" from packed 64-bit integers in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -IF k[0] THEN - dst[63:0] := (1.0/SQRT(b[63:0])); -ELSE - dst[63:0] := src[63:0]; -FI -dst[127:64] := a[127:64]; -dst[MAX:128] := 0; +FOR j := 0 to 3 + i := j*64 + IF k[j] + dst[i+63:i] := a[i+63:i] - b[i+63:i] + ELSE + dst[i+63:i] := 0 + FI +ENDFOR +dst[MAX:256] := 0 - +
immintrin.h
 - - Floating Point - AVX512ER - Elementary Math Functions - - - - - Compute the approximate reciprocal square root of the lower double-precision (64-bit) floating-point element in "b", store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". The maximum relative error for this approximation is less than 2^-28. [round_note] - -IF k[0] THEN - dst[63:0] := (1.0/SQRT(b[63:0])); -ELSE - dst[63:0] := 0; -FI -dst[127:64] := a[127:64]; -dst[MAX:128] := 0; + + Integer + AVX512VL + AVX512F + Arithmetic + + + + + + Subtract packed 64-bit integers in "b" from packed 64-bit integers in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +FOR j := 0 to 1 + i := j*64 + IF k[j] + dst[i+63:i] := a[i+63:i] - b[i+63:i] + ELSE + dst[i+63:i] := src[i+63:i] + FI +ENDFOR +dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point - AVX512ER - Elementary Math Functions - - - - Compute the approximate reciprocal square root of the lower double-precision (64-bit) floating-point element in "b", store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". The maximum relative error for this approximation is less than 2^-28. + + Integer + AVX512VL + AVX512F + Arithmetic + + + + + Subtract packed 64-bit integers in "b" from packed 64-bit integers in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -IF k[0] THEN - dst[63:0] := (1.0/SQRT(b[63:0])); -ELSE - dst[63:0] := 0; -FI -dst[127:64] := a[127:64]; -dst[MAX:128] := 0; +FOR j := 0 to 1 + i := j*64 + IF k[j] + dst[i+63:i] := a[i+63:i] - b[i+63:i] + ELSE + dst[i+63:i] := 0 + FI +ENDFOR +dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point - AVX512ER - Elementary Math Functions - - - - Compute the approximate reciprocal square root of the lower single-precision (32-bit) floating-point element in "b", store the result in the lower element of "dst", and copy the upper 3 packed elements from "a" to the upper elements of "dst". The maximum relative error for this approximation is less than 2^-28. [round_note] + + Integer + AVX512VL + AVX512F + Logical + + + + + + + Bitwise ternary logic that provides the capability to implement any three-operand binary function; the specific binary function is specified by value in "imm8". For each bit in each packed 32-bit integer, the corresponding bit from "src", "a", and "b" are used to form a 3 bit index into "imm8", and the value at that bit in "imm8" is written to the corresponding bit in "dst" using writemask "k" at 32-bit granularity (32-bit elements are copied from "src" when the corresponding mask bit is not set). -dst[31:0] := (1.0/SQRT(b[31:0])); -dst[127:32] := a[127:32]; -dst[MAX:128] := 0; +FOR j := 0 to 7 + i := j*32 + IF k[j] + FOR h := 0 to 31 + index[2:0] := (src[i+h] << 2) OR (a[i+h] << 1) OR b[i+h] + dst[i+h] := imm8[index[2:0]] + ENDFOR + ELSE + dst[i+31:i] := src[i+31:i] + FI +ENDFOR +dst[MAX:256] := 0 - +
immintrin.h
 - - Floating Point - AVX512ER - Elementary Math Functions - - - Compute the approximate reciprocal square root of the lower single-precision (32-bit) floating-point element in "b", store the result in the lower element of "dst", and copy the upper 3 packed elements from "a" to the upper elements of "dst". The maximum relative error for this approximation is less than 2^-28. + + Integer + AVX512VL + AVX512F + Logical + + + + + + + Bitwise ternary logic that provides the capability to implement any three-operand binary function; the specific binary function is specified by value in "imm8". For each bit in each packed 32-bit integer, the corresponding bit from "a", "b", and "c" are used to form a 3 bit index into "imm8", and the value at that bit in "imm8" is written to the corresponding bit in "dst" using zeromask "k" at 32-bit granularity (32-bit elements are zeroed out when the corresponding mask bit is not set). -dst[31:0] := (1.0/SQRT(b[31:0])); -dst[127:32] := a[127:32]; -dst[MAX:128] := 0; +FOR j := 0 to 7 + i := j*32 + IF k[j] + FOR h := 0 to 31 + index[2:0] := (a[i+h] << 2) OR (b[i+h] << 1) OR c[i+h] + dst[i+h] := imm8[index[2:0]] + ENDFOR + ELSE + dst[i+31:i] := 0 + FI +ENDFOR +dst[MAX:256] := 0 - +
immintrin.h
 - - Floating Point - AVX512ER - Elementary Math Functions - - - - - - Compute the approximate reciprocal square root of the lower single-precision (32-bit) floating-point element in "b", store the result in the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". The maximum relative error for this approximation is less than 2^-28. [round_note] - -IF k[0] THEN - dst[31:0] := (1.0/SQRT(b[31:0])); -ELSE - dst[31:0] := src[31:0]; -FI -dst[127:32] := a[127:32]; -dst[MAX:128] := 0; + + Integer + AVX512VL + AVX512F + Logical + + + + + + Bitwise ternary logic that provides the capability to implement any three-operand binary function; the specific binary function is specified by value in "imm8". For each bit in each packed 32-bit integer, the corresponding bit from "a", "b", and "c" are used to form a 3 bit index into "imm8", and the value at that bit in "imm8" is written to the corresponding bit in "dst". + +FOR j := 0 to 7 + i := j*32 + FOR h := 0 to 31 + index[2:0] := (a[i+h] << 2) OR (b[i+h] << 1) OR c[i+h] + dst[i+h] := imm8[index[2:0]] + ENDFOR +ENDFOR +dst[MAX:256] := 0 - +
immintrin.h
 - - Floating Point - AVX512ER - Elementary Math Functions - - - - - Compute the approximate reciprocal square root of the lower single-precision (32-bit) floating-point element in "b", store the result in the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". The maximum relative error for this approximation is less than 2^-28. + + Integer + AVX512VL + AVX512F + Logical + + + + + + + Bitwise ternary logic that provides the capability to implement any three-operand binary function; the specific binary function is specified by value in "imm8". For each bit in each packed 32-bit integer, the corresponding bit from "src", "a", and "b" are used to form a 3 bit index into "imm8", and the value at that bit in "imm8" is written to the corresponding bit in "dst" using writemask "k" at 32-bit granularity (32-bit elements are copied from "src" when the corresponding mask bit is not set). -IF k[0] THEN - dst[31:0] := (1.0/SQRT(b[31:0])); -ELSE - dst[31:0] := src[31:0]; -FI -dst[127:32] := a[127:32]; -dst[MAX:128] := 0; +FOR j := 0 to 3 + i := j*32 + IF k[j] + FOR h := 0 to 31 + index[2:0] := (src[i+h] << 2) OR (a[i+h] << 1) OR b[i+h] + dst[i+h] := imm8[index[2:0]] + ENDFOR + ELSE + dst[i+31:i] := src[i+31:i] + FI +ENDFOR +dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point - AVX512ER - Elementary Math Functions - - - - - Compute the approximate reciprocal square root of the lower single-precision (32-bit) floating-point element in "b", store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". The maximum relative error for this approximation is less than 2^-28. [round_note] - -IF k[0] THEN - dst[31:0] := (1.0/SQRT(b[31:0])); -ELSE - dst[31:0] := 0; -FI -dst[127:32] := a[127:32]; -dst[MAX:128] := 0; + + Integer + AVX512VL + AVX512F + Logical + + + + + + + Bitwise ternary logic that provides the capability to implement any three-operand binary function; the specific binary function is specified by value in "imm8". For each bit in each packed 32-bit integer, the corresponding bit from "a", "b", and "c" are used to form a 3 bit index into "imm8", and the value at that bit in "imm8" is written to the corresponding bit in "dst" using zeromask "k" at 32-bit granularity (32-bit elements are zeroed out when the corresponding mask bit is not set). + +FOR j := 0 to 3 + i := j*32 + IF k[j] + FOR h := 0 to 31 + index[2:0] := (a[i+h] << 2) OR (b[i+h] << 1) OR c[i+h] + dst[i+h] := imm8[index[2:0]] + ENDFOR + ELSE + dst[i+31:i] := 0 + FI +ENDFOR +dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point - AVX512ER - Elementary Math Functions - - - - Compute the approximate reciprocal square root of the lower single-precision (32-bit) floating-point element in "b", store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". The maximum relative error for this approximation is less than 2^-28. + + Integer + AVX512VL + AVX512F + Logical + + + + + + Bitwise ternary logic that provides the capability to implement any three-operand binary function; the specific binary function is specified by value in "imm8". For each bit in each packed 32-bit integer, the corresponding bit from "a", "b", and "c" are used to form a 3 bit index into "imm8", and the value at that bit in "imm8" is written to the corresponding bit in "dst". -IF k[0] THEN - dst[31:0] := (1.0/SQRT(b[31:0])); -ELSE - dst[31:0] := 0; -FI -dst[127:32] := a[127:32]; -dst[MAX:128] := 0; +FOR j := 0 to 3 + i := j*32 + FOR h := 0 to 31 + index[2:0] := (a[i+h] << 2) OR (b[i+h] << 1) OR c[i+h] + dst[i+h] := imm8[index[2:0]] + ENDFOR +ENDFOR +dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point - AVX512ER - Elementary Math Functions - - - Compute the approximate reciprocal square root of packed single-precision (32-bit) floating-point elements in "a", store the results in "dst". The maximum relative error for this approximation is less than 2^-28. [round_note] + + Integer + AVX512VL + AVX512F + Logical + + + + + + + Bitwise ternary logic that provides the capability to implement any three-operand binary function; the specific binary function is specified by value in "imm8". For each bit in each packed 64-bit integer, the corresponding bit from "src", "a", and "b" are used to form a 3 bit index into "imm8", and the value at that bit in "imm8" is written to the corresponding bit in "dst" using writemask "k" at 64-bit granularity (64-bit elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*32; - dst[i+31:i] := (1.0/SQRT(a[i+31:i])); -ENDFOR; +FOR j := 0 to 3 + i := j*64 + IF k[j] + FOR h := 0 to 63 + index[2:0] := (src[i+h] << 2) OR (a[i+h] << 1) OR b[i+h] + dst[i+h] := imm8[index[2:0]] + ENDFOR + ELSE + dst[i+63:i] := src[i+63:i] + FI +ENDFOR +dst[MAX:256] := 0 - +
immintrin.h
 - - Floating Point - AVX512ER - Elementary Math Functions - - Compute the approximate reciprocal square root of packed single-precision (32-bit) floating-point elements in "a", store the results in "dst". The maximum relative error for this approximation is less than 2^-28. + + Integer + AVX512VL + AVX512F + Logical + + + + + + + Bitwise ternary logic that provides the capability to implement any three-operand binary function; the specific binary function is specified by value in "imm8". For each bit in each packed 64-bit integer, the corresponding bit from "a", "b", and "c" are used to form a 3 bit index into "imm8", and the value at that bit in "imm8" is written to the corresponding bit in "dst" using zeromask "k" at 64-bit granularity (64-bit elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*32; - dst[i+31:i] := (1.0/SQRT(a[i+31:i])); -ENDFOR; +FOR j := 0 to 3 + i := j*64 + IF k[j] + FOR h := 0 to 63 + index[2:0] := (a[i+h] << 2) OR (b[i+h] << 1) OR c[i+h] + dst[i+h] := imm8[index[2:0]] + ENDFOR + ELSE + dst[i+63:i] := 0 + FI +ENDFOR +dst[MAX:256] := 0 - +
immintrin.h
 - - Floating Point - AVX512ER - Elementary Math Functions - - - - - Compute the approximate reciprocal square root of packed single-precision (32-bit) floating-point elements in "a", store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). The maximum relative error for this approximation is less than 2^-28. [round_note] + + Integer + AVX512VL + AVX512F + Logical + + + + + + Bitwise ternary logic that provides the capability to implement any three-operand binary function; the specific binary function is specified by value in "imm8". For each bit in each packed 64-bit integer, the corresponding bit from "a", "b", and "c" are used to form a 3 bit index into "imm8", and the value at that bit in "imm8" is written to the corresponding bit in "dst". -FOR j := 0 to 15 - i := j*32; - IF k[j] THEN - dst[i+31:i] := (1.0/SQRT(a[i+31:i])); - ELSE - dst[i+31:i] := src[i+31:i]; - FI -ENDFOR; +FOR j := 0 to 3 + i := j*64 + FOR h := 0 to 63 + index[2:0] := (a[i+h] << 2) OR (b[i+h] << 1) OR c[i+h] + dst[i+h] := imm8[index[2:0]] + ENDFOR +ENDFOR +dst[MAX:256] := 0 - +
immintrin.h
 - - Floating Point - AVX512ER - Elementary Math Functions - - - - Compute the approximate reciprocal square root of packed single-precision (32-bit) floating-point elements in "a", store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). The maximum relative error for this approximation is less than 2^-28. + + Integer + AVX512VL + AVX512F + Logical + + + + + + + Bitwise ternary logic that provides the capability to implement any three-operand binary function; the specific binary function is specified by value in "imm8". For each bit in each packed 64-bit integer, the corresponding bit from "src", "a", and "b" are used to form a 3 bit index into "imm8", and the value at that bit in "imm8" is written to the corresponding bit in "dst" using writemask "k" at 64-bit granularity (64-bit elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*32; - IF k[j] THEN - dst[i+31:i] := (1.0/SQRT(a[i+31:i])); +FOR j := 0 to 1 + i := j*64 + IF k[j] + FOR h := 0 to 63 + index[2:0] := (src[i+h] << 2) OR (a[i+h] << 1) OR b[i+h] + dst[i+h] := imm8[index[2:0]] + ENDFOR ELSE - dst[i+31:i] := src[i+31:i]; + dst[i+63:i] := src[i+63:i] FI -ENDFOR; +ENDFOR +dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point - AVX512ER - Elementary Math Functions - - - - Compute the approximate reciprocal square root of packed single-precision (32-bit) floating-point elements in "a", store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). The maximum relative error for this approximation is less than 2^-28. [round_note] + + Integer + AVX512VL + AVX512F + Logical + + + + + + + Bitwise ternary logic that provides the capability to implement any three-operand binary function; the specific binary function is specified by value in "imm8". For each bit in each packed 64-bit integer, the corresponding bit from "a", "b", and "c" are used to form a 3 bit index into "imm8", and the value at that bit in "imm8" is written to the corresponding bit in "dst" using zeromask "k" at 64-bit granularity (64-bit elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*32; - IF k[j] THEN - dst[i+31:i] := (1.0/SQRT(a[i+31:i])); +FOR j := 0 to 1 + i := j*64 + IF k[j] + FOR h := 0 to 63 + index[2:0] := (a[i+h] << 2) OR (b[i+h] << 1) OR c[i+h] + dst[i+h] := imm8[index[2:0]] + ENDFOR ELSE - dst[i+31:i] := 0; + dst[i+63:i] := 0 FI -ENDFOR; +ENDFOR +dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point - AVX512ER - Elementary Math Functions - - - Compute the approximate reciprocal square root of packed single-precision (32-bit) floating-point elements in "a", store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). The maximum relative error for this approximation is less than 2^-28. + + Integer + AVX512VL + AVX512F + Logical + + + + + + Bitwise ternary logic that provides the capability to implement any three-operand binary function; the specific binary function is specified by value in "imm8". For each bit in each packed 64-bit integer, the corresponding bit from "a", "b", and "c" are used to form a 3 bit index into "imm8", and the value at that bit in "imm8" is written to the corresponding bit in "dst". -FOR j := 0 to 15 - i := j*32; - IF k[j] THEN - dst[i+31:i] := (1.0/SQRT(a[i+31:i])); - ELSE - dst[i+31:i] := 0; - FI -ENDFOR; +FOR j := 0 to 1 + i := j*64 + FOR h := 0 to 63 + index[2:0] := (a[i+h] << 2) OR (b[i+h] << 1) OR c[i+h] + dst[i+h] := imm8[index[2:0]] + ENDFOR +ENDFOR +dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point - AVX512ER - Elementary Math Functions - - - Compute the approximate reciprocal square root of packed double-precision (64-bit) floating-point elements in "a", store the results in "dst". The maximum relative error for this approximation is less than 2^-28. [round_note] + + Integer + Mask + AVX512VL + AVX512F + Compare + + + + + Compute the bitwise AND of packed 32-bit integers in "a" and "b", producing intermediate 32-bit values, and set the corresponding bit in result mask "k" (subject to writemask "k") if the intermediate value is non-zero. FOR j := 0 to 7 - i := j*64; - dst[i+63:i] := (1.0/SQRT(a[i+63:i])); -ENDFOR; + i := j*32 + IF k1[j] + k[j] := ((a[i+31:i] AND b[i+31:i]) != 0) ? 1 : 0 + ELSE + k[j] := 0 + FI +ENDFOR +k[MAX:8] := 0 - +
immintrin.h
 - - Floating Point - AVX512ER - Elementary Math Functions - - Compute the approximate reciprocal square root of packed double-precision (64-bit) floating-point elements in "a", store the results in "dst". The maximum relative error for this approximation is less than 2^-28. + + Integer + Mask + AVX512VL + AVX512F + Compare + + + + Compute the bitwise AND of packed 32-bit integers in "a" and "b", producing intermediate 32-bit values, and set the corresponding bit in result mask "k" if the intermediate value is non-zero. FOR j := 0 to 7 - i := j*64; - dst[i+63:i] := (1.0/SQRT(a[i+63:i])); -ENDFOR; + i := j*32 + k[j] := ((a[i+31:i] AND b[i+31:i]) != 0) ? 1 : 0 +ENDFOR +k[MAX:8] := 0 - +
immintrin.h
 - - Floating Point - AVX512ER - Elementary Math Functions - - - - - Compute the approximate reciprocal square root of packed double-precision (64-bit) floating-point elements in "a", store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). The maximum relative error for this approximation is less than 2^-28. [round_note] + + Integer + Mask + AVX512VL + AVX512F + Compare + + + + + Compute the bitwise AND of packed 32-bit integers in "a" and "b", producing intermediate 32-bit values, and set the corresponding bit in result mask "k" (subject to writemask "k") if the intermediate value is non-zero. -FOR j := 0 to 7 - i := j*64; - IF k[j] THEN - dst[i+63:i] := (1.0/SQRT(a[i+63:i])); +FOR j := 0 to 3 + i := j*32 + IF k1[j] + k[j] := ((a[i+31:i] AND b[i+31:i]) != 0) ? 1 : 0 ELSE - dst[i+63:i] := src[i+63:i]; + k[j] := 0 FI -ENDFOR; +ENDFOR +k[MAX:4] := 0 - +
immintrin.h
 - - Floating Point - AVX512ER - Elementary Math Functions - - - - Compute the approximate reciprocal square root of packed double-precision (64-bit) floating-point elements in "a", store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). The maximum relative error for this approximation is less than 2^-28. + + Integer + Mask + AVX512VL + AVX512F + Compare + + + + Compute the bitwise AND of packed 32-bit integers in "a" and "b", producing intermediate 32-bit values, and set the corresponding bit in result mask "k" if the intermediate value is non-zero. -FOR j := 0 to 7 - i := j*64; - IF k[j] THEN - dst[i+63:i] := (1.0/SQRT(a[i+63:i])); +FOR j := 0 to 3 + i := j*32 + k[j] := ((a[i+31:i] AND b[i+31:i]) != 0) ? 1 : 0 +ENDFOR +k[MAX:4] := 0 + + +
immintrin.h
+ + + Integer + Mask + AVX512VL + AVX512F + Compare + + + + + Compute the bitwise AND of packed 64-bit integers in "a" and "b", producing intermediate 64-bit values, and set the corresponding bit in result mask "k" (subject to writemask "k") if the intermediate value is non-zero. + +FOR j := 0 to 3 + i := j*64 + IF k1[j] + k[j] := ((a[i+63:i] AND b[i+63:i]) != 0) ? 1 : 0 ELSE - dst[i+63:i] := src[i+63:i]; + k[j] := 0 FI -ENDFOR; +ENDFOR +k[MAX:4] := 0 - +
immintrin.h
 - - Floating Point - AVX512ER - Elementary Math Functions - - - - Compute the approximate reciprocal square root of packed double-precision (64-bit) floating-point elements in "a", store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). The maximum relative error for this approximation is less than 2^-28. [round_note] + + Integer + Mask + AVX512VL + AVX512F + Compare + + + + Compute the bitwise AND of packed 64-bit integers in "a" and "b", producing intermediate 64-bit values, and set the corresponding bit in result mask "k" if the intermediate value is non-zero. -FOR j := 0 to 7 - i := j*64; - IF k[j] THEN - dst[i+63:i] := (1.0/SQRT(a[i+63:i])); +FOR j := 0 to 3 + i := j*64 + k[j] := ((a[i+63:i] AND b[i+63:i]) != 0) ? 1 : 0 +ENDFOR +k[MAX:4] := 0 + + +
immintrin.h
+ + + Integer + Mask + AVX512VL + AVX512F + Compare + + + + + Compute the bitwise AND of packed 64-bit integers in "a" and "b", producing intermediate 64-bit values, and set the corresponding bit in result mask "k" (subject to writemask "k") if the intermediate value is non-zero. + +FOR j := 0 to 1 + i := j*64 + IF k1[j] + k[j] := ((a[i+63:i] AND b[i+63:i]) != 0) ? 1 : 0 ELSE - dst[i+63:i] := 0; + k[j] := 0 FI -ENDFOR; +ENDFOR +k[MAX:2] := 0 - +
immintrin.h
 - - Floating Point - AVX512ER - Elementary Math Functions - - - Compute the approximate reciprocal square root of packed double-precision (64-bit) floating-point elements in "a", store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). The maximum relative error for this approximation is less than 2^-28. + + Integer + Mask + AVX512VL + AVX512F + Compare + + + + Compute the bitwise AND of packed 64-bit integers in "a" and "b", producing intermediate 64-bit values, and set the corresponding bit in result mask "k" if the intermediate value is non-zero. + +FOR j := 0 to 1 + i := j*64 + k[j] := ((a[i+63:i] AND b[i+63:i]) != 0) ? 1 : 0 +ENDFOR +k[MAX:2] := 0 + + +
immintrin.h
+ + + Integer + Mask + AVX512VL + AVX512F + Compare + + + + + Compute the bitwise NAND of packed 32-bit integers in "a" and "b", producing intermediate 32-bit values, and set the corresponding bit in result mask "k" (subject to writemask "k") if the intermediate value is zero. FOR j := 0 to 7 - i := j*64; - IF k[j] THEN - dst[i+63:i] := (1.0/SQRT(a[i+63:i])); + i := j*32 + IF k1[j] + k[j] := ((a[i+31:i] AND b[i+31:i]) == 0) ? 1 : 0 ELSE - dst[i+63:i] := 0; + k[j] := 0 FI -ENDFOR; +ENDFOR +k[MAX:8] := 0 - +
immintrin.h
 - + Integer - BMI1 - Bit Manipulation - - Count the number of trailing zero bits in unsigned 32-bit integer "a", and return that count in "dst". + Mask + AVX512VL + AVX512F + Compare + + + + Compute the bitwise NAND of packed 32-bit integers in "a" and "b", producing intermediate 32-bit values, and set the corresponding bit in result mask "k" if the intermediate value is zero. -tmp := 0 -dst := 0 -DO WHILE ((tmp < 32) AND a[tmp] = 0) - tmp := tmp + 1 - dst := dst + 1 -OD +FOR j := 0 to 7 + i := j*32 + k[j] := ((a[i+31:i] AND b[i+31:i]) == 0) ? 1 : 0 +ENDFOR +k[MAX:8] := 0 - +
immintrin.h
 - + Integer - BMI1 - Bit Manipulation - - Count the number of trailing zero bits in unsigned 64-bit integer "a", and return that count in "dst". + Mask + AVX512VL + AVX512F + Compare + + + + + Compute the bitwise NAND of packed 32-bit integers in "a" and "b", producing intermediate 32-bit values, and set the corresponding bit in result mask "k" (subject to writemask "k") if the intermediate value is zero. -tmp := 0 -dst := 0 -DO WHILE ((tmp < 64) AND a[tmp] = 0) - tmp := tmp + 1 - dst := dst + 1 -OD +FOR j := 0 to 3 + i := j*32 + IF k1[j] + k[j] := ((a[i+31:i] AND b[i+31:i]) == 0) ? 1 : 0 + ELSE + k[j] := 0 + FI +ENDFOR +k[MAX:4] := 0 - +
immintrin.h
 - - Floating Point - KNCNI - Load - - - - - Depending on "bc", loads 1, 4, or 16 elements of type and size determined by "conv" from memory address "mt" and converts all elements to single-precision (32-bit) floating-point elements, storing the results in "dst". "hint" indicates to the processor whether the data is non-temporal. - -addr = MEM[mt] -FOR j := 0 to 15 + + Integer + Mask + AVX512VL + AVX512F + Compare + + + + Compute the bitwise NAND of packed 32-bit integers in "a" and "b", producing intermediate 32-bit values, and set the corresponding bit in result mask "k" if the intermediate value is zero. + +FOR j := 0 to 3 i := j*32 - CASE bc OF - _MM_BROADCAST32_NONE: - CASE conv OF - _MM_UPCONV_PS_NONE: - n := j*32 - dst[i+31:i] := addr[n+31:n] - _MM_UPCONV_PS_FLOAT16: - n := j*16 - dst[i+31:i] := Float16ToFloat32(addr[n+15:n]) - _MM_UPCONV_PS_UINT8: - n := j*8 - dst[i+31:i] := UInt8ToFloat32(addr[n+7:n]) - _MM_UPCONV_PS_SINT8: - n := j*8 - dst[i+31:i] := SInt8ToFloat32(addr[n+7:n]) - _MM_UPCONV_PS_UINT16: - n := j*16 - dst[i+31:i] := UInt16ToFloat32(addr[n+15:n]) - _MM_UPCONV_PS_SINT16: - n := j*16 - dst[i+31:i] := SInt16ToFloat32(addr[n+15:n]) - ESAC - _MM_BROADCAST_1X16: - CASE conv OF - _MM_UPCONV_PS_NONE: - n := j*32 - dst[i+31:i] := addr[31:0] - _MM_UPCONV_PS_FLOAT16: - n := j*16 - dst[i+31:i] := Float16ToFloat32(addr[15:0]) - _MM_UPCONV_PS_UINT8: - n := j*8 - dst[i+31:i] := UInt8ToFloat32(addr[7:0]) - _MM_UPCONV_PS_SINT8: - n := j*8 - dst[i+31:i] := SInt8ToFloat32(addr[7:0]) - _MM_UPCONV_PS_UINT16: - n := j*16 - dst[i+31:i] := UInt16ToFloat32(addr[15:0]) - _MM_UPCONV_PS_SINT16: - n := j*16 - dst[i+31:i] := SInt16ToFloat32(addr[15:0]) - ESAC - _MM_BROADCAST_4X16: - mod := j%4 - CASE conv OF - _MM_UPCONV_PS_NONE: - n := mod*32 - dst[i+31:i] := addr[n+31:n] - _MM_UPCONV_PS_FLOAT16: - n := mod*16 - dst[i+31:i] := Float16ToFloat32(addr[n+15:n]) - _MM_UPCONV_PS_UINT8: - n := mod*8 - dst[i+31:i] := UInt8ToFloat32(addr[n+7:n]) - _MM_UPCONV_PS_SINT8: - n := mod*8 - dst[i+31:i] := SInt8ToFloat32(addr[n+7:n]) - _MM_UPCONV_PS_UINT16: - n := mod*16 - dst[i+31:i] := UInt16ToFloat32(addr[n+15:n]) - _MM_UPCONV_PS_SINT16: - n := mod*16 - dst[i+31:i] := SInt16ToFloat32(addr[n+15:n]) - ESAC - ESAC + k[j] := ((a[i+31:i] AND b[i+31:i]) == 0) ? 1 : 0 ENDFOR -dst[MAX:512] := 0 +k[MAX:4] := 0 - - - +
immintrin.h
 - - Floating Point - KNCNI - Load - - - - - - - Depending on "bc", loads 1, 4, or 16 elements of type and size determined by "conv" from memory address "mt" and converts all elements to single-precision (32-bit) floating-point elements, storing the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). "hint" indicates to the processor whether the data is non-temporal. + + Integer + Mask + AVX512VL + AVX512F + Compare + + + + + Compute the bitwise NAND of packed 64-bit integers in "a" and "b", producing intermediate 64-bit values, and set the corresponding bit in result mask "k" (subject to writemask "k") if the intermediate value is zero. -addr = MEM[mt] -FOR j := 0 to 15 +FOR j := 0 to 3 + i := j*64 + IF k1[j] + k[j] := ((a[i+63:i] AND b[i+63:i]) == 0) ? 1 : 0 + ELSE + k[j] := 0 + FI +ENDFOR +k[MAX:4] := 0 + + +
immintrin.h
+ + + Integer + Mask + AVX512VL + AVX512F + Compare + + + + Compute the bitwise NAND of packed 64-bit integers in "a" and "b", producing intermediate 64-bit values, and set the corresponding bit in result mask "k" if the intermediate value is zero. + +FOR j := 0 to 3 + i := j*64 + k[j] := ((a[i+63:i] AND b[i+63:i]) == 0) ? 1 : 0 +ENDFOR +k[MAX:4] := 0 + + +
immintrin.h
+ + + Integer + Mask + AVX512VL + AVX512F + Compare + + + + + Compute the bitwise NAND of packed 64-bit integers in "a" and "b", producing intermediate 64-bit values, and set the corresponding bit in result mask "k" (subject to writemask "k") if the intermediate value is zero. + +FOR j := 0 to 1 + i := j*64 + IF k1[j] + k[j] := ((a[i+63:i] AND b[i+63:i]) == 0) ? 1 : 0 + ELSE + k[j] := 0 + FI +ENDFOR +k[MAX:2] := 0 + + +
immintrin.h
+ + + Integer + Mask + AVX512VL + AVX512F + Compare + + + + Compute the bitwise NAND of packed 64-bit integers in "a" and "b", producing intermediate 64-bit values, and set the corresponding bit in result mask "k" if the intermediate value is zero. + +FOR j := 0 to 1 + i := j*64 + k[j] := ((a[i+63:i] AND b[i+63:i]) == 0) ? 1 : 0 +ENDFOR +k[MAX:2] := 0 + + +
immintrin.h
+ + + Integer + AVX512VL + AVX512F + Miscellaneous + + + + + + Unpack and interleave 32-bit integers from the high half of each 128-bit lane in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +DEFINE INTERLEAVE_HIGH_DWORDS(src1[127:0], src2[127:0]) { + dst[31:0] := src1[95:64] + dst[63:32] := src2[95:64] + dst[95:64] := src1[127:96] + dst[127:96] := src2[127:96] + RETURN dst[127:0] +} +tmp_dst[127:0] := INTERLEAVE_HIGH_DWORDS(a[127:0], b[127:0]) +tmp_dst[255:128] := INTERLEAVE_HIGH_DWORDS(a[255:128], b[255:128]) +FOR j := 0 to 7 i := j*32 IF k[j] - CASE bc OF - _MM_BROADCAST32_NONE: - CASE conv OF - _MM_UPCONV_PS_NONE: - n := j*32 - dst[i+31:i] := addr[n+31:n] - _MM_UPCONV_PS_FLOAT16: - n := j*16 - dst[i+31:i] := Float16ToFloat32(addr[n+15:n]) - _MM_UPCONV_PS_UINT8: - n := j*8 - dst[i+31:i] := UInt8ToFloat32(addr[n+7:n]) - _MM_UPCONV_PS_SINT8: - n := j*8 - dst[i+31:i] := SInt8ToFloat32(addr[n+7:n]) - _MM_UPCONV_PS_UINT16: - n := j*16 - dst[i+31:i] := UInt16ToFloat32(addr[n+15:n]) - _MM_UPCONV_PS_SINT16: - n := j*16 - dst[i+31:i] := SInt16ToFloat32(addr[n+15:n]) - ESAC - _MM_BROADCAST_1X16: - CASE conv OF - _MM_UPCONV_PS_NONE: - n := j*32 - dst[i+31:i] := addr[31:0] - _MM_UPCONV_PS_FLOAT16: - n := j*16 - dst[i+31:i] := Float16ToFloat32(addr[15:0]) - _MM_UPCONV_PS_UINT8: - n := j*8 - dst[i+31:i] := UInt8ToFloat32(addr[7:0]) - _MM_UPCONV_PS_SINT8: - n := j*8 - dst[i+31:i] := SInt8ToFloat32(addr[7:0]) - _MM_UPCONV_PS_UINT16: - n := j*16 - dst[i+31:i] := UInt16ToFloat32(addr[15:0]) - _MM_UPCONV_PS_SINT16: - n := j*16 - dst[i+31:i] := SInt16ToFloat32(addr[15:0]) - ESAC - _MM_BROADCAST_4X16: - mod := j%4 - CASE conv OF - _MM_UPCONV_PS_NONE: - n := mod*32 - dst[i+31:i] := addr[n+31:n] - _MM_UPCONV_PS_FLOAT16: - n := mod*16 - dst[i+31:i] := Float16ToFloat32(addr[n+15:n]) - _MM_UPCONV_PS_UINT8: - n := mod*8 - dst[i+31:i] := UInt8ToFloat32(addr[n+7:n]) - _MM_UPCONV_PS_SINT8: - n := mod*8 - dst[i+31:i] := SInt8ToFloat32(addr[n+7:n]) - _MM_UPCONV_PS_UINT16: - n := mod*16 - dst[i+31:i] := UInt16ToFloat32(addr[n+15:n]) - _MM_UPCONV_PS_SINT16: - n := mod*16 - dst[i+31:i] := SInt16ToFloat32(addr[n+15:n]) - ESAC - ESAC + dst[i+31:i] := tmp_dst[i+31:i] ELSE dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - - - +
immintrin.h
 - + Integer - KNCNI - Load - - - - - Depending on "bc", loads 1, 4, or 16 elements of type and size determined by "conv" from memory address "mt" and converts all elements to 32-bit integer elements, storing the results in "dst". "hint" indicates to the processor whether the data is non-temporal. + AVX512VL + AVX512F + Miscellaneous + + + + + Unpack and interleave 32-bit integers from the high half of each 128-bit lane in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -addr = MEM[mt] -FOR j := 0 to 15 +DEFINE INTERLEAVE_HIGH_DWORDS(src1[127:0], src2[127:0]) { + dst[31:0] := src1[95:64] + dst[63:32] := src2[95:64] + dst[95:64] := src1[127:96] + dst[127:96] := src2[127:96] + RETURN dst[127:0] +} +tmp_dst[127:0] := INTERLEAVE_HIGH_DWORDS(a[127:0], b[127:0]) +tmp_dst[255:128] := INTERLEAVE_HIGH_DWORDS(a[255:128], b[255:128]) +FOR j := 0 to 7 i := j*32 - CASE bc OF - _MM_BROADCAST32_NONE: - CASE conv OF - _MM_UPCONV_EPI32_NONE: - n := j*32 - dst[i+31:i] := addr[n+31:n] - _MM_UPCONV_EPI32_UINT8: - n := j*8 - dst[i+31:i] := UInt8ToInt32(addr[n+7:n]) - _MM_UPCONV_EPI32_SINT8: - n := j*8 - dst[i+31:i] := SInt8ToInt32(addr[n+7:n]) - _MM_UPCONV_EPI32_UINT16: - n := j*16 - dst[i+31:i] := UInt16ToInt32(addr[n+15:n]) - _MM_UPCONV_EPI32_SINT16: - n := j*16 - dst[i+31:i] := SInt16ToInt32(addr[n+15:n]) - ESAC - _MM_BROADCAST_1X16: - CASE conv OF - _MM_UPCONV_EPI32_NONE: - n := j*32 - dst[i+31:i] := addr[31:0] - _MM_UPCONV_EPI32_UINT8: - n := j*8 - dst[i+31:i] := UInt8ToInt32(addr[7:0]) - _MM_UPCONV_EPI32_SINT8: - n := j*8 - dst[i+31:i] := SInt8ToInt32(addr[7:0]) - _MM_UPCONV_EPI32_UINT16: - n := j*16 - dst[i+31:i] := UInt16ToInt32(addr[15:0]) - _MM_UPCONV_EPI32_SINT16: - n := j*16 - dst[i+31:i] := SInt16ToInt32(addr[15:0]) - ESAC - _MM_BROADCAST_4X16: - mod := j%4 - CASE conv OF - _MM_UPCONV_EPI32_NONE: - n := mod*32 - dst[i+31:i] := addr[n+31:n] - _MM_UPCONV_EPI32_UINT8: - n := mod*8 - dst[i+31:i] := UInt8ToInt32(addr[n+7:n]) - _MM_UPCONV_EPI32_SINT8: - n := mod*8 - dst[i+31:i] := SInt8ToInt32(addr[n+7:n]) - _MM_UPCONV_EPI32_UINT16: - n := mod*16 - dst[i+31:i] := UInt16ToInt32(addr[n+15:n]) - _MM_UPCONV_EPI32_SINT16: - n := mod*16 - dst[i+31:i] := SInt16ToInt32(addr[n+15:n]) - ESAC - ESAC + IF k[j] + dst[i+31:i] := tmp_dst[i+31:i] + ELSE + dst[i+31:i] := 0 + FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - - - +
immintrin.h
 - + Integer - KNCNI - Load - - - - - - - Depending on "bc", loads 1, 4, or 16 elements of type and size determined by "conv" from memory address "mt" and converts all elements to 32-bit integer elements, storing the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). "hint" indicates to the processor whether the data is non-temporal. + AVX512VL + AVX512F + Miscellaneous + + + + + + Unpack and interleave 32-bit integers from the high half of "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -addr = MEM[mt] -FOR j := 0 to 15 +DEFINE INTERLEAVE_HIGH_DWORDS(src1[127:0], src2[127:0]) { + dst[31:0] := src1[95:64] + dst[63:32] := src2[95:64] + dst[95:64] := src1[127:96] + dst[127:96] := src2[127:96] + RETURN dst[127:0] +} +tmp_dst[127:0] := INTERLEAVE_HIGH_DWORDS(a[127:0], b[127:0]) +FOR j := 0 to 3 i := j*32 IF k[j] - CASE bc OF - _MM_BROADCAST32_NONE: - CASE conv OF - _MM_UPCONV_EPI32_NONE: - n := j*32 - dst[i+31:i] := addr[n+31:n] - _MM_UPCONV_EPI32_UINT8: - n := j*8 - dst[i+31:i] := UInt8ToInt32(addr[n+7:n]) - _MM_UPCONV_EPI32_SINT8: - n := j*8 - dst[i+31:i] := SInt8ToInt32(addr[n+7:n]) - _MM_UPCONV_EPI32_UINT16: - n := j*16 - dst[i+31:i] := UInt16ToInt32(addr[n+15:n]) - _MM_UPCONV_EPI32_SINT16: - n := j*16 - dst[i+31:i] := SInt16ToInt32(addr[n+15:n]) - ESAC - _MM_BROADCAST_1X16: - CASE conv OF - _MM_UPCONV_EPI32_NONE: - n := j*32 - dst[i+31:i] := addr[31:0] - _MM_UPCONV_EPI32_UINT8: - n := j*8 - dst[i+31:i] := UInt8ToInt32(addr[7:0]) - _MM_UPCONV_EPI32_SINT8: - n := j*8 - dst[i+31:i] := SInt8ToInt32(addr[7:0]) - _MM_UPCONV_EPI32_UINT16: - n := j*16 - dst[i+31:i] := UInt16ToInt32(addr[15:0]) - _MM_UPCONV_EPI32_SINT16: - n := j*16 - dst[i+31:i] := SInt16ToInt32(addr[15:0]) - ESAC - _MM_BROADCAST_4X16: - mod := j%4 - CASE conv OF - _MM_UPCONV_EPI32_NONE: - n := mod*32 - dst[i+31:i] := addr[n+31:n] - _MM_UPCONV_EPI32_UINT8: - n := mod*8 - dst[i+31:i] := UInt8ToInt32(addr[n+7:n]) - _MM_UPCONV_EPI32_SINT8: - n := mod*8 - dst[i+31:i] := SInt8ToInt32(addr[n+7:n]) - _MM_UPCONV_EPI32_UINT16: - n := mod*16 - dst[i+31:i] := UInt16ToInt32(addr[n+15:n]) - _MM_UPCONV_EPI32_SINT16: - n := mod*16 - dst[i+31:i] := SInt16ToInt32(addr[n+15:n]) - ESAC - ESAC + dst[i+31:i] := tmp_dst[i+31:i] ELSE dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - - - +
immintrin.h
 - - Floating Point - KNCNI - Load - - - - - Depending on "bc", loads 1, 4, or 8 elements of type and size determined by "conv" from memory address "mt" and converts all elements to double-precision (64-bit) floating-point elements, storing the results in "dst". "hint" indicates to the processor whether the data is non-temporal. + + Integer + AVX512VL + AVX512F + Miscellaneous + + + + + Unpack and interleave 32-bit integers from the high half of "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -addr = MEM[mt] -FOR j := 0 to 7 - i := j*64 - CASE bc OF - _MM_BROADCAST64_NONE: - CASE conv OF - _MM_UPCONV_PD_NONE: - n := j*64 - dst[i+63:i] := addr[n+63:n] - ESAC - _MM_BROADCAST_1X8: - CASE conv OF - _MM_UPCONV_PD_NONE: - n := j*64 - dst[i+63:i] := addr[63:0] - ESAC - _MM_BROADCAST_4X8: - mod := j%4 - CASE conv OF - _MM_UPCONV_PD_NONE: - n := mod*64 - dst[i+63:i] := addr[n+63:n] - ESAC - ESAC +DEFINE INTERLEAVE_HIGH_DWORDS(src1[127:0], src2[127:0]) { + dst[31:0] := src1[95:64] + dst[63:32] := src2[95:64] + dst[95:64] := src1[127:96] + dst[127:96] := src2[127:96] + RETURN dst[127:0] +} +tmp_dst[127:0] := INTERLEAVE_HIGH_DWORDS(a[127:0], b[127:0]) +FOR j := 0 to 3 + i := j*32 + IF k[j] + dst[i+31:i] := tmp_dst[i+31:i] + ELSE + dst[i+31:i] := 0 + FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - - - +
immintrin.h
 - - Floating Point - KNCNI - Load - - - - - - - Depending on "bc", loads 1, 4, or 8 elements of type and size determined by "conv" from memory address "mt" and converts all elements to double-precision (64-bit) floating-point elements, storing the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). "hint" indicates to the processor whether the data is non-temporal. + + Integer + AVX512VL + AVX512F + Miscellaneous + + + + + + Unpack and interleave 64-bit integers from the high half of each 128-bit lane in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -addr = MEM[mt] -FOR j := 0 to 7 +DEFINE INTERLEAVE_HIGH_QWORDS(src1[127:0], src2[127:0]) { + dst[63:0] := src1[127:64] + dst[127:64] := src2[127:64] + RETURN dst[127:0] +} +tmp_dst[127:0] := INTERLEAVE_HIGH_QWORDS(a[127:0], b[127:0]) +tmp_dst[255:128] := INTERLEAVE_HIGH_QWORDS(a[255:128], b[255:128]) +FOR j := 0 to 3 i := j*64 IF k[j] - CASE bc OF - _MM_BROADCAST64_NONE: - CASE conv OF - _MM_UPCONV_PD_NONE: - n := j*64 - dst[i+63:i] := addr[n+63:n] - ESAC - _MM_BROADCAST_1X8: - CASE conv OF - _MM_UPCONV_PD_NONE: - n := j*64 - dst[i+63:i] := addr[63:0] - ESAC - _MM_BROADCAST_4X8: - mod := j%4 - CASE conv OF - _MM_UPCONV_PD_NONE: - n := mod*64 - dst[i+63:i] := addr[n+63:n] - ESAC - ESAC + dst[i+63:i] := tmp_dst[i+63:i] ELSE dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - - - +
immintrin.h
 - + Integer - KNCNI - Load - - - - - Depending on "bc", loads 1, 4, or 8 elements of type and size determined by "conv" from memory address "mt" and converts all elements to 64-bit integer elements, storing the results in "dst". "hint" indicates to the processor whether the data is non-temporal. + AVX512VL + AVX512F + Miscellaneous + + + + + Unpack and interleave 64-bit integers from the high half of each 128-bit lane in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -addr = MEM[mt] -FOR j := 0 to 7 +DEFINE INTERLEAVE_HIGH_QWORDS(src1[127:0], src2[127:0]) { + dst[63:0] := src1[127:64] + dst[127:64] := src2[127:64] + RETURN dst[127:0] +} +tmp_dst[127:0] := INTERLEAVE_HIGH_QWORDS(a[127:0], b[127:0]) +tmp_dst[255:128] := INTERLEAVE_HIGH_QWORDS(a[255:128], b[255:128]) +FOR j := 0 to 3 i := j*64 - CASE bc OF - _MM_BROADCAST64_NONE: - CASE conv OF - _MM_UPCONV_EPI64_NONE: - n := j*64 - dst[i+63:i] := addr[n+63:n] - ESAC - _MM_BROADCAST_1X8: - CASE conv OF - _MM_UPCONV_EPI64_NONE: - n := j*64 - dst[i+63:i] := addr[63:0] - ESAC - _MM_BROADCAST_4X8: - mod := j%4 - CASE conv OF - _MM_UPCONV_EPI64_NONE: - n := mod*64 - dst[i+63:i] := addr[n+63:n] - ESAC - ESAC + IF k[j] + dst[i+63:i] := tmp_dst[i+63:i] + ELSE + dst[i+63:i] := 0 + FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - - - +
immintrin.h
 - + Integer - KNCNI - Load - - - - - - - Depending on "bc", loads 1, 4, or 8 elements of type and size determined by "conv" from memory address "mt" and converts all elements to 64-bit integer elements, storing the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). "hint" indicates to the processor whether the data is non-temporal. + AVX512VL + AVX512F + Miscellaneous + + + + + + Unpack and interleave 64-bit integers from the high half of "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -addr = MEM[mt] -FOR j := 0 to 7 +DEFINE INTERLEAVE_HIGH_QWORDS(src1[127:0], src2[127:0]) { + dst[63:0] := src1[127:64] + dst[127:64] := src2[127:64] + RETURN dst[127:0] +} +tmp_dst[127:0] := INTERLEAVE_HIGH_QWORDS(a[127:0], b[127:0]) +FOR j := 0 to 1 i := j*64 IF k[j] - CASE bc OF - _MM_BROADCAST64_NONE: - CASE conv OF - _MM_UPCONV_EPI64_NONE: - n := j*64 - dst[i+63:i] := addr[n+63:n] - ESAC - _MM_BROADCAST_1X8: - CASE conv OF - _MM_UPCONV_EPI64_NONE: - n := j*64 - dst[i+63:i] := addr[63:0] - ESAC - _MM_BROADCAST_4X8: - mod := j%4 - CASE conv OF - _MM_UPCONV_EPI64_NONE: - n := mod*64 - dst[i+63:i] := addr[n+63:n] - ESAC - ESAC + dst[i+63:i] := tmp_dst[i+63:i] ELSE dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - - - +
immintrin.h
 - - Floating Point - KNCNI - Swizzle - - - Performs a swizzle transformation of each of the four groups of packed 4xsingle-precision (32-bit) floating-point elements in "v" using swizzle parameter "s", storing the results in "dst". + + Integer + AVX512VL + AVX512F + Miscellaneous + + + + + Unpack and interleave 64-bit integers from the high half of "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -CASE s OF -_MM_SWIZ_REG_NONE: - dst[511:0] := v[511:0] -_MM_SWIZ_REG_DCBA: - dst[511:0] := v[511:0] -_MM_SWIZ_REG_CDAB: - FOR j := 0 to 7 - i := j*64 - dst[i+31:i] := v[i+63:i+32] - dst[i+63:i+32] := v[i+31:i] - ENDFOR -_MM_SWIZ_REG_BADC: - FOR j := 0 to 3 - i := j*128 - dst[i+31:i] := v[i+95:i+64] - dst[i+63:i+32] := v[i+127:i+96] - dst[i+95:i+64] := v[i+31:i] - dst[i+127:i+96] := v[i+63:i+32] - ENDFOR -_MM_SWIZ_REG_AAAA: - FOR j := 0 to 3 - i := j*128 - dst[i+31:i] := v[i+31:i] - dst[i+63:i+32] := v[i+31:i] - dst[i+95:i+64] := v[i+31:i] - dst[i+127:i+96] := v[i+31:i] - ENDFOR -_MM_SWIZ_REG_BBBB: - FOR j := 0 to 3 - i := j*128 - dst[i+31:i] := v[i+63:i+32] - dst[i+63:i+32] := v[i+63:i+32] - dst[i+95:i+64] := v[i+63:i+32] - dst[i+127:i+96] := v[i+63:i+32] - ENDFOR -_MM_SWIZ_REG_CCCC: - FOR j := 0 to 3 - i := j*128 - dst[i+31:i] := v[i+95:i+64] - dst[i+63:i+32] := v[i+95:i+64] - dst[i+95:i+64] := v[i+95:i+64] - dst[i+127:i+96] := v[i+95:i+64] - ENDFOR -_MM_SWIZ_REG_DDDD: - FOR j := 0 to 3 - i := j*128 - dst[i+31:i] := v[i+127:i+96] - dst[i+63:i+32] := v[i+127:i+96] - dst[i+95:i+64] := v[i+127:i+96] - dst[i+127:i+96] := v[i+127:i+96] - ENDFOR -_MM_SWIZ_REG_DACB: - FOR j := 0 to 3 - i := j*128 - dst[i+31:i] := v[i+63:i+32] - dst[i+63:i+32] := v[i+95:i+64] - dst[i+95:i+64] := v[i+31:i] - dst[i+127:i+96] := v[i+127:i+96] - ENDFOR -ESAC -dst[MAX:512] := 0 +DEFINE INTERLEAVE_HIGH_QWORDS(src1[127:0], src2[127:0]) { + dst[63:0] := src1[127:64] + dst[127:64] := src2[127:64] + RETURN dst[127:0] +} +tmp_dst[127:0] := INTERLEAVE_HIGH_QWORDS(a[127:0], b[127:0]) +FOR j := 0 to 1 + i := j*64 + IF k[j] + dst[i+63:i] := tmp_dst[i+63:i] + ELSE + dst[i+63:i] := 0 + FI +ENDFOR +dst[MAX:128] := 0 +
immintrin.h
 - - Floating Point - KNCNI - Swizzle - - - Performs a swizzle transformation of each of the two groups of packed 4x double-precision (64-bit) floating-point elements in "v" using swizzle parameter "s", storing the results in "dst". + + Integer + AVX512VL + AVX512F + Miscellaneous + + + + + + Unpack and interleave 32-bit integers from the low half of each 128-bit lane in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -CASE s OF -_MM_SWIZ_REG_NONE: - dst[511:0] := v[511:0] -_MM_SWIZ_REG_DCBA: - dst[511:0] := v[511:0] -_MM_SWIZ_REG_CDAB: - FOR j := 0 to 3 - i := j*64 - dst[i+63:i] := v[i+127:i+64] - dst[i+127:i+64] := v[i+63:i] - ENDFOR -_MM_SWIZ_REG_BADC: - FOR j := 0 to 1 - i := j*256 - dst[i+63:i] := v[i+191:i+128] - dst[i+127:i+64] := v[i+255:i+192] - dst[i+191:i+128] := v[i+63:i] - dst[i+255:i+192] := v[i+127:i+64] - ENDFOR -_MM_SWIZ_REG_AAAA: - FOR j := 0 to 1 - i := j*256 - dst[i+63:i] := v[i+63:i] - dst[i+127:i+64] := v[i+63:i] - dst[i+191:i+128] := v[i+63:i] - dst[i+255:i+192] := v[i+63:i] - ENDFOR -_MM_SWIZ_REG_BBBB: - FOR j := 0 to 1 - i := j*256 - dst[i+63:i] := v[i+127:i+63] - dst[i+127:i+64] := v[i+127:i+63] - dst[i+191:i+128] := v[i+127:i+63] - dst[i+255:i+192] := v[i+127:i+63] - ENDFOR -_MM_SWIZ_REG_CCCC: - FOR j := 0 to 1 - i := j*256 - dst[i+63:i] := v[i+191:i+128] - dst[i+127:i+64] := v[i+191:i+128] - dst[i+191:i+128] := v[i+191:i+128] - dst[i+255:i+192] := v[i+191:i+128] - ENDFOR -_MM_SWIZ_REG_DDDD: - FOR j := 0 to 1 - i := j*256 - dst[i+63:i] := v[i+255:i+192] - dst[i+127:i+64] := v[i+255:i+192] - dst[i+191:i+128] := v[i+255:i+192] - dst[i+255:i+192] := v[i+255:i+192] - ENDFOR -_MM_SWIZ_REG_DACB: - FOR j := 0 to 1 - i := j*256 - dst[i+63:i] := v[i+127:i+64] - dst[i+127:i+64] := v[i+191:i+128] - dst[i+191:i+128] := v[i+63:i] - dst[i+255:i+192] := v[i+255:i+192] - ENDFOR -ESAC -dst[MAX:512] := 0 +DEFINE INTERLEAVE_DWORDS(src1[127:0], src2[127:0]) { + dst[31:0] := src1[31:0] + dst[63:32] := src2[31:0] + dst[95:64] := src1[63:32] + dst[127:96] := src2[63:32] + RETURN dst[127:0] +} +tmp_dst[127:0] := INTERLEAVE_DWORDS(a[127:0], b[127:0]) +tmp_dst[255:128] := INTERLEAVE_DWORDS(a[255:128], b[255:128]) +FOR j := 0 to 7 + i := j*32 + IF k[j] + dst[i+31:i] := tmp_dst[i+31:i] + ELSE + dst[i+31:i] := src[i+31:i] + FI +ENDFOR +dst[MAX:256] := 0 +
immintrin.h
 - + Integer - KNCNI - Swizzle - - - Performs a swizzle transformation of each of the four groups of packed 4x 32-bit integer elements in "v" using swizzle parameter "s", storing the results in "dst". + AVX512VL + AVX512F + Miscellaneous + + + + + Unpack and interleave 32-bit integers from the low half of each 128-bit lane in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -CASE s OF -_MM_SWIZ_REG_NONE: - dst[511:0] := v[511:0] -_MM_SWIZ_REG_DCBA: - dst[511:0] := v[511:0] -_MM_SWIZ_REG_CDAB: - FOR j := 0 to 7 - i := j*64 - dst[i+31:i] := v[i+63:i+32] - dst[i+63:i+32] := v[i+31:i] - ENDFOR -_MM_SWIZ_REG_BADC: - FOR j := 0 to 3 - i := j*128 - dst[i+31:i] := v[i+95:i+64] - dst[i+63:i+32] := v[i+127:i+96] - dst[i+95:i+64] := v[i+31:i] - dst[i+127:i+96] := v[i+63:i+32] - ENDFOR -_MM_SWIZ_REG_AAAA: - FOR j := 0 to 3 - i := j*128 - dst[i+31:i] := v[i+31:i] - dst[i+63:i+32] := v[i+31:i] - dst[i+95:i+64] := v[i+31:i] - dst[i+127:i+96] := v[i+31:i] - ENDFOR -_MM_SWIZ_REG_BBBB: - FOR j := 0 to 3 - i := j*128 - dst[i+31:i] := v[i+63:i+32] - dst[i+63:i+32] := v[i+63:i+32] - dst[i+95:i+64] := v[i+63:i+32] - dst[i+127:i+96] := v[i+63:i+32] - ENDFOR -_MM_SWIZ_REG_CCCC: - FOR j := 0 to 3 - i := j*128 - dst[i+31:i] := v[i+95:i+64] - dst[i+63:i+32] := v[i+95:i+64] - dst[i+95:i+64] := v[i+95:i+64] - dst[i+127:i+96] := v[i+95:i+64] - ENDFOR -_MM_SWIZ_REG_DDDD: - FOR j := 0 to 3 - i := j*128 - dst[i+31:i] := v[i+127:i+96] - dst[i+63:i+32] := v[i+127:i+96] - dst[i+95:i+64] := v[i+127:i+96] - dst[i+127:i+96] := v[i+127:i+96] - ENDFOR -_MM_SWIZ_REG_DACB: - FOR j := 0 to 3 - i := j*128 - dst[i+31:i] := v[i+63:i+32] - dst[i+63:i+32] := v[i+95:i+64] - dst[i+95:i+64] := v[i+31:i] - dst[i+127:i+96] := v[i+127:i+96] - ENDFOR -ESAC -dst[MAX:512] := 0 +DEFINE INTERLEAVE_DWORDS(src1[127:0], src2[127:0]) { + dst[31:0] := src1[31:0] + dst[63:32] := src2[31:0] + dst[95:64] := src1[63:32] + dst[127:96] := src2[63:32] + RETURN dst[127:0] +} +tmp_dst[127:0] := INTERLEAVE_DWORDS(a[127:0], b[127:0]) +tmp_dst[255:128] := INTERLEAVE_DWORDS(a[255:128], b[255:128]) +FOR j := 0 to 7 + i := j*32 + IF k[j] + dst[i+31:i] := tmp_dst[i+31:i] + ELSE + dst[i+31:i] := 0 + FI +ENDFOR +dst[MAX:256] := 0 +
immintrin.h
 - + Integer - KNCNI - Swizzle - - - Performs a swizzle transformation of each of the two groups of packed 4x64-bit integer elements in "v" using swizzle parameter "s", storing the results in "dst". + AVX512VL + AVX512F + Miscellaneous + + + + + + Unpack and interleave 32-bit integers from the low half of "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -CASE s OF -_MM_SWIZ_REG_NONE: - dst[511:0] := v[511:0] -_MM_SWIZ_REG_DCBA: - dst[511:0] := v[511:0] -_MM_SWIZ_REG_CDAB: - FOR j := 0 to 3 - i := j*64 - dst[i+63:i] := v[i+127:i+64] - dst[i+127:i+64] := v[i+63:i] - ENDFOR -_MM_SWIZ_REG_BADC: - FOR j := 0 to 1 - i := j*256 - dst[i+63:i] := v[i+191:i+128] - dst[i+127:i+64] := v[i+255:i+192] - dst[i+191:i+128] := v[i+63:i] - dst[i+255:i+192] := v[i+127:i+64] - ENDFOR -_MM_SWIZ_REG_AAAA: - FOR j := 0 to 1 - i := j*256 - dst[i+63:i] := v[i+63:i] - dst[i+127:i+64] := v[i+63:i] - dst[i+191:i+128] := v[i+63:i] - dst[i+255:i+192] := v[i+63:i] - ENDFOR -_MM_SWIZ_REG_BBBB: - FOR j := 0 to 1 - i := j*256 - dst[i+63:i] := v[i+127:i+63] - dst[i+127:i+64] := v[i+127:i+63] - dst[i+191:i+128] := v[i+127:i+63] - dst[i+255:i+192] := v[i+127:i+63] - ENDFOR -_MM_SWIZ_REG_CCCC: - FOR j := 0 to 1 - i := j*256 - dst[i+63:i] := v[i+191:i+128] - dst[i+127:i+64] := v[i+191:i+128] - dst[i+191:i+128] := v[i+191:i+128] - dst[i+255:i+192] := v[i+191:i+128] - ENDFOR -_MM_SWIZ_REG_DDDD: - FOR j := 0 to 1 - i := j*256 - dst[i+63:i] := v[i+255:i+192] - dst[i+127:i+64] := v[i+255:i+192] - dst[i+191:i+128] := v[i+255:i+192] - dst[i+255:i+192] := v[i+255:i+192] - ENDFOR -_MM_SWIZ_REG_DACB: - FOR j := 0 to 1 - i := j*256 - dst[i+63:i] := v[i+127:i+64] - dst[i+127:i+64] := v[i+191:i+128] - dst[i+191:i+128] := v[i+63:i] - dst[i+255:i+192] := v[i+255:i+192] - ENDFOR -ESAC -dst[MAX:512] := 0 +DEFINE INTERLEAVE_DWORDS(src1[127:0], src2[127:0]) { + dst[31:0] := src1[31:0] + dst[63:32] := src2[31:0] + dst[95:64] := src1[63:32] + dst[127:96] := src2[63:32] + RETURN dst[127:0] +} +tmp_dst[127:0] := INTERLEAVE_DWORDS(a[127:0], b[127:0]) +FOR j := 0 to 3 + i := j*32 + IF k[j] + dst[i+31:i] := tmp_dst[i+31:i] + ELSE + dst[i+31:i] := src[i+31:i] + FI +ENDFOR +dst[MAX:128] := 0 +
immintrin.h
 - - Floating Point - KNCNI - Swizzle - - - - - Performs a swizzle transformation of each of the four groups of packed 4x single-precision (32-bit) floating-point elements in "v" using swizzle parameter "s", storing the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Integer + AVX512VL + AVX512F + Miscellaneous + + + + + Unpack and interleave 32-bit integers from the low half of "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -CASE s OF -_MM_SWIZ_REG_NONE: - dst[511:0] := v[511:0] -_MM_SWIZ_REG_DCBA: - dst[511:0] := v[511:0] -_MM_SWIZ_REG_CDAB: - FOR j := 0 to 7 - i := j*64 - IF k[j*2] - dst[i+31:i] := v[i+63:i+32] - ELSE - dst[i+31:i] := src[i+31:i] - FI - IF k[j*2+1] - dst[i+63:i+32] := v[i+31:i] - ELSE - dst[i+63:i+32] := src[i+63:i+32] - FI - ENDFOR -_MM_SWIZ_REG_BADC: - FOR j := 0 to 3 - i := j*128 - IF k[j*4] - dst[i+31:i] := v[i+95:i+64] - ELSE - dst[i+31:i] := src[i+31:i] - FI - IF k[j*4+1] - dst[i+63:i+32] := v[i+127:i+96] - ELSE - dst[i+63:i+32] := src[i+63:i+32] - FI - IF k[j*4+2] - dst[i+95:i+64] := v[i+31:i] - ELSE - dst[i+95:i+64] := src[i+95:i+64] - FI - IF k[j*4+3] - dst[i+127:i+96] := v[i+63:i+32] - ELSE - dst[i+127:i+96] := src[i+127:i+96] - FI - ENDFOR -_MM_SWIZ_REG_AAAA: - FOR j := 0 to 3 - i := j*128 - IF k[j*4] - dst[i+31:i] := v[i+31:i] - ELSE - dst[i+31:i] := src[i+31:i] - FI - IF k[j*4+1] - dst[i+63:i+32] := v[i+31:i] - ELSE - dst[i+63:i+32] := src[i+63:i+32] - FI - IF k[j*4+2] - dst[i+95:i+64] := v[i+31:i] - ELSE - dst[i+95:i+64] := src[i+95:i+64] - FI - IF k[j*4+3] - dst[i+127:i+96] := v[i+31:i] - ELSE - dst[i+127:i+96] := src[i+127:i+96] - FI - ENDFOR -_MM_SWIZ_REG_BBBB: - FOR j := 0 to 3 - i := j*128 - IF k[j*4] - dst[i+31:i] := v[i+63:i+32] - ELSE - dst[i+31:i] := src[i+31:i] - FI - IF k[j*4+1] - dst[i+63:i+32] := v[i+63:i+32] - ELSE - dst[i+63:i+32] := src[i+63:i+32] - FI - IF k[j*4+2] - dst[i+95:i+64] := v[i+63:i+32] - ELSE - dst[i+95:i+64] := src[i+95:i+64] - FI - IF k[j*4+3] - dst[i+127:i+96] := v[i+63:i+32] - ELSE - dst[i+127:i+96] := src[i+127:i+96] - FI - ENDFOR -_MM_SWIZ_REG_CCCC: - FOR j := 0 to 3 - i := j*128 - IF k[j*4] - dst[i+31:i] := v[i+95:i+64] - ELSE - dst[i+31:i] := src[i+31:i] - FI - IF k[j*4+1] - dst[i+63:i+32] := v[i+95:i+64] - ELSE - dst[i+63:i+32] := src[i+63:i+32] - FI - IF k[j*4+2] - dst[i+95:i+64] := v[i+95:i+64] - ELSE - dst[i+95:i+64] := src[i+95:i+64] - FI - IF k[j*4+3] - dst[i+127:i+96] := v[i+95:i+64] - ELSE - dst[i+127:i+96] := src[i+127:i+96] - FI - ENDFOR -_MM_SWIZ_REG_DDDD: - FOR j := 0 to 3 - i := j*128 - IF k[j*4] - dst[i+31:i] := v[i+127:i+96] - ELSE - dst[i+31:i] := src[i+31:i] - FI - IF k[j*4+1] - dst[i+63:i+32] := v[i+127:i+96] - ELSE - dst[i+63:i+32] := src[i+63:i+32] - FI - IF k[j*4+2] - dst[i+95:i+64] := v[i+127:i+96] - ELSE - dst[i+95:i+64] := src[i+95:i+64] - FI - IF k[j*4+3] - dst[i+127:i+96] := v[i+127:i+96] - ELSE - dst[i+127:i+96] := src[i+127:i+96] - FI - ENDFOR -_MM_SWIZ_REG_DACB: - FOR j := 0 to 3 - i := j*128 - IF k[j*4] - dst[i+31:i] := v[i+63:i+32] - ELSE - dst[i+31:i] := src[i+31:i] - FI - IF k[j*4+1] - dst[i+63:i+32] := v[i+95:i+64] - ELSE - dst[i+63:i+32] := src[i+63:i+32] - FI - IF k[j*4+2] - dst[i+95:i+64] := v[i+31:i] - ELSE - dst[i+95:i+64] := src[i+95:i+64] - FI - IF k[j*4+3] - dst[i+127:i+96] := v[i+127:i+96] - ELSE - dst[i+127:i+96] := src[i+127:i+96] - FI - ENDFOR -ESAC -dst[MAX:512] := 0 +DEFINE INTERLEAVE_DWORDS(src1[127:0], src2[127:0]) { + dst[31:0] := src1[31:0] + dst[63:32] := src2[31:0] + dst[95:64] := src1[63:32] + dst[127:96] := src2[63:32] + RETURN dst[127:0] +} +tmp_dst[127:0] := INTERLEAVE_DWORDS(a[127:0], b[127:0]) +FOR j := 0 to 3 + i := j*32 + IF k[j] + dst[i+31:i] := tmp_dst[i+31:i] + ELSE + dst[i+31:i] := 0 + FI +ENDFOR +dst[MAX:128] := 0 +
immintrin.h
 - - Floating Point - KNCNI - Swizzle - - - - - Performs a swizzle transformation of each of the two groups of packed 4x double-precision (64-bit) floating-point elements in "v" using swizzle parameter "s", storing the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Integer + AVX512VL + AVX512F + Miscellaneous + + + + + + Unpack and interleave 64-bit integers from the low half of each 128-bit lane in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +DEFINE INTERLEAVE_QWORDS(src1[127:0], src2[127:0]) { + dst[63:0] := src1[63:0] + dst[127:64] := src2[63:0] + RETURN dst[127:0] +} +tmp_dst[127:0] := INTERLEAVE_QWORDS(a[127:0], b[127:0]) +tmp_dst[255:128] := INTERLEAVE_QWORDS(a[255:128], b[255:128]) +FOR j := 0 to 3 + i := j*64 + IF k[j] + dst[i+63:i] := tmp_dst[i+63:i] + ELSE + dst[i+63:i] := src[i+63:i] + FI +ENDFOR +dst[MAX:256] := 0 + + +
immintrin.h
+ + + Integer + AVX512VL + AVX512F + Miscellaneous + + + + + Unpack and interleave 64-bit integers from the low half of each 128-bit lane in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + +DEFINE INTERLEAVE_QWORDS(src1[127:0], src2[127:0]) { + dst[63:0] := src1[63:0] + dst[127:64] := src2[63:0] + RETURN dst[127:0] +} +tmp_dst[127:0] := INTERLEAVE_QWORDS(a[127:0], b[127:0]) +tmp_dst[255:128] := INTERLEAVE_QWORDS(a[255:128], b[255:128]) +FOR j := 0 to 3 + i := j*64 + IF k[j] + dst[i+63:i] := tmp_dst[i+63:i] + ELSE + dst[i+63:i] := 0 + FI +ENDFOR +dst[MAX:256] := 0 + + +
immintrin.h
+ + + Integer + AVX512VL + AVX512F + Miscellaneous + + + + + + Unpack and interleave 64-bit integers from the low half of "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +DEFINE INTERLEAVE_QWORDS(src1[127:0], src2[127:0]) { + dst[63:0] := src1[63:0] + dst[127:64] := src2[63:0] + RETURN dst[127:0] +} +tmp_dst[127:0] := INTERLEAVE_QWORDS(a[127:0], b[127:0]) +FOR j := 0 to 1 + i := j*64 + IF k[j] + dst[i+63:i] := tmp_dst[i+63:i] + ELSE + dst[i+63:i] := src[i+63:i] + FI +ENDFOR +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Integer + AVX512VL + AVX512F + Miscellaneous + + + + + Unpack and interleave 64-bit integers from the low half of "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + +DEFINE INTERLEAVE_QWORDS(src1[127:0], src2[127:0]) { + dst[63:0] := src1[63:0] + dst[127:64] := src2[63:0] + RETURN dst[127:0] +} +tmp_dst[127:0] := INTERLEAVE_QWORDS(a[127:0], b[127:0]) +FOR j := 0 to 1 + i := j*64 + IF k[j] + dst[i+63:i] := tmp_dst[i+63:i] + ELSE + dst[i+63:i] := 0 + FI +ENDFOR +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Integer + AVX512VL + AVX512F + Logical + + + + + + Compute the bitwise XOR of packed 32-bit integers in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +FOR j := 0 to 7 + i := j*32 + IF k[j] + dst[i+31:i] := a[i+31:i] XOR b[i+31:i] + ELSE + dst[i+31:i] := src[i+31:i] + FI +ENDFOR +dst[MAX:256] := 0 + + +
immintrin.h
+ + + Integer + AVX512VL + AVX512F + Logical + + + + + Compute the bitwise XOR of packed 32-bit integers in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + +FOR j := 0 to 7 + i := j*32 + IF k[j] + dst[i+31:i] := a[i+31:i] XOR b[i+31:i] + ELSE + dst[i+31:i] := 0 + FI +ENDFOR +dst[MAX:256] := 0 + + +
immintrin.h
+ + + Integer + AVX512VL + AVX512F + Logical + + + + + + Compute the bitwise XOR of packed 32-bit integers in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +FOR j := 0 to 3 + i := j*32 + IF k[j] + dst[i+31:i] := a[i+31:i] XOR b[i+31:i] + ELSE + dst[i+31:i] := src[i+31:i] + FI +ENDFOR +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Integer + AVX512VL + AVX512F + Logical + + + + + Compute the bitwise XOR of packed 32-bit integers in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + +FOR j := 0 to 3 + i := j*32 + IF k[j] + dst[i+31:i] := a[i+31:i] XOR b[i+31:i] + ELSE + dst[i+31:i] := 0 + FI +ENDFOR +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Integer + AVX512VL + AVX512F + Logical + + + + + + Compute the bitwise XOR of packed 64-bit integers in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +FOR j := 0 to 3 + i := j*64 + IF k[j] + dst[i+63:i] := a[i+63:i] XOR b[i+63:i] + ELSE + dst[i+63:i] := src[i+63:i] + FI +ENDFOR +dst[MAX:256] := 0 + + +
immintrin.h
+ + + Integer + AVX512VL + AVX512F + Logical + + + + + Compute the bitwise XOR of packed 64-bit integers in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + +FOR j := 0 to 3 + i := j*64 + IF k[j] + dst[i+63:i] := a[i+63:i] XOR b[i+63:i] + ELSE + dst[i+63:i] := 0 + FI +ENDFOR +dst[MAX:256] := 0 + + +
immintrin.h
+ + + Integer + AVX512VL + AVX512F + Logical + + + + + + Compute the bitwise XOR of packed 64-bit integers in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +FOR j := 0 to 1 + i := j*64 + IF k[j] + dst[i+63:i] := a[i+63:i] XOR b[i+63:i] + ELSE + dst[i+63:i] := src[i+63:i] + FI +ENDFOR +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Integer + AVX512VL + AVX512F + Logical + + + + + Compute the bitwise XOR of packed 64-bit integers in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + +FOR j := 0 to 1 + i := j*64 + IF k[j] + dst[i+63:i] := a[i+63:i] XOR b[i+63:i] + ELSE + dst[i+63:i] := 0 + FI +ENDFOR +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512VL + AVX512F + Arithmetic + + + + + Compute the approximate reciprocal of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). The maximum relative error for this approximation is less than 2^-14. + +FOR j := 0 to 3 + i := j*64 + IF k[j] + dst[i+63:i] := (1.0 / a[i+63:i]) + ELSE + dst[i+63:i] := src[i+63:i] + FI +ENDFOR +dst[MAX:256] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512VL + AVX512F + Arithmetic + + + + Compute the approximate reciprocal of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). The maximum relative error for this approximation is less than 2^-14. + +FOR j := 0 to 3 + i := j*64 + IF k[j] + dst[i+63:i] := (1.0 / a[i+63:i]) + ELSE + dst[i+63:i] := 0 + FI +ENDFOR +dst[MAX:256] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512VL + AVX512F + Arithmetic + + + Compute the approximate reciprocal of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst". The maximum relative error for this approximation is less than 2^-14. + +FOR j := 0 to 3 + i := j*64 + dst[i+63:i] := (1.0 / a[i+63:i]) +ENDFOR +dst[MAX:256] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512VL + AVX512F + Arithmetic + + + + + Compute the approximate reciprocal of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). The maximum relative error for this approximation is less than 2^-14. + +FOR j := 0 to 1 + i := j*64 + IF k[j] + dst[i+63:i] := (1.0 / a[i+63:i]) + ELSE + dst[i+63:i] := src[i+63:i] + FI +ENDFOR +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512VL + AVX512F + Arithmetic + + + + Compute the approximate reciprocal of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). The maximum relative error for this approximation is less than 2^-14. + +FOR j := 0 to 1 + i := j*64 + IF k[j] + dst[i+63:i] := (1.0 / a[i+63:i]) + ELSE + dst[i+63:i] := 0 + FI +ENDFOR +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512VL + AVX512F + Arithmetic + + + Compute the approximate reciprocal of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst". The maximum relative error for this approximation is less than 2^-14. + +FOR j := 0 to 1 + i := j*64 + dst[i+63:i] := (1.0 / a[i+63:i]) +ENDFOR +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512VL + AVX512F + Arithmetic + + + + + Compute the approximate reciprocal of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). The maximum relative error for this approximation is less than 2^-14. + +FOR j := 0 to 7 + i := j*32 + IF k[j] + dst[i+31:i] := (1.0 / a[i+31:i]) + ELSE + dst[i+31:i] := src[i+31:i] + FI +ENDFOR +dst[MAX:256] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512VL + AVX512F + Arithmetic + + + + Compute the approximate reciprocal of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). The maximum relative error for this approximation is less than 2^-14. + +FOR j := 0 to 7 + i := j*32 + IF k[j] + dst[i+31:i] := (1.0 / a[i+31:i]) + ELSE + dst[i+31:i] := 0 + FI +ENDFOR +dst[MAX:256] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512VL + AVX512F + Arithmetic + + + Compute the approximate reciprocal of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst". The maximum relative error for this approximation is less than 2^-14. + +FOR j := 0 to 7 + i := j*32 + dst[i+31:i] := (1.0 / a[i+31:i]) +ENDFOR +dst[MAX:256] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512VL + AVX512F + Arithmetic + + + + + Compute the approximate reciprocal of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). The maximum relative error for this approximation is less than 2^-14. + +FOR j := 0 to 3 + i := j*32 + IF k[j] + dst[i+31:i] := (1.0 / a[i+31:i]) + ELSE + dst[i+31:i] := src[i+31:i] + FI +ENDFOR +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512VL + AVX512F + Arithmetic + + + + Compute the approximate reciprocal of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). The maximum relative error for this approximation is less than 2^-14. + +FOR j := 0 to 3 + i := j*32 + IF k[j] + dst[i+31:i] := (1.0 / a[i+31:i]) + ELSE + dst[i+31:i] := 0 + FI +ENDFOR +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512VL + AVX512F + Arithmetic + + + Compute the approximate reciprocal of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst". The maximum relative error for this approximation is less than 2^-14. + +FOR j := 0 to 3 + i := j*32 + dst[i+31:i] := (1.0 / a[i+31:i]) +ENDFOR +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512VL + AVX512F + Miscellaneous + + + + + + Round packed double-precision (64-bit) floating-point elements in "a" to the number of fraction bits specified by "imm8", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). [round_imm_note] + +DEFINE RoundScaleFP64(src1[63:0], imm8[7:0]) { + m[63:0] := FP64(imm8[7:4]) // number of fraction bits after the binary point to be preserved + tmp[63:0] := POW(2.0, -m) * ROUND(POW(2.0, m) * src1[63:0], imm8[3:0]) + IF IsInf(tmp[63:0]) + tmp[63:0] := src1[63:0] + FI + RETURN tmp[63:0] +} +FOR j := 0 to 3 + i := j*64 + IF k[j] + dst[i+63:i] := RoundScaleFP64(a[i+63:i], imm8[7:0]) + ELSE + dst[i+63:i] := src[i+63:i] + FI +ENDFOR +dst[MAX:256] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512VL + AVX512F + Miscellaneous + + + + + Round packed double-precision (64-bit) floating-point elements in "a" to the number of fraction bits specified by "imm8", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). [round_imm_note] + +DEFINE RoundScaleFP64(src1[63:0], imm8[7:0]) { + m[63:0] := FP64(imm8[7:4]) // number of fraction bits after the binary point to be preserved + tmp[63:0] := POW(2.0, -m) * ROUND(POW(2.0, m) * src1[63:0], imm8[3:0]) + IF IsInf(tmp[63:0]) + tmp[63:0] := src1[63:0] + FI + RETURN tmp[63:0] +} +FOR j := 0 to 3 + i := j*64 + IF k[j] + dst[i+63:i] := RoundScaleFP64(a[i+63:i], imm8[7:0]) + ELSE + dst[i+63:i] := 0 + FI +ENDFOR +dst[MAX:256] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512VL + AVX512F + Miscellaneous + + + + Round packed double-precision (64-bit) floating-point elements in "a" to the number of fraction bits specified by "imm8", and store the results in "dst". [round_imm_note] + +DEFINE RoundScaleFP64(src1[63:0], imm8[7:0]) { + m[63:0] := FP64(imm8[7:4]) // number of fraction bits after the binary point to be preserved + tmp[63:0] := POW(2.0, -m) * ROUND(POW(2.0, m) * src1[63:0], imm8[3:0]) + IF IsInf(tmp[63:0]) + tmp[63:0] := src1[63:0] + FI + RETURN tmp[63:0] +} +FOR j := 0 to 3 + i := j*64 + dst[i+63:i] := RoundScaleFP64(a[i+63:i], imm8[7:0]) +ENDFOR +dst[MAX:256] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512VL + AVX512F + Miscellaneous + + + + + + Round packed double-precision (64-bit) floating-point elements in "a" to the number of fraction bits specified by "imm8", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). [round_imm_note] + +DEFINE RoundScaleFP64(src1[63:0], imm8[7:0]) { + m[63:0] := FP64(imm8[7:4]) // number of fraction bits after the binary point to be preserved + tmp[63:0] := POW(2.0, -m) * ROUND(POW(2.0, m) * src1[63:0], imm8[3:0]) + IF IsInf(tmp[63:0]) + tmp[63:0] := src1[63:0] + FI + RETURN tmp[63:0] +} +FOR j := 0 to 1 + i := j*64 + IF k[j] + dst[i+63:i] := RoundScaleFP64(a[i+63:i], imm8[7:0]) + ELSE + dst[i+63:i] := src[i+63:i] + FI +ENDFOR +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512VL + AVX512F + Miscellaneous + + + + + Round packed double-precision (64-bit) floating-point elements in "a" to the number of fraction bits specified by "imm8", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). [round_imm_note] + +DEFINE RoundScaleFP64(src1[63:0], imm8[7:0]) { + m[63:0] := FP64(imm8[7:4]) // number of fraction bits after the binary point to be preserved + tmp[63:0] := POW(2.0, -m) * ROUND(POW(2.0, m) * src1[63:0], imm8[3:0]) + IF IsInf(tmp[63:0]) + tmp[63:0] := src1[63:0] + FI + RETURN tmp[63:0] +} +FOR j := 0 to 1 + i := j*64 + IF k[j] + dst[i+63:i] := RoundScaleFP64(a[i+63:i], imm8[7:0]) + ELSE + dst[i+63:i] := 0 + FI +ENDFOR +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512VL + AVX512F + Miscellaneous + + + + Round packed double-precision (64-bit) floating-point elements in "a" to the number of fraction bits specified by "imm8", and store the results in "dst". [round_imm_note] + +DEFINE RoundScaleFP64(src1[63:0], imm8[7:0]) { + m[63:0] := FP64(imm8[7:4]) // number of fraction bits after the binary point to be preserved + tmp[63:0] := POW(2.0, -m) * ROUND(POW(2.0, m) * src1[63:0], imm8[3:0]) + IF IsInf(tmp[63:0]) + tmp[63:0] := src1[63:0] + FI + RETURN tmp[63:0] +} +FOR j := 0 to 1 + i := j*64 + dst[i+63:i] := RoundScaleFP64(a[i+63:i], imm8[7:0]) +ENDFOR +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512VL + AVX512F + Miscellaneous + + + + + + Round packed single-precision (32-bit) floating-point elements in "a" to the number of fraction bits specified by "imm8", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). [round_imm_note] + +DEFINE RoundScaleFP32(src1[31:0], imm8[7:0]) { + m[31:0] := FP32(imm8[7:4]) // number of fraction bits after the binary point to be preserved + tmp[31:0] := POW(FP32(2.0), -m) * ROUND(POW(FP32(2.0), m) * src1[31:0], imm8[3:0]) + IF IsInf(tmp[31:0]) + tmp[31:0] := src1[31:0] + FI + RETURN tmp[31:0] +} +FOR j := 0 to 7 + i := j*32 + IF k[j] + dst[i+31:i] := RoundScaleFP32(a[i+31:i], imm8[7:0]) + ELSE + dst[i+31:i] := src[i+31:i] + FI +ENDFOR +dst[MAX:256] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512VL + AVX512F + Miscellaneous + + + + + Round packed single-precision (32-bit) floating-point elements in "a" to the number of fraction bits specified by "imm8", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). [round_imm_note] + +DEFINE RoundScaleFP32(src1[31:0], imm8[7:0]) { + m[31:0] := FP32(imm8[7:4]) // number of fraction bits after the binary point to be preserved + tmp[31:0] := POW(FP32(2.0), -m) * ROUND(POW(FP32(2.0), m) * src1[31:0], imm8[3:0]) + IF IsInf(tmp[31:0]) + tmp[31:0] := src1[31:0] + FI + RETURN tmp[31:0] +} +FOR j := 0 to 7 + i := j*32 + IF k[j] + dst[i+31:i] := RoundScaleFP32(a[i+31:i], imm8[7:0]) + ELSE + dst[i+31:i] := 0 + FI +ENDFOR +dst[MAX:256] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512VL + AVX512F + Miscellaneous + + + + Round packed single-precision (32-bit) floating-point elements in "a" to the number of fraction bits specified by "imm8", and store the results in "dst". [round_imm_note] + +DEFINE RoundScaleFP32(src1[31:0], imm8[7:0]) { + m[31:0] := FP32(imm8[7:4]) // number of fraction bits after the binary point to be preserved + tmp[31:0] := POW(FP32(2.0), -m) * ROUND(POW(FP32(2.0), m) * src1[31:0], imm8[3:0]) + IF IsInf(tmp[31:0]) + tmp[31:0] := src1[31:0] + FI + RETURN tmp[31:0] +} +FOR j := 0 to 7 + i := j*32 + dst[i+31:i] := RoundScaleFP32(a[i+31:i], imm8[7:0]) +ENDFOR +dst[MAX:256] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512VL + AVX512F + Miscellaneous + + + + + + Round packed single-precision (32-bit) floating-point elements in "a" to the number of fraction bits specified by "imm8", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). [round_imm_note] + +DEFINE RoundScaleFP32(src1[31:0], imm8[7:0]) { + m[31:0] := FP32(imm8[7:4]) // number of fraction bits after the binary point to be preserved + tmp[31:0] := POW(FP32(2.0), -m) * ROUND(POW(FP32(2.0), m) * src1[31:0], imm8[3:0]) + IF IsInf(tmp[31:0]) + tmp[31:0] := src1[31:0] + FI + RETURN tmp[31:0] +} +FOR j := 0 to 3 + i := j*32 + IF k[j] + dst[i+31:i] := RoundScaleFP32(a[i+31:i], imm8[7:0]) + ELSE + dst[i+31:i] := src[i+31:i] + FI +ENDFOR +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512VL + AVX512F + Miscellaneous + + + + + Round packed single-precision (32-bit) floating-point elements in "a" to the number of fraction bits specified by "imm8", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). [round_imm_note] + +DEFINE RoundScaleFP32(src1[31:0], imm8[7:0]) { + m[31:0] := FP32(imm8[7:4]) // number of fraction bits after the binary point to be preserved + tmp[31:0] := POW(FP32(2.0), -m) * ROUND(POW(FP32(2.0), m) * src1[31:0], imm8[3:0]) + IF IsInf(tmp[31:0]) + tmp[31:0] := src1[31:0] + FI + RETURN tmp[31:0] +} +FOR j := 0 to 3 + i := j*32 + IF k[j] + dst[i+31:i] := RoundScaleFP32(a[i+31:i], imm8[7:0]) + ELSE + dst[i+31:i] := 0 + FI +ENDFOR +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512VL + AVX512F + Miscellaneous + + + + Round packed single-precision (32-bit) floating-point elements in "a" to the number of fraction bits specified by "imm8", and store the results in "dst". [round_imm_note] + +DEFINE RoundScaleFP32(src1[31:0], imm8[7:0]) { + m[31:0] := FP32(imm8[7:4]) // number of fraction bits after the binary point to be preserved + tmp[31:0] := POW(FP32(2.0), -m) * ROUND(POW(FP32(2.0), m) * src1[31:0], imm8[3:0]) + IF IsInf(tmp[31:0]) + tmp[31:0] := src1[31:0] + FI + RETURN tmp[31:0] +} +FOR j := 0 to 3 + i := j*32 + dst[i+31:i] := RoundScaleFP32(a[i+31:i], imm8[7:0]) +ENDFOR +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512VL + AVX512F + Arithmetic + + + + + Compute the approximate reciprocal square root of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). The maximum relative error for this approximation is less than 2^-14. + +FOR j := 0 to 3 + i := j*64 + IF k[j] + dst[i+63:i] := (1.0 / SQRT(a[i+63:i])) + ELSE + dst[i+63:i] := src[i+63:i] + FI +ENDFOR +dst[MAX:256] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512VL + AVX512F + Arithmetic + + + + Compute the approximate reciprocal square root of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). The maximum relative error for this approximation is less than 2^-14. + +FOR j := 0 to 3 + i := j*64 + IF k[j] + dst[i+63:i] := (1.0 / SQRT(a[i+63:i])) + ELSE + dst[i+63:i] := 0 + FI +ENDFOR +dst[MAX:256] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512VL + AVX512F + Arithmetic + + + + + Compute the approximate reciprocal square root of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). The maximum relative error for this approximation is less than 2^-14. + +FOR j := 0 to 1 + i := j*64 + IF k[j] + dst[i+63:i] := (1.0 / SQRT(a[i+63:i])) + ELSE + dst[i+63:i] := src[i+63:i] + FI +ENDFOR +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512VL + AVX512F + Arithmetic + + + + Compute the approximate reciprocal square root of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). The maximum relative error for this approximation is less than 2^-14. + +FOR j := 0 to 1 + i := j*64 + IF k[j] + dst[i+63:i] := (1.0 / SQRT(a[i+63:i])) + ELSE + dst[i+63:i] := 0 + FI +ENDFOR +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512VL + AVX512F + Arithmetic + + + + + Compute the approximate reciprocal square root of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). The maximum relative error for this approximation is less than 2^-14. + +FOR j := 0 to 7 + i := j*32 + IF k[j] + dst[i+31:i] := (1.0 / SQRT(a[i+31:i])) + ELSE + dst[i+31:i] := src[i+31:i] + FI +ENDFOR +dst[MAX:256] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512VL + AVX512F + Arithmetic + + + + Compute the approximate reciprocal square root of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). The maximum relative error for this approximation is less than 2^-14. + +FOR j := 0 to 7 + i := j*32 + IF k[j] + dst[i+31:i] := (1.0 / SQRT(a[i+31:i])) + ELSE + dst[i+31:i] := 0 + FI +ENDFOR +dst[MAX:256] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512VL + AVX512F + Arithmetic + + + + + Compute the approximate reciprocal square root of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). The maximum relative error for this approximation is less than 2^-14. + +FOR j := 0 to 3 + i := j*32 + IF k[j] + dst[i+31:i] := (1.0 / SQRT(a[i+31:i])) + ELSE + dst[i+31:i] := src[i+31:i] + FI +ENDFOR +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512VL + AVX512F + Arithmetic + + + + Compute the approximate reciprocal square root of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). The maximum relative error for this approximation is less than 2^-14. + +FOR j := 0 to 3 + i := j*32 + IF k[j] + dst[i+31:i] := (1.0 / SQRT(a[i+31:i])) + ELSE + dst[i+31:i] := 0 + FI +ENDFOR +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512VL + AVX512F + Miscellaneous + + + + + + Scale the packed double-precision (64-bit) floating-point elements in "a" using values from "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + DEFINE SCALE(src1, src2) { + IF (src2 == NaN) + IF (src2 == SNaN) + RETURN QNAN(src2) + FI + ELSE IF (src1 == NaN) + IF (src1 == SNaN) + RETURN QNAN(src1) + FI + IF (src2 != INF) + RETURN QNAN(src1) + FI + ELSE + tmp_src2 := src2 + tmp_src1 := src1 + IF (IS_DENORMAL(src2) AND MXCSR.DAZ) + tmp_src2 := 0 + FI + IF (IS_DENORMAL(src1) AND MXCSR.DAZ) + tmp_src1 := 0 + FI + FI + dst[63:0] := tmp_src1[63:0] * POW(2.0, FLOOR(tmp_src2[63:0])) + RETURN dst[63:0] +} +FOR j := 0 to 3 + i := j*64 + IF k[j] + dst[i+63:i] := SCALE(a[i+63:0], b[i+63:i]) + ELSE + dst[i+63:i] := src[i+63:i] + FI +ENDFOR +dst[MAX:256] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512VL + AVX512F + Miscellaneous + + + + + Scale the packed double-precision (64-bit) floating-point elements in "a" using values from "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + DEFINE SCALE(src1, src2) { + IF (src2 == NaN) + IF (src2 == SNaN) + RETURN QNAN(src2) + FI + ELSE IF (src1 == NaN) + IF (src1 == SNaN) + RETURN QNAN(src1) + FI + IF (src2 != INF) + RETURN QNAN(src1) + FI + ELSE + tmp_src2 := src2 + tmp_src1 := src1 + IF (IS_DENORMAL(src2) AND MXCSR.DAZ) + tmp_src2 := 0 + FI + IF (IS_DENORMAL(src1) AND MXCSR.DAZ) + tmp_src1 := 0 + FI + FI + dst[63:0] := tmp_src1[63:0] * POW(2.0, FLOOR(tmp_src2[63:0])) + RETURN dst[63:0] +} +FOR j := 0 to 3 + i := j*64 + IF k[j] + dst[i+63:i] := SCALE(a[i+63:0], b[i+63:i]) + ELSE + dst[i+63:i] := 0 + FI +ENDFOR +dst[MAX:256] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512VL + AVX512F + Miscellaneous + + + + Scale the packed double-precision (64-bit) floating-point elements in "a" using values from "b", and store the results in "dst". + DEFINE SCALE(src1, src2) { + IF (src2 == NaN) + IF (src2 == SNaN) + RETURN QNAN(src2) + FI + ELSE IF (src1 == NaN) + IF (src1 == SNaN) + RETURN QNAN(src1) + FI + IF (src2 != INF) + RETURN QNAN(src1) + FI + ELSE + tmp_src2 := src2 + tmp_src1 := src1 + IF (IS_DENORMAL(src2) AND MXCSR.DAZ) + tmp_src2 := 0 + FI + IF (IS_DENORMAL(src1) AND MXCSR.DAZ) + tmp_src1 := 0 + FI + FI + dst[63:0] := tmp_src1[63:0] * POW(2.0, FLOOR(tmp_src2[63:0])) + RETURN dst[63:0] +} +FOR j := 0 to 3 + i := j*64 + dst[i+63:i] := SCALE(a[i+63:0], b[i+63:i]) +ENDFOR +dst[MAX:256] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512VL + AVX512F + Miscellaneous + + + + + + Scale the packed double-precision (64-bit) floating-point elements in "a" using values from "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + DEFINE SCALE(src1, src2) { + IF (src2 == NaN) + IF (src2 == SNaN) + RETURN QNAN(src2) + FI + ELSE IF (src1 == NaN) + IF (src1 == SNaN) + RETURN QNAN(src1) + FI + IF (src2 != INF) + RETURN QNAN(src1) + FI + ELSE + tmp_src2 := src2 + tmp_src1 := src1 + IF (IS_DENORMAL(src2) AND MXCSR.DAZ) + tmp_src2 := 0 + FI + IF (IS_DENORMAL(src1) AND MXCSR.DAZ) + tmp_src1 := 0 + FI + FI + dst[63:0] := tmp_src1[63:0] * POW(2.0, FLOOR(tmp_src2[63:0])) + RETURN dst[63:0] +} +FOR j := 0 to 1 + i := j*64 + IF k[j] + dst[i+63:i] := SCALE(a[i+63:0], b[i+63:i]) + ELSE + dst[i+63:i] := src[i+63:i] + FI +ENDFOR +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512VL + AVX512F + Miscellaneous + + + + + Scale the packed double-precision (64-bit) floating-point elements in "a" using values from "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + DEFINE SCALE(src1, src2) { + IF (src2 == NaN) + IF (src2 == SNaN) + RETURN QNAN(src2) + FI + ELSE IF (src1 == NaN) + IF (src1 == SNaN) + RETURN QNAN(src1) + FI + IF (src2 != INF) + RETURN QNAN(src1) + FI + ELSE + tmp_src2 := src2 + tmp_src1 := src1 + IF (IS_DENORMAL(src2) AND MXCSR.DAZ) + tmp_src2 := 0 + FI + IF (IS_DENORMAL(src1) AND MXCSR.DAZ) + tmp_src1 := 0 + FI + FI + dst[63:0] := tmp_src1[63:0] * POW(2.0, FLOOR(tmp_src2[63:0])) + RETURN dst[63:0] +} +FOR j := 0 to 1 + i := j*64 + IF k[j] + dst[i+63:i] := SCALE(a[i+63:0], b[i+63:i]) + ELSE + dst[i+63:i] := 0 + FI +ENDFOR +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512VL + AVX512F + Miscellaneous + + + + Scale the packed double-precision (64-bit) floating-point elements in "a" using values from "b", and store the results in "dst". + DEFINE SCALE(src1, src2) { + IF (src2 == NaN) + IF (src2 == SNaN) + RETURN QNAN(src2) + FI + ELSE IF (src1 == NaN) + IF (src1 == SNaN) + RETURN QNAN(src1) + FI + IF (src2 != INF) + RETURN QNAN(src1) + FI + ELSE + tmp_src2 := src2 + tmp_src1 := src1 + IF (IS_DENORMAL(src2) AND MXCSR.DAZ) + tmp_src2 := 0 + FI + IF (IS_DENORMAL(src1) AND MXCSR.DAZ) + tmp_src1 := 0 + FI + FI + dst[63:0] := tmp_src1[63:0] * POW(2.0, FLOOR(tmp_src2[63:0])) + RETURN dst[63:0] +} +FOR j := 0 to 1 + i := j*64 + dst[i+63:i] := SCALE(a[i+63:0], b[i+63:i]) +ENDFOR +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512VL + AVX512F + Miscellaneous + + + + + + Scale the packed single-precision (32-bit) floating-point elements in "a" using values from "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + DEFINE SCALE(src1, src2) { + IF (src2 == NaN) + IF (src2 == SNaN) + RETURN QNAN(src2) + FI + ELSE IF (src1 == NaN) + IF (src1 == SNaN) + RETURN QNAN(src1) + FI + IF (src2 != INF) + RETURN QNAN(src1) + FI + ELSE + tmp_src2 := src2 + tmp_src1 := src1 + IF (IS_DENORMAL(src2) AND MXCSR.DAZ) + tmp_src2 := 0 + FI + IF (IS_DENORMAL(src1) AND MXCSR.DAZ) + tmp_src1 := 0 + FI + FI + dst[31:0] := tmp_src1[31:0] * POW(2.0, FLOOR(tmp_src2[31:0])) + RETURN dst[31:0] +} +FOR j := 0 to 7 + i := j*32 + IF k[j] + dst[i+31:i] := SCALE(a[i+31:0], b[i+31:i]) + ELSE + dst[i+31:i] := src[i+31:i] + FI +ENDFOR +dst[MAX:256] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512VL + AVX512F + Miscellaneous + + + + + Scale the packed single-precision (32-bit) floating-point elements in "a" using values from "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + DEFINE SCALE(src1, src2) { + IF (src2 == NaN) + IF (src2 == SNaN) + RETURN QNAN(src2) + FI + ELSE IF (src1 == NaN) + IF (src1 == SNaN) + RETURN QNAN(src1) + FI + IF (src2 != INF) + RETURN QNAN(src1) + FI + ELSE + tmp_src2 := src2 + tmp_src1 := src1 + IF (IS_DENORMAL(src2) AND MXCSR.DAZ) + tmp_src2 := 0 + FI + IF (IS_DENORMAL(src1) AND MXCSR.DAZ) + tmp_src1 := 0 + FI + FI + dst[31:0] := tmp_src1[31:0] * POW(2.0, FLOOR(tmp_src2[31:0])) + RETURN dst[31:0] +} +FOR j := 0 to 7 + i := j*32 + IF k[j] + dst[i+31:i] := SCALE(a[i+31:0], b[i+31:i]) + ELSE + dst[i+31:i] := 0 + FI +ENDFOR +dst[MAX:256] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512VL + AVX512F + Miscellaneous + + + + Scale the packed single-precision (32-bit) floating-point elements in "a" using values from "b", and store the results in "dst". + DEFINE SCALE(src1, src2) { + IF (src2 == NaN) + IF (src2 == SNaN) + RETURN QNAN(src2) + FI + ELSE IF (src1 == NaN) + IF (src1 == SNaN) + RETURN QNAN(src1) + FI + IF (src2 != INF) + RETURN QNAN(src1) + FI + ELSE + tmp_src2 := src2 + tmp_src1 := src1 + IF (IS_DENORMAL(src2) AND MXCSR.DAZ) + tmp_src2 := 0 + FI + IF (IS_DENORMAL(src1) AND MXCSR.DAZ) + tmp_src1 := 0 + FI + FI + dst[31:0] := tmp_src1[31:0] * POW(2.0, FLOOR(tmp_src2[31:0])) + RETURN dst[31:0] +} +FOR j := 0 to 7 + i := j*32 + dst[i+31:i] := SCALE(a[i+31:0], b[i+31:i]) +ENDFOR +dst[MAX:256] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512VL + AVX512F + Miscellaneous + + + + + + Scale the packed single-precision (32-bit) floating-point elements in "a" using values from "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + DEFINE SCALE(src1, src2) { + IF (src2 == NaN) + IF (src2 == SNaN) + RETURN QNAN(src2) + FI + ELSE IF (src1 == NaN) + IF (src1 == SNaN) + RETURN QNAN(src1) + FI + IF (src2 != INF) + RETURN QNAN(src1) + FI + ELSE + tmp_src2 := src2 + tmp_src1 := src1 + IF (IS_DENORMAL(src2) AND MXCSR.DAZ) + tmp_src2 := 0 + FI + IF (IS_DENORMAL(src1) AND MXCSR.DAZ) + tmp_src1 := 0 + FI + FI + dst[31:0] := tmp_src1[31:0] * POW(2.0, FLOOR(tmp_src2[31:0])) + RETURN dst[31:0] +} +FOR j := 0 to 3 + i := j*32 + IF k[j] + dst[i+31:i] := SCALE(a[i+31:0], b[i+31:i]) + ELSE + dst[i+31:i] := src[i+31:i] + FI +ENDFOR +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512VL + AVX512F + Miscellaneous + + + + + Scale the packed single-precision (32-bit) floating-point elements in "a" using values from "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + DEFINE SCALE(src1, src2) { + IF (src2 == NaN) + IF (src2 == SNaN) + RETURN QNAN(src2) + FI + ELSE IF (src1 == NaN) + IF (src1 == SNaN) + RETURN QNAN(src1) + FI + IF (src2 != INF) + RETURN QNAN(src1) + FI + ELSE + tmp_src2 := src2 + tmp_src1 := src1 + IF (IS_DENORMAL(src2) AND MXCSR.DAZ) + tmp_src2 := 0 + FI + IF (IS_DENORMAL(src1) AND MXCSR.DAZ) + tmp_src1 := 0 + FI + FI + dst[31:0] := tmp_src1[31:0] * POW(2.0, FLOOR(tmp_src2[31:0])) + RETURN dst[31:0] +} +FOR j := 0 to 3 + i := j*32 + IF k[j] + dst[i+31:i] := SCALE(a[i+31:0], b[i+31:i]) + ELSE + dst[i+31:i] := 0 + FI +ENDFOR +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512VL + AVX512F + Miscellaneous + + + + Scale the packed single-precision (32-bit) floating-point elements in "a" using values from "b", and store the results in "dst". + DEFINE SCALE(src1, src2) { + IF (src2 == NaN) + IF (src2 == SNaN) + RETURN QNAN(src2) + FI + ELSE IF (src1 == NaN) + IF (src1 == SNaN) + RETURN QNAN(src1) + FI + IF (src2 != INF) + RETURN QNAN(src1) + FI + ELSE + tmp_src2 := src2 + tmp_src1 := src1 + IF (IS_DENORMAL(src2) AND MXCSR.DAZ) + tmp_src2 := 0 + FI + IF (IS_DENORMAL(src1) AND MXCSR.DAZ) + tmp_src1 := 0 + FI + FI + dst[31:0] := tmp_src1[31:0] * POW(2.0, FLOOR(tmp_src2[31:0])) + RETURN dst[31:0] +} +FOR j := 0 to 3 + i := j*32 + dst[i+31:i] := SCALE(a[i+31:0], b[i+31:i]) +ENDFOR +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512VL + AVX512F + Store + + + + + + Scatter double-precision (64-bit) floating-point elements from "a" into memory using 32-bit indices. 64-bit elements are stored at addresses starting at "base_addr" and offset by each 32-bit element in "vindex" (each index is scaled by the factor in "scale"). "scale" should be 1, 2, 4 or 8. + +FOR j := 0 to 3 + i := j*64 + m := j*32 + addr := base_addr + SignExtend64(vindex[m+31:m]) * ZeroExtend64(scale) * 8 + MEM[addr+63:addr] := a[i+63:i] +ENDFOR + + +
immintrin.h
+ + + Floating Point + AVX512VL + AVX512F + Store + + + + + + + Scatter double-precision (64-bit) floating-point elements from "a" into memory using 32-bit indices. 64-bit elements are stored at addresses starting at "base_addr" and offset by each 32-bit element in "vindex" (each index is scaled by the factor in "scale") subject to mask "k" (elements are not stored when the corresponding mask bit is not set). "scale" should be 1, 2, 4 or 8. + +FOR j := 0 to 3 + i := j*64 + m := j*32 + IF k[j] + addr := base_addr + SignExtend64(vindex[m+31:m]) * ZeroExtend64(scale) * 8 + MEM[addr+63:addr] := a[i+63:i] + FI +ENDFOR + + +
immintrin.h
+ + + Floating Point + AVX512VL + AVX512F + Store + + + + + + Scatter double-precision (64-bit) floating-point elements from "a" into memory using 32-bit indices. 64-bit elements are stored at addresses starting at "base_addr" and offset by each 32-bit element in "vindex" (each index is scaled by the factor in "scale"). "scale" should be 1, 2, 4 or 8. + +FOR j := 0 to 1 + i := j*64 + m := j*32 + addr := base_addr + SignExtend64(vindex[m+31:m]) * ZeroExtend64(scale) * 8 + MEM[addr+63:addr] := a[i+63:i] +ENDFOR + + +
immintrin.h
+ + + Floating Point + AVX512VL + AVX512F + Store + + + + + + + Scatter double-precision (64-bit) floating-point elements from "a" into memory using 32-bit indices. 64-bit elements are stored at addresses starting at "base_addr" and offset by each 32-bit element in "vindex" (each index is scaled by the factor in "scale") subject to mask "k" (elements are not stored when the corresponding mask bit is not set). "scale" should be 1, 2, 4 or 8. + +FOR j := 0 to 1 + i := j*64 + m := j*32 + IF k[j] + addr := base_addr + SignExtend64(vindex[m+31:m]) * ZeroExtend64(scale) * 8 + MEM[addr+63:addr] := a[i+63:i] + FI +ENDFOR + + +
immintrin.h
+ + + Floating Point + AVX512VL + AVX512F + Store + + + + + + Scatter single-precision (32-bit) floating-point elements from "a" into memory using 32-bit indices. 32-bit elements are stored at addresses starting at "base_addr" and offset by each 32-bit element in "vindex" (each index is scaled by the factor in "scale"). "scale" should be 1, 2, 4 or 8. + +FOR j := 0 to 7 + i := j*32 + m := j*32 + addr := base_addr + SignExtend64(vindex[m+31:m]) * ZeroExtend64(scale) * 8 + MEM[addr+31:addr] := a[i+31:i] +ENDFOR + + +
immintrin.h
+ + + Floating Point + AVX512VL + AVX512F + Store + + + + + + + Scatter single-precision (32-bit) floating-point elements from "a" into memory using 32-bit indices. 32-bit elements are stored at addresses starting at "base_addr" and offset by each 32-bit element in "vindex" (each index is scaled by the factor in "scale") subject to mask "k" (elements are not stored when the corresponding mask bit is not set). "scale" should be 1, 2, 4 or 8. + +FOR j := 0 to 7 + i := j*32 + m := j*32 + IF k[j] + addr := base_addr + SignExtend64(vindex[m+31:m]) * ZeroExtend64(scale) * 8 + MEM[addr+31:addr] := a[i+31:i] + FI +ENDFOR + + +
immintrin.h
+ + + Floating Point + AVX512VL + AVX512F + Store + + + + + + Scatter single-precision (32-bit) floating-point elements from "a" into memory using 32-bit indices. 32-bit elements are stored at addresses starting at "base_addr" and offset by each 32-bit element in "vindex" (each index is scaled by the factor in "scale"). "scale" should be 1, 2, 4 or 8. + +FOR j := 0 to 3 + i := j*32 + m := j*32 + addr := base_addr + SignExtend64(vindex[m+31:m]) * ZeroExtend64(scale) * 8 + MEM[addr+31:addr] := a[i+31:i] +ENDFOR + + +
immintrin.h
+ + + Floating Point + AVX512VL + AVX512F + Store + + + + + + + Scatter single-precision (32-bit) floating-point elements from "a" into memory using 32-bit indices. 32-bit elements are stored at addresses starting at "base_addr" and offset by each 32-bit element in "vindex" (each index is scaled by the factor in "scale") subject to mask "k" (elements are not stored when the corresponding mask bit is not set). "scale" should be 1, 2, 4 or 8. + +FOR j := 0 to 3 + i := j*32 + m := j*32 + IF k[j] + addr := base_addr + SignExtend64(vindex[m+31:m]) * ZeroExtend64(scale) * 8 + MEM[addr+31:addr] := a[i+31:i] + FI +ENDFOR + + +
immintrin.h
+ + + Floating Point + AVX512VL + AVX512F + Store + + + + + + Scatter double-precision (64-bit) floating-point elements from "a" into memory using 64-bit indices. 64-bit elements are stored at addresses starting at "base_addr" and offset by each 64-bit element in "vindex" (each index is scaled by the factor in "scale"). "scale" should be 1, 2, 4 or 8. + +FOR j := 0 to 3 + i := j*64 + m := j*64 + addr := base_addr + vindex[m+63:m] * ZeroExtend64(scale) * 8 + MEM[addr+63:addr] := a[i+63:i] +ENDFOR + + +
immintrin.h
+ + + Floating Point + AVX512VL + AVX512F + Store + + + + + + + Scatter double-precision (64-bit) floating-point elements from "a" into memory using 64-bit indices. 64-bit elements are stored at addresses starting at "base_addr" and offset by each 64-bit element in "vindex" (each index is scaled by the factor in "scale") subject to mask "k" (elements are not stored when the corresponding mask bit is not set). "scale" should be 1, 2, 4 or 8. + +FOR j := 0 to 3 + i := j*64 + m := j*64 + IF k[j] + addr := base_addr + vindex[m+63:m] * ZeroExtend64(scale) * 8 + MEM[addr+63:addr] := a[i+63:i] + FI +ENDFOR + + +
immintrin.h
+ + + Floating Point + AVX512VL + AVX512F + Store + + + + + + Scatter double-precision (64-bit) floating-point elements from "a" into memory using 64-bit indices. 64-bit elements are stored at addresses starting at "base_addr" and offset by each 64-bit element in "vindex" (each index is scaled by the factor in "scale"). "scale" should be 1, 2, 4 or 8. + +FOR j := 0 to 1 + i := j*64 + m := j*64 + addr := base_addr + vindex[m+63:m] * ZeroExtend64(scale) * 8 + MEM[addr+63:addr] := a[i+63:i] +ENDFOR + + +
immintrin.h
+ + + Floating Point + AVX512VL + AVX512F + Store + + + + + + + Scatter double-precision (64-bit) floating-point elements from "a" into memory using 64-bit indices. 64-bit elements are stored at addresses starting at "base_addr" and offset by each 64-bit element in "vindex" (each index is scaled by the factor in "scale") subject to mask "k" (elements are not stored when the corresponding mask bit is not set). "scale" should be 1, 2, 4 or 8. + +FOR j := 0 to 1 + i := j*64 + m := j*64 + IF k[j] + addr := base_addr + vindex[m+63:m] * ZeroExtend64(scale) * 8 + MEM[addr+63:addr] := a[i+63:i] + FI +ENDFOR + + +
immintrin.h
+ + + Floating Point + AVX512VL + AVX512F + Store + + + + + + Scatter single-precision (32-bit) floating-point elements from "a" into memory using 64-bit indices. 32-bit elements are stored at addresses starting at "base_addr" and offset by each 64-bit element in "vindex" (each index is scaled by the factor in "scale") subject to mask "k" (elements are not stored when the corresponding mask bit is not set). "scale" should be 1, 2, 4 or 8. + +FOR j := 0 to 3 + i := j*32 + m := j*64 + addr := base_addr + vindex[m+63:m] * ZeroExtend64(scale) * 8 + MEM[addr+31:addr] := a[i+31:i] +ENDFOR + + +
immintrin.h
+ + + Floating Point + AVX512VL + AVX512F + Store + + + + + + + Scatter single-precision (32-bit) floating-point elements from "a" into memory using 64-bit indices. 32-bit elements are stored at addresses starting at "base_addr" and offset by each 64-bit element in "vindex" (each index is scaled by the factor in "scale") subject to mask "k" (elements are not stored when the corresponding mask bit is not set). "scale" should be 1, 2, 4 or 8. + +FOR j := 0 to 3 + i := j*32 + m := j*64 + IF k[j] + addr := base_addr + vindex[m+63:m] * ZeroExtend64(scale) * 8 + MEM[addr+31:addr] := a[i+31:i] + FI +ENDFOR + + +
immintrin.h
+ + + Floating Point + AVX512VL + AVX512F + Store + + + + + + Scatter single-precision (32-bit) floating-point elements from "a" into memory using 64-bit indices. 32-bit elements are stored at addresses starting at "base_addr" and offset by each 64-bit element in "vindex" (each index is scaled by the factor in "scale") subject to mask "k" (elements are not stored when the corresponding mask bit is not set). "scale" should be 1, 2, 4 or 8. + +FOR j := 0 to 1 + i := j*32 + m := j*64 + addr := base_addr + vindex[m+63:m] * ZeroExtend64(scale) * 8 + MEM[addr+31:addr] := a[i+31:i] +ENDFOR + + +
immintrin.h
+ + + Floating Point + AVX512VL + AVX512F + Store + + + + + + + Scatter single-precision (32-bit) floating-point elements from "a" into memory using 64-bit indices. 32-bit elements are stored at addresses starting at "base_addr" and offset by each 64-bit element in "vindex" (each index is scaled by the factor in "scale") subject to mask "k" (elements are not stored when the corresponding mask bit is not set). "scale" should be 1, 2, 4 or 8. + +FOR j := 0 to 1 + i := j*32 + m := j*64 + IF k[j] + addr := base_addr + vindex[m+63:m] * ZeroExtend64(scale) * 8 + MEM[addr+31:addr] := a[i+31:i] + FI +ENDFOR + + +
immintrin.h
+ + + Floating Point + AVX512VL + AVX512F + Miscellaneous + + + + + + + Shuffle 128-bits (composed of 4 single-precision (32-bit) floating-point elements) selected by "imm8" from "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +tmp_dst.m128[0] := a.m128[imm8[0]] +tmp_dst.m128[1] := b.m128[imm8[1]] +FOR j := 0 to 7 + i := j*32 + IF k[j] + dst[i+31:i] := tmp_dst[i+31:i] + ELSE + dst[i+31:i] := src[i+31:i] + FI +ENDFOR +dst[MAX:256] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512VL + AVX512F + Miscellaneous + + + + + + Shuffle 128-bits (composed of 4 single-precision (32-bit) floating-point elements) selected by "imm8" from "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + +tmp_dst.m128[0] := a.m128[imm8[0]] +tmp_dst.m128[1] := b.m128[imm8[1]] +FOR j := 0 to 7 + i := j*32 + IF k[j] + dst[i+31:i] := tmp_dst[i+31:i] + ELSE + dst[i+31:i] := 0 + FI +ENDFOR +dst[MAX:256] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512VL + AVX512F + Miscellaneous + + + + + Shuffle 128-bits (composed of 4 single-precision (32-bit) floating-point elements) selected by "imm8" from "a" and "b", and store the results in "dst". + +dst.m128[0] := a.m128[imm8[0]] +dst.m128[1] := b.m128[imm8[1]] +dst[MAX:256] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512VL + AVX512F + Miscellaneous + + + + + + + Shuffle 128-bits (composed of 2 double-precision (64-bit) floating-point elements) selected by "imm8" from "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +tmp_dst.m128[0] := a.m128[imm8[0]] +tmp_dst.m128[1] := b.m128[imm8[1]] +FOR j := 0 to 3 + i := j*64 + IF k[j] + dst[i+63:i] := tmp_dst[i+63:i] + ELSE + dst[i+63:i] := src[i+63:i] + FI +ENDFOR +dst[MAX:256] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512VL + AVX512F + Miscellaneous + + + + + + Shuffle 128-bits (composed of 2 double-precision (64-bit) floating-point elements) selected by "imm8" from "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + +tmp_dst.m128[0] := a.m128[imm8[0]] +tmp_dst.m128[1] := b.m128[imm8[1]] +FOR j := 0 to 3 + i := j*64 + IF k[j] + dst[i+63:i] := tmp_dst[i+63:i] + ELSE + dst[i+63:i] := 0 + FI +ENDFOR +dst[MAX:256] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512VL + AVX512F + Miscellaneous + + + + + Shuffle 128-bits (composed of 2 double-precision (64-bit) floating-point elements) selected by "imm8" from "a" and "b", and store the results in "dst". + +dst.m128[0] := a.m128[imm8[0]] +dst.m128[1] := b.m128[imm8[1]] +dst[MAX:256] := 0 + + +
immintrin.h
+ + + AVX512VL + AVX512F + Miscellaneous + + + + + + + Shuffle 128-bits (composed of 4 32-bit integers) selected by "imm8" from "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +tmp_dst.m128[0] := a.m128[imm8[0]] +tmp_dst.m128[1] := b.m128[imm8[1]] +FOR j := 0 to 7 + i := j*32 + IF k[j] + dst[i+31:i] := tmp_dst[i+31:i] + ELSE + dst[i+31:i] := src[i+31:i] + FI +ENDFOR +dst[MAX:256] := 0 + + +
immintrin.h
+ + + AVX512VL + AVX512F + Miscellaneous + + + + + + Shuffle 128-bits (composed of 4 32-bit integers) selected by "imm8" from "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + +tmp_dst.m128[0] := a.m128[imm8[0]] +tmp_dst.m128[1] := b.m128[imm8[1]] +FOR j := 0 to 7 + i := j*32 + IF k[j] + dst[i+31:i] := tmp_dst[i+31:i] + ELSE + dst[i+31:i] := 0 + FI +ENDFOR +dst[MAX:256] := 0 + + +
immintrin.h
+ + + AVX512VL + AVX512F + Miscellaneous + + + + + Shuffle 128-bits (composed of 4 32-bit integers) selected by "imm8" from "a" and "b", and store the results in "dst". + +dst.m128[0] := a.m128[imm8[0]] +dst.m128[1] := b.m128[imm8[1]] +dst[MAX:256] := 0 + + +
immintrin.h
+ + + AVX512VL + AVX512F + Miscellaneous + + + + + + + Shuffle 128-bits (composed of 2 64-bit integers) selected by "imm8" from "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +tmp_dst.m128[0] := a.m128[imm8[0]] +tmp_dst.m128[1] := b.m128[imm8[1]] +FOR j := 0 to 3 + i := j*64 + IF k[j] + dst[i+63:i] := tmp_dst[i+63:i] + ELSE + dst[i+63:i] := src[i+63:i] + FI +ENDFOR +dst[MAX:256] := 0 + + +
immintrin.h
+ + + AVX512VL + AVX512F + Miscellaneous + + + + + + Shuffle 128-bits (composed of 2 64-bit integers) selected by "imm8" from "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + +tmp_dst.m128[0] := a.m128[imm8[0]] +tmp_dst.m128[1] := b.m128[imm8[1]] +FOR j := 0 to 3 + i := j*64 + IF k[j] + dst[i+63:i] := tmp_dst[i+63:i] + ELSE + dst[i+63:i] := 0 + FI +ENDFOR +dst[MAX:256] := 0 + + +
immintrin.h
+ + + AVX512VL + AVX512F + Miscellaneous + + + + + Shuffle 128-bits (composed of 2 64-bit integers) selected by "imm8" from "a" and "b", and store the results in "dst". + +dst.m128[0] := a.m128[imm8[0]] +dst.m128[1] := b.m128[imm8[1]] +dst[MAX:256] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512VL + AVX512F + Miscellaneous + + + + + + + Shuffle double-precision (64-bit) floating-point elements within 128-bit lanes using the control in "imm8", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +tmp_dst[63:0] := (imm8[0] == 0) ? a[63:0] : a[127:64] +tmp_dst[127:64] := (imm8[1] == 0) ? b[63:0] : b[127:64] +tmp_dst[191:128] := (imm8[2] == 0) ? a[191:128] : a[255:192] +tmp_dst[255:192] := (imm8[3] == 0) ? b[191:128] : b[255:192] +FOR j := 0 to 3 + i := j*64 + IF k[j] + dst[i+63:i] := tmp_dst[i+63:i] + ELSE + dst[i+63:i] := src[i+63:i] + FI +ENDFOR +dst[MAX:256] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512VL + AVX512F + Miscellaneous + + + + + + Shuffle double-precision (64-bit) floating-point elements within 128-bit lanes using the control in "imm8", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + +tmp_dst[63:0] := (imm8[0] == 0) ? a[63:0] : a[127:64] +tmp_dst[127:64] := (imm8[1] == 0) ? b[63:0] : b[127:64] +tmp_dst[191:128] := (imm8[2] == 0) ? a[191:128] : a[255:192] +tmp_dst[255:192] := (imm8[3] == 0) ? b[191:128] : b[255:192] +FOR j := 0 to 3 + i := j*64 + IF k[j] + dst[i+63:i] := tmp_dst[i+63:i] + ELSE + dst[i+63:i] := 0 + FI +ENDFOR +dst[MAX:256] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512VL + AVX512F + Miscellaneous + + + + + + + Shuffle double-precision (64-bit) floating-point elements using the control in "imm8", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +tmp_dst[63:0] := (imm8[0] == 0) ? a[63:0] : a[127:64] +tmp_dst[127:64] := (imm8[1] == 0) ? b[63:0] : b[127:64] +FOR j := 0 to 1 + i := j*64 + IF k[j] + dst[i+63:i] := tmp_dst[i+63:i] + ELSE + dst[i+63:i] := src[i+63:i] + FI +ENDFOR +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512VL + AVX512F + Miscellaneous + + + + + + Shuffle double-precision (64-bit) floating-point elements using the control in "imm8", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + +tmp_dst[63:0] := (imm8[0] == 0) ? a[63:0] : a[127:64] +tmp_dst[127:64] := (imm8[1] == 0) ? b[63:0] : b[127:64] +FOR j := 0 to 1 + i := j*64 + IF k[j] + dst[i+63:i] := tmp_dst[i+63:i] + ELSE + dst[i+63:i] := 0 + FI +ENDFOR +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512VL + AVX512F + Miscellaneous + + + + + + + Shuffle single-precision (32-bit) floating-point elements in "a" within 128-bit lanes using the control in "imm8", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +DEFINE SELECT4(src, control) { + CASE(control[1:0]) OF + 0: tmp[31:0] := src[31:0] + 1: tmp[31:0] := src[63:32] + 2: tmp[31:0] := src[95:64] + 3: tmp[31:0] := src[127:96] + ESAC + RETURN tmp[31:0] +} +tmp_dst[31:0] := SELECT4(a[127:0], imm8[1:0]) +tmp_dst[63:32] := SELECT4(a[127:0], imm8[3:2]) +tmp_dst[95:64] := SELECT4(b[127:0], imm8[5:4]) +tmp_dst[127:96] := SELECT4(b[127:0], imm8[7:6]) +tmp_dst[159:128] := SELECT4(a[255:128], imm8[1:0]) +tmp_dst[191:160] := SELECT4(a[255:128], imm8[3:2]) +tmp_dst[223:192] := SELECT4(b[255:128], imm8[5:4]) +tmp_dst[255:224] := SELECT4(b[255:128], imm8[7:6]) +FOR j := 0 to 7 + i := j*32 + IF k[j] + dst[i+31:i] := tmp_dst[i+31:i] + ELSE + dst[i+31:i] := src[i+31:i] + FI +ENDFOR +dst[MAX:256] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512VL + AVX512F + Miscellaneous + + + + + + Shuffle single-precision (32-bit) floating-point elements in "a" within 128-bit lanes using the control in "imm8", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + +DEFINE SELECT4(src, control) { + CASE(control[1:0]) OF + 0: tmp[31:0] := src[31:0] + 1: tmp[31:0] := src[63:32] + 2: tmp[31:0] := src[95:64] + 3: tmp[31:0] := src[127:96] + ESAC + RETURN tmp[31:0] +} +tmp_dst[31:0] := SELECT4(a[127:0], imm8[1:0]) +tmp_dst[63:32] := SELECT4(a[127:0], imm8[3:2]) +tmp_dst[95:64] := SELECT4(b[127:0], imm8[5:4]) +tmp_dst[127:96] := SELECT4(b[127:0], imm8[7:6]) +tmp_dst[159:128] := SELECT4(a[255:128], imm8[1:0]) +tmp_dst[191:160] := SELECT4(a[255:128], imm8[3:2]) +tmp_dst[223:192] := SELECT4(b[255:128], imm8[5:4]) +tmp_dst[255:224] := SELECT4(b[255:128], imm8[7:6]) +FOR j := 0 to 7 + i := j*32 + IF k[j] + dst[i+31:i] := tmp_dst[i+31:i] + ELSE + dst[i+31:i] := 0 + FI +ENDFOR +dst[MAX:256] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512VL + AVX512F + Miscellaneous + + + + + + + Shuffle single-precision (32-bit) floating-point elements in "a" using the control in "imm8", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +DEFINE SELECT4(src, control) { + CASE(control[1:0]) OF + 0: tmp[31:0] := src[31:0] + 1: tmp[31:0] := src[63:32] + 2: tmp[31:0] := src[95:64] + 3: tmp[31:0] := src[127:96] + ESAC + RETURN tmp[31:0] +} +tmp_dst[31:0] := SELECT4(a[127:0], imm8[1:0]) +tmp_dst[63:32] := SELECT4(a[127:0], imm8[3:2]) +tmp_dst[95:64] := SELECT4(b[127:0], imm8[5:4]) +tmp_dst[127:96] := SELECT4(b[127:0], imm8[7:6]) +FOR j := 0 to 3 + i := j*32 + IF k[j] + dst[i+31:i] := tmp_dst[i+31:i] + ELSE + dst[i+31:i] := src[i+31:i] + FI +ENDFOR +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512VL + AVX512F + Miscellaneous + + + + + + Shuffle single-precision (32-bit) floating-point elements in "a" using the control in "imm8", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + +DEFINE SELECT4(src, control) { + CASE(control[1:0]) OF + 0: tmp[31:0] := src[31:0] + 1: tmp[31:0] := src[63:32] + 2: tmp[31:0] := src[95:64] + 3: tmp[31:0] := src[127:96] + ESAC + RETURN tmp[31:0] +} +tmp_dst[31:0] := SELECT4(a[127:0], imm8[1:0]) +tmp_dst[63:32] := SELECT4(a[127:0], imm8[3:2]) +tmp_dst[95:64] := SELECT4(b[127:0], imm8[5:4]) +tmp_dst[127:96] := SELECT4(b[127:0], imm8[7:6]) +FOR j := 0 to 3 + i := j*32 + IF k[j] + dst[i+31:i] := tmp_dst[i+31:i] + ELSE + dst[i+31:i] := 0 + FI +ENDFOR +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512VL + AVX512F + Elementary Math Functions + + + + + Compute the square root of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +FOR j := 0 to 3 + i := j*64 + IF k[j] + dst[i+63:i] := SQRT(a[i+63:i]) + ELSE + dst[i+63:i] := src[i+63:i] + FI +ENDFOR +dst[MAX:256] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512VL + AVX512F + Elementary Math Functions + + + + Compute the square root of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + +FOR j := 0 to 3 + i := j*64 + IF k[j] + dst[i+63:i] := SQRT(a[i+63:i]) + ELSE + dst[i+63:i] := 0 + FI +ENDFOR +dst[MAX:256] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512VL + AVX512F + Elementary Math Functions + + + + + Compute the square root of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +FOR j := 0 to 1 + i := j*64 + IF k[j] + dst[i+63:i] := SQRT(a[i+63:i]) + ELSE + dst[i+63:i] := src[i+63:i] + FI +ENDFOR +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512VL + AVX512F + Elementary Math Functions + + + + Compute the square root of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + +FOR j := 0 to 1 + i := j*64 + IF k[j] + dst[i+63:i] := SQRT(a[i+63:i]) + ELSE + dst[i+63:i] := 0 + FI +ENDFOR +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512VL + AVX512F + Elementary Math Functions + + + + + Compute the square root of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +FOR j := 0 to 7 + i := j*32 + IF k[j] + dst[i+31:i] := SQRT(a[i+31:i]) + ELSE + dst[i+31:i] := src[i+31:i] + FI +ENDFOR +dst[MAX:256] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512VL + AVX512F + Elementary Math Functions + + + + Compute the square root of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + +FOR j := 0 to 7 + i := j*32 + IF k[j] + dst[i+31:i] := SQRT(a[i+31:i]) + ELSE + dst[i+31:i] := 0 + FI +ENDFOR +dst[MAX:256] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512VL + AVX512F + Elementary Math Functions + + + + + Compute the square root of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +FOR j := 0 to 3 + i := j*32 + IF k[j] + dst[i+31:i] := SQRT(a[i+31:i]) + ELSE + dst[i+31:i] := src[i+31:i] + FI +ENDFOR +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512VL + AVX512F + Elementary Math Functions + + + + Compute the square root of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + +FOR j := 0 to 3 + i := j*32 + IF k[j] + dst[i+31:i] := SQRT(a[i+31:i]) + ELSE + dst[i+31:i] := 0 + FI +ENDFOR +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512VL + AVX512F + Arithmetic + + + + + + Subtract packed double-precision (64-bit) floating-point elements in "b" from packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +FOR j := 0 to 3 + i := j*64 + IF k[j] + dst[i+63:i] := a[i+63:i] - b[i+63:i] + ELSE + dst[i+63:i] := src[i+63:i] + FI +ENDFOR +dst[MAX:256] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512VL + AVX512F + Arithmetic + + + + + Subtract packed double-precision (64-bit) floating-point elements in "b" from packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + +FOR j := 0 to 3 + i := j*64 + IF k[j] + dst[i+63:i] := a[i+63:i] - b[i+63:i] + ELSE + dst[i+63:i] := 0 + FI +ENDFOR +dst[MAX:256] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512VL + AVX512F + Arithmetic + + + + + + Subtract packed double-precision (64-bit) floating-point elements in "b" from packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +FOR j := 0 to 1 + i := j*64 + IF k[j] + dst[i+63:i] := a[i+63:i] - b[i+63:i] + ELSE + dst[i+63:i] := src[i+63:i] + FI +ENDFOR +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512VL + AVX512F + Arithmetic + + + + + Subtract packed double-precision (64-bit) floating-point elements in "b" from packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + +FOR j := 0 to 1 + i := j*64 + IF k[j] + dst[i+63:i] := a[i+63:i] - b[i+63:i] + ELSE + dst[i+63:i] := 0 + FI +ENDFOR +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512VL + AVX512F + Arithmetic + + + + + + Subtract packed single-precision (32-bit) floating-point elements in "b" from packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +FOR j := 0 to 7 + i := j*32 + IF k[j] + dst[i+31:i] := a[i+31:i] - b[i+31:i] + ELSE + dst[i+31:i] := src[i+31:i] + FI +ENDFOR +dst[MAX:256] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512VL + AVX512F + Arithmetic + + + + + Subtract packed single-precision (32-bit) floating-point elements in "b" from packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + +FOR j := 0 to 7 + i := j*32 + IF k[j] + dst[i+31:i] := a[i+31:i] - b[i+31:i] + ELSE + dst[i+31:i] := 0 + FI +ENDFOR +dst[MAX:256] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512VL + AVX512F + Arithmetic + + + + + + Subtract packed single-precision (32-bit) floating-point elements in "b" from packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +FOR j := 0 to 3 + i := j*32 + IF k[j] + dst[i+31:i] := a[i+31:i] - b[i+31:i] + ELSE + dst[i+31:i] := src[i+31:i] + FI +ENDFOR +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512VL + AVX512F + Arithmetic + + + + + Subtract packed single-precision (32-bit) floating-point elements in "b" from packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + +FOR j := 0 to 3 + i := j*32 + IF k[j] + dst[i+31:i] := a[i+31:i] - b[i+31:i] + ELSE + dst[i+31:i] := 0 + FI +ENDFOR +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512VL + AVX512F + Miscellaneous + + + + + + Unpack and interleave double-precision (64-bit) floating-point elements from the high half of each 128-bit lane in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +DEFINE INTERLEAVE_HIGH_QWORDS(src1[127:0], src2[127:0]) { + dst[63:0] := src1[127:64] + dst[127:64] := src2[127:64] + RETURN dst[127:0] +} +tmp_dst[127:0] := INTERLEAVE_HIGH_QWORDS(a[127:0], b[127:0]) +tmp_dst[255:128] := INTERLEAVE_HIGH_QWORDS(a[255:128], b[255:128]) +FOR j := 0 to 3 + i := j*64 + IF k[j] + dst[i+63:i] := tmp_dst[i+63:i] + ELSE + dst[i+63:i] := src[i+63:i] + FI +ENDFOR +dst[MAX:256] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512VL + AVX512F + Miscellaneous + + + + + Unpack and interleave double-precision (64-bit) floating-point elements from the high half of each 128-bit lane in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + +DEFINE INTERLEAVE_HIGH_QWORDS(src1[127:0], src2[127:0]) { + dst[63:0] := src1[127:64] + dst[127:64] := src2[127:64] + RETURN dst[127:0] +} +tmp_dst[127:0] := INTERLEAVE_HIGH_QWORDS(a[127:0], b[127:0]) +tmp_dst[255:128] := INTERLEAVE_HIGH_QWORDS(a[255:128], b[255:128]) +FOR j := 0 to 3 + i := j*64 + IF k[j] + dst[i+63:i] := tmp_dst[i+63:i] + ELSE + dst[i+63:i] := 0 + FI +ENDFOR +dst[MAX:256] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512VL + AVX512F + Miscellaneous + + + + + + Unpack and interleave double-precision (64-bit) floating-point elements from the high half of "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +DEFINE INTERLEAVE_HIGH_QWORDS(src1[127:0], src2[127:0]) { + dst[63:0] := src1[127:64] + dst[127:64] := src2[127:64] + RETURN dst[127:0] +} +tmp_dst[127:0] := INTERLEAVE_HIGH_QWORDS(a[127:0], b[127:0]) +FOR j := 0 to 1 + i := j*64 + IF k[j] + dst[i+63:i] := tmp_dst[i+63:i] + ELSE + dst[i+63:i] := src[i+63:i] + FI +ENDFOR +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512VL + AVX512F + Miscellaneous + + + + + Unpack and interleave double-precision (64-bit) floating-point elements from the high half of "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + +DEFINE INTERLEAVE_HIGH_QWORDS(src1[127:0], src2[127:0]) { + dst[63:0] := src1[127:64] + dst[127:64] := src2[127:64] + RETURN dst[127:0] +} +tmp_dst[127:0] := INTERLEAVE_HIGH_QWORDS(a[127:0], b[127:0]) +FOR j := 0 to 1 + i := j*64 + IF k[j] + dst[i+63:i] := tmp_dst[i+63:i] + ELSE + dst[i+63:i] := 0 + FI +ENDFOR +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512VL + AVX512F + Miscellaneous + + + + + + Unpack and interleave single-precision (32-bit) floating-point elements from the high half of each 128-bit lane in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +DEFINE INTERLEAVE_HIGH_DWORDS(src1[127:0], src2[127:0]) { + dst[31:0] := src1[95:64] + dst[63:32] := src2[95:64] + dst[95:64] := src1[127:96] + dst[127:96] := src2[127:96] + RETURN dst[127:0] +} +tmp_dst[127:0] := INTERLEAVE_HIGH_DWORDS(a[127:0], b[127:0]) +tmp_dst[255:128] := INTERLEAVE_HIGH_DWORDS(a[255:128], b[255:128]) +FOR j := 0 to 7 + i := j*32 + IF k[j] + dst[i+31:i] := tmp_dst[i+31:i] + ELSE + dst[i+31:i] := src[i+31:i] + FI +ENDFOR +dst[MAX:256] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512VL + AVX512F + Miscellaneous + + + + + Unpack and interleave single-precision (32-bit) floating-point elements from the high half of each 128-bit lane in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + +DEFINE INTERLEAVE_HIGH_DWORDS(src1[127:0], src2[127:0]) { + dst[31:0] := src1[95:64] + dst[63:32] := src2[95:64] + dst[95:64] := src1[127:96] + dst[127:96] := src2[127:96] + RETURN dst[127:0] +} +tmp_dst[127:0] := INTERLEAVE_HIGH_DWORDS(a[127:0], b[127:0]) +tmp_dst[255:128] := INTERLEAVE_HIGH_DWORDS(a[255:128], b[255:128]) +FOR j := 0 to 7 + i := j*32 + IF k[j] + dst[i+31:i] := tmp_dst[i+31:i] + ELSE + dst[i+31:i] := 0 + FI +ENDFOR +dst[MAX:256] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512VL + AVX512F + Miscellaneous + + + + + + Unpack and interleave single-precision (32-bit) floating-point elements from the high half of "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +DEFINE INTERLEAVE_HIGH_DWORDS(src1[127:0], src2[127:0]) { + dst[31:0] := src1[95:64] + dst[63:32] := src2[95:64] + dst[95:64] := src1[127:96] + dst[127:96] := src2[127:96] + RETURN dst[127:0] +} +tmp_dst[127:0] := INTERLEAVE_HIGH_DWORDS(a[127:0], b[127:0]) +FOR j := 0 to 3 + i := j*32 + IF k[j] + dst[i+31:i] := tmp_dst[i+31:i] + ELSE + dst[i+31:i] := src[i+31:i] + FI +ENDFOR +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512VL + AVX512F + Miscellaneous + + + + + Unpack and interleave single-precision (32-bit) floating-point elements from the high half of "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + +DEFINE INTERLEAVE_HIGH_DWORDS(src1[127:0], src2[127:0]) { + dst[31:0] := src1[95:64] + dst[63:32] := src2[95:64] + dst[95:64] := src1[127:96] + dst[127:96] := src2[127:96] + RETURN dst[127:0] +} +tmp_dst[127:0] := INTERLEAVE_HIGH_DWORDS(a[127:0], b[127:0]) +FOR j := 0 to 3 + i := j*32 + IF k[j] + dst[i+31:i] := tmp_dst[i+31:i] + ELSE + dst[i+31:i] := 0 + FI +ENDFOR +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512VL + AVX512F + Miscellaneous + + + + + + Unpack and interleave double-precision (64-bit) floating-point elements from the low half of each 128-bit lane in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +DEFINE INTERLEAVE_QWORDS(src1[127:0], src2[127:0]) { + dst[63:0] := src1[63:0] + dst[127:64] := src2[63:0] + RETURN dst[127:0] +} +tmp_dst[127:0] := INTERLEAVE_QWORDS(a[127:0], b[127:0]) +tmp_dst[255:128] := INTERLEAVE_QWORDS(a[255:128], b[255:128]) +FOR j := 0 to 3 + i := j*64 + IF k[j] + dst[i+63:i] := tmp_dst[i+63:i] + ELSE + dst[i+63:i] := src[i+63:i] + FI +ENDFOR +dst[MAX:256] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512VL + AVX512F + Miscellaneous + + + + + Unpack and interleave double-precision (64-bit) floating-point elements from the low half of each 128-bit lane in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + +DEFINE INTERLEAVE_QWORDS(src1[127:0], src2[127:0]) { + dst[63:0] := src1[63:0] + dst[127:64] := src2[63:0] + RETURN dst[127:0] +} +tmp_dst[127:0] := INTERLEAVE_QWORDS(a[127:0], b[127:0]) +tmp_dst[255:128] := INTERLEAVE_QWORDS(a[255:128], b[255:128]) +FOR j := 0 to 3 + i := j*64 + IF k[j] + dst[i+63:i] := tmp_dst[i+63:i] + ELSE + dst[i+63:i] := 0 + FI +ENDFOR +dst[MAX:256] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512VL + AVX512F + Miscellaneous + + + + + + Unpack and interleave double-precision (64-bit) floating-point elements from the low half of "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +DEFINE INTERLEAVE_QWORDS(src1[127:0], src2[127:0]) { + dst[63:0] := src1[63:0] + dst[127:64] := src2[63:0] + RETURN dst[127:0] +} +tmp_dst[127:0] := INTERLEAVE_QWORDS(a[127:0], b[127:0]) +FOR j := 0 to 1 + i := j*64 + IF k[j] + dst[i+63:i] := tmp_dst[i+63:i] + ELSE + dst[i+63:i] := src[i+63:i] + FI +ENDFOR +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512VL + AVX512F + Miscellaneous + + + + + Unpack and interleave double-precision (64-bit) floating-point elements from the low half of "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + +DEFINE INTERLEAVE_QWORDS(src1[127:0], src2[127:0]) { + dst[63:0] := src1[63:0] + dst[127:64] := src2[63:0] + RETURN dst[127:0] +} +tmp_dst[127:0] := INTERLEAVE_QWORDS(a[127:0], b[127:0]) +FOR j := 0 to 1 + i := j*64 + IF k[j] + dst[i+63:i] := tmp_dst[i+63:i] + ELSE + dst[i+63:i] := 0 + FI +ENDFOR +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512VL + AVX512F + Miscellaneous + + + + + + Unpack and interleave single-precision (32-bit) floating-point elements from the low half of each 128-bit lane in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +DEFINE INTERLEAVE_DWORDS(src1[127:0], src2[127:0]) { + dst[31:0] := src1[31:0] + dst[63:32] := src2[31:0] + dst[95:64] := src1[63:32] + dst[127:96] := src2[63:32] + RETURN dst[127:0] +} +tmp_dst[127:0] := INTERLEAVE_DWORDS(a[127:0], b[127:0]) +tmp_dst[255:128] := INTERLEAVE_DWORDS(a[255:128], b[255:128]) +FOR j := 0 to 7 + i := j*32 + IF k[j] + dst[i+31:i] := tmp_dst[i+31:i] + ELSE + dst[i+31:i] := src[i+31:i] + FI +ENDFOR +dst[MAX:256] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512VL + AVX512F + Miscellaneous + + + + + Unpack and interleave single-precision (32-bit) floating-point elements from the low half of each 128-bit lane in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + +DEFINE INTERLEAVE_DWORDS(src1[127:0], src2[127:0]) { + dst[31:0] := src1[31:0] + dst[63:32] := src2[31:0] + dst[95:64] := src1[63:32] + dst[127:96] := src2[63:32] + RETURN dst[127:0] +} +tmp_dst[127:0] := INTERLEAVE_DWORDS(a[127:0], b[127:0]) +tmp_dst[255:128] := INTERLEAVE_DWORDS(a[255:128], b[255:128]) +FOR j := 0 to 7 + i := j*32 + IF k[j] + dst[i+31:i] := tmp_dst[i+31:i] + ELSE + dst[i+31:i] := 0 + FI +ENDFOR +dst[MAX:256] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512VL + AVX512F + Miscellaneous + + + + + + Unpack and interleave single-precision (32-bit) floating-point elements from the low half of "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +DEFINE INTERLEAVE_DWORDS(src1[127:0], src2[127:0]) { + dst[31:0] := src1[31:0] + dst[63:32] := src2[31:0] + dst[95:64] := src1[63:32] + dst[127:96] := src2[63:32] + RETURN dst[127:0] +} +tmp_dst[127:0] := INTERLEAVE_DWORDS(a[127:0], b[127:0]) +FOR j := 0 to 3 + i := j*32 + IF k[j] + dst[i+31:i] := tmp_dst[i+31:i] + ELSE + dst[i+31:i] := src[i+31:i] + FI +ENDFOR +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512VL + AVX512F + Miscellaneous + + + + + Unpack and interleave single-precision (32-bit) floating-point elements from the low half of "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + +DEFINE INTERLEAVE_DWORDS(src1[127:0], src2[127:0]) { + dst[31:0] := src1[31:0] + dst[63:32] := src2[31:0] + dst[95:64] := src1[63:32] + dst[127:96] := src2[63:32] + RETURN dst[127:0] +} +tmp_dst[127:0] := INTERLEAVE_DWORDS(a[127:0], b[127:0]) +FOR j := 0 to 3 + i := j*32 + IF k[j] + dst[i+31:i] := tmp_dst[i+31:i] + ELSE + dst[i+31:i] := 0 + FI +ENDFOR +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Integer + AVX512F + Store + + + + Store 512-bits (composed of 8 packed 64-bit integers) from "a" into memory. + "mem_addr" does not need to be aligned on any particular boundary. + +MEM[mem_addr+511:mem_addr] := a[511:0] + + +
immintrin.h
+ + + Integer + AVX512F + Store + + + + Store 512-bits (composed of 16 packed 32-bit integers) from "a" into memory. + "mem_addr" does not need to be aligned on any particular boundary. + +MEM[mem_addr+511:mem_addr] := a[511:0] + + +
immintrin.h
+ + + Integer + AVX512VL + AVX512F + Store + + + + Store 256-bits (composed of 4 packed 64-bit integers) from "a" into memory. + "mem_addr" does not need to be aligned on any particular boundary. + +MEM[mem_addr+255:mem_addr] := a[255:0] + + +
immintrin.h
+ + + Integer + AVX512VL + AVX512F + Store + + + + Store 256-bits (composed of 8 packed 32-bit integers) from "a" into memory. + "mem_addr" does not need to be aligned on any particular boundary. + +MEM[mem_addr+255:mem_addr] := a[255:0] + + +
immintrin.h
+ + + Integer + AVX512VL + AVX512F + Store + + + + Store 128-bits (composed of 2 packed 64-bit integers) from "a" into memory. + "mem_addr" does not need to be aligned on any particular boundary. + +MEM[mem_addr+127:mem_addr] := a[127:0] + + +
immintrin.h
+ + + Integer + AVX512VL + AVX512F + Store + + + + Store 128-bits (composed of 4 packed 32-bit integers) from "a" into memory. + "mem_addr" does not need to be aligned on any particular boundary. + +MEM[mem_addr+127:mem_addr] := a[127:0] + + +
immintrin.h
+ + + Integer + AVX512VL + AVX512F + Store + + + + Store 256-bits (composed of 4 packed 64-bit integers) from "a" into memory. + "mem_addr" must be aligned on a 32-byte boundary or a general-protection exception may be generated. + +MEM[mem_addr+255:mem_addr] := a[255:0] + + +
immintrin.h
+ + + Integer + AVX512VL + AVX512F + Store + + + + Store 256-bits (composed of 8 packed 32-bit integers) from "a" into memory. + "mem_addr" must be aligned on a 32-byte boundary or a general-protection exception may be generated. + +MEM[mem_addr+255:mem_addr] := a[255:0] + + +
immintrin.h
+ + + Integer + AVX512VL + AVX512F + Store + + + + Store 128-bits (composed of 2 packed 64-bit integers) from "a" into memory. + "mem_addr" must be aligned on a 16-byte boundary or a general-protection exception may be generated. + +MEM[mem_addr+127:mem_addr] := a[127:0] + + +
immintrin.h
+ + + Integer + AVX512VL + AVX512F + Store + + + + Store 128-bits (composed of 4 packed 32-bit integers) from "a" into memory. + "mem_addr" must be aligned on a 16-byte boundary or a general-protection exception may be generated. + +MEM[mem_addr+127:mem_addr] := a[127:0] + + +
immintrin.h
+ + + Integer + AVX512F + Load + + + Load 512-bits (composed of 8 packed 64-bit integers) from memory into "dst". + "mem_addr" does not need to be aligned on any particular boundary. + +dst[511:0] := MEM[mem_addr+511:mem_addr] +dst[MAX:512] := 0 + + +
immintrin.h
+ + + Integer + AVX512F + Load + + + Load 512-bits (composed of 16 packed 32-bit integers) from memory into "dst". + "mem_addr" does not need to be aligned on any particular boundary. + +dst[511:0] := MEM[mem_addr+511:mem_addr] +dst[MAX:512] := 0 + + +
immintrin.h
+ + + Integer + AVX512VL + AVX512F + Load + + + Load 256-bits (composed of 4 packed 64-bit integers) from memory into "dst". + "mem_addr" does not need to be aligned on any particular boundary. + +dst[255:0] := MEM[mem_addr+255:mem_addr] +dst[MAX:256] := 0 + + +
immintrin.h
+ + + Integer + AVX512VL + AVX512F + Load + + + Load 256-bits (composed of 8 packed 32-bit integers) from memory into "dst". + "mem_addr" does not need to be aligned on any particular boundary. + +dst[255:0] := MEM[mem_addr+255:mem_addr] +dst[MAX:256] := 0 + + +
immintrin.h
+ + + Integer + AVX512VL + AVX512F + Load + + + Load 128-bits (composed of 2 packed 64-bit integers) from memory into "dst". + "mem_addr" does not need to be aligned on any particular boundary. + +dst[127:0] := MEM[mem_addr+127:mem_addr] +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Integer + AVX512VL + AVX512F + Load + + + Load 128-bits (composed of 4 packed 32-bit integers) from memory into "dst". + "mem_addr" does not need to be aligned on any particular boundary. + +dst[127:0] := MEM[mem_addr+127:mem_addr] +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Integer + AVX512VL + AVX512F + Load + + + Load 256-bits (composed of 4 packed 64-bit integers) from memory into "dst". + "mem_addr" must be aligned on a 32-byte boundary or a general-protection exception may be generated. + +dst[255:0] := MEM[mem_addr+255:mem_addr] +dst[MAX:256] := 0 + + +
immintrin.h
+ + + Integer + AVX512VL + AVX512F + Load + + + Load 256-bits (composed of 8 packed 32-bit integers) from memory into "dst". + "mem_addr" must be aligned on a 32-byte boundary or a general-protection exception may be generated. + +dst[255:0] := MEM[mem_addr+255:mem_addr] +dst[MAX:256] := 0 + + +
immintrin.h
+ + + Integer + AVX512VL + AVX512F + Load + + + Load 128-bits (composed of 2 packed 64-bit integers) from memory into "dst". + "mem_addr" must be aligned on a 16-byte boundary or a general-protection exception may be generated. + +dst[127:0] := MEM[mem_addr+127:mem_addr] +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Integer + AVX512VL + AVX512F + Load + + + Load 128-bits (composed of 4 packed 32-bit integers) from memory into "dst". + "mem_addr" must be aligned on a 16-byte boundary or a general-protection exception may be generated. + +dst[127:0] := MEM[mem_addr+127:mem_addr] +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Integer + AVX512VL + AVX512F + Logical + + + + Compute the bitwise XOR of packed 64-bit integers in "a" and "b", and store the results in "dst". + +FOR j := 0 to 3 + i := j*64 + dst[i+63:i] := a[i+63:i] XOR b[i+63:i] +ENDFOR +dst[MAX:256] := 0 + + +
immintrin.h
+ + + Integer + AVX512VL + AVX512F + Logical + + + + Compute the bitwise XOR of packed 32-bit integers in "a" and "b", and store the results in "dst". + +FOR j := 0 to 7 + i := j*32 + dst[i+31:i] := a[i+31:i] XOR b[i+31:i] +ENDFOR +dst[MAX:256] := 0 + + +
immintrin.h
+ + + Integer + AVX512VL + AVX512F + Logical + + + + Compute the bitwise XOR of packed 64-bit integers in "a" and "b", and store the results in "dst". + +FOR j := 0 to 1 + i := j*64 + dst[i+63:i] := a[i+63:i] XOR b[i+63:i] +ENDFOR +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Integer + AVX512VL + AVX512F + Logical + + + + Compute the bitwise XOR of packed 32-bit integers in "a" and "b", and store the results in "dst". + +FOR j := 0 to 3 + i := j*32 + dst[i+31:i] := a[i+31:i] XOR b[i+31:i] +ENDFOR +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Integer + AVX512VL + AVX512F + Logical + + + + Compute the bitwise OR of packed 64-bit integers in "a" and "b", and store the results in "dst". + +FOR j := 0 to 3 + i := j*64 + dst[i+63:i] := a[i+63:i] OR b[i+63:i] +ENDFOR +dst[MAX:256] := 0 + + +
immintrin.h
+ + + Integer + AVX512VL + AVX512F + Logical + + + + Compute the bitwise OR of packed 32-bit integers in "a" and "b", and store the results in "dst". + +FOR j := 0 to 7 + i := j*32 + dst[i+31:i] := a[i+31:i] OR b[i+31:i] +ENDFOR +dst[MAX:256] := 0 + + +
immintrin.h
+ + + Integer + AVX512VL + AVX512F + Logical + + + + Compute the bitwise OR of packed 64-bit integers in "a" and "b", and store the results in "dst". + +FOR j := 0 to 1 + i := j*64 + dst[i+63:i] := a[i+63:i] OR b[i+63:i] +ENDFOR +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Integer + AVX512VL + AVX512F + Logical + + + + Compute the bitwise OR of packed 32-bit integers in "a" and "b", and store the results in "dst". + +FOR j := 0 to 3 + i := j*32 + dst[i+31:i] := a[i+31:i] OR b[i+31:i] +ENDFOR +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Integer + AVX512F + VAES + Cryptography + + + + Perform the last round of an AES encryption flow on data (state) in "a" using the round key in "RoundKey", and store the results in "dst"." + FOR j := 0 to 3 + i := j*128 + a[i+127:i] := ShiftRows(a[i+127:i]) + a[i+127:i] := SubBytes(a[i+127:i]) + dst[i+127:i] := a[i+127:i] XOR RoundKey[i+127:i] +ENDFOR +dst[MAX:512] := 0 + + +
immintrin.h
+ + + Integer + AVX512F + VAES + Cryptography + + + + Perform one round of an AES encryption flow on data (state) in "a" using the round key in "RoundKey", and store the results in "dst"." + FOR j := 0 to 3 + i := j*128 + a[i+127:i] := ShiftRows(a[i+127:i]) + a[i+127:i] := SubBytes(a[i+127:i]) + a[i+127:i] := MixColumns(a[i+127:i]) + dst[i+127:i] := a[i+127:i] XOR RoundKey[i+127:i] +ENDFOR +dst[MAX:512] := 0 + + +
immintrin.h
+ + + Integer + AVX512F + VAES + Cryptography + + + + Perform the last round of an AES decryption flow on data (state) in "a" using the round key in "RoundKey", and store the results in "dst". + FOR j := 0 to 3 + i := j*128 + a[i+127:i] := InvShiftRows(a[i+127:i]) + a[i+127:i] := InvSubBytes(a[i+127:i]) + dst[i+127:i] := a[i+127:i] XOR RoundKey[i+127:i] +ENDFOR +dst[MAX:512] := 0 + + +
immintrin.h
+ + + Integer + AVX512F + VAES + Cryptography + + + + Perform one round of an AES decryption flow on data (state) in "a" using the round key in "RoundKey", and store the results in "dst". + FOR j := 0 to 3 + i := j*128 + a[i+127:i] := InvShiftRows(a[i+127:i]) + a[i+127:i] := InvSubBytes(a[i+127:i]) + a[i+127:i] := InvMixColumns(a[i+127:i]) + dst[i+127:i] := a[i+127:i] XOR RoundKey[i+127:i] +ENDFOR +dst[MAX:512] := 0 + + +
immintrin.h
+ + + Mask + AVX512F + Mask + + + + Compute the bitwise AND of 16-bit masks "a" and "b", and store the result in "k". + +k[15:0] := a[15:0] AND b[15:0] +k[MAX:16] := 0 + + +
immintrin.h
+ + + Mask + AVX512F + Mask + + + + Compute the bitwise NOT of 16-bit masks "a" and then AND with "b", and store the result in "k". + +k[15:0] := (NOT a[15:0]) AND b[15:0] +k[MAX:16] := 0 + + +
immintrin.h
+ + + Mask + AVX512F + Mask + + + Compute the bitwise NOT of 16-bit mask "a", and store the result in "k". + +k[15:0] := NOT a[15:0] +k[MAX:16] := 0 + + +
immintrin.h
+ + + Mask + AVX512F + Mask + + + + Compute the bitwise OR of 16-bit masks "a" and "b", and store the result in "k". + +k[15:0] := a[15:0] OR b[15:0] +k[MAX:16] := 0 + + +
immintrin.h
+ + + Mask + AVX512F + Mask + + + + Compute the bitwise XNOR of 16-bit masks "a" and "b", and store the result in "k". + +k[15:0] := NOT (a[15:0] XOR b[15:0]) +k[MAX:16] := 0 + + +
immintrin.h
+ + + Mask + AVX512F + Mask + + + + Compute the bitwise XOR of 16-bit masks "a" and "b", and store the result in "k". + +k[15:0] := a[15:0] XOR b[15:0] +k[MAX:16] := 0 + + +
immintrin.h
+ + + Mask + AVX512F + Mask + + + + Shift the bits of 16-bit mask "a" left by "count" while shifting in zeros, and store the least significant 16 bits of the result in "k". + +k[MAX:0] := 0 +IF count[7:0] <= 15 + k[15:0] := a[15:0] << count[7:0] +FI + + +
immintrin.h
+ + + Mask + AVX512F + Mask + + + + Shift the bits of 16-bit mask "a" right by "count" while shifting in zeros, and store the least significant 16 bits of the result in "k". + +k[MAX:0] := 0 +IF count[7:0] <= 15 + k[15:0] := a[15:0] >> count[7:0] +FI + + +
immintrin.h
+ + + Mask + AVX512F + Load + + + Load 16-bit mask from memory into "k". + +k[15:0] := MEM[mem_addr+15:mem_addr] + + +
immintrin.h
+ + + Mask + AVX512F + Store + + + + Store 16-bit mask from "a" into memory. + +MEM[mem_addr+15:mem_addr] := a[15:0] + + +
immintrin.h
+ + + Mask + AVX512F + Mask + + + + + Compute the bitwise OR of 16-bit masks "a" and "b". If the result is all zeros, store 1 in "dst", otherwise store 0 in "dst". If the result is all ones, store 1 in "all_ones", otherwise store 0 in "all_ones". + +tmp[15:0] := a[15:0] OR b[15:0] +IF tmp[15:0] == 0x0 + dst := 1 +ELSE + dst := 0 +FI +IF tmp[15:0] == 0xFFFF + MEM[all_ones+7:all_ones] := 1 +ELSE + MEM[all_ones+7:all_ones] := 0 +FI + + +
immintrin.h
+ + + Mask + AVX512F + Mask + + + + Compute the bitwise OR of 16-bit masks "a" and "b". If the result is all zeroes, store 1 in "dst", otherwise store 0 in "dst". + +tmp[15:0] := a[15:0] OR b[15:0] +IF tmp[15:0] == 0x0 + dst := 1 +ELSE + dst := 0 +FI + + +
immintrin.h
+ + + Mask + AVX512F + Mask + + + + Compute the bitwise OR of 16-bit masks "a" and "b". If the result is all ones, store 1 in "dst", otherwise store 0 in "dst". + +tmp[15:0] := a[15:0] OR b[15:0] +IF tmp[15:0] == 0xFFFF + dst := 1 +ELSE + dst := 0 +FI + + +
immintrin.h
+ + + AVX512F + Mask + + + Convert 16-bit mask "a" into an integer value, and store the result in "dst". + +dst := ZeroExtend32(a[15:0]) + + +
immintrin.h
+ + + AVX512F + Mask + + + Convert integer value "a" into an 16-bit mask, and store the result in "k". + +k := ZeroExtend16(a[15:0]) + + +
immintrin.h
+ + + Mask + AVX512F + Mask + + + + Compute the bitwise NOT of 16-bit masks "a" and then AND with "b", and store the result in "k". + +k[15:0] := (NOT a[15:0]) AND b[15:0] +k[MAX:16] := 0 + + +
immintrin.h
+ + + Mask + AVX512F + Mask + + + + Compute the bitwise AND of 16-bit masks "a" and "b", and store the result in "k". + +k[15:0] := a[15:0] AND b[15:0] +k[MAX:16] := 0 + + +
immintrin.h
+ + + Mask + AVX512F + Mask + + + Copy 16-bit mask "a" to "k". + +k[15:0] := a[15:0] +k[MAX:16] := 0 + + +
immintrin.h
+ + + Mask + AVX512F + Mask + + + Compute the bitwise NOT of 16-bit mask "a", and store the result in "k". + +k[15:0] := NOT a[15:0] +k[MAX:16] := 0 + + +
immintrin.h
+ + + Mask + AVX512F + Mask + + + + Compute the bitwise OR of 16-bit masks "a" and "b", and store the result in "k". + +k[15:0] := a[15:0] OR b[15:0] +k[MAX:16] := 0 + + +
immintrin.h
+ + + Mask + AVX512F + Mask + + + + Unpack and interleave 8 bits from masks "a" and "b", and store the 16-bit result in "k". + +k[7:0] := b[7:0] +k[15:8] := a[7:0] +k[MAX:16] := 0 + + +
immintrin.h
+ + + Mask + AVX512F + Mask + + + + Compute the bitwise XNOR of 16-bit masks "a" and "b", and store the result in "k". + +k[15:0] := NOT (a[15:0] XOR b[15:0]) +k[MAX:16] := 0 + + +
immintrin.h
+ + + Mask + AVX512F + Mask + + + + Compute the bitwise XOR of 16-bit masks "a" and "b", and store the result in "k". + +k[15:0] := a[15:0] XOR b[15:0] +k[MAX:16] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512F + Arithmetic + + + + + Add packed double-precision (64-bit) floating-point elements in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + +FOR j := 0 to 7 + i := j*64 + IF k[j] + dst[i+63:i] := a[i+63:i] + b[i+63:i] + ELSE + dst[i+63:i] := 0 + FI +ENDFOR +dst[MAX:512] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512F + Arithmetic + + + + + + Add packed double-precision (64-bit) floating-point elements in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + [round_note] + +FOR j := 0 to 7 + i := j*64 + IF k[j] + dst[i+63:i] := a[i+63:i] + b[i+63:i] + ELSE + dst[i+63:i] := 0 + FI +ENDFOR +dst[MAX:512] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512F + Arithmetic + + + + + Add packed single-precision (32-bit) floating-point elements in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + +FOR j := 0 to 15 + i := j*32 + IF k[j] + dst[i+31:i] := a[i+31:i] + b[i+31:i] + ELSE + dst[i+31:i] := 0 + FI +ENDFOR +dst[MAX:512] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512F + Arithmetic + + + + + + Add packed single-precision (32-bit) floating-point elements in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + [round_note] + +FOR j := 0 to 15 + i := j*32 + IF k[j] + dst[i+31:i] := a[i+31:i] + b[i+31:i] + ELSE + dst[i+31:i] := 0 + FI +ENDFOR +dst[MAX:512] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512F + Arithmetic + + + + + Add the lower double-precision (64-bit) floating-point element in "a" and "b", store the result in the lower element of "dst", and copy the upper element from "a" to the upper element of "dst". + [round_note] + +dst[63:0] := a[63:0] + b[63:0] +dst[127:64] := a[127:64] +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512F + Arithmetic + + + + + + + Add the lower double-precision (64-bit) floating-point element in "a" and "b", store the result in the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". + [round_note] + +IF k[0] + dst[63:0] := a[63:0] + b[63:0] +ELSE + dst[63:0] := src[63:0] +FI +dst[127:64] := a[127:64] +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512F + Arithmetic + + + + + + Add the lower double-precision (64-bit) floating-point element in "a" and "b", store the result in the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". + +IF k[0] + dst[63:0] := a[63:0] + b[63:0] +ELSE + dst[63:0] := src[63:0] +FI +dst[127:64] := a[127:64] +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512F + Arithmetic + + + + + + Add the lower double-precision (64-bit) floating-point element in "a" and "b", store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". + [round_note] + +IF k[0] + dst[63:0] := a[63:0] + b[63:0] +ELSE + dst[63:0] := 0 +FI +dst[127:64] := a[127:64] +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512F + Arithmetic + + + + + Add the lower double-precision (64-bit) floating-point element in "a" and "b", store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". + +IF k[0] + dst[63:0] := a[63:0] + b[63:0] +ELSE + dst[63:0] := 0 +FI +dst[127:64] := a[127:64] +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512F + Arithmetic + + + + + Add the lower single-precision (32-bit) floating-point element in "a" and "b", store the result in the lower element of "dst", and copy the upper 3 packed elements from "a" to the upper elements of "dst". + [round_note] + +dst[31:0] := a[31:0] + b[31:0] +dst[127:32] := a[127:32] +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512F + Arithmetic + + + + + + + Add the lower single-precision (32-bit) floating-point element in "a" and "b", store the result in the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". + [round_note] + +IF k[0] + dst[31:0] := a[31:0] + b[31:0] +ELSE + dst[31:0] := src[31:0] +FI +dst[127:32] := a[127:32] +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512F + Arithmetic + + + + + + Add the lower single-precision (32-bit) floating-point element in "a" and "b", store the result in the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". + +IF k[0] + dst[31:0] := a[31:0] + b[31:0] +ELSE + dst[31:0] := src[31:0] +FI +dst[127:32] := a[127:32] +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512F + Arithmetic + + + + + + Add the lower single-precision (32-bit) floating-point element in "a" and "b", store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". + [round_note] + +IF k[0] + dst[31:0] := a[31:0] + b[31:0] +ELSE + dst[31:0] := 0 +FI +dst[127:32] := a[127:32] +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512F + Arithmetic + + + + + Add the lower single-precision (32-bit) floating-point element in "a" and "b", store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". + +IF k[0] + dst[31:0] := a[31:0] + b[31:0] +ELSE + dst[31:0] := 0 +FI +dst[127:32] := a[127:32] +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Integer + AVX512F + Miscellaneous + + + + + + Concatenate "a" and "b" into a 128-byte immediate result, shift the result right by "imm8" 32-bit elements, and stores the low 64 bytes (16 elements) in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + +temp[1023:512] := a[511:0] +temp[511:0] := b[511:0] +temp[1023:0] := temp[1023:0] >> (32*imm8[3:0]) +FOR j := 0 to 15 + i := j*32 + IF k[j] + dst[i+31:i] := temp[i+31:i] + ELSE + dst[i+31:i] := 0 + FI +ENDFOR +dst[MAX:512] := 0 + + +
immintrin.h
+ + + Integer + AVX512F + Miscellaneous + + + + + Concatenate "a" and "b" into a 128-byte immediate result, shift the result right by "imm8" 64-bit elements, and store the low 64 bytes (8 elements) in "dst". + +temp[1023:512] := a[511:0] +temp[511:0] := b[511:0] +temp[1023:0] := temp[1023:0] >> (64*imm8[2:0]) +dst[511:0] := temp[511:0] +dst[MAX:512] := 0 + + +
immintrin.h
+ + + Integer + AVX512F + Miscellaneous + + + + + + + Concatenate "a" and "b" into a 128-byte immediate result, shift the result right by "imm8" 64-bit elements, and store the low 64 bytes (8 elements) in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +temp[1023:512] := a[511:0] +temp[511:0] := b[511:0] +temp[1023:0] := temp[1023:0] >> (64*imm8[2:0]) +FOR j := 0 to 7 + i := j*64 + IF k[j] + dst[i+63:i] := temp[i+63:i] + ELSE + dst[i+63:i] := src[i+63:i] + FI +ENDFOR +dst[MAX:512] := 0 + + +
immintrin.h
+ + + Integer + AVX512F + Miscellaneous + + + + + + Concatenate "a" and "b" into a 128-byte immediate result, shift the result right by "imm8" 64-bit elements, and stores the low 64 bytes (8 elements) in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + +temp[1023:512] := a[511:0] +temp[511:0] := b[511:0] +temp[1023:0] := temp[1023:0] >> (64*imm8[2:0]) +FOR j := 0 to 7 + i := j*64 + IF k[j] + dst[i+63:i] := temp[i+63:i] + ELSE + dst[i+63:i] := 0 + FI +ENDFOR +dst[MAX:512] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512F + Swizzle + + + Broadcast the 4 packed single-precision (32-bit) floating-point elements from "a" to all elements of "dst". + +FOR j := 0 to 15 + i := j*32 + n := (j % 4)*32 + dst[i+31:i] := a[n+31:n] +ENDFOR +dst[MAX:512] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512F + Swizzle + + + + + Broadcast the 4 packed single-precision (32-bit) floating-point elements from "a" to all elements of "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +FOR j := 0 to 15 + i := j*32 + n := (j % 4)*32 + IF k[j] + dst[i+31:i] := a[n+31:n] + ELSE + dst[i+31:i] := src[i+31:i] + FI +ENDFOR +dst[MAX:512] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512F + Swizzle + + + + Broadcast the 4 packed single-precision (32-bit) floating-point elements from "a" to all elements of "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + +FOR j := 0 to 15 + i := j*32 + n := (j % 4)*32 + IF k[j] + dst[i+31:i] := a[n+31:n] + ELSE + dst[i+31:i] := 0 + FI +ENDFOR +dst[MAX:512] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512F + Swizzle + + + Broadcast the 4 packed double-precision (64-bit) floating-point elements from "a" to all elements of "dst". + +FOR j := 0 to 7 + i := j*64 + n := (j % 4)*64 + dst[i+63:i] := a[n+63:n] +ENDFOR +dst[MAX:512] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512F + Swizzle + + + + + Broadcast the 4 packed double-precision (64-bit) floating-point elements from "a" to all elements of "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +FOR j := 0 to 7 + i := j*64 + n := (j % 4)*64 + IF k[j] + dst[i+63:i] := a[n+63:n] + ELSE + dst[i+63:i] := src[i+63:i] + FI +ENDFOR +dst[MAX:512] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512F + Swizzle + + + + Broadcast the 4 packed double-precision (64-bit) floating-point elements from "a" to all elements of "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + +FOR j := 0 to 7 + i := j*64 + n := (j % 4)*64 + IF k[j] + dst[i+63:i] := a[n+63:n] + ELSE + dst[i+63:i] := 0 + FI +ENDFOR +dst[MAX:512] := 0 + + +
immintrin.h
+ + + Integer + AVX512F + Swizzle + + + Broadcast the 4 packed 32-bit integers from "a" to all elements of "dst". + +FOR j := 0 to 15 + i := j*32 + n := (j % 4)*32 + dst[i+31:i] := a[n+31:n] +ENDFOR +dst[MAX:512] := 0 + + +
immintrin.h
+ + + Integer + AVX512F + Swizzle + + + + + Broadcast the 4 packed 32-bit integers from "a" to all elements of "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +FOR j := 0 to 15 + i := j*32 + n := (j % 4)*32 + IF k[j] + dst[i+31:i] := a[n+31:n] + ELSE + dst[i+31:i] := src[i+31:i] + FI +ENDFOR +dst[MAX:512] := 0 + + +
immintrin.h
+ + + Integer + AVX512F + Swizzle + + + + Broadcast the 4 packed 32-bit integers from "a" to all elements of "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + +FOR j := 0 to 15 + i := j*32 + n := (j % 4)*32 + IF k[j] + dst[i+31:i] := a[n+31:n] + ELSE + dst[i+31:i] := 0 + FI +ENDFOR +dst[MAX:512] := 0 + + +
immintrin.h
+ + + Integer + AVX512F + Swizzle + + + Broadcast the 4 packed 64-bit integers from "a" to all elements of "dst". + +FOR j := 0 to 7 + i := j*64 + n := (j % 4)*64 + dst[i+63:i] := a[n+63:n] +ENDFOR +dst[MAX:512] := 0 + + +
immintrin.h
+ + + Integer + AVX512F + Swizzle + + + + + Broadcast the 4 packed 64-bit integers from "a" to all elements of "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +FOR j := 0 to 7 + i := j*64 + n := (j % 4)*64 + IF k[j] + dst[i+63:i] := a[n+63:n] + ELSE + dst[i+63:i] := src[i+63:i] + FI +ENDFOR +dst[MAX:512] := 0 + + +
immintrin.h
+ + + Integer + AVX512F + Swizzle + + + + Broadcast the 4 packed 64-bit integers from "a" to all elements of "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + +FOR j := 0 to 7 + i := j*64 + n := (j % 4)*64 + IF k[j] + dst[i+63:i] := a[n+63:n] + ELSE + dst[i+63:i] := 0 + FI +ENDFOR +dst[MAX:512] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512F + Swizzle + + + Broadcast the low double-precision (64-bit) floating-point element from "a" to all elements of "dst". + +FOR j := 0 to 7 + i := j*64 + dst[i+63:i] := a[63:0] +ENDFOR +dst[MAX:512] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512F + Swizzle + + + + + Broadcast the low double-precision (64-bit) floating-point element from "a" to all elements of "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +FOR j := 0 to 7 + i := j*64 + IF k[j] + dst[i+63:i] := a[63:0] + ELSE + dst[i+63:i] := src[i+63:i] + FI +ENDFOR +dst[MAX:512] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512F + Swizzle + + + + Broadcast the low double-precision (64-bit) floating-point element from "a" to all elements of "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + +FOR j := 0 to 7 + i := j*64 + IF k[j] + dst[i+63:i] := a[63:0] + ELSE + dst[i+63:i] := 0 + FI +ENDFOR +dst[MAX:512] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512F + Swizzle + + + Broadcast the low single-precision (32-bit) floating-point element from "a" to all elements of "dst". + +FOR j := 0 to 15 + i := j*32 + dst[i+31:i] := a[31:0] +ENDFOR +dst[MAX:512] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512F + Swizzle + + + + + Broadcast the low single-precision (32-bit) floating-point element from "a" to all elements of "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +FOR j := 0 to 15 + i := j*32 + IF k[j] + dst[i+31:i] := a[31:0] + ELSE + dst[i+31:i] := src[i+31:i] + FI +ENDFOR +dst[MAX:512] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512F + Swizzle + + + + Broadcast the low single-precision (32-bit) floating-point element from "a" to all elements of "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + +FOR j := 0 to 15 + i := j*32 + IF k[j] + dst[i+31:i] := a[31:0] + ELSE + dst[i+31:i] := 0 + FI +ENDFOR +dst[MAX:512] := 0 + + +
immintrin.h
+ + + Floating Point + Mask + AVX512F + Compare + + + + + + Compare the lower double-precision (64-bit) floating-point element in "a" and "b" based on the comparison operand specified by "imm8", and store the result in mask vector "k". [sae_note] + CASE (imm8[4:0]) OF +0: OP := _CMP_EQ_OQ +1: OP := _CMP_LT_OS +2: OP := _CMP_LE_OS +3: OP := _CMP_UNORD_Q +4: OP := _CMP_NEQ_UQ +5: OP := _CMP_NLT_US +6: OP := _CMP_NLE_US +7: OP := _CMP_ORD_Q +8: OP := _CMP_EQ_UQ +9: OP := _CMP_NGE_US +10: OP := _CMP_NGT_US +11: OP := _CMP_FALSE_OQ +12: OP := _CMP_NEQ_OQ +13: OP := _CMP_GE_OS +14: OP := _CMP_GT_OS +15: OP := _CMP_TRUE_UQ +16: OP := _CMP_EQ_OS +17: OP := _CMP_LT_OQ +18: OP := _CMP_LE_OQ +19: OP := _CMP_UNORD_S +20: OP := _CMP_NEQ_US +21: OP := _CMP_NLT_UQ +22: OP := _CMP_NLE_UQ +23: OP := _CMP_ORD_S +24: OP := _CMP_EQ_US +25: OP := _CMP_NGE_UQ +26: OP := _CMP_NGT_UQ +27: OP := _CMP_FALSE_OS +28: OP := _CMP_NEQ_OS +29: OP := _CMP_GE_OQ +30: OP := _CMP_GT_OQ +31: OP := _CMP_TRUE_US +ESAC +k[0] := ( a[63:0] OP b[63:0] ) ? 1 : 0 +k[MAX:1] := 0 + + +
immintrin.h
+ + + Floating Point + Mask + AVX512F + Compare + + + + + Compare the lower double-precision (64-bit) floating-point element in "a" and "b" based on the comparison operand specified by "imm8", and store the result in mask vector "k". + CASE (imm8[4:0]) OF +0: OP := _CMP_EQ_OQ +1: OP := _CMP_LT_OS +2: OP := _CMP_LE_OS +3: OP := _CMP_UNORD_Q +4: OP := _CMP_NEQ_UQ +5: OP := _CMP_NLT_US +6: OP := _CMP_NLE_US +7: OP := _CMP_ORD_Q +8: OP := _CMP_EQ_UQ +9: OP := _CMP_NGE_US +10: OP := _CMP_NGT_US +11: OP := _CMP_FALSE_OQ +12: OP := _CMP_NEQ_OQ +13: OP := _CMP_GE_OS +14: OP := _CMP_GT_OS +15: OP := _CMP_TRUE_UQ +16: OP := _CMP_EQ_OS +17: OP := _CMP_LT_OQ +18: OP := _CMP_LE_OQ +19: OP := _CMP_UNORD_S +20: OP := _CMP_NEQ_US +21: OP := _CMP_NLT_UQ +22: OP := _CMP_NLE_UQ +23: OP := _CMP_ORD_S +24: OP := _CMP_EQ_US +25: OP := _CMP_NGE_UQ +26: OP := _CMP_NGT_UQ +27: OP := _CMP_FALSE_OS +28: OP := _CMP_NEQ_OS +29: OP := _CMP_GE_OQ +30: OP := _CMP_GT_OQ +31: OP := _CMP_TRUE_US +ESAC +k[0] := ( a[63:0] OP b[63:0] ) ? 1 : 0 +k[MAX:1] := 0 + + +
immintrin.h
+ + + Floating Point + Mask + AVX512F + Compare + + + + + + + Compare the lower double-precision (64-bit) floating-point element in "a" and "b" based on the comparison operand specified by "imm8", and store the result in mask vector "k" using zeromask "k1" (the element is zeroed out when mask bit 0 is not set). [sae_note] + CASE (imm8[4:0]) OF +0: OP := _CMP_EQ_OQ +1: OP := _CMP_LT_OS +2: OP := _CMP_LE_OS +3: OP := _CMP_UNORD_Q +4: OP := _CMP_NEQ_UQ +5: OP := _CMP_NLT_US +6: OP := _CMP_NLE_US +7: OP := _CMP_ORD_Q +8: OP := _CMP_EQ_UQ +9: OP := _CMP_NGE_US +10: OP := _CMP_NGT_US +11: OP := _CMP_FALSE_OQ +12: OP := _CMP_NEQ_OQ +13: OP := _CMP_GE_OS +14: OP := _CMP_GT_OS +15: OP := _CMP_TRUE_UQ +16: OP := _CMP_EQ_OS +17: OP := _CMP_LT_OQ +18: OP := _CMP_LE_OQ +19: OP := _CMP_UNORD_S +20: OP := _CMP_NEQ_US +21: OP := _CMP_NLT_UQ +22: OP := _CMP_NLE_UQ +23: OP := _CMP_ORD_S +24: OP := _CMP_EQ_US +25: OP := _CMP_NGE_UQ +26: OP := _CMP_NGT_UQ +27: OP := _CMP_FALSE_OS +28: OP := _CMP_NEQ_OS +29: OP := _CMP_GE_OQ +30: OP := _CMP_GT_OQ +31: OP := _CMP_TRUE_US +ESAC +IF k1[0] + k[0] := ( a[63:0] OP b[63:0] ) ? 1 : 0 +ELSE + k[0] := 0 +FI +k[MAX:1] := 0 + + +
immintrin.h
+ + + Floating Point + Mask + AVX512F + Compare + + + + + + Compare the lower double-precision (64-bit) floating-point element in "a" and "b" based on the comparison operand specified by "imm8", and store the result in mask vector "k" using zeromask "k1" (the element is zeroed out when mask bit 0 is not set). + CASE (imm8[4:0]) OF +0: OP := _CMP_EQ_OQ +1: OP := _CMP_LT_OS +2: OP := _CMP_LE_OS +3: OP := _CMP_UNORD_Q +4: OP := _CMP_NEQ_UQ +5: OP := _CMP_NLT_US +6: OP := _CMP_NLE_US +7: OP := _CMP_ORD_Q +8: OP := _CMP_EQ_UQ +9: OP := _CMP_NGE_US +10: OP := _CMP_NGT_US +11: OP := _CMP_FALSE_OQ +12: OP := _CMP_NEQ_OQ +13: OP := _CMP_GE_OS +14: OP := _CMP_GT_OS +15: OP := _CMP_TRUE_UQ +16: OP := _CMP_EQ_OS +17: OP := _CMP_LT_OQ +18: OP := _CMP_LE_OQ +19: OP := _CMP_UNORD_S +20: OP := _CMP_NEQ_US +21: OP := _CMP_NLT_UQ +22: OP := _CMP_NLE_UQ +23: OP := _CMP_ORD_S +24: OP := _CMP_EQ_US +25: OP := _CMP_NGE_UQ +26: OP := _CMP_NGT_UQ +27: OP := _CMP_FALSE_OS +28: OP := _CMP_NEQ_OS +29: OP := _CMP_GE_OQ +30: OP := _CMP_GT_OQ +31: OP := _CMP_TRUE_US +ESAC +IF k1[0] + k[0] := ( a[63:0] OP b[63:0] ) ? 1 : 0 +ELSE + k[0] := 0 +FI +k[MAX:1] := 0 + + +
immintrin.h
+ + + Floating Point + Mask + AVX512F + Compare + + + + + + Compare the lower single-precision (32-bit) floating-point element in "a" and "b" based on the comparison operand specified by "imm8", and store the result in mask vector "k". [sae_note] + CASE (imm8[4:0]) OF +0: OP := _CMP_EQ_OQ +1: OP := _CMP_LT_OS +2: OP := _CMP_LE_OS +3: OP := _CMP_UNORD_Q +4: OP := _CMP_NEQ_UQ +5: OP := _CMP_NLT_US +6: OP := _CMP_NLE_US +7: OP := _CMP_ORD_Q +8: OP := _CMP_EQ_UQ +9: OP := _CMP_NGE_US +10: OP := _CMP_NGT_US +11: OP := _CMP_FALSE_OQ +12: OP := _CMP_NEQ_OQ +13: OP := _CMP_GE_OS +14: OP := _CMP_GT_OS +15: OP := _CMP_TRUE_UQ +16: OP := _CMP_EQ_OS +17: OP := _CMP_LT_OQ +18: OP := _CMP_LE_OQ +19: OP := _CMP_UNORD_S +20: OP := _CMP_NEQ_US +21: OP := _CMP_NLT_UQ +22: OP := _CMP_NLE_UQ +23: OP := _CMP_ORD_S +24: OP := _CMP_EQ_US +25: OP := _CMP_NGE_UQ +26: OP := _CMP_NGT_UQ +27: OP := _CMP_FALSE_OS +28: OP := _CMP_NEQ_OS +29: OP := _CMP_GE_OQ +30: OP := _CMP_GT_OQ +31: OP := _CMP_TRUE_US +ESAC +k[0] := ( a[31:0] OP b[31:0] ) ? 1 : 0 +k[MAX:1] := 0 + + +
immintrin.h
+ + + Floating Point + Mask + AVX512F + Compare + + + + + Compare the lower single-precision (32-bit) floating-point element in "a" and "b" based on the comparison operand specified by "imm8", and store the result in mask vector "k". + CASE (imm8[4:0]) OF +0: OP := _CMP_EQ_OQ +1: OP := _CMP_LT_OS +2: OP := _CMP_LE_OS +3: OP := _CMP_UNORD_Q +4: OP := _CMP_NEQ_UQ +5: OP := _CMP_NLT_US +6: OP := _CMP_NLE_US +7: OP := _CMP_ORD_Q +8: OP := _CMP_EQ_UQ +9: OP := _CMP_NGE_US +10: OP := _CMP_NGT_US +11: OP := _CMP_FALSE_OQ +12: OP := _CMP_NEQ_OQ +13: OP := _CMP_GE_OS +14: OP := _CMP_GT_OS +15: OP := _CMP_TRUE_UQ +16: OP := _CMP_EQ_OS +17: OP := _CMP_LT_OQ +18: OP := _CMP_LE_OQ +19: OP := _CMP_UNORD_S +20: OP := _CMP_NEQ_US +21: OP := _CMP_NLT_UQ +22: OP := _CMP_NLE_UQ +23: OP := _CMP_ORD_S +24: OP := _CMP_EQ_US +25: OP := _CMP_NGE_UQ +26: OP := _CMP_NGT_UQ +27: OP := _CMP_FALSE_OS +28: OP := _CMP_NEQ_OS +29: OP := _CMP_GE_OQ +30: OP := _CMP_GT_OQ +31: OP := _CMP_TRUE_US +ESAC +k[0] := ( a[31:0] OP b[31:0] ) ? 1 : 0 +k[MAX:1] := 0 + + +
immintrin.h
+ + + Floating Point + Mask + AVX512F + Compare + + + + + + + Compare the lower single-precision (32-bit) floating-point element in "a" and "b" based on the comparison operand specified by "imm8", and store the result in mask vector "k" using zeromask "k1" (the element is zeroed out when mask bit 0 is not set). [sae_note] + CASE (imm8[4:0]) OF +0: OP := _CMP_EQ_OQ +1: OP := _CMP_LT_OS +2: OP := _CMP_LE_OS +3: OP := _CMP_UNORD_Q +4: OP := _CMP_NEQ_UQ +5: OP := _CMP_NLT_US +6: OP := _CMP_NLE_US +7: OP := _CMP_ORD_Q +8: OP := _CMP_EQ_UQ +9: OP := _CMP_NGE_US +10: OP := _CMP_NGT_US +11: OP := _CMP_FALSE_OQ +12: OP := _CMP_NEQ_OQ +13: OP := _CMP_GE_OS +14: OP := _CMP_GT_OS +15: OP := _CMP_TRUE_UQ +16: OP := _CMP_EQ_OS +17: OP := _CMP_LT_OQ +18: OP := _CMP_LE_OQ +19: OP := _CMP_UNORD_S +20: OP := _CMP_NEQ_US +21: OP := _CMP_NLT_UQ +22: OP := _CMP_NLE_UQ +23: OP := _CMP_ORD_S +24: OP := _CMP_EQ_US +25: OP := _CMP_NGE_UQ +26: OP := _CMP_NGT_UQ +27: OP := _CMP_FALSE_OS +28: OP := _CMP_NEQ_OS +29: OP := _CMP_GE_OQ +30: OP := _CMP_GT_OQ +31: OP := _CMP_TRUE_US +ESAC +IF k1[0] + k[0] := ( a[31:0] OP b[31:0] ) ? 1 : 0 +ELSE + k[0] := 0 +FI +k[MAX:1] := 0 + + +
immintrin.h
+ + + Floating Point + Mask + AVX512F + Compare + + + + + + Compare the lower single-precision (32-bit) floating-point element in "a" and "b" based on the comparison operand specified by "imm8", and store the result in mask vector "k" using zeromask "k1" (the element is zeroed out when mask bit 0 is not set). + CASE (imm8[4:0]) OF +0: OP := _CMP_EQ_OQ +1: OP := _CMP_LT_OS +2: OP := _CMP_LE_OS +3: OP := _CMP_UNORD_Q +4: OP := _CMP_NEQ_UQ +5: OP := _CMP_NLT_US +6: OP := _CMP_NLE_US +7: OP := _CMP_ORD_Q +8: OP := _CMP_EQ_UQ +9: OP := _CMP_NGE_US +10: OP := _CMP_NGT_US +11: OP := _CMP_FALSE_OQ +12: OP := _CMP_NEQ_OQ +13: OP := _CMP_GE_OS +14: OP := _CMP_GT_OS +15: OP := _CMP_TRUE_UQ +16: OP := _CMP_EQ_OS +17: OP := _CMP_LT_OQ +18: OP := _CMP_LE_OQ +19: OP := _CMP_UNORD_S +20: OP := _CMP_NEQ_US +21: OP := _CMP_NLT_UQ +22: OP := _CMP_NLE_UQ +23: OP := _CMP_ORD_S +24: OP := _CMP_EQ_US +25: OP := _CMP_NGE_UQ +26: OP := _CMP_NGT_UQ +27: OP := _CMP_FALSE_OS +28: OP := _CMP_NEQ_OS +29: OP := _CMP_GE_OQ +30: OP := _CMP_GT_OQ +31: OP := _CMP_TRUE_US +ESAC +IF k1[0] + k[0] := ( a[31:0] OP b[31:0] ) ? 1 : 0 +ELSE + k[0] := 0 +FI +k[MAX:1] := 0 + + +
immintrin.h
+ + + Floating Point + Flag + AVX512F + Compare + + + + + + Compare the lower double-precision (64-bit) floating-point element in "a" and "b" based on the comparison operand specified by "imm8", and return the boolean result (0 or 1). [sae_note] + CASE (imm8[4:0]) OF +0: OP := _CMP_EQ_OQ +1: OP := _CMP_LT_OS +2: OP := _CMP_LE_OS +3: OP := _CMP_UNORD_Q +4: OP := _CMP_NEQ_UQ +5: OP := _CMP_NLT_US +6: OP := _CMP_NLE_US +7: OP := _CMP_ORD_Q +8: OP := _CMP_EQ_UQ +9: OP := _CMP_NGE_US +10: OP := _CMP_NGT_US +11: OP := _CMP_FALSE_OQ +12: OP := _CMP_NEQ_OQ +13: OP := _CMP_GE_OS +14: OP := _CMP_GT_OS +15: OP := _CMP_TRUE_UQ +16: OP := _CMP_EQ_OS +17: OP := _CMP_LT_OQ +18: OP := _CMP_LE_OQ +19: OP := _CMP_UNORD_S +20: OP := _CMP_NEQ_US +21: OP := _CMP_NLT_UQ +22: OP := _CMP_NLE_UQ +23: OP := _CMP_ORD_S +24: OP := _CMP_EQ_US +25: OP := _CMP_NGE_UQ +26: OP := _CMP_NGT_UQ +27: OP := _CMP_FALSE_OS +28: OP := _CMP_NEQ_OS +29: OP := _CMP_GE_OQ +30: OP := _CMP_GT_OQ +31: OP := _CMP_TRUE_US +ESAC +RETURN ( a[63:0] OP b[63:0] ) ? 1 : 0 + + +
immintrin.h
+ + + Floating Point + Flag + AVX512F + Compare + + + + + + Compare the lower single-precision (32-bit) floating-point element in "a" and "b" based on the comparison operand specified by "imm8", and return the boolean result (0 or 1). [sae_note] + CASE (imm8[4:0]) OF +0: OP := _CMP_EQ_OQ +1: OP := _CMP_LT_OS +2: OP := _CMP_LE_OS +3: OP := _CMP_UNORD_Q +4: OP := _CMP_NEQ_UQ +5: OP := _CMP_NLT_US +6: OP := _CMP_NLE_US +7: OP := _CMP_ORD_Q +8: OP := _CMP_EQ_UQ +9: OP := _CMP_NGE_US +10: OP := _CMP_NGT_US +11: OP := _CMP_FALSE_OQ +12: OP := _CMP_NEQ_OQ +13: OP := _CMP_GE_OS +14: OP := _CMP_GT_OS +15: OP := _CMP_TRUE_UQ +16: OP := _CMP_EQ_OS +17: OP := _CMP_LT_OQ +18: OP := _CMP_LE_OQ +19: OP := _CMP_UNORD_S +20: OP := _CMP_NEQ_US +21: OP := _CMP_NLT_UQ +22: OP := _CMP_NLE_UQ +23: OP := _CMP_ORD_S +24: OP := _CMP_EQ_US +25: OP := _CMP_NGE_UQ +26: OP := _CMP_NGT_UQ +27: OP := _CMP_FALSE_OS +28: OP := _CMP_NEQ_OS +29: OP := _CMP_GE_OQ +30: OP := _CMP_GT_OQ +31: OP := _CMP_TRUE_US +ESAC +RETURN ( a[31:0] OP b[31:0] ) ? 1 : 0 + + + +
immintrin.h
+ + + Floating Point + AVX512F + Swizzle + + + + + Contiguously store the active double-precision (64-bit) floating-point elements in "a" (those with their respective bit set in writemask "k") to "dst", and pass through the remaining elements from "src". + +size := 64 +m := 0 +FOR j := 0 to 7 + i := j*64 + IF k[j] + dst[m+size-1:m] := a[i+63:i] + m := m + size + FI +ENDFOR +dst[511:m] := src[511:m] +dst[MAX:512] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512F + Store + Swizzle + + + + + Contiguously store the active double-precision (64-bit) floating-point elements in "a" (those with their respective bit set in writemask "k") to unaligned memory at "base_addr". + +size := 64 +m := base_addr +FOR j := 0 to 7 + i := j*64 + IF k[j] + MEM[m+size-1:m] := a[i+63:i] + m := m + size + FI +ENDFOR + + +
immintrin.h
+ + + Floating Point + AVX512F + Swizzle + + + + Contiguously store the active double-precision (64-bit) floating-point elements in "a" (those with their respective bit set in zeromask "k") to "dst", and set the remaining elements to zero. + +size := 64 +m := 0 +FOR j := 0 to 7 + i := j*64 + IF k[j] + dst[m+size-1:m] := a[i+63:i] + m := m + size + FI +ENDFOR +dst[511:m] := 0 +dst[MAX:512] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512F + Swizzle + + + + + Contiguously store the active single-precision (32-bit) floating-point elements in "a" (those with their respective bit set in writemask "k") to "dst", and pass through the remaining elements from "src". + +size := 32 +m := 0 +FOR j := 0 to 15 + i := j*32 + IF k[j] + dst[m+size-1:m] := a[i+31:i] + m := m + size + FI +ENDFOR +dst[511:m] := src[511:m] +dst[MAX:512] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512F + Store + Swizzle + + + + + Contiguously store the active single-precision (32-bit) floating-point elements in "a" (those with their respective bit set in writemask "k") to unaligned memory at "base_addr". + +size := 32 +m := base_addr +FOR j := 0 to 15 + i := j*32 + IF k[j] + MEM[m+size-1:m] := a[i+31:i] + m := m + size + FI +ENDFOR + + +
immintrin.h
+ + + Floating Point + AVX512F + Swizzle + + + + Contiguously store the active single-precision (32-bit) floating-point elements in "a" (those with their respective bit set in zeromask "k") to "dst", and set the remaining elements to zero. + +size := 32 +m := 0 +FOR j := 0 to 15 + i := j*32 + IF k[j] + dst[m+size-1:m] := a[i+31:i] + m := m + size + FI +ENDFOR +dst[511:m] := 0 +dst[MAX:512] := 0 + + +
immintrin.h
+ + + Floating Point + Integer + AVX512F + Convert + + + Convert packed signed 32-bit integers in "a" to packed double-precision (64-bit) floating-point elements, and store the results in "dst". + +FOR j := 0 to 7 + i := j*32 + m := j*64 + dst[m+63:m] := Convert_Int32_To_FP64(a[i+31:i]) +ENDFOR +dst[MAX:512] := 0 + + +
immintrin.h
+ + + Floating Point + Integer + AVX512F + Convert + + + + + Convert packed signed 32-bit integers in "a" to packed double-precision (64-bit) floating-point elements, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +FOR j := 0 to 7 + i := j*32 + m := j*64 + IF k[j] + dst[m+63:m] := Convert_Int32_To_FP64(a[i+31:i]) + ELSE + dst[m+63:m] := src[m+63:m] + FI +ENDFOR +dst[MAX:512] := 0 + + +
immintrin.h
+ + + Floating Point + Integer + AVX512F + Convert + + + + Convert packed signed 32-bit integers in "a" to packed double-precision (64-bit) floating-point elements, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + +FOR j := 0 to 7 + i := j*32 + m := j*64 + IF k[j] + dst[m+63:m] := Convert_Int32_To_FP64(a[i+31:i]) + ELSE + dst[m+63:m] := 0 + FI +ENDFOR +dst[MAX:512] := 0 + + +
immintrin.h
+ + + Floating Point + Integer + AVX512F + Convert + + + + Convert packed signed 32-bit integers in "a" to packed single-precision (32-bit) floating-point elements, and store the results in "dst". + [round_note] + +FOR j := 0 to 15 + i := 32*j + dst[i+31:i] := Convert_Int32_To_FP32(a[i+31:i]) +ENDFOR +dst[MAX:512] := 0 + + +
immintrin.h
+ + + Floating Point + Integer + AVX512F + Convert + + + Convert packed signed 32-bit integers in "a" to packed single-precision (32-bit) floating-point elements, and store the results in "dst". + +FOR j := 0 to 15 + i := 32*j + dst[i+31:i] := Convert_Int32_To_FP32(a[i+31:i]) +ENDFOR +dst[MAX:512] := 0 + + +
immintrin.h
+ + + Floating Point + Integer + AVX512F + Convert + + + + + + Convert packed signed 32-bit integers in "a" to packed single-precision (32-bit) floating-point elements, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + [round_note] + +FOR j := 0 to 15 + i := j*32 + IF k[j] + dst[i+31:i] := Convert_Int32_To_FP32(a[i+31:i]) + ELSE + dst[i+31:i] := src[i+31:i] + FI +ENDFOR +dst[MAX:512] := 0 + + +
immintrin.h
+ + + Floating Point + Integer + AVX512F + Convert + + + + + Convert packed signed 32-bit integers in "a" to packed single-precision (32-bit) floating-point elements, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +FOR j := 0 to 15 + i := j*32 + IF k[j] + dst[i+31:i] := Convert_Int32_To_FP32(a[i+31:i]) + ELSE + dst[i+31:i] := src[i+31:i] + FI +ENDFOR +dst[MAX:512] := 0 + + +
immintrin.h
+ + + Floating Point + Integer + AVX512F + Convert + + + + + Convert packed signed 32-bit integers in "a" to packed single-precision (32-bit) floating-point elements, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + [round_note] + +FOR j := 0 to 15 + i := 32*j + IF k[j] + dst[i+31:i] := Convert_Int32_To_FP32(a[i+31:i]) + ELSE + dst[i+31:i] := 0 + FI +ENDFOR +dst[MAX:512] := 0 + + +
immintrin.h
+ + + Floating Point + Integer + AVX512F + Convert + + + + Convert packed signed 32-bit integers in "a" to packed single-precision (32-bit) floating-point elements, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + +FOR j := 0 to 15 + i := 32*j + IF k[j] + dst[i+31:i] := Convert_Int32_To_FP32(a[i+31:i]) + ELSE + dst[i+31:i] := 0 + FI +ENDFOR +dst[MAX:512] := 0 + + +
immintrin.h
+ + + Floating Point + Integer + AVX512F + Convert + + + + Convert packed double-precision (64-bit) floating-point elements in "a" to packed 32-bit integers, and store the results in "dst". + [round_note] + +FOR j := 0 to 7 + i := 32*j + k := 64*j + dst[i+31:i] := Convert_FP64_To_Int32(a[k+63:k]) +ENDFOR +dst[MAX:256] := 0 + + +
immintrin.h
+ + + Floating Point + Integer + AVX512F + Convert + + + Convert packed double-precision (64-bit) floating-point elements in "a" to packed 32-bit integers, and store the results in "dst". + +FOR j := 0 to 7 + i := 32*j + k := 64*j + dst[i+31:i] := Convert_FP64_To_Int32(a[k+63:k]) +ENDFOR +dst[MAX:256] := 0 + + +
immintrin.h
+ + + Floating Point + Integer + AVX512F + Convert + + + + + + Convert packed double-precision (64-bit) floating-point elements in "a" to packed 32-bit integers, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + [round_note] + +FOR j := 0 to 7 + i := j*32 + l := j*64 + IF k[j] + dst[i+31:i] := Convert_FP64_To_Int32(a[l+63:l]) + ELSE + dst[i+31:i] := src[i+31:i] + FI +ENDFOR +dst[MAX:256] := 0 + + +
immintrin.h
+ + + Floating Point + Integer + AVX512F + Convert + + + + + Convert packed double-precision (64-bit) floating-point elements in "a" to packed 32-bit integers, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +FOR j := 0 to 7 + i := j*32 + l := j*64 + IF k[j] + dst[i+31:i] := Convert_FP64_To_Int32(a[l+63:l]) + ELSE + dst[i+31:i] := src[i+31:i] + FI +ENDFOR +dst[MAX:256] := 0 + + +
immintrin.h
+ + + Floating Point + Integer + AVX512F + Convert + + + + + Convert packed double-precision (64-bit) floating-point elements in "a" to packed 32-bit integers, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + [round_note] + +FOR j := 0 to 7 + i := 32*j + l := 64*j + IF k[j] + dst[i+31:i] := Convert_FP64_To_Int32(a[l+63:l]) + ELSE + dst[i+31:i] := 0 + FI +ENDFOR +dst[MAX:256] := 0 + + +
immintrin.h
+ + + Floating Point + Integer + AVX512F + Convert + + + + Convert packed double-precision (64-bit) floating-point elements in "a" to packed 32-bit integers, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + +FOR j := 0 to 7 + i := 32*j + l := 64*j + IF k[j] + dst[i+31:i] := Convert_FP64_To_Int32(a[l+63:l]) + ELSE + dst[i+31:i] := 0 + FI +ENDFOR +dst[MAX:256] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512F + Convert + + + + Convert packed double-precision (64-bit) floating-point elements in "a" to packed single-precision (32-bit) floating-point elements, and store the results in "dst". + [round_note] + +FOR j := 0 to 7 + i := 32*j + k := 64*j + dst[i+31:i] := Convert_FP64_To_FP32(a[k+63:k]) +ENDFOR +dst[MAX:256] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512F + Convert + + + Convert packed double-precision (64-bit) floating-point elements in "a" to packed single-precision (32-bit) floating-point elements, and store the results in "dst". + +FOR j := 0 to 7 + i := 32*j + k := 64*j + dst[i+31:i] := Convert_FP64_To_FP32(a[k+63:k]) +ENDFOR +dst[MAX:256] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512F + Convert + + + + + + Convert packed double-precision (64-bit) floating-point elements in "a" to packed single-precision (32-bit) floating-point elements, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + [round_note] + +FOR j := 0 to 7 + i := j*32 + l := j*64 + IF k[j] + dst[i+31:i] := Convert_FP64_To_FP32(a[l+63:l]) + ELSE + dst[i+31:i] := src[i+31:i] + FI +ENDFOR +dst[MAX:256] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512F + Convert + + + + + Convert packed double-precision (64-bit) floating-point elements in "a" to packed single-precision (32-bit) floating-point elements, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +FOR j := 0 to 7 + i := 32*j + l := 64*j + IF k[j] + dst[i+31:i] := Convert_FP64_To_FP32(a[l+63:l]) + ELSE + dst[i+31:i] := src[i+31:i] + FI +ENDFOR +dst[MAX:256] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512F + Convert + + + + + Convert packed double-precision (64-bit) floating-point elements in "a" to packed single-precision (32-bit) floating-point elements, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + [round_note] + +FOR j := 0 to 7 + i := j*32 + l := j*64 + IF k[j] + dst[i+31:i] := Convert_FP64_To_FP32(a[l+63:l]) + ELSE + dst[i+31:i] := 0 + FI +ENDFOR +dst[MAX:256] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512F + Convert + + + + Convert packed double-precision (64-bit) floating-point elements in "a" to packed single-precision (32-bit) floating-point elements, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + +FOR j := 0 to 7 + i := j*32 + l := j*64 + IF k[j] + dst[i+31:i] := Convert_FP64_To_FP32(a[l+63:l]) + ELSE + dst[i+31:i] := 0 + FI +ENDFOR +dst[MAX:256] := 0 + + +
immintrin.h
+ + + Floating Point + Integer + AVX512F + Convert + + + + Convert packed double-precision (64-bit) floating-point elements in "a" to packed unsigned 32-bit integers, and store the results in "dst". + [round_note] + +FOR j := 0 to 7 + i := 32*j + k := 64*j + dst[i+31:i] := Convert_FP64_To_UInt32(a[k+63:k]) +ENDFOR +dst[MAX:256] := 0 + + +
immintrin.h
+ + + Floating Point + Integer + AVX512F + Convert + + + Convert packed double-precision (64-bit) floating-point elements in "a" to packed unsigned 32-bit integers, and store the results in "dst". + +FOR j := 0 to 7 + i := 32*j + k := 64*j + dst[i+31:i] := Convert_FP64_To_UInt32(a[k+63:k]) +ENDFOR +dst[MAX:256] := 0 + + +
immintrin.h
+ + + Floating Point + Integer + AVX512F + Convert + + + + + + Convert packed double-precision (64-bit) floating-point elements in "a" to packed unsigned 32-bit integers, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + [round_note] + +FOR j := 0 to 7 + i := j*32 + l := j*64 + IF k[j] + dst[i+31:i] := Convert_FP64_To_UInt32(a[l+63:l]) + ELSE + dst[i+31:i] := src[i+31:i] + FI +ENDFOR +dst[MAX:256] := 0 + + +
immintrin.h
+ + + Floating Point + Integer + AVX512F + Convert + + + + + Convert packed double-precision (64-bit) floating-point elements in "a" to packed unsigned 32-bit integers, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +FOR j := 0 to 7 + i := j*32 + l := j*64 + IF k[j] + dst[i+31:i] := Convert_FP64_To_UInt32(a[l+63:l]) + ELSE + dst[i+31:i] := src[i+31:i] + FI +ENDFOR +dst[MAX:256] := 0 + + +
immintrin.h
+ + + Floating Point + Integer + AVX512F + Convert + + + + + Convert packed double-precision (64-bit) floating-point elements in "a" to packed unsigned 32-bit integers, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + [round_note] + +FOR j := 0 to 7 + i := 32*j + l := 64*j + IF k[j] + dst[i+31:i] := Convert_FP64_To_UInt32(a[l+63:l]) + ELSE + dst[i+31:i] := 0 + FI +ENDFOR +dst[MAX:256] := 0 + + +
immintrin.h
+ + + Floating Point + Integer + AVX512F + Convert + + + + Convert packed double-precision (64-bit) floating-point elements in "a" to packed unsigned 32-bit integers, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + +FOR j := 0 to 7 + i := 32*j + l := 64*j + IF k[j] + dst[i+31:i] := Convert_FP64_To_UInt32(a[l+63:l]) + ELSE + dst[i+31:i] := 0 + FI +ENDFOR +dst[MAX:256] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512F + Convert + + + + Convert packed half-precision (16-bit) floating-point elements in "a" to packed single-precision (32-bit) floating-point elements, and store the results in "dst". [sae_note] + +FOR j := 0 to 15 + i := j*32 + m := j*16 + dst[i+31:i] := Convert_FP16_To_FP32(a[m+15:m]) +ENDFOR +dst[MAX:512] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512F + Convert + + + Convert packed half-precision (16-bit) floating-point elements in "a" to packed single-precision (32-bit) floating-point elements, and store the results in "dst". + +FOR j := 0 to 15 + i := j*32 + m := j*16 + dst[i+31:i] := Convert_FP16_To_FP32(a[m+15:m]) +ENDFOR +dst[MAX:512] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512F + Convert + + + + + + Convert packed half-precision (16-bit) floating-point elements in "a" to packed single-precision (32-bit) floating-point elements, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). [sae_note] + +FOR j := 0 to 15 + i := j*32 + m := j*16 + IF k[j] + dst[i+31:i] := Convert_FP16_To_FP32(a[m+15:m]) + ELSE + dst[i+31:i] := src[i+31:i] + FI +ENDFOR +dst[MAX:512] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512F + Convert + + + + + Convert packed half-precision (16-bit) floating-point elements in "a" to packed single-precision (32-bit) floating-point elements, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +FOR j := 0 to 15 + i := j*32 + m := j*16 + IF k[j] + dst[i+31:i] := Convert_FP16_To_FP32(a[m+15:m]) + ELSE + dst[i+31:i] := src[i+31:i] + FI +ENDFOR +dst[MAX:512] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512F + Convert + + + + + Convert packed half-precision (16-bit) floating-point elements in "a" to packed single-precision (32-bit) floating-point elements, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). [sae_note] + +FOR j := 0 to 15 + i := j*32 + m := j*16 + IF k[j] + dst[i+31:i] := Convert_FP16_To_FP32(a[m+15:m]) + ELSE + dst[i+31:i] := 0 + FI +ENDFOR +dst[MAX:512] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512F + Convert + + + + Convert packed half-precision (16-bit) floating-point elements in "a" to packed single-precision (32-bit) floating-point elements, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + +FOR j := 0 to 15 + i := j*32 + m := j*16 + IF k[j] + dst[i+31:i] := Convert_FP16_To_FP32(a[m+15:m]) + ELSE + dst[i+31:i] := 0 + FI +ENDFOR +dst[MAX:512] := 0 + + +
immintrin.h
+ + + Floating Point + Integer + AVX512F + Convert + + + + Convert packed single-precision (32-bit) floating-point elements in "a" to packed 32-bit integers, and store the results in "dst". + [round_note] -CASE s OF -_MM_SWIZ_REG_NONE: - dst[511:0] := v[511:0] -_MM_SWIZ_REG_DCBA: - dst[511:0] := v[511:0] -_MM_SWIZ_REG_CDAB: - FOR j := 0 to 3 - i := j*64 - IF k[j*2] - dst[i+63:i] := v[i+127:i+64] - ELSE - dst[i+63:i] := src[i+63:i] - FI - IF k[j*2+1] - dst[i+127:i+64] := v[i+63:i] - ELSE - dst[i+127:i+64] := src[i+127:i+64] - FI - ENDFOR -_MM_SWIZ_REG_BADC: - FOR j := 0 to 1 - i := j*256 - IF k[j*4] - dst[i+63:i] := v[i+191:i+128] - ELSE - dst[i+63:i] := src[i+63:i] - FI - IF k[j*4+1] - dst[i+127:i+64] := v[i+255:i+192] - ELSE - dst[i+127:i+64] := src[i+127:i+64] - FI - IF k[j*4+2] - dst[i+191:i+128] := v[i+63:i] - ELSE - dst[i+191:i+128] := src[i+191:i+128] - FI - IF k[j*4+3] - dst[i+255:i+192] := v[i+127:i+64] - ELSE - dst[i+255:i+192] := src[i+255:i+192] - FI - ENDFOR -_MM_SWIZ_REG_AAAA: - FOR j := 0 to 1 - i := j*256 - IF k[j*4] - dst[i+63:i] := v[i+63:i] - ELSE - dst[i+63:i] := src[i+63:i] - FI - IF k[j*4+1] - dst[i+127:i+64] := v[i+63:i] - ELSE - dst[i+127:i+64] := src[i+127:i+64] - FI - IF k[j*4+2] - dst[i+191:i+128] := v[i+63:i] - ELSE - dst[i+191:i+128] := src[i+191:i+128] - FI - IF k[j*4+3] - dst[i+255:i+192] := v[i+63:i] - ELSE - dst[i+255:i+192] := src[i+255:i+192] - FI - ENDFOR -_MM_SWIZ_REG_BBBB: - FOR j := 0 to 1 - i := j*256 - IF k[j*4] - dst[i+63:i] := v[i+127:i+63] - ELSE - dst[i+63:i] := src[i+63:i] - FI - IF k[j*4+1] - dst[i+127:i+64] := v[i+127:i+63] - ELSE - dst[i+127:i+64] := src[i+127:i+64] - FI - IF k[j*4+2] - dst[i+191:i+128] := v[i+127:i+63] - ELSE - dst[i+191:i+128] := src[i+191:i+128] - FI - IF k[j*4+3] - dst[i+255:i+192] := v[i+127:i+63] - ELSE - dst[i+255:i+192] := src[i+255:i+192] - FI - ENDFOR -_MM_SWIZ_REG_CCCC: - FOR j := 0 to 1 - i := j*256 - IF k[j*4] - dst[i+63:i] := v[i+191:i+128] - ELSE - dst[i+63:i] := src[i+63:i] - FI - IF k[j*4+1] - dst[i+127:i+64] := v[i+191:i+128] - ELSE - dst[i+127:i+64] := src[i+127:i+64] - FI - IF k[j*4+2] - dst[i+191:i+128] := v[i+191:i+128] - ELSE - dst[i+191:i+128] := src[i+191:i+128] - FI - IF k[j*4+3] - dst[i+255:i+192] := v[i+191:i+128] - ELSE - dst[i+255:i+192] := src[i+255:i+192] - FI - ENDFOR -_MM_SWIZ_REG_DDDD: - FOR j := 0 to 1 - i := j*256 - IF k[j*4] - dst[i+63:i] := v[i+255:i+192] - ELSE - dst[i+63:i] := src[i+63:i] - FI - IF k[j*4+1] - dst[i+127:i+64] := v[i+255:i+192] - ELSE - dst[i+127:i+64] := src[i+127:i+64] - FI - IF k[j*4+2] - dst[i+191:i+128] := v[i+255:i+192] - ELSE - dst[i+191:i+128] := src[i+191:i+128] - FI - IF k[j*4+3] - dst[i+255:i+192] := v[i+255:i+192] - ELSE - dst[i+255:i+192] := src[i+255:i+192] - FI - ENDFOR -_MM_SWIZ_REG_DACB: - FOR j := 0 to 1 - i := j*256 - IF k[j*4] - dst[i+63:i] := v[i+127:i+64] - ELSE - dst[i+63:i] := src[i+63:i] - FI - IF k[j*4+1] - dst[i+127:i+64] := v[i+191:i+128] - ELSE - dst[i+127:i+64] := src[i+127:i+64] - FI - IF k[j*4+2] - dst[i+191:i+128] := v[i+63:i] - ELSE - dst[i+191:i+128] := src[i+191:i+128] - FI - IF k[j*4+3] - dst[i+255:i+192] := v[i+255:i+192] - ELSE - dst[i+255:i+192] := src[i+255:i+192] - FI - ENDFOR -ESAC +FOR j := 0 to 15 + i := 32*j + dst[i+31:i] := Convert_FP32_To_Int32(a[i+31:i]) +ENDFOR +dst[MAX:512] := 0 + + +
immintrin.h
+ + + Floating Point + Integer + AVX512F + Convert + + + Convert packed single-precision (32-bit) floating-point elements in "a" to packed 32-bit integers, and store the results in "dst". + +FOR j := 0 to 15 + i := 32*j + dst[i+31:i] := Convert_FP32_To_Int32(a[i+31:i]) +ENDFOR +dst[MAX:512] := 0 + + +
immintrin.h
+ + + Floating Point + Integer + AVX512F + Convert + + + + + + Convert packed single-precision (32-bit) floating-point elements in "a" to packed 32-bit integers, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + [round_note] + +FOR j := 0 to 15 + i := j*32 + IF k[j] + dst[i+31:i] := Convert_FP32_To_Int32(a[i+31:i]) + ELSE + dst[i+31:i] := src[i+31:i] + FI +ENDFOR +dst[MAX:512] := 0 + + +
immintrin.h
+ + + Floating Point + Integer + AVX512F + Convert + + + + + Convert packed single-precision (32-bit) floating-point elements in "a" to packed 32-bit integers, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +FOR j := 0 to 15 + i := j*32 + IF k[j] + dst[i+31:i] := Convert_FP32_To_Int32(a[i+31:i]) + ELSE + dst[i+31:i] := src[i+31:i] + FI +ENDFOR +dst[MAX:512] := 0 + + +
immintrin.h
+ + + Floating Point + Integer + AVX512F + Convert + + + + + Convert packed single-precision (32-bit) floating-point elements in "a" to packed 32-bit integers, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + [round_note] + +FOR j := 0 to 15 + i := 32*j + IF k[j] + dst[i+31:i] := Convert_FP32_To_Int32(a[i+31:i]) + ELSE + dst[i+31:i] := 0 + FI +ENDFOR +dst[MAX:512] := 0 + + +
immintrin.h
+ + + Floating Point + Integer + AVX512F + Convert + + + + Convert packed single-precision (32-bit) floating-point elements in "a" to packed 32-bit integers, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + +FOR j := 0 to 15 + i := 32*j + IF k[j] + dst[i+31:i] := Convert_FP32_To_Int32(a[i+31:i]) + ELSE + dst[i+31:i] := 0 + FI +ENDFOR +dst[MAX:512] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512F + Convert + + + + Convert packed single-precision (32-bit) floating-point elements in "a" to packed double-precision (64-bit) floating-point elements, and store the results in "dst". [sae_note] + +FOR j := 0 to 7 + i := 64*j + k := 32*j + dst[i+63:i] := Convert_FP32_To_FP64(a[k+31:k]) +ENDFOR +dst[MAX:512] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512F + Convert + + + Convert packed single-precision (32-bit) floating-point elements in "a" to packed double-precision (64-bit) floating-point elements, and store the results in "dst". + +FOR j := 0 to 7 + i := 64*j + k := 32*j + dst[i+63:i] := Convert_FP32_To_FP64(a[k+31:k]) +ENDFOR +dst[MAX:512] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512F + Convert + + + + + + Convert packed single-precision (32-bit) floating-point elements in "a" to packed double-precision (64-bit) floating-point elements, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). [sae_note] + +FOR j := 0 to 7 + i := 64*j + l := 32*j + IF k[j] + dst[i+63:i] := Convert_FP32_To_FP64(a[l+31:l]) + ELSE + dst[i+63:i] := src[i+63:i] + FI +ENDFOR +dst[MAX:512] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512F + Convert + + + + + Convert packed single-precision (32-bit) floating-point elements in "a" to packed double-precision (64-bit) floating-point elements, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +FOR j := 0 to 7 + i := 64*j + l := 32*j + IF k[j] + dst[i+63:i] := Convert_FP32_To_FP64(a[l+31:l]) + ELSE + dst[i+63:i] := src[i+63:i] + FI +ENDFOR +dst[MAX:512] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512F + Convert + + + + + Convert packed single-precision (32-bit) floating-point elements in "a" to packed double-precision (64-bit) floating-point elements, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). [sae_note] + +FOR j := 0 to 7 + i := 64*j + l := 32*j + IF k[j] + dst[i+63:i] := Convert_FP32_To_FP64(a[l+31:l]) + ELSE + dst[i+63:i] := 0 + FI +ENDFOR +dst[MAX:512] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512F + Convert + + + + Convert packed single-precision (32-bit) floating-point elements in "a" to packed double-precision (64-bit) floating-point elements, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + +FOR j := 0 to 7 + i := 64*j + l := 32*j + IF k[j] + dst[i+63:i] := Convert_FP32_To_FP64(a[l+31:l]) + ELSE + dst[i+63:i] := 0 + FI +ENDFOR +dst[MAX:512] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512F + Convert + + + + Convert packed single-precision (32-bit) floating-point elements in "a" to packed half-precision (16-bit) floating-point elements, and store the results in "dst". [sae_note] + +FOR j := 0 to 15 + i := 16*j + l := 32*j + dst[i+15:i] := Convert_FP32_To_FP16(a[l+31:l]) +ENDFOR +dst[MAX:256] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512F + Convert + + + + Convert packed single-precision (32-bit) floating-point elements in "a" to packed half-precision (16-bit) floating-point elements, and store the results in "dst". [sae_note] + +FOR j := 0 to 15 + i := 16*j + l := 32*j + dst[i+15:i] := Convert_FP32_To_FP16(a[l+31:l]) +ENDFOR +dst[MAX:256] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512F + Convert + + + + + + Convert packed single-precision (32-bit) floating-point elements in "a" to packed half-precision (16-bit) floating-point elements, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). [sae_note] + +FOR j := 0 to 15 + i := 16*j + l := 32*j + IF k[j] + dst[i+15:i] := Convert_FP32_To_FP16(a[l+31:l]) + ELSE + dst[i+15:i] := src[i+15:i] + FI +ENDFOR +dst[MAX:256] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512F + Convert + + + + + + Convert packed single-precision (32-bit) floating-point elements in "a" to packed half-precision (16-bit) floating-point elements, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). [sae_note] + +FOR j := 0 to 15 + i := 16*j + l := 32*j + IF k[j] + dst[i+15:i] := Convert_FP32_To_FP16(a[l+31:l]) + ELSE + dst[i+15:i] := src[i+15:i] + FI +ENDFOR +dst[MAX:256] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512F + Convert + + + + + Convert packed single-precision (32-bit) floating-point elements in "a" to packed half-precision (16-bit) floating-point elements, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). [sae_note] + +FOR j := 0 to 15 + i := 16*j + l := 32*j + IF k[j] + dst[i+15:i] := Convert_FP32_To_FP16(a[l+31:l]) + ELSE + dst[i+15:i] := 0 + FI +ENDFOR +dst[MAX:256] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512F + Convert + + + + + Convert packed single-precision (32-bit) floating-point elements in "a" to packed half-precision (16-bit) floating-point elements, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). [sae_note] + +FOR j := 0 to 15 + i := 16*j + l := 32*j + IF k[j] + dst[i+15:i] := Convert_FP32_To_FP16(a[l+31:l]) + ELSE + dst[i+15:i] := 0 + FI +ENDFOR +dst[MAX:256] := 0 + + +
immintrin.h
+ + + Floating Point + Integer + AVX512F + Convert + + + + Convert packed single-precision (32-bit) floating-point elements in "a" to packed unsigned 32-bit integers, and store the results in "dst". + [round_note] + +FOR j := 0 to 15 + i := 32*j + dst[i+31:i] := Convert_FP32_To_UInt32(a[i+31:i]) +ENDFOR +dst[MAX:512] := 0 + + +
immintrin.h
+ + + Floating Point + Integer + AVX512F + Convert + + + Convert packed single-precision (32-bit) floating-point elements in "a" to packed unsigned 32-bit integers, and store the results in "dst". + +FOR j := 0 to 15 + i := 32*j + dst[i+31:i] := Convert_FP32_To_UInt32(a[i+31:i]) +ENDFOR +dst[MAX:512] := 0 + + +
immintrin.h
+ + + Floating Point + Integer + AVX512F + Convert + + + + + + Convert packed single-precision (32-bit) floating-point elements in "a" to packed unsigned 32-bit integers, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + [round_note] + +FOR j := 0 to 15 + i := j*32 + IF k[j] + dst[i+31:i] := Convert_FP32_To_UInt32(a[i+31:i]) + ELSE + dst[i+31:i] := src[i+31:i] + FI +ENDFOR +dst[MAX:512] := 0 + + +
immintrin.h
+ + + Floating Point + Integer + AVX512F + Convert + + + + + Convert packed single-precision (32-bit) floating-point elements in "a" to packed unsigned 32-bit integers, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +FOR j := 0 to 15 + i := 32*j + IF k[j] + dst[i+31:i] := Convert_FP32_To_UInt32(a[i+31:i]) + ELSE + dst[i+31:i] := src[i+31:i] + FI +ENDFOR dst[MAX:512] := 0 +
immintrin.h
 - + + Floating Point Integer - KNCNI - Swizzle - - - - - Performs a swizzle transformation of each of the four groups of packed 4x32-bit integer elements in "v" using swizzle parameter "s", storing the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + AVX512F + Convert + + + + + Convert packed single-precision (32-bit) floating-point elements in "a" to packed unsigned 32-bit integers, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + [round_note] -CASE s OF -_MM_SWIZ_REG_NONE: - dst[511:0] := v[511:0] -_MM_SWIZ_REG_DCBA: - dst[511:0] := v[511:0] -_MM_SWIZ_REG_CDAB: - FOR j := 0 to 7 - i := j*64 - IF k[j*2] - dst[i+31:i] := v[i+63:i+32] - ELSE - dst[i+31:i] := src[i+31:i] - FI - IF k[j*2+1] - dst[i+63:i+32] := v[i+31:i] - ELSE - dst[i+63:i+32] := src[i+63:i+32] - FI - ENDFOR -_MM_SWIZ_REG_BADC: - FOR j := 0 to 3 - i := j*128 - IF k[j*4] - dst[i+31:i] := v[i+95:i+64] - ELSE - dst[i+31:i] := src[i+31:i] - FI - IF k[j*4+1] - dst[i+63:i+32] := v[i+127:i+96] - ELSE - dst[i+63:i+32] := src[i+63:i+32] - FI - IF k[j*4+2] - dst[i+95:i+64] := v[i+31:i] - ELSE - dst[i+95:i+64] := src[i+95:i+64] - FI - IF k[j*4+3] - dst[i+127:i+96] := v[i+63:i+32] - ELSE - dst[i+127:i+96] := src[i+127:i+96] - FI - ENDFOR -_MM_SWIZ_REG_AAAA: - FOR j := 0 to 3 - i := j*128 - IF k[j*4] - dst[i+31:i] := v[i+31:i] - ELSE - dst[i+31:i] := src[i+31:i] - FI - IF k[j*4+1] - dst[i+63:i+32] := v[i+31:i] - ELSE - dst[i+63:i+32] := src[i+63:i+32] - FI - IF k[j*4+2] - dst[i+95:i+64] := v[i+31:i] - ELSE - dst[i+95:i+64] := src[i+95:i+64] - FI - IF k[j*4+3] - dst[i+127:i+96] := v[i+31:i] - ELSE - dst[i+127:i+96] := src[i+127:i+96] - FI - ENDFOR -_MM_SWIZ_REG_BBBB: - FOR j := 0 to 3 - i := j*128 - IF k[j*4] - dst[i+31:i] := v[i+63:i+32] - ELSE - dst[i+31:i] := src[i+31:i] - FI - IF k[j*4+1] - dst[i+63:i+32] := v[i+63:i+32] - ELSE - dst[i+63:i+32] := src[i+63:i+32] - FI - IF k[j*4+2] - dst[i+95:i+64] := v[i+63:i+32] - ELSE - dst[i+95:i+64] := src[i+95:i+64] - FI - IF k[j*4+3] - dst[i+127:i+96] := v[i+63:i+32] - ELSE - dst[i+127:i+96] := src[i+127:i+96] - FI - ENDFOR -_MM_SWIZ_REG_CCCC: - FOR j := 0 to 3 - i := j*128 - IF k[j*4] - dst[i+31:i] := v[i+95:i+64] - ELSE - dst[i+31:i] := src[i+31:i] - FI - IF k[j*4+1] - dst[i+63:i+32] := v[i+95:i+64] - ELSE - dst[i+63:i+32] := src[i+63:i+32] - FI - IF k[j*4+2] - dst[i+95:i+64] := v[i+95:i+64] - ELSE - dst[i+95:i+64] := src[i+95:i+64] - FI - IF k[j*4+3] - dst[i+127:i+96] := v[i+95:i+64] - ELSE - dst[i+127:i+96] := src[i+127:i+96] - FI - ENDFOR -_MM_SWIZ_REG_DDDD: - FOR j := 0 to 3 - i := j*128 - IF k[j*4] - dst[i+31:i] := v[i+127:i+96] - ELSE - dst[i+31:i] := src[i+31:i] - FI - IF k[j*4+1] - dst[i+63:i+32] := v[i+127:i+96] - ELSE - dst[i+63:i+32] := src[i+63:i+32] - FI - IF k[j*4+2] - dst[i+95:i+64] := v[i+127:i+96] - ELSE - dst[i+95:i+64] := src[i+95:i+64] - FI - IF k[j*4+3] - dst[i+127:i+96] := v[i+127:i+96] - ELSE - dst[i+127:i+96] := src[i+127:i+96] - FI - ENDFOR -_MM_SWIZ_REG_DACB: - FOR j := 0 to 3 - i := j*128 - IF k[j*4] - dst[i+31:i] := v[i+63:i+32] - ELSE - dst[i+31:i] := src[i+31:i] - FI - IF k[j*4+1] - dst[i+63:i+32] := v[i+95:i+64] - ELSE - dst[i+63:i+32] := src[i+63:i+32] - FI - IF k[j*4+2] - dst[i+95:i+64] := v[i+31:i] - ELSE - dst[i+95:i+64] := src[i+95:i+64] - FI - IF k[j*4+3] - dst[i+127:i+96] := v[i+127:i+96] - ELSE - dst[i+127:i+96] := src[i+127:i+96] - FI - ENDFOR -ESAC +FOR j := 0 to 15 + i := 32*j + IF k[j] + dst[i+31:i] := Convert_FP32_To_UInt32(a[i+31:i]) + ELSE + dst[i+31:i] := 0 + FI +ENDFOR +dst[MAX:512] := 0 + + +
immintrin.h
+ + + Floating Point + Integer + AVX512F + Convert + + + + Convert packed single-precision (32-bit) floating-point elements in "a" to packed unsigned 32-bit integers, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + +FOR j := 0 to 15 + i := 32*j + IF k[j] + dst[i+31:i] := Convert_FP32_To_UInt32(a[i+31:i]) + ELSE + dst[i+31:i] := 0 + FI +ENDFOR dst[MAX:512] := 0 + +
immintrin.h
+ + + Floating Point + AVX512F + Convert + + + + Convert the lower double-precision (64-bit) floating-point element in "a" to a 32-bit integer, and store the result in "dst". + [round_note] + +dst[31:0] := Convert_FP64_To_Int32(a[63:0]) + + +
immintrin.h
+ + + Floating Point + AVX512F + Convert + + + + Convert the lower double-precision (64-bit) floating-point element in "a" to a 64-bit integer, and store the result in "dst". + [round_note] + +dst[63:0] := Convert_FP64_To_Int64(a[63:0]) + + +
immintrin.h
+ + + Floating Point + Integer + AVX512F + Convert + + + + Convert the lower double-precision (64-bit) floating-point element in "a" to a 32-bit integer, and store the result in "dst". + [round_note] + +dst[31:0] := Convert_FP64_To_Int32(a[63:0]) + + +
immintrin.h
+ + + Floating Point + Integer + AVX512F + Convert + + + + Convert the lower double-precision (64-bit) floating-point element in "a" to a 64-bit integer, and store the result in "dst". + [round_note] + +dst[63:0] := Convert_FP64_To_Int64(a[63:0]) + + +
immintrin.h
+ + + Floating Point + AVX512F + Convert + + + Convert the lower double-precision (64-bit) floating-point element in "a" to a 32-bit integer, and store the result in "dst". + +dst[31:0] := Convert_FP64_To_Int32(a[63:0]) + + +
immintrin.h
+ + + Floating Point + AVX512F + Convert + + + Convert the lower double-precision (64-bit) floating-point element in "a" to a 64-bit integer, and store the result in "dst". + +dst[63:0] := Convert_FP64_To_Int64(a[63:0]) + + +
immintrin.h
+ + + Floating Point + AVX512F + Convert + + + + + Convert the lower double-precision (64-bit) floating-point element in "b" to a single-precision (32-bit) floating-point element, store the result in the lower element of "dst", and copy the upper 3 packed elements from "a" to the upper elements of "dst". + [round_note] + +dst[31:0] := Convert_FP64_To_FP32(b[63:0]) +dst[127:32] := a[127:32] +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512F + Convert + + + + + + + Convert the lower double-precision (64-bit) floating-point element in "b" to a single-precision (32-bit) floating-point element, store the result in the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". + [round_note] + +IF k[0] + dst[31:0] := Convert_FP64_To_FP32(b[63:0]) +ELSE + dst[31:0] := src[31:0] +FI +dst[127:32] := a[127:32] +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512F + Convert + + + + + + Convert the lower double-precision (64-bit) floating-point element in "b" to a single-precision (32-bit) floating-point element, store the result in the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". + +IF k[0] + dst[31:0] := Convert_FP64_To_FP32(b[63:0]) +ELSE + dst[31:0] := src[31:0] +FI +dst[127:32] := a[127:32] +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512F + Convert + + + + + + Convert the lower double-precision (64-bit) floating-point element in "b" to a single-precision (32-bit) floating-point element, store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". + [round_note] + +IF k[0] + dst[31:0] := Convert_FP64_To_FP32(b[63:0]) +ELSE + dst[31:0] := 0 +FI +dst[127:32] := a[127:32] +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512F + Convert + + + + + Convert the lower double-precision (64-bit) floating-point element in "b" to a single-precision (32-bit) floating-point element, store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". + +IF k[0] + dst[31:0] := Convert_FP64_To_FP32(b[63:0]) +ELSE + dst[31:0] := 0 +FI +dst[127:32] := a[127:32] +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Floating Point + Integer + AVX512F + Convert + + + + Convert the lower double-precision (64-bit) floating-point element in "a" to an unsigned 32-bit integer, and store the result in "dst". + [round_note] + +dst[31:0] := Convert_FP64_To_UInt32(a[63:0]) + + +
immintrin.h
+ + + Floating Point + Integer + AVX512F + Convert + + + + Convert the lower double-precision (64-bit) floating-point element in "a" to an unsigned 64-bit integer, and store the result in "dst". + [round_note] + +dst[63:0] := Convert_FP64_To_UInt64(a[63:0]) + + +
immintrin.h
+ + + Floating Point + Integer + AVX512F + Convert + + + Convert the lower double-precision (64-bit) floating-point element in "a" to an unsigned 32-bit integer, and store the result in "dst". + +dst[31:0] := Convert_FP64_To_UInt32(a[63:0]) + + +
immintrin.h
+ + + Floating Point + Integer + AVX512F + Convert + + + Convert the lower double-precision (64-bit) floating-point element in "a" to an unsigned 64-bit integer, and store the result in "dst". + +dst[63:0] := Convert_FP64_To_UInt64(a[63:0]) + + +
immintrin.h
+ + + Floating Point + AVX512F + Convert + + + + + Convert the signed 64-bit integer "b" to a double-precision (64-bit) floating-point element, store the result in the lower element of "dst", and copy the upper element from "a" to the upper element of "dst". + [round_note] + +dst[63:0] := Convert_Int64_To_FP64(b[63:0]) +dst[127:64] := a[127:64] +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Floating Point + Integer + AVX512F + Convert + + + + + Convert the signed 64-bit integer "b" to a double-precision (64-bit) floating-point element, store the result in the lower element of "dst", and copy the upper element from "a" to the upper element of "dst". + [round_note] + +dst[63:0] := Convert_Int64_To_FP64(b[63:0]) +dst[127:64] := a[127:64] +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512F + Convert + + + + Convert the signed 32-bit integer "b" to a double-precision (64-bit) floating-point element, store the result in the lower element of "dst", and copy the upper element from "a" to the upper element of "dst". + +dst[63:0] := Convert_Int32_To_FP64(b[31:0]) +dst[127:64] := a[127:64] +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512F + Convert + + + + Convert the signed 64-bit integer "b" to a double-precision (64-bit) floating-point element, store the result in the lower element of "dst", and copy the upper element from "a" to the upper element of "dst". + +dst[63:0] := Convert_Int64_To_FP64(b[63:0]) +dst[127:64] := a[127:64] +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512F + Convert + + + + + Convert the signed 32-bit integer "b" to a single-precision (32-bit) floating-point element, store the result in the lower element of "dst", and copy the upper 3 packed elements from "a" to the upper elements of "dst". + [round_note] + +dst[31:0] := Convert_Int32_To_FP32(b[31:0]) +dst[127:32] := a[127:32] +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512F + Convert + + + + + Convert the signed 64-bit integer "b" to a single-precision (32-bit) floating-point element, store the result in the lower element of "dst", and copy the upper 3 packed elements from "a" to the upper elements of "dst". + [round_note] + +dst[31:0] := Convert_Int64_To_FP32(b[63:0]) +dst[127:32] := a[127:32] +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512F + Convert + + + + + Convert the signed 32-bit integer "b" to a single-precision (32-bit) floating-point element, store the result in the lower element of "dst", and copy the upper 3 packed elements from "a" to the upper elements of "dst". + [round_note] + +dst[31:0] := Convert_Int32_To_FP32(b[31:0]) +dst[127:32] := a[127:32] +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Floating Point + Integer + AVX512F + Convert + + + + + Convert the signed 64-bit integer "b" to a single-precision (32-bit) floating-point element, store the result in the lower element of "dst", and copy the upper 3 packed elements from "a" to the upper elements of "dst". + [round_note] + +dst[31:0] := Convert_Int64_To_FP32(b[63:0]) +dst[127:32] := a[127:32] +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512F + Convert + + + + Convert the signed 32-bit integer "b" to a single-precision (32-bit) floating-point element, store the result in the lower element of "dst", and copy the upper 3 packed elements from "a" to the upper elements of "dst". + +dst[31:0] := Convert_Int32_To_FP32(b[31:0]) +dst[127:32] := a[127:32] +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512F + Convert + + + + Convert the signed 64-bit integer "b" to a single-precision (32-bit) floating-point element, store the result in the lower element of "dst", and copy the upper 3 packed elements from "a" to the upper elements of "dst". + +dst[31:0] := Convert_Int64_To_FP32(b[63:0]) +dst[127:32] := a[127:32] +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512F + Convert + + + + + Convert the lower single-precision (32-bit) floating-point element in "b" to a double-precision (64-bit) floating-point element, store the result in the lower element of "dst", and copy the upper element from "a" to the upper element of "dst". + [sae_note] + +dst[63:0] := Convert_FP32_To_FP64(b[31:0]) +dst[127:64] := a[127:64] +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512F + Convert + + + + + + + Convert the lower single-precision (32-bit) floating-point element in "b" to a double-precision (64-bit) floating-point element, store the result in the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". + [sae_note] + +IF k[0] + dst[63:0] := Convert_FP32_To_FP64(b[31:0]) +ELSE + dst[63:0] := src[63:0] +FI +dst[127:64] := a[127:64] +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512F + Convert + + + + + + Convert the lower single-precision (32-bit) floating-point element in "b" to a double-precision (64-bit) floating-point element, store the result in the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". + +IF k[0] + dst[63:0] := Convert_FP32_To_FP64(b[31:0]) +ELSE + dst[63:0] := src[63:0] +FI +dst[127:64] := a[127:64] +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512F + Convert + + + + + + Convert the lower single-precision (32-bit) floating-point element in "b" to a double-precision (64-bit) floating-point element, store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". + [sae_note] + +IF k[0] + dst[63:0] := Convert_FP32_To_FP64(b[31:0]) +ELSE + dst[63:0] := 0 +FI +dst[127:64] := a[127:64] +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512F + Convert + + + + + Convert the lower single-precision (32-bit) floating-point element in "b" to a double-precision (64-bit) floating-point element, store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". + +IF k[0] + dst[63:0] := Convert_FP32_To_FP64(b[31:0]) +ELSE + dst[63:0] := 0 +FI +dst[127:64] := a[127:64] +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512F + Convert + + + + Convert the lower single-precision (32-bit) floating-point element in "a" to a 32-bit integer, and store the result in "dst". + [round_note] + +dst[31:0] := Convert_FP32_To_Int32(a[31:0]) + + +
immintrin.h
+ + + Floating Point + AVX512F + Convert + + + + Convert the lower single-precision (32-bit) floating-point element in "a" to a 64-bit integer, and store the result in "dst". + [round_note] + +dst[63:0] := Convert_FP32_To_Int64(a[31:0]) + +
immintrin.h
 - + + Floating Point Integer - KNCNI - Swizzle - - - - - Performs a swizzle transformation of each of the four groups of packed 4x64-bit integer elements in "v" using swizzle parameter "s", storing the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + AVX512F + Convert + + + + Convert the lower single-precision (32-bit) floating-point element in "a" to a 32-bit integer, and store the result in "dst". + [round_note] -CASE s OF -_MM_SWIZ_REG_NONE: - dst[511:0] := v[511:0] -_MM_SWIZ_REG_DCBA: - dst[511:0] := v[511:0] -_MM_SWIZ_REG_CDAB: - FOR j := 0 to 3 - i := j*64 - IF k[j*2] - dst[i+63:i] := v[i+127:i+64] - ELSE - dst[i+63:i] := src[i+63:i] - FI - IF k[j*2+1] - dst[i+127:i+64] := v[i+63:i] - ELSE - dst[i+127:i+64] := src[i+127:i+64] - FI - ENDFOR -_MM_SWIZ_REG_BADC: - FOR j := 0 to 1 - i := j*256 - IF k[j*4] - dst[i+63:i] := v[i+191:i+128] - ELSE - dst[i+63:i] := src[i+63:i] - FI - IF k[j*4+1] - dst[i+127:i+64] := v[i+255:i+192] - ELSE - dst[i+127:i+64] := src[i+127:i+64] - FI - IF k[j*4+2] - dst[i+191:i+128] := v[i+63:i] - ELSE - dst[i+191:i+128] := src[i+191:i+128] - FI - IF k[j*4+3] - dst[i+255:i+192] := v[i+127:i+64] - ELSE - dst[i+255:i+192] := src[i+255:i+192] - FI - ENDFOR -_MM_SWIZ_REG_AAAA: - FOR j := 0 to 1 - i := j*256 - IF k[j*4] - dst[i+63:i] := v[i+63:i] - ELSE - dst[i+63:i] := src[i+63:i] - FI - IF k[j*4+1] - dst[i+127:i+64] := v[i+63:i] - ELSE - dst[i+127:i+64] := src[i+127:i+64] - FI - IF k[j*4+2] - dst[i+191:i+128] := v[i+63:i] - ELSE - dst[i+191:i+128] := src[i+191:i+128] - FI - IF k[j*4+3] - dst[i+255:i+192] := v[i+63:i] - ELSE - dst[i+255:i+192] := src[i+255:i+192] - FI - ENDFOR -_MM_SWIZ_REG_BBBB: - FOR j := 0 to 1 - i := j*256 - IF k[j*4] - dst[i+63:i] := v[i+127:i+63] - ELSE - dst[i+63:i] := src[i+63:i] - FI - IF k[j*4+1] - dst[i+127:i+64] := v[i+127:i+63] - ELSE - dst[i+127:i+64] := src[i+127:i+64] - FI - IF k[j*4+2] - dst[i+191:i+128] := v[i+127:i+63] - ELSE - dst[i+191:i+128] := src[i+191:i+128] - FI - IF k[j*4+3] - dst[i+255:i+192] := v[i+127:i+63] - ELSE - dst[i+255:i+192] := src[i+255:i+192] - FI - ENDFOR -_MM_SWIZ_REG_CCCC: - FOR j := 0 to 1 - i := j*256 - IF k[j*4] - dst[i+63:i] := v[i+191:i+128] - ELSE - dst[i+63:i] := src[i+63:i] - FI - IF k[j*4+1] - dst[i+127:i+64] := v[i+191:i+128] - ELSE - dst[i+127:i+64] := src[i+127:i+64] - FI - IF k[j*4+2] - dst[i+191:i+128] := v[i+191:i+128] - ELSE - dst[i+191:i+128] := src[i+191:i+128] - FI - IF k[j*4+3] - dst[i+255:i+192] := v[i+191:i+128] - ELSE - dst[i+255:i+192] := src[i+255:i+192] - FI - ENDFOR -_MM_SWIZ_REG_DDDD: - FOR j := 0 to 1 - i := j*256 - IF k[j*4] - dst[i+63:i] := v[i+255:i+192] - ELSE - dst[i+63:i] := src[i+63:i] - FI - IF k[j*4+1] - dst[i+127:i+64] := v[i+255:i+192] - ELSE - dst[i+127:i+64] := src[i+127:i+64] - FI - IF k[j*4+2] - dst[i+191:i+128] := v[i+255:i+192] - ELSE - dst[i+191:i+128] := src[i+191:i+128] - FI - IF k[j*4+3] - dst[i+255:i+192] := v[i+255:i+192] - ELSE - dst[i+255:i+192] := src[i+255:i+192] - FI - ENDFOR -_MM_SWIZ_REG_DACB: - FOR j := 0 to 1 - i := j*256 - IF k[j*4] - dst[i+63:i] := v[i+127:i+64] - ELSE - dst[i+63:i] := src[i+63:i] - FI - IF k[j*4+1] - dst[i+127:i+64] := v[i+191:i+128] - ELSE - dst[i+127:i+64] := src[i+127:i+64] - FI - IF k[j*4+2] - dst[i+191:i+128] := v[i+63:i] - ELSE - dst[i+191:i+128] := src[i+191:i+128] - FI - IF k[j*4+3] - dst[i+255:i+192] := v[i+255:i+192] - ELSE - dst[i+255:i+192] := src[i+255:i+192] - FI - ENDFOR -ESAC -dst[MAX:512] := 0 +dst[31:0] := Convert_FP32_To_Int32(a[31:0]) +
immintrin.h
 - + Floating Point - KNCNI - Store - - - - - Downconverts packed single-precision (32-bit) floating-point elements stored in "v" to a smaller type depending on "conv" and stores them in memory location "mt". "hint" indicates to the processor whether the data is non-temporal. + Integer + AVX512F + Convert + + + + Convert the lower single-precision (32-bit) floating-point element in "a" to a 64-bit integer, and store the result in "dst". + [round_note] -addr := MEM[mt] -FOR j := 0 to 15 - i := j*32 - CASE conv OF - _MM_DOWNCONV_PS_NONE: - addr[i+31:i] := v[i+31:i] - _MM_DOWNCONV_PS_FLOAT16: - n := j*16 - addr[n+15:n] := Float32ToFloat16(v[i+31:i]) - _MM_DOWNCONV_PS_UINT8: - n := j*8 - addr[n+7:n] := Float32ToUInt8(v[i+31:i]) - _MM_DOWNCONV_PS_SINT8: - n := j*8 - addr[n+7:n] := Float32ToSInt8(v[i+31:i]) - _MM_DOWNCONV_PS_UINT16: - n := j*16 - addr[n+15:n] := Float32ToUInt16(v[i+31:i]) - _MM_DOWNCONV_PS_SINT16: - n := j*16 - addr[n+15:n] := Float32ToSInt16(v[i+31:i]) - ESAC -ENDFOR +dst[63:0] := Convert_FP32_To_Int64(a[31:0]) + + +
immintrin.h
+ + + Floating Point + AVX512F + Convert + + + Convert the lower single-precision (32-bit) floating-point element in "a" to a 32-bit integer, and store the result in "dst". + +dst[31:0] := Convert_FP32_To_Int32(a[31:0]) + + +
immintrin.h
+ + + Floating Point + AVX512F + Convert + + + Convert the lower single-precision (32-bit) floating-point element in "a" to a 64-bit integer, and store the result in "dst". + +dst[63:0] := Convert_FP32_To_Int64(a[31:0]) - +
immintrin.h
 - + + Floating Point Integer - KNCNI - Store - - - - - Downconverts packed 32-bit integer elements stored in "v" to a smaller type depending on "conv" and stores them in memory location "mt". "hint" indicates to the processor whether the data is non-temporal. + AVX512F + Convert + + + + Convert the lower single-precision (32-bit) floating-point element in "a" to an unsigned 32-bit integer, and store the result in "dst". + [round_note] -addr := MEM[mt] -FOR j := 0 to 15 - i := j*32 - CASE conv OF - _MM_DOWNCONV_EPI32_NONE: - addr[i+31:i] := v[i+31:i] - _MM_DOWNCONV_EPI32_UINT8: - n := j*8 - addr[n+7:n] := Int32ToUInt8(v[i+31:i]) - _MM_DOWNCONV_EPI32_SINT8: - n := j*8 - addr[n+7:n] := Int32ToSInt8(v[i+31:i]) - _MM_DOWNCONV_EPI32_UINT16: - n := j*16 - addr[n+15:n] := Int32ToUInt16(v[i+31:i]) - _MM_DOWNCONV_EPI32_SINT16: - n := j*16 - addr[n+15:n] := Int32ToSInt16(v[i+31:i]) - ESAC -ENDFOR +dst[31:0] := Convert_FP32_To_UInt32(a[31:0]) - +
immintrin.h
 - + Floating Point - KNCNI - Store - - - - - Downconverts packed double-precision (64-bit) floating-point elements stored in "v" to a smaller type depending on "conv" and stores them in memory location "mt". "hint" indicates to the processor whether the data is non-temporal. + Integer + AVX512F + Convert + + + + Convert the lower single-precision (32-bit) floating-point element in "a" to an unsigned 64-bit integer, and store the result in "dst". + [round_note] + +dst[63:0] := Convert_FP32_To_UInt64(a[31:0]) + + +
immintrin.h
+ + + Floating Point + Integer + AVX512F + Convert + + + Convert the lower single-precision (32-bit) floating-point element in "a" to an unsigned 32-bit integer, and store the result in "dst". + +dst[31:0] := Convert_FP32_To_UInt32(a[31:0]) + + +
immintrin.h
+ + + Floating Point + Integer + AVX512F + Convert + + + Convert the lower single-precision (32-bit) floating-point element in "a" to an unsigned 64-bit integer, and store the result in "dst". + +dst[63:0] := Convert_FP32_To_UInt64(a[31:0]) + + +
immintrin.h
+ + + Floating Point + Integer + AVX512F + Convert + + + + Convert packed double-precision (64-bit) floating-point elements in "a" to packed 32-bit integers with truncation, and store the results in "dst". [sae_note] -addr := MEM[mt] FOR j := 0 to 7 - i := j*64 - CASE conv OF - _MM_DOWNCONV_PS_NONE: - addr[i+63:i] := v[i+63:i] - ESAC + i := 32*j + k := 64*j + dst[i+31:i] := Convert_FP64_To_Int32_Truncate(a[k+63:k]) ENDFOR +dst[MAX:256] := 0 - +
immintrin.h
 - + + Floating Point Integer - KNCNI - Store - - - - - Downconverts packed 64-bit integer elements stored in "v" to a smaller type depending on "conv" and stores them in memory location "mt". "hint" indicates to the processor whether the data is non-temporal. + AVX512F + Convert + + + Convert packed double-precision (64-bit) floating-point elements in "a" to packed 32-bit integers with truncation, and store the results in "dst". -addr := MEM[mt] FOR j := 0 to 7 - i := j*64 - CASE conv OF - _MM_DOWNCONV_EPI64_NONE: addr[i+63:i] := v[i+63:i] - ESAC + i := 32*j + k := 64*j + dst[i+31:i] := Convert_FP64_To_Int32_Truncate(a[k+63:k]) ENDFOR +dst[MAX:256] := 0 - +
immintrin.h
 - + Floating Point - KNCNI - Store - - - - - - Downconverts packed single-precision (32-bit) floating-point elements stored in "v" to a smaller type depending on "conv" and stores them in memory location "mt" using writemask "k" (elements are not written to memory when the corresponding mask bit is not set). "hint" indicates to the processor whether the data is non-temporal. + Integer + AVX512F + Convert + + + + + + Convert packed double-precision (64-bit) floating-point elements in "a" to packed 32-bit integers with truncation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). [sae_note] -FOR j := 0 to 15 - i := j*32 +FOR j := 0 to 7 + i := 32*j + l := 64*j IF k[j] - CASE conv OF - _MM_DOWNCONV_PS_NONE: - mt[i+31:i] := v[i+31:i] - _MM_DOWNCONV_PS_FLOAT16: - n := j*16 - mt[n+15:n] := Float32ToFloat16(v[i+31:i]) - _MM_DOWNCONV_PS_UINT8: - n := j*8 - mt[n+7:n] := Float32ToUInt8(v[i+31:i]) - _MM_DOWNCONV_PS_SINT8: - n := j*8 - mt[n+7:n] := Float32ToSInt8(v[i+31:i]) - _MM_DOWNCONV_PS_UINT16: - n := j*16 - mt[n+15:n] := Float32ToUInt16(v[i+31:i]) - _MM_DOWNCONV_PS_SINT16: - n := j*16 - mt[n+15:n] := Float32ToSInt16(v[i+31:i]) - ESAC - FI + dst[i+31:i] := Convert_FP64_To_Int32_Truncate(a[l+63:l]) + ELSE + dst[i+31:i] := src[i+31:i] + FI ENDFOR +dst[MAX:256] := 0 - +
immintrin.h
 - + Floating Point - KNCNI - Store - - - - - - Downconverts packed double-precision (64-bit) floating-point elements stored in "v" to a smaller type depending on "conv" and stores them in memory location "mt" (elements in "mt" are unaltered when the corresponding mask bit is not set). "hint" indicates to the processor whether the data is non-temporal. + Integer + AVX512F + Convert + + + + + Convert packed double-precision (64-bit) floating-point elements in "a" to packed 32-bit integers with truncation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -addr := MEM[mt] FOR j := 0 to 7 - i := j*64 - CASE conv OF - _MM_DOWNCONV_PD_NONE: - IF k[j] - mt[i+63:i] := v[i+63:i] - FI - ESAC + i := 32*j + l := 64*j + IF k[j] + dst[i+31:i] := Convert_FP64_To_Int32_Truncate(a[l+63:l]) + ELSE + dst[i+31:i] := src[i+31:i] + FI ENDFOR +dst[MAX:256] := 0 - +
immintrin.h
 - + + Floating Point Integer - KNCNI - Store - - - - - - Downconverts packed 32-bit integer elements stored in "v" to a smaller type depending on "conv" and stores them in memory location "mt" (elements in "mt" are unaltered when the corresponding mask bit is not set). "hint" indicates to the processor whether the data is non-temporal. + AVX512F + Convert + + + + + Convert packed double-precision (64-bit) floating-point elements in "a" to packed 32-bit integers with truncation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). [sae_note] -addr := MEM[mt] -FOR j := 0 to 15 - i := j*32 +FOR j := 0 to 7 + i := 32*j + l := 64*j IF k[j] - CASE conv OF - _MM_DOWNCONV_EPI32_NONE: - addr[i+31:i] := v[i+31:i] - _MM_DOWNCONV_EPI32_UINT8: - n := j*8 - addr[n+7:n] := Int32ToUInt8(v[i+31:i]) - _MM_DOWNCONV_EPI32_SINT8: - n := j*8 - addr[n+7:n] := Int32ToSInt8(v[i+31:i]) - _MM_DOWNCONV_EPI32_UINT16: - n := j*16 - addr[n+15:n] := Int32ToUInt16(v[i+31:i]) - _MM_DOWNCONV_EPI32_SINT16: - n := j*16 - addr[n+15:n] := Int32ToSInt16(v[i+31:i]) - ESAC + dst[i+31:i] := Convert_FP64_To_Int32_Truncate(a[l+63:l]) + ELSE + dst[i+31:i] := 0 FI ENDFOR +dst[MAX:256] := 0 - +
immintrin.h
 - + + Floating Point Integer - KNCNI - Store - - - - - - Downconverts packed 64-bit integer elements stored in "v" to a smaller type depending on "conv" and stores them in memory location "mt" (elements in "mt" are unaltered when the corresponding mask bit is not set). "hint" indicates to the processor whether the data is non-temporal. + AVX512F + Convert + + + + Convert packed double-precision (64-bit) floating-point elements in "a" to packed 32-bit integers with truncation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -addr := MEM[mt] FOR j := 0 to 7 - i := j*64 + i := 32*j + l := 64*j IF k[j] - CASE conv OF - _MM_DOWNCONV_EPI64_NONE: addr[i+63:i] := v[i+63:i] - ESAC + dst[i+31:i] := Convert_FP64_To_Int32_Truncate(a[l+63:l]) + ELSE + dst[i+31:i] := 0 FI ENDFOR +dst[MAX:256] := 0 - +
immintrin.h
 - + Floating Point - KNCNI - Store - - - Stores packed single-precision (32-bit) floating-point elements from "v" to memory address "mt" with a no-read hint to the processor. + Integer + AVX512F + Convert + + + + Convert packed double-precision (64-bit) floating-point elements in "a" to packed unsigned 32-bit integers with truncation, and store the results in "dst". [sae_note] -addr := MEM[mt] -FOR j := 0 to 15 - i := j*32 - addr[i+31:i] := v[i+31:i] +FOR j := 0 to 7 + i := 32*j + k := 64*j + dst[i+31:i] := Convert_FP64_To_UInt32_Truncate(a[k+63:k]) ENDFOR +dst[MAX:256] := 0 - +
immintrin.h
 - + Floating Point - KNCNI - Store - - - Stores packed double-precision (64-bit) floating-point elements from "v" to memory address "mt" with a no-read hint to the processor. + Integer + AVX512F + Convert + + + Convert packed double-precision (64-bit) floating-point elements in "a" to packed unsigned 32-bit integers with truncation, and store the results in "dst". + +FOR j := 0 to 7 + i := 32*j + k := 64*j + dst[i+31:i] := Convert_FP64_To_UInt32_Truncate(a[k+63:k]) +ENDFOR +dst[MAX:256] := 0 + + +
immintrin.h
+ + + Floating Point + Integer + AVX512F + Convert + + + + + + Convert packed double-precision (64-bit) floating-point elements in "a" to packed unsigned 32-bit integers with truncation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). [sae_note] + +FOR j := 0 to 7 + i := 32*j + l := 64*j + IF k[j] + dst[i+31:i] := Convert_FP64_To_UInt32_Truncate(a[l+63:l]) + ELSE + dst[i+31:i] := src[i+31:i] + FI +ENDFOR +dst[MAX:256] := 0 + + +
immintrin.h
+ + + Floating Point + Integer + AVX512F + Convert + + + + + Convert packed double-precision (64-bit) floating-point elements in "a" to packed unsigned 32-bit integers with truncation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -addr := MEM[mt] FOR j := 0 to 7 - i := j*64 - addr[i+63:i] := v[i+63:i] + i := 32*j + l := 64*j + IF k[j] + dst[i+31:i] := Convert_FP64_To_UInt32_Truncate(a[l+63:l]) + ELSE + dst[i+31:i] := src[i+31:i] + FI ENDFOR +dst[MAX:256] := 0 - +
immintrin.h
 - + Floating Point - KNCNI - Store - - - Stores packed single-precision (32-bit) floating-point elements from "v" to memory address "mt" with a no-read hint and using a weakly-ordered memory consistency model (stores performed with this function are not globally ordered, and subsequent stores from the same thread can be observed before them). + Integer + AVX512F + Convert + + + + + Convert packed double-precision (64-bit) floating-point elements in "a" to packed unsigned 32-bit integers with truncation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). [sae_note] -addr := MEM[mt] -FOR j := 0 to 15 - i := j*32 - addr[i+31:i] := v[i+31:i] +FOR j := 0 to 7 + i := 32*j + l := 64*j + IF k[j] + dst[i+31:i] := Convert_FP64_To_UInt32_Truncate(a[l+63:l]) + ELSE + dst[i+31:i] := 0 + FI ENDFOR +dst[MAX:256] := 0 - +
immintrin.h
 - + Floating Point - KNCNI - Store - - - Stores packed double-precision (64-bit) floating-point elements from "v" to memory address "mt" with a no-read hint and using a weakly-ordered memory consistency model (stores performed with this function are not globally ordered, and subsequent stores from the same thread can be observed before them). + Integer + AVX512F + Convert + + + + Convert packed double-precision (64-bit) floating-point elements in "a" to packed unsigned 32-bit integers with truncation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -addr := MEM[mt] FOR j := 0 to 7 - i := j*64 - addr[i+63:i] := v[i+63:i] + i := 32*j + l := 64*j + IF k[j] + dst[i+31:i] := Convert_FP64_To_UInt32_Truncate(a[l+63:l]) + ELSE + dst[i+31:i] := 0 + FI ENDFOR +dst[MAX:256] := 0 - +
immintrin.h
 - + + Floating Point Integer - KNCNI - Arithmetic - - - - - Performs element-by-element addition of packed 32-bit integers in "v2" and "v3" and the corresponding bit in "k2", storing the result of the addition in "dst" and the result of the carry in "k2_res". + AVX512F + Convert + + + + Convert packed single-precision (32-bit) floating-point elements in "a" to packed 32-bit integers with truncation, and store the results in "dst". [sae_note] FOR j := 0 to 15 - i := j*32 - k2_res[j] := Carry(v2[i+31:i] + v3[i+31:i] + k2[j]) - dst[i+31:i] := v2[i+31:i] + v3[i+31:i] + k2[j] + i := 32*j + dst[i+31:i] := Convert_FP32_To_Int32_Truncate(a[i+31:i]) ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + + Floating Point Integer - KNCNI - Arithmetic - - - - - - Performs element-by-element addition of packed 32-bit integers in "v2" and "v3" and the corresponding bit in "k2", storing the result of the addition in "dst" and the result of the carry in "k2_res" using writemask "k1" (elements are copied from "v2" when the corresponding mask bit is not set). + AVX512F + Convert + + + Convert packed single-precision (32-bit) floating-point elements in "a" to packed 32-bit integers with truncation, and store the results in "dst". FOR j := 0 to 15 - i := j*32 - IF k1[j] - k2_res[j] := Carry(v2[i+31:i] + v3[i+31:i] + k2[j]) - dst[i+31:i] := v2[i+31:i] + v3[i+31:i] + k2[j] - ELSE - dst[i+31:i] := v2[i+31:i] - FI + i := 32*j + dst[i+31:i] := Convert_FP32_To_Int32_Truncate(a[i+31:i]) ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - KNCNI - Arithmetic - - - Performs element-by-element addition between packed double-precision (64-bit) floating-point elements in "v2" and "v3" and negates their sum, storing the results in "dst". + Integer + AVX512F + Convert + + + + + + Convert packed single-precision (32-bit) floating-point elements in "a" to packed 32-bit integers with truncation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). [sae_note] -FOR j := 0 to 7 - i := j*64 - dst[i+63:i] := -(v2[i+63:i] + v3[i+63:i]) +FOR j := 0 to 15 + i := 32*j + IF k[j] + dst[i+31:i] := Convert_FP32_To_Int32_Truncate(a[i+31:i]) + ELSE + dst[i+31:i] := src[i+31:i] + FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - KNCNI - Arithmetic - - - - - Performs element-by-element addition between packed double-precision (64-bit) floating-point elements in "v2" and "v3" and negates their sum, storing the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Integer + AVX512F + Convert + + + + + Convert packed single-precision (32-bit) floating-point elements in "a" to packed 32-bit integers with truncation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := j*64 +FOR j := 0 to 15 + i := 32*j IF k[j] - dst[i+63:i] := -(v2[i+63:i] + v3[i+63:i]) + dst[i+31:i] := Convert_FP32_To_Int32_Truncate(a[i+31:i]) ELSE - dst[i+63:i] := src[i+63:i] + dst[i+31:i] := src[i+31:i] FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - KNCNI - Arithmetic - - - Performs element-by-element addition between packed single-precision (32-bit) floating-point elements in "v2" and "v3" and negates their sum, storing the results in "dst". + Integer + AVX512F + Convert + + + + + Convert packed single-precision (32-bit) floating-point elements in "a" to packed 32-bit integers with truncation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). [sae_note] FOR j := 0 to 15 - i := j*32 - dst[i+31:i] := -(v2[i+31:i] + v3[i+31:i]) + i := 32*j + IF k[j] + dst[i+31:i] := Convert_FP32_To_Int32_Truncate(a[i+31:i]) + ELSE + dst[i+31:i] := 0 + FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - KNCNI - Arithmetic - - - - - Performs element-by-element addition between packed single-precision (32-bit) floating-point elements in "v2" and "v3" and negates their sum, storing the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Integer + AVX512F + Convert + + + + Convert packed single-precision (32-bit) floating-point elements in "a" to packed 32-bit integers with truncation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). FOR j := 0 to 15 - i := j*32 + i := 32*j IF k[j] - dst[i+31:i] := -(v2[i+31:i] + v3[i+31:i]) + dst[i+31:i] := Convert_FP32_To_Int32_Truncate(a[i+31:i]) ELSE - dst[i+31:i] := src[i+31:i] + dst[i+31:i] := 0 FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - KNCNI - Arithmetic - - - - Performs element by element addition between packed double-precision (64-bit) floating-point elements in "v2" and "v3" and negates the sum, storing the result in "dst". - [round_note] - + Integer + AVX512F + Convert + + + + Convert packed single-precision (32-bit) floating-point elements in "a" to packed unsigned 32-bit integers with truncation, and store the results in "dst". [sae_note] -FOR j := 0 to 7 - i := j*64 - dst[i+63:i] := -(v2[i+63:i] + v3[i+63:i]) +FOR j := 0 to 15 + i := 32*j + dst[i+31:i] := Convert_FP32_To_UInt32_Truncate(a[i+31:i]) ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - KNCNI - Arithmetic - - - - - - Performs element by element addition between packed double-precision (64-bit) floating-point elements in "v2" and "v3" and negates the sum, storing the result in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - [round_note] - + Integer + AVX512F + Convert + + + Convert packed single-precision (32-bit) floating-point elements in "a" to packed unsigned 32-bit integers with truncation, and store the results in "dst". -FOR j := 0 to 7 - i := j*64 - IF k[j] - dst[i+63:i] := -(v2[i+63:i] + v3[i+63:i]) - ELSE - dst[i+63:i] := src[i+63:i] - FI +FOR j := 0 to 15 + i := 32*j + dst[i+31:i] := Convert_FP32_To_UInt32_Truncate(a[i+31:i]) ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - KNCNI - Arithmetic - - - - Performs element by element addition between packed single-precision (32-bit) floating-point elements in "v2" and "v3" and negates the sum, storing the result in "dst". - [round_note] - + Integer + AVX512F + Convert + + + + + + Convert packed single-precision (32-bit) floating-point elements in "a" to packed unsigned 32-bit integers with truncation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). [sae_note] FOR j := 0 to 15 - i := j*32 - dst[i+31:i] := -(v2[i+31:i] + v3[i+31:i]) + i := 32*j + IF k[j] + dst[i+31:i] := Convert_FP32_To_UInt32_Truncate(a[i+31:i]) + ELSE + dst[i+31:i] := src[i+31:i] + FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - KNCNI - Arithmetic - - - - - - Performs element by element addition between packed single-precision (32-bit) floating-point elements in "v2" and "v3" and negates the sum, storing the result in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - [round_note] - + Integer + AVX512F + Convert + + + + + Convert packed double-precision (32-bit) floating-point elements in "a" to packed unsigned 32-bit integers with truncation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). FOR j := 0 to 15 - i := j*32 + i := 32*j IF k[j] - dst[i+31:i] := -(v2[i+31:i] + v3[i+31:i]) + dst[i+31:i] := Convert_FP64_To_UInt32_Truncate(a[i+31:i]) ELSE dst[i+31:i] := src[i+31:i] FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - KNCNI - Arithmetic - - - Performs element-by-element subtraction of packed double-precision (64-bit) floating-point elements in "v2" from "v3" storing the results in "dst". + Integer + AVX512F + Convert + + + + + Convert packed single-precision (32-bit) floating-point elements in "a" to packed unsigned 32-bit integers with truncation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). [sae_note] -FOR j := 0 to 7 - i := j*64 - dst[i+63:i] := v3[i+63:i] - v2[i+63:i] +FOR j := 0 to 15 + i := 32*j + IF k[j] + dst[i+31:i] := Convert_FP32_To_UInt32_Truncate(a[i+31:i]) + ELSE + dst[i+31:i] := 0 + FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - KNCNI - Arithmetic - - - - - Performs element-by-element subtraction of packed double-precision (64-bit) floating-point elements in "v2" from "v3" storing the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Integer + AVX512F + Convert + + + + Convert packed double-precision (32-bit) floating-point elements in "a" to packed unsigned 32-bit integers with truncation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := j*64 +FOR j := 0 to 15 + i := 32*j IF k[j] - dst[i+63:i] := v3[i+63:i] - v2[i+63:i] + dst[i+31:i] := Convert_FP64_To_UInt32_Truncate(a[i+31:i]) ELSE - dst[i+63:i] := src[i+63:i] + dst[i+31:i] := 0 FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - KNCNI - Arithmetic - - - Performs element-by-element subtraction of packed single-precision (32-bit) floating-point elements in "v2" from "v3" storing the results in "dst". + AVX512F + Convert + + + + Convert the lower double-precision (64-bit) floating-point element in "a" to a 32-bit integer with truncation, and store the result in "dst". + [sae_note] -FOR j := 0 to 15 - i := j*32 - dst[i+31:i] := v3[i+31:i] - v2[i+31:i] -ENDFOR -dst[MAX:512] := 0 +dst[31:0] := Convert_FP64_To_Int32_Truncate(a[63:0]) - +
immintrin.h
 - + Floating Point - KNCNI - Arithmetic - - - - - Performs element-by-element subtraction of packed single-precision (32-bit) floating-point elements in "v2" from "v3" storing the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + AVX512F + Convert + + + + Convert the lower double-precision (64-bit) floating-point element in "a" to a 64-bit integer with truncation, and store the result in "dst". + [sae_note] -FOR j := 0 to 15 - i := j*32 - IF k[j] - dst[i+31:i] := v3[i+31:i] - v2[i+31:i] - ELSE - dst[i+31:i] := src[i+31:i] - FI -ENDFOR -dst[MAX:512] := 0 +dst[63:0] := Convert_FP64_To_Int64_Truncate(a[63:0]) - +
immintrin.h
 - + Floating Point - KNCNI - Arithmetic - - - - Performs element-by-element subtraction of packed double-precision (64-bit) floating-point elements in "v2" from "v3" storing the results in "dst". - [round_note] - + Integer + AVX512F + Convert + + + + Convert the lower double-precision (64-bit) floating-point element in "a" to a 32-bit integer with truncation, and store the result in "dst". + [sae_note] -FOR j := 0 to 7 - i := j*64 - dst[i+63:i] := v3[i+63:i] - v2[i+63:i] -ENDFOR -dst[MAX:512] := 0 +dst[31:0] := Convert_FP64_To_Int32_Truncate(a[63:0]) - +
immintrin.h
 - + Floating Point - KNCNI - Arithmetic - - - - - - Performs element-by-element subtraction of packed double-precision (64-bit) floating-point elements in "v2" from "v3" storing the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - [round_note] - + Integer + AVX512F + Convert + + + + Convert the lower double-precision (64-bit) floating-point element in "a" to a 64-bit integer with truncation, and store the result in "dst". + [sae_note] -FOR j := 0 to 7 - i := j*64 - IF k[j] - dst[i+63:i] := v3[i+63:i] - v2[i+63:i] - ELSE - dst[i+63:i] := src[i+63:i] - FI -ENDFOR -dst[MAX:512] := 0 +dst[63:0] := Convert_FP64_To_Int64_Truncate(a[63:0]) - +
immintrin.h
 - + Floating Point - KNCNI - Arithmetic - - - - Performs element-by-element subtraction of packed single-precision (32-bit) floating-point elements in "v2" from "v3" storing the results in "dst". - [round_note] - + AVX512F + Convert + + + Convert the lower double-precision (64-bit) floating-point element in "a" to a 32-bit integer with truncation, and store the result in "dst". -FOR j := 0 to 15 - i := j*32 - dst[i+31:i] := v3[i+31:i] - v2[i+31:i]) -ENDFOR -dst[MAX:512] := 0 +dst[31:0] := Convert_FP64_To_Int32_Truncate(a[63:0]) - +
immintrin.h
 - + Floating Point - KNCNI - Arithmetic - - - - - - Performs element-by-element subtraction of packed single-precision (32-bit) floating-point elements in "v2" from "v3" storing the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - [round_note] - + AVX512F + Convert + + + Convert the lower double-precision (64-bit) floating-point element in "a" to a 64-bit integer with truncation, and store the result in "dst". -FOR j := 0 to 15 - i := j*32 - IF k[j] - dst[i+31:i] := v3[i+31:i] - v2[i+31:i] - ELSE - dst[i+31:i] := src[i+31:i] - FI -ENDFOR -dst[MAX:512] := 0 +dst[63:0] := Convert_FP64_To_Int64_Truncate(a[63:0]) - +
immintrin.h
 - + + Floating Point Integer - KNCNI - Arithmetic - - - Performs element-by-element subtraction of packed 32-bit integer elements in "v2" from "v3" storing the results in "dst". + AVX512F + Convert + + + + Convert the lower double-precision (64-bit) floating-point element in "a" to an unsigned 32-bit integer with truncation, and store the result in "dst". + [sae_note] -FOR j := 0 to 15 - i := j*32 - dst[i+31:i] := v3[i+31:i] - v2[i+31:i] -ENDFOR -dst[MAX:512] := 0 +dst[31:0] := Convert_FP64_To_UInt32_Truncate(a[63:0]) - +
immintrin.h
 - + + Floating Point Integer - KNCNI - Arithmetic - - - - - Performs element-by-element subtraction of packed 32-bit integer elements in "v2" from "v3" storing the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set) + AVX512F + Convert + + + + Convert the lower double-precision (64-bit) floating-point element in "a" to an unsigned 64-bit integer with truncation, and store the result in "dst". + [sae_note] -FOR j := 0 to 15 - i := j*32 - IF k[j] - dst[i+31:i] := v3[i+31:i] - v2[i+31:i] - ELSE - dst[i+31:i] := src[i+31:i] - FI -ENDFOR -dst[MAX:512] := 0 +dst[63:0] := Convert_FP64_To_UInt64_Truncate(a[63:0]) - +
immintrin.h
 - + + Floating Point Integer - KNCNI - Arithmetic - - - - Performs element-by-element addition of packed 32-bit integer elements in "v2" and "v3", storing the resultant carry in "k2_res" (carry flag) and the addition results in "dst". + AVX512F + Convert + + + Convert the lower double-precision (64-bit) floating-point element in "a" to an unsigned 32-bit integer with truncation, and store the result in "dst". -FOR j := 0 to 15 - i := j*32 - dst[i+31:i] := v2[i+31:i] + v3[i+31:i] - k2_res[j] := Carry(v2[i+31:i] + v3[i+31:i]) -ENDFOR -dst[MAX:512] := 0 +dst[31:0] := Convert_FP64_To_UInt32_Truncate(a[63:0]) - +
immintrin.h
 - + + Floating Point Integer - KNCNI - Arithmetic - - - - - - Performs element-by-element addition of packed 32-bit integer elements in "v2" and "v3", storing the resultant carry in "k2_res" (carry flag) and the addition results in "dst" using writemask "k" (elements are copied from "v2" and "k_old" when the corresponding mask bit is not set). + AVX512F + Convert + + + Convert the lower double-precision (64-bit) floating-point element in "a" to an unsigned 64-bit integer with truncation, and store the result in "dst". -FOR j := 0 to 15 - i := j*32 - IF k[j] - dst[i+31:i] := v2[i+31:i] + v3[i+31:i] - ELSE - dst[i+31:i] := v2[i+31:i] - k2_res[j] := k_old[j] - FI -ENDFOR -dst[MAX:512] := 0 +dst[63:0] := Convert_FP64_To_UInt64_Truncate(a[63:0]) + + +
immintrin.h
+ + + Floating Point + AVX512F + Convert + + + + Convert the lower single-precision (32-bit) floating-point element in "a" to a 32-bit integer with truncation, and store the result in "dst". + [sae_note] + +dst[31:0] := Convert_FP32_To_Int32_Truncate(a[31:0]) + + +
immintrin.h
+ + + Floating Point + AVX512F + Convert + + + + Convert the lower single-precision (32-bit) floating-point element in "a" to a 64-bit integer with truncation, and store the result in "dst". + [sae_note] + +dst[63:0] := Convert_FP32_To_Int64_Truncate(a[31:0]) - +
immintrin.h
 - + + Floating Point Integer - KNCNI - Arithmetic - - - - Performs an element-by-element addition of packed 32-bit integer elements in "v2" and "v3", storing the results in "dst" and the sign of the sum in "sign" (sign flag). + AVX512F + Convert + + + + Convert the lower single-precision (32-bit) floating-point element in "a" to a 32-bit integer with truncation, and store the result in "dst". + [sae_note] -FOR j := 0 to 15 - i := j*32 - dst[i+31:i] := v2[i+31:i] + v3[i+31:i] - sign[j] := v2[i+31:i] & v3[i+31:i] & 0x80000000 -ENDFOR -dst[MAX:512] := 0 +dst[31:0] := Convert_FP32_To_Int32_Truncate(a[31:0]) - +
immintrin.h
 - + + Floating Point Integer - KNCNI - Arithmetic - - - - - - Performs an element-by-element addition of packed 32-bit integer elements in "v2" and "v3", storing the results in "dst" and the sign of the sum in "sign" (sign flag). Results are stored using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + AVX512F + Convert + + + + Convert the lower single-precision (32-bit) floating-point element in "a" to a 64-bit integer with truncation, and store the result in "dst". + [sae_note] -FOR j := 0 to 15 - i := j*32 - IF k[j] - dst[i+31:i] := v2[i+31:i] + v3[i+31:i] - sign[j] := v2[i+31:i] & v3[i+31:i] & 0x80000000 - ELSE - dst[i+31:i] := src[i+31:i] - FI -ENDFOR -dst[MAX:512] := 0 +dst[63:0] := Convert_FP32_To_Int64_Truncate(a[31:0]) - +
immintrin.h
 - + Floating Point - KNCNI - Arithmetic - - - - Performs an element-by-element addition of packed single-precision (32-bit) floating-point elements in "v2" and "v3", storing the results in "dst" and the sign of the sum in "sign" (sign flag). + AVX512F + Convert + + + Convert the lower single-precision (32-bit) floating-point element in "a" to a 32-bit integer with truncation, and store the result in "dst". -FOR j := 0 to 15 - i := j*32 - dst[i+31:i] := v2[i+31:i] + v3[i+31:i] - sign[j] := v2[i+31:i] & v3[i+31:i] & 0x80000000 -ENDFOR -dst[MAX:512] := 0 +dst[31:0] := Convert_FP32_To_Int32_Truncate(a[31:0]) - +
immintrin.h
 - + Floating Point - KNCNI - Arithmetic - - - - - - Performs an element-by-element addition of packed single-precision (32-bit) floating-point elements in "v2" and "v3", storing the results in "dst" and the sign of the sum in "sign" (sign flag). Results are stored using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + AVX512F + Convert + + + Convert the lower single-precision (32-bit) floating-point element in "a" to a 64-bit integer with truncation, and store the result in "dst". -FOR j := 0 to 15 - i := j*32 - IF k[j] - dst[i+31:i] := v2[i+31:i] + v3[i+31:i] - sign[j] := v2[i+31:i] & v3[i+31:i] & 0x80000000 - ELSE - dst[i+31:i] := src[i+31:i] - FI -ENDFOR -dst[MAX:512] := 0 +dst[63:0] := Convert_FP32_To_Int64_Truncate(a[31:0]) - +
immintrin.h
 - + Floating Point - KNCNI - Arithmetic - - - - - Performs an element-by-element addition of packed single-precision (32-bit) floating-point elements in "v2" and "v3", storing the results in "dst" and the sign of the sum in "sign" (sign flag). - [round_note] - + Integer + AVX512F + Convert + + + + Convert the lower single-precision (32-bit) floating-point element in "a" to an unsigned 32-bit integer with truncation, and store the result in "dst". + [sae_note] -FOR j := 0 to 15 - i := j*32 - dst[i+31:i] := v2[i+31:i] + v3[i+31:i] - sign[j] := v2[i+31:i] & v3[i+31:i] & 0x80000000 -ENDFOR -dst[MAX:512] := 0 +dst[31:0] := Convert_FP32_To_UInt32_Truncate(a[31:0]) - +
immintrin.h
 - + Floating Point - KNCNI - Arithmetic - - - - - - - Performs an element-by-element addition of packed single-precision (32-bit) floating-point elements in "v2" and "v3", storing the results in "dst" and the sign of the sum in "sign" (sign flag). Results are stored using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - [round_note] - + Integer + AVX512F + Convert + + + + Convert the lower single-precision (32-bit) floating-point element in "a" to an unsigned 64-bit integer with truncation, and store the result in "dst". + [sae_note] -FOR j := 0 to 15 - i := j*32 - IF k[j] - dst[i+31:i] := v2[i+31:i] + v3[i+31:i] - sign[j] := v2[i+31:i] & v3[i+31:i] & 0x80000000 - ELSE - dst[i+31:i] := src[i+31:i] - FI -ENDFOR -dst[MAX:512] := 0 +dst[63:0] := Convert_FP32_To_UInt64_Truncate(a[31:0]) - +
immintrin.h
 - + + Floating Point Integer - KNCNI - Arithmetic - - - - Performs element-by-element subtraction of packed 32-bit integer elements in "v3" from "v2", storing the results in "dst" and the nth borrow bit in the nth position of "borrow" (borrow flag). + AVX512F + Convert + + + Convert the lower single-precision (32-bit) floating-point element in "a" to an unsigned 32-bit integer with truncation, and store the result in "dst". -FOR j := 0 to 15 - i := j*32 - dst[i+31:i] := v2[i+31:i] - v3[i+31:i] - borrow[j] := Borrow(v2[i+31:i] - v3[i+31:i]) -ENDFOR -dst[MAX:512] := 0 +dst[31:0] := Convert_FP32_To_UInt32_Truncate(a[31:0]) - +
immintrin.h
 - + + Floating Point Integer - KNCNI - Arithmetic - - - - - - Performs element-by-element subtraction of packed 32-bit integer elements in "v3" from "v2", storing the results in "dst" and the nth borrow bit in the nth position of "borrow" (borrow flag). Results are stored using writemask "k" (elements are copied from "v2" and "k_old" when the corresponding mask bit is not set). + AVX512F + Convert + + + Convert the lower single-precision (32-bit) floating-point element in "a" to an unsigned 64-bit integer with truncation, and store the result in "dst". -FOR j := 0 to 15 - i := j*32 - IF k[j] - dst[i+31:i] := v2[i+31:i] - v3[i+31:i] - borrow[j] := Borrow(v2[i+31:i] - v3[i+31:i]) - ELSE - dst[i+31:i] := src[i+31:i] - borrow[j] := k_old[j] - FI +dst[63:0] := Convert_FP32_To_UInt64_Truncate(a[31:0]) + + +
immintrin.h
+ + + Floating Point + Integer + AVX512F + Convert + + + Convert packed unsigned 32-bit integers in "a" to packed double-precision (64-bit) floating-point elements, and store the results in "dst". + +FOR j := 0 to 7 + i := j*64 + l := j*32 + dst[i+63:i] := Convert_Int64_To_FP64(a[l+31:l]) ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + + Floating Point Integer - KNCNI - Arithmetic - - - - Performs element-by-element subtraction of packed 32-bit integer elements in "v2" from "v3", storing the results in "dst" and "v2". The borrowed value from the subtraction difference for the nth element is written to the nth bit of "borrow" (borrow flag). + AVX512F + Convert + + + + + Convert packed unsigned 32-bit integers in "a" to packed double-precision (64-bit) floating-point elements, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*32 - dst[i+31:i] := v3[i+31:i] - v2[i+31:i] - borrow[j] := Borrow(v3[i+31:i] - v2[i+31:i]) +FOR j := 0 to 7 + i := j*64 + l := j*32 + IF k[j] + dst[i+63:i] := Convert_Int64_To_FP64(a[l+31:l]) + ELSE + dst[i+63:i] := src[i+63:i] + FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + + Floating Point Integer - KNCNI - Arithmetic - - - - - - Performs element-by-element subtraction of packed 32-bit integer elements in "v2" from "v3", storing the results in "dst" and "v2". The borrowed value from the subtraction difference for the nth element is written to the nth bit of "borrow" (borrow flag). Results are written using writemask "k" (elements are copied from "k" to "k_old" when the corresponding mask bit is not set). + AVX512F + Convert + + + + Convert packed unsigned 32-bit integers in "a" to packed double-precision (64-bit) floating-point elements, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*32 +FOR j := 0 to 7 + i := j*64 + l := j*32 IF k[j] - diff := v3[i+31:i] - v2[i+31:i] - borrow[j] := Borrow(v3[i+31:i] - v2[i+31:i]) - dst[i+31:i] := diff - v2[i+31:i] := diff + dst[i+63:i] := Convert_Int64_To_FP64(a[l+31:l]) ELSE - borrow[j] := k_old[j] - FI + dst[i+63:i] := 0 + FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + + Floating Point Integer - KNCNI - Arithmetic - - - - - Performs element-by-element three-input subtraction of packed 32-bit integer elements of "v3" as well as the corresponding bit from "k" from "v2". The borrowed value from the subtraction difference for the nth element is written to the nth bit of "borrow" (borrow flag). Results are stored in "dst". + AVX512F + Convert + + + + Convert packed unsigned 32-bit integers in "a" to packed single-precision (32-bit) floating-point elements, and store the results in "dst". + [round_note] FOR j := 0 to 15 - i := j*32 - dst[i+31:i] := v2[i+31:i] - v3[i+31:i] - k[j] - borrow[j] := Borrow(v2[i+31:i] - v3[i+31:i] - k[j]) + i := 32*j + dst[i+31:i] := Convert_Int32_To_FP32(a[i+31:i]) ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + + Floating Point Integer - KNCNI - Arithmetic - - - - - - Performs element-by-element three-input subtraction of packed 32-bit integer elements of "v3" as well as the corresponding bit from "k2" from "v2". The borrowed value from the subtraction difference for the nth element is written to the nth bit of "borrow" (borrow flag). Results are stored in "dst" using writemask "k1" (elements are copied from "v2" when the corresponding mask bit is not set). + AVX512F + Convert + + + Convert packed unsigned 32-bit integers in "a" to packed single-precision (32-bit) floating-point elements, and store the results in "dst". FOR j := 0 to 15 - i := j*32 - IF k1[j] - dst[i+31:i] := v2[i+31:i] - v3[i+31:i] - k2[j] - borrow[j] := Borrow(v2[i+31:i] - v3[i+31:i] - k2[j]) - ELSE - dst[i+31:i] := v2[i+31:i] - FI + i := 32*j + dst[i+31:i] := Convert_Int32_To_FP32(a[i+31:i]) ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + + Floating Point Integer - KNCNI - Arithmetic - - - - - Performs element-by-element three-input subtraction of packed 32-bit integer elements of "v2" as well as the corresponding bit from "k" from "v3". The borrowed value from the subtraction difference for the nth element is written to the nth bit of "borrow" (borrow flag). Results are stored in "dst". + AVX512F + Convert + + + + + + Convert packed unsigned 32-bit integers in "a" to packed single-precision (32-bit) floating-point elements, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + [round_note] FOR j := 0 to 15 i := j*32 - dst[i+31:i] := v3[i+31:i] - v2[i+31:i] - k[j] - borrow[j] := Borrow(v2[i+31:i] - v3[i+31:i] - k[j]) + IF k[j] + dst[i+31:i] := Convert_Int32_To_FP32(a[i+31:i]) + ELSE + dst[i+31:i] := src[i+31:i] + FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + + Floating Point Integer - KNCNI - Arithmetic - - - - - - Performs element-by-element three-input subtraction of packed 32-bit integer elements of "v2" as well as the corresponding bit from "k2" from "v3". The borrowed value from the subtraction difference for the nth element is written to the nth bit of "borrow" (borrow flag). Results are stored in "dst" using writemask "k1" (elements are copied from "v2" when the corresponding mask bit is not set). + AVX512F + Convert + + + + + Convert packed unsigned 32-bit integers in "a" to packed single-precision (32-bit) floating-point elements, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). FOR j := 0 to 15 i := j*32 - IF k1[j] - dst[i+31:i] := v3[i+31:i] - v2[i+31:i] - k2[j] - borrow[j] := Borrow(v2[i+31:i] - v3[i+31:i] - k[j]) + IF k[j] + dst[i+31:i] := Convert_Int32_To_FP32(a[i+31:i]) ELSE - dst[i+31:i] := v2[i+31:i] + dst[i+31:i] := src[i+31:i] FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + + Floating Point Integer - AVX512F/KNCNI - Logical - - - - - Performs element-by-element bitwise AND between packed 32-bit integer elements of "v2" and "v3", storing the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + AVX512F + Convert + + + + + Convert packed unsigned 32-bit integers in "a" to packed single-precision (32-bit) floating-point elements, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + [round_note] FOR j := 0 to 15 - i := j*32 + i := 32*j IF k[j] - dst[i+31:i] := v2[i+31:i] & v3[i+31:i] + dst[i+31:i] := Convert_Int32_To_FP32(a[i+31:i]) ELSE - dst[i+31:i] := src[i+31:i] + dst[i+31:i] := 0 FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - KNCNI + Integer + AVX512F Convert - - - Performs element-by-element conversion of packed double-precision (64-bit) floating-point elements in "v2" to packed single-precision (32-bit) floating-point elements, storing the results in "dst". Results are written to the lower half of "dst", and the upper half locations are set to '0'. - [round_note] - + + + + Convert packed unsigned 32-bit integers in "a" to packed single-precision (32-bit) floating-point elements, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := j*64 - k := j*32 - dst[k+31:k] := Float64ToFloat32(v2[i+63:i]) +FOR j := 0 to 15 + i := 32*j + IF k[j] + dst[i+31:i] := Convert_Int32_To_FP32(a[i+31:i]) + ELSE + dst[i+31:i] := 0 + FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - KNCNI + Integer + AVX512F Convert - - - - - Performs element-by-element conversion of packed double-precision (64-bit) floating-point elements in "v2" to packed single-precision (32-bit) floating-point elements, storing the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). Results are written to the lower half of "dst", and the upper half locations are set to '0'. - [round_note] - + + + + + Convert the unsigned 64-bit integer "b" to a double-precision (64-bit) floating-point element, store the result in the lower element of "dst", and copy the upper element from "a" to the upper element of "dst". + [round_note] -FOR j := 0 to 7 - i := j*64 - l := j*32 - IF k[j] - dst[l+31:l] := Float64ToFloat32(v2[i+63:i]) - ELSE - dst[l+31:l] := src[l+31:i] - FI -ENDFOR -dst[MAX:512] := 0 +dst[63:0] := Convert_Int64_To_FP64(b[63:0]) +dst[127:64] := a[127:64] +dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point Integer - KNCNI + AVX512F Convert - - - Performs element-by-element conversion of packed double-precision (64-bit) floating-point elements in "v2" to packed 32-bit unsigned integer elements, storing the results in "dst". Results are written to the lower half of "dst", and the upper half locations are set to '0'. - [round_note] - + + + + Convert the unsigned 32-bit integer "b" to a double-precision (64-bit) floating-point element, store the result in the lower element of "dst", and copy the upper element from "a" to the upper element of "dst". -FOR j := 0 to 7 - i := j*64 - k := j*32 - dst[k+31:k] := Float64ToInt32(v2[i+63:i]) -ENDFOR -dst[MAX:512] := 0 +dst[63:0] := Convert_Int32_To_FP64(b[31:0]) +dst[127:64] := a[127:64] +dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point Integer - KNCNI + AVX512F Convert - - - - - Performs element-by-element conversion of packed double-precision (64-bit) floating-point elements in "v2" to packed 32-bit unsigned integer elements, storing the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). Results are written to the lower half of "dst", and the upper half locations are set to '0'. - [round_note] - + + + + Convert the unsigned 64-bit integer "b" to a double-precision (64-bit) floating-point element, store the result in the lower element of "dst", and copy the upper element from "a" to the upper element of "dst". -FOR j := 0 to 7 - i := j*64 - l := j*32 - IF k[j] - dst[l+31:l] := Float64ToInt32(v2[i+63:i]) - ELSE - dst[l+31:l] := src[l+31:i] - FI -ENDFOR -dst[MAX:512] := 0 +dst[63:0] := Convert_Int64_To_FP64(b[63:0]) +dst[127:64] := a[127:64] +dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point - AVX512F/KNCNI + Integer + AVX512F Convert - - Performs element-by-element conversion of the lower half of packed single-precision (32-bit) floating-point elements in "v2" to packed double-precision (64-bit) floating-point elements, storing the results in "dst". + + + + + Convert the unsigned 32-bit integer "b" to a single-precision (32-bit) floating-point element, store the result in the lower element of "dst", and copy the upper 3 packed elements from "a" to the upper elements of "dst". + [round_note] -FOR j := 0 to 7 - i := j*32 - k := j*64 - dst[k+63:k] := Float32ToFloat64(v2[i+31:i]) -ENDFOR -dst[MAX:512] := 0 +dst[31:0] := Convert_Int32_To_FP32(b[31:0]) +dst[127:32] := a[127:32] +dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point - AVX512F/KNCNI + Integer + AVX512F Convert - - - - Performs element-by-element conversion of the lower half of packed single-precision (32-bit) floating-point elements in "v2" to packed double-precision (64-bit) floating-point elements, storing the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + + + + Convert the unsigned 64-bit integer "b" to a single-precision (32-bit) floating-point element, store the result in the lower element of "dst", and copy the upper 3 packed elements from "a" to the upper elements of "dst". + [round_note] -FOR j := 0 to 7 - i := j*32 - l := j*64 - IF k[j] - dst[l+63:l] := Float32ToFloat64(v2[i+31:i]) - ELSE - dst[l+63:l] := src[l+63:l]: - FI -ENDFOR -dst[MAX:512] := 0 +dst[31:0] := Convert_Int64_To_FP32(b[63:0]) +dst[127:32] := a[127:32] +dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point Integer - KNCNI + AVX512F Convert - - - - Performs element-by-element conversion of packed single-precision (32-bit) floating-point elements in "v2" to packed 32-bit integer elements and performs an optional exponent adjust using "expadj", storing the results in "dst". - [round_note] - + + + + Convert the unsigned 32-bit integer "b" to a single-precision (32-bit) floating-point element, store the result in the lower element of "dst", and copy the upper 3 packed elements from "a" to the upper elements of "dst". -FOR j := 0 to 15 - i := j*32 - dst[i+31:i] := Float32ToInt32(v2[i+31:i]) - CASE expadj OF - _MM_EXPADJ_NONE: dst[i+31:i] = dst[i+31:i] * 2**0 - _MM_EXPADJ_4: dst[i+31:i] = dst[i+31:i] * 2**4 - _MM_EXPADJ_5: dst[i+31:i] = dst[i+31:i] * 2**5 - _MM_EXPADJ_8: dst[i+31:i] = dst[i+31:i] * 2**8 - _MM_EXPADJ_16: dst[i+31:i] = dst[i+31:i] * 2**16 - _MM_EXPADJ_24: dst[i+31:i] = dst[i+31:i] * 2**24 - _MM_EXPADJ_31: dst[i+31:i] = dst[i+31:i] * 2**31 - _MM_EXPADJ_32: dst[i+31:i] = dst[i+31:i] * 2**32 - ESAC -ENDFOR -dst[MAX:512] := 0 +dst[31:0] := Convert_Int32_To_FP32(b[31:0]) +dst[127:32] := a[127:32] +dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point Integer - KNCNI + AVX512F Convert - - - - Performs element-by-element conversion of packed single-precision (32-bit) floating-point elements in "v2" to packed 32-bit unsigned integer elements and performing an optional exponent adjust using "expadj", storing the results in "dst". - [round_note] - + + + + Convert the unsigned 64-bit integer "b" to a single-precision (32-bit) floating-point element, store the result in the lower element of "dst", and copy the upper 3 packed elements from "a" to the upper elements of "dst". -FOR j := 0 to 15 - i := j*32 - dst[i+31:i] := Float32ToUInt32(v2[i+31:i]) - CASE expadj OF - _MM_EXPADJ_NONE: dst[i+31:i] = dst[i+31:i] 0 - _MM_EXPADJ_4: dst[i+31:i] = dst[i+31:i] 4 - _MM_EXPADJ_5: dst[i+31:i] = dst[i+31:i] 5 - _MM_EXPADJ_8: dst[i+31:i] = dst[i+31:i] 8 - _MM_EXPADJ_16: dst[i+31:i] = dst[i+31:i] 16 - _MM_EXPADJ_24: dst[i+31:i] = dst[i+31:i] 24 - _MM_EXPADJ_31: dst[i+31:i] = dst[i+31:i] 31 - _MM_EXPADJ_32: dst[i+31:i] = dst[i+31:i] 32 - ESAC +dst[31:0] := Convert_Int64_To_FP32(b[63:0]) +dst[127:32] := a[127:32] +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512F + Arithmetic + + + + Divide packed double-precision (64-bit) floating-point elements in "a" by packed elements in "b", and store the results in "dst". + +FOR j := 0 to 7 + i := 64*j + dst[i+63:i] := a[i+63:i] / b[i+63:i] ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - Integer - AVX512F/KNCNI - Convert - - Performs element-by-element conversion of the lower half of packed 32-bit integer elements in "v2" to packed double-precision (64-bit) floating-point elements, storing the results in "dst". + AVX512F + Arithmetic + + + + + Divide packed double-precision (64-bit) floating-point elements in "a" by packed elements in "b", =and store the results in "dst". + [round_note] FOR j := 0 to 7 - i := j*32 - l := j*64 - dst[l+63:l] := Int32ToFloat64(v2[i+31:i]) + i := 64*j + dst[i+63:i] := a[i+63:i] / b[i+63:i] ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - Integer - AVX512F/KNCNI - Convert - - - - Performs element-by-element conversion of the lower half of packed 32-bit integer elements in "v2" to packed double-precision (64-bit) floating-point elements, storing the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + AVX512F + Arithmetic + + + + + + Divide packed double-precision (64-bit) floating-point elements in "a" by packed elements in "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). FOR j := 0 to 7 - i := j*32 - n := j*64 + i := 64*j IF k[j] - dst[k+63:k] := Int32ToFloat64(v2[i+31:i]) + dst[i+63:i] := a[i+63:i] / b[i+63:i] ELSE - dst[n+63:n] := src[n+63:n] + dst[i+63:i] := src[i+63:i] FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - Integer - AVX512F/KNCNI - Convert - - Performs element-by-element conversion of the lower half of packed 32-bit unsigned integer elements in "v2" to packed double-precision (64-bit) floating-point elements, storing the results in "dst". + AVX512F + Arithmetic + + + + + + + Divide packed double-precision (64-bit) floating-point elements in "a" by packed elements in "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + [round_note] FOR j := 0 to 7 - i := j*32 - k := j*64 - dst[k+63:k] := UInt32ToFloat64(v2[i+31:i]) + i := 64*j + IF k[j] + dst[i+63:i] := a[i+63:i] / b[i+63:i] + ELSE + dst[i+63:i] := src[i+63:i] + FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - Integer - AVX512F/KNCNI - Convert - - - - Performs element-by-element conversion of the lower half of 32-bit unsigned integer elements in "v2" to packed double-precision (64-bit) floating-point elements, storing the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + AVX512F + Arithmetic + + + + + Divide packed double-precision (64-bit) floating-point elements in "a" by packed elements in "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). FOR j := 0 to 7 - i := j*32 - l := j*64 + i := 64*j IF k[j] - dst[l+63:l] := UInt32ToFloat64(v2[i+31:i]) + dst[i+63:i] := a[i+63:i] / b[i+63:i] ELSE - dst[l+63:l] := src[l+63:l] + dst[i+63:i] := 0 FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - Integer - KNCNI - Convert - - - - Performs element-by-element conversion of packed 32-bit unsigned integer elements in "v2" to packed single-precision (32-bit) floating-point elements and performing an optional exponent adjust using "expadj", storing the results in "dst". - [round_note] - + AVX512F + Arithmetic + + + + + + Divide packed double-precision (64-bit) floating-point elements in "a" by packed elements in "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + [round_note] -FOR j := 0 to 15 - i := j*32 - dst[i+31:i] := UInt32ToFloat32(v2[i+31:i]) - CASE expadj OF - _MM_EXPADJ_NONE: dst[i+31:i] = dst[i+31:i] * 2**0 - _MM_EXPADJ_4: dst[i+31:i] = dst[i+31:i] * 2**4 - _MM_EXPADJ_5: dst[i+31:i] = dst[i+31:i] * 2**5 - _MM_EXPADJ_8: dst[i+31:i] = dst[i+31:i] * 2**8 - _MM_EXPADJ_16: dst[i+31:i] = dst[i+31:i] * 2**16 - _MM_EXPADJ_24: dst[i+31:i] = dst[i+31:i] * 2**24 - _MM_EXPADJ_31: dst[i+31:i] = dst[i+31:i] * 2**31 - _MM_EXPADJ_32: dst[i+31:i] = dst[i+31:i] * 2**32 - ESAC +FOR j := 0 to 7 + i := 64*j + IF k[j] + dst[i+63:i] := a[i+63:i] / b[i+63:i] + ELSE + dst[i+63:i] := 0 + FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - Integer - KNCNI - Convert - - - - - - Performs element-by-element conversion of packed 32-bit unsigned integer elements in "v2" to packed single-precision (32-bit) floating-point elements and performing an optional exponent adjust using "expadj", storing the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - [round_note] - + AVX512F + Arithmetic + + + + Divide packed single-precision (32-bit) floating-point elements in "a" by packed elements in "b", and store the results in "dst". FOR j := 0 to 15 - i := j*32 - IF k[j] - dst[i+31:i] := Int32ToFloat32(v2[i+31:i]) - CASE expadj OF - _MM_EXPADJ_NONE: dst[i+31:i] = dst[i+31:i] * 2**0 - _MM_EXPADJ_4: dst[i+31:i] = dst[i+31:i] * 2**4 - _MM_EXPADJ_5: dst[i+31:i] = dst[i+31:i] * 2**5 - _MM_EXPADJ_8: dst[i+31:i] = dst[i+31:i] * 2**8 - _MM_EXPADJ_16: dst[i+31:i] = dst[i+31:i] * 2**16 - _MM_EXPADJ_24: dst[i+31:i] = dst[i+31:i] * 2**24 - _MM_EXPADJ_31: dst[i+31:i] = dst[i+31:i] * 2**31 - _MM_EXPADJ_32: dst[i+31:i] = dst[i+31:i] * 2**32 - ESAC - ELSE - dst[i+31:i] := src[i+31:i] - FI + i := 32*j + dst[i+31:i] := a[i+31:i] / b[i+31:i] ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - KNCNI - Elementary Math Functions - - Approximates the base-2 exponent of the packed single-precision (32-bit) floating-point elements in "v2" with eight bits for sign and magnitude and 24 bits for the fractional part. Results are stored in "dst". + AVX512F + Arithmetic + + + + + Divide packed single-precision (32-bit) floating-point elements in "a" by packed elements in "b", and store the results in "dst". + [round_note] FOR j := 0 to 15 - i := j*32 - dst[i+31:i] := exp223(v2[i+31:i]) + i := 32*j + dst[i+31:i] := a[i+31:i] / b[i+31:i] ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - KNCNI - Elementary Math Functions - - - - Approximates the base-2 exponent of the packed single-precision (32-bit) floating-point elements in "v2" with eight bits for sign and magnitude and 24 bits for the fractional part. Results are stored in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + AVX512F + Arithmetic + + + + + + Divide packed single-precision (32-bit) floating-point elements in "a" by packed elements in "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). FOR j := 0 to 15 - i := j*32 + i := 32*j IF k[j] - dst[i+31:i] := exp223(v2[i+31:i]) + dst[i+31:i] := a[i+31:i] / b[i+31:i] ELSE dst[i+31:i] := src[i+31:i] FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - KNCNI - Miscellaneous - - - - Fixes up NaN's from packed double-precision (64-bit) floating-point elements in "v1" and "v2", storing the results in "dst" and storing the quietized NaN's from "v1" in "v3". + AVX512F + Arithmetic + + + + + + + Divide packed single-precision (32-bit) floating-point elements in "a" by packed elements in "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + [round_note] -FOR j := 0 to 7 - i := j*64 - dst[i+63:i] := FixupNaNs(v1[i+63:i], v2[i+63:i]) - v3[i+63:i] := QuietizeNaNs(v1[i+63:i]) +FOR j := 0 to 15 + i := 32*j + IF k[j] + dst[i+31:i] := a[i+31:i] / b[i+31:i] + ELSE + dst[i+31:i] := src[i+31:i] + FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - KNCNI - Miscellaneous - - - - - Fixes up NaN's from packed double-precision (64-bit) floating-point elements in "v1" and "v2", storing the results in "dst" using writemask "k" (only elements whose corresponding mask bit is set are used in the computation). Quietized NaN's from "v1" are stored in "v3". + AVX512F + Arithmetic + + + + + Divide packed single-precision (32-bit) floating-point elements in "a" by packed elements in "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := j*64 +FOR j := 0 to 15 + i := 32*j IF k[j] - dst[i+63:i] := FixupNaNs(v1[i+63:i], v2[i+63:i]) - v3[i+63:i] := QuietizeNaNs(v1[i+63:i]) + dst[i+31:i] := a[i+31:i] / b[i+31:i] + ELSE + dst[i+31:i] := 0 FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - KNCNI - Miscellaneous - - - - Fixes up NaN's from packed single-precision (32-bit) floating-point elements in "v1" and "v2", storing the results in "dst" and storing the quietized NaN's from "v1" in "v3". + AVX512F + Arithmetic + + + + + + Divide packed single-precision (32-bit) floating-point elements in "a" by packed elements in "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + [round_note] FOR j := 0 to 15 - i := j*32 - dst[i+31:i] := FixupNaNs(v1[i+31:i], v2[i+31:i]) - v3[i+31:i] := QuietizeNaNs(v1[i+31:i]) + i := 32*j + IF k[j] + dst[i+31:i] := a[i+31:i] / b[i+31:i] + ELSE + dst[i+31:i] := 0 + FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - KNCNI - Miscellaneous - - - - - Fixes up NaN's from packed single-precision (32-bit) floating-point elements in "v1" and "v2", storing the results in "dst" using writemask "k" (only elements whose corresponding mask bit is set are used in the computation). Quietized NaN's from "v1" are stored in "v3". + AVX512F + Arithmetic + + + + + Divide the lower double-precision (64-bit) floating-point element in "a" by the lower double-precision (64-bit) floating-point element in "b", store the result in the lower element of "dst", and copy the upper element from "a" to the upper element of "dst". + [round_note] -FOR j := 0 to 15 - i := j*32 - IF k[j] - dst[i+31:i] := FixupNaNs(v1[i+31:i], v2[i+31:i]) - v3[i+31:i] := QuietizeNaNs(v1[i+31:i]) - FI -ENDFOR -dst[MAX:512] := 0 +dst[63:0] := a[63:0] / b[63:0] +dst[127:64] := a[127:64] +dst[MAX:128] := 0 - +
immintrin.h
 - - Integer - AVX512F/KNCNI - Load - - - - - - Up-converts 16 memory locations starting at location "mv" at packed 32-bit integer indices stored in "index" scaled by "scale" using "conv" to 32-bit integer elements and stores them in "dst". AVX512 supports _MM_UPCONV_EPI32_NONE. + + Floating Point + AVX512F + Arithmetic + + + + + + + Divide the lower double-precision (64-bit) floating-point element in "a" by the lower double-precision (64-bit) floating-point element in "b", store the result in the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". + [round_note] -FOR j := 0 to 15 - addr := MEM[mv + index[j] * scale] - i := j*32 - CASE conv OF - _MM_UPCONV_EPI32_NONE: - dst[i+31:i] := addr[i+31:i] - _MM_UPCONV_EPI32_UINT8: - n := j*7 - dst[i+31:i] := UInt8ToUInt32(addr[n+7:n]) - _MM_UPCONV_EPI32_SINT8: - n := j*7 - dst[i+31:i] := Int8ToInt32(addr[n+7:n]) - _MM_UPCONV_EPI32_UINT16: - n := j*16 - dst[i+31:i] := UInt16ToUInt32(addr[n+15:n]) - _MM_UPCONV_EPI32_SINT16: - n := j*16 - dst[i+31:i] := Int16ToInt32(addr[n+15:n]) - ESAC -ENDFOR -dst[MAX:512] := 0 +IF k[0] + dst[63:0] := a[63:0] / b[63:0] +ELSE + dst[63:0] := src[63:0] +FI +dst[127:64] := a[127:64] +dst[MAX:128] := 0 - +
immintrin.h
 - - Integer - AVX512F/KNCNI - Load - - - - - - - - Up-converts 16 single-precision (32-bit) memory locations starting at location "mv" at packed 32-bit integer indices stored in "index" scaled by "scale" using "conv" to 32-bit integer elements and stores them in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). AVX512 supports _MM_UPCONV_EPI32_NONE. + + Floating Point + AVX512F + Arithmetic + + + + + + Divide the lower double-precision (64-bit) floating-point element in "a" by the lower double-precision (64-bit) floating-point element in "b", store the result in the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". -FOR j := 0 to 15 - addr := MEM[mv + index[j] * scale] - i := j*32 - IF k[j] - CASE conv OF - _MM_UPCONV_EPI32_NONE: - dst[i+31:i] := addr[i+31:i] - _MM_UPCONV_EPI32_UINT8: - n := j*7 - dst[i+31:i] := UInt8ToUInt32(addr[n+7:n]) - _MM_UPCONV_EPI32_SINT8: - n := j*7 - dst[i+31:i] := Int8ToInt32(addr[n+7:n]) - _MM_UPCONV_EPI32_UINT16: - n := j*16 - dst[i+31:i] := UInt16ToUInt32(addr[n+15:n]) - _MM_UPCONV_EPI32_SINT16: - n := j*16 - dst[i+31:i] := Int16ToInt32(addr[n+15:n]) - ESAC - ELSE - dst[i+31:i] := src[i+31:i] - FI -ENDFOR -dst[MAX:512] := 0 +IF k[0] + dst[63:0] := a[63:0] / b[63:0] +ELSE + dst[63:0] := src[63:0] +FI +dst[127:64] := a[127:64] +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512F + Arithmetic + + + + + + Divide the lower double-precision (64-bit) floating-point element in "a" by the lower double-precision (64-bit) floating-point element in "b", store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". + [round_note] + +IF k[0] + dst[63:0] := a[63:0] / b[63:0] +ELSE + dst[63:0] := 0 +FI +dst[127:64] := a[127:64] +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512F + Arithmetic + + + + + Divide the lower double-precision (64-bit) floating-point element in "a" by the lower double-precision (64-bit) floating-point element in "b", store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". + +IF k[0] + dst[63:0] := a[63:0] / b[63:0] +ELSE + dst[63:0] := 0 +FI +dst[127:64] := a[127:64] +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512F + Arithmetic + + + + + Divide the lower single-precision (32-bit) floating-point element in "a" by the lower single-precision (32-bit) floating-point element in "b", store the result in the lower element of "dst", and copy the upper 3 packed elements from "a" to the upper elements of "dst". + [round_note] + +dst[31:0] := a[31:0] / b[31:0] +dst[127:32] := a[127:32] +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512F + Arithmetic + + + + + + + Divide the lower single-precision (32-bit) floating-point element in "a" by the lower single-precision (32-bit) floating-point element in "b", store the result in the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". + [round_note] + +IF k[0] + dst[31:0] := a[31:0] / b[31:0] +ELSE + dst[31:0] := src[31:0] +FI +dst[127:32] := a[127:32] +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512F + Arithmetic + + + + + + Divide the lower single-precision (32-bit) floating-point element in "a" by the lower single-precision (32-bit) floating-point element in "b", store the result in the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". + +IF k[0] + dst[31:0] := a[31:0] / b[31:0] +ELSE + dst[31:0] := src[31:0] +FI +dst[127:32] := a[127:32] +dst[MAX:128] := 0 - +
immintrin.h
 - - Integer - AVX512F/KNCNI - Load - - - - - - Up-converts 8 double-precision (64-bit) memory locations starting at location "mv" at packed 32-bit integer indices stored in the lower half of "index" scaled by "scale" using "conv" to 64-bit integer elements and stores them in "dst". + + Floating Point + AVX512F + Arithmetic + + + + + + Divide the lower single-precision (32-bit) floating-point element in "a" by the lower single-precision (32-bit) floating-point element in "b", store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". + [round_note] -FOR j := 0 to 7 - addr := MEM[mv + index[j] * scale] - i := j*64 - CASE conv OF - _MM_UPCONV_EPI64_NONE: dst[i+63:i] := addr[i+63:i] - ESAC -ENDFOR -dst[MAX:512] := 0 +IF k[0] + dst[31:0] := a[31:0] / b[31:0] +ELSE + dst[31:0] := 0 +FI +dst[127:32] := a[127:32] +dst[MAX:128] := 0 - +
immintrin.h
 - - Integer - AVX512F/KNCNI - Load - - - - - - - - Up-converts 8 double-precision (64-bit) memory locations starting at location "mv" at packed 32-bit integer indices stored in the lower half of "index" scaled by "scale" using "conv" to 64-bit integer elements and stores them in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Floating Point + AVX512F + Arithmetic + + + + + Divide the lower single-precision (32-bit) floating-point element in "a" by the lower single-precision (32-bit) floating-point element in "b", store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". + +IF k[0] + dst[31:0] := a[31:0] / b[31:0] +ELSE + dst[31:0] := 0 +FI +dst[127:32] := a[127:32] +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512F + Swizzle + + + + + Load contiguous active double-precision (64-bit) floating-point elements from "a" (those with their respective bit set in mask "k"), and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). +m := 0 FOR j := 0 to 7 - addr := MEM[mv + index[j] * scale] i := j*64 IF k[j] - CASE conv OF - _MM_UPCONV_EPI64_NONE: dst[i+63:i] := addr[i+63:i] - ESAC + dst[i+63:i] := a[m+63:m] + m := m + 64 ELSE dst[i+63:i] := src[i+63:i] FI ENDFOR dst[MAX:512] := 0 - -
immintrin.h
- - - Floating Point - AVX512F/KNCNI - Load - - - - - - Up-converts 16 memory locations starting at location "mv" at packed 32-bit integer indices stored in "index" scaled by "scale" using "conv" to single-precision (32-bit) floating-point elements and stores them in "dst". - -FOR j := 0 to 15 - addr := MEM[mv + index[j] * scale] - i := j*32 - CASE conv OF - _MM_UPCONV_PS_NONE: - dst[i+31:i] := addr[i+31:i] - _MM_UPCONV_PS_FLOAT16: - n := j*16 - dst[i+31:i] := Float16ToFloat32(addr[n+15:n]) - _MM_UPCONV_PS_UINT8: - n := j*8 - dst[i+31:i] := UInt8ToFloat32(addr[n+7:n]) - _MM_UPCONV_PS_SINT8: - n := j*8 - dst[i+31:i] := SInt8ToFloat32(addr[n+7:n]) - _MM_UPCONV_PS_UINT16: - n := j*16 - dst[i+31:i] := UInt16ToFloat32(addr[n+15:n]) - _MM_UPCONV_PS_SINT16: - n := j*16 - dst[i+31:i] := SInt16ToFloat32(addr[n+15:n]) - ESAC -ENDFOR -dst[MAX:512] := 0 - - +
immintrin.h
 - + Floating Point - AVX512F/KNCNI + AVX512F Load - - - - - - - - Up-converts 16 single-precision (32-bit) memory locations starting at location "mv" at packed 32-bit integer indices stored in "index" scaled by "scale" using "conv" to single-precision (32-bit) floating-point elements and stores them in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Swizzle + + + + + Load contiguous active double-precision (64-bit) floating-point elements from unaligned memory at "mem_addr" (those with their respective bit set in mask "k"), and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 15 - addr := MEM[mv + index[j] * scale] - i := j*32 +m := 0 +FOR j := 0 to 7 + i := j*64 IF k[j] - CASE conv OF - _MM_UPCONV_PS_NONE: - dst[i+31:i] := addr[i+31:i] - _MM_UPCONV_PS_FLOAT16: - n := j*16 - dst[i+31:i] := Float16ToFloat32(addr[n+15:n]) - _MM_UPCONV_PS_UINT8: - n := j*8 - dst[i+31:i] := UInt8ToFloat32(addr[n+7:n]) - _MM_UPCONV_PS_SINT8: - n := j*8 - dst[i+31:i] := SInt8ToFloat32(addr[n+7:n]) - _MM_UPCONV_PS_UINT16: - n := j*16 - dst[i+31:i] := UInt16ToFloat32(addr[n+15:n]) - _MM_UPCONV_PS_SINT16: - n := j*16 - dst[i+31:i] := SInt16ToFloat32(addr[n+15:n]) - ESAC + dst[i+63:i] := MEM[mem_addr+m+63:mem_addr+m] + m := m + 64 ELSE - dst[i+31:i] := src[i+31:i] + dst[i+63:i] := src[i+63:i] FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - AVX512F/KNCNI - Load - - - - - - Up-converts 8 double-precision (64-bit) floating-point elements in memory locations starting at location "mv" at packed 32-bit integer indices stored in the lower half of "index" scaled by "scale" using "conv" to 64-bit floating-point elements and stores them in "dst". + AVX512F + Swizzle + + + + Load contiguous active double-precision (64-bit) floating-point elements from "a" (those with their respective bit set in mask "k"), and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). +m := 0 FOR j := 0 to 7 - addr := MEM[mv + index[j] * scale] i := j*64 - CASE conv OF - _MM_UPCONV_PD_NONE: dst[i+63:i] := addr[i+63:i] - ESAC + IF k[j] + dst[i+63:i] := a[m+63:m] + m := m + 64 + ELSE + dst[i+63:i] := 0 + FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - AVX512F/KNCNI + AVX512F Load - - - - - - - - Up-converts 8 double-precision (64-bit) floating-point elements in memory locations starting at location "mv" at packed 32-bit integer indices stored in the lower half of "index" scaled by "scale" using "conv" to 64-bit floating-point elements and stores them in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Swizzle + + + + Load contiguous active double-precision (64-bit) floating-point elements from unaligned memory at "mem_addr" (those with their respective bit set in mask "k"), and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). +m := 0 FOR j := 0 to 7 - addr := MEM[mv + index[j] * scale] i := j*64 IF k[j] - CASE conv OF - _MM_UPCONV_PD_NONE: dst[i+63:i] := addr[i+63:i] - ESAC + dst[i+63:i] := MEM[mem_addr+m+63:mem_addr+m] + m := m + 64 ELSE - dst[i+63:i] := src[i+63:i] + dst[i+63:i] := 0 FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - AVX512PF/KNCNI - Load - - - - - - Prefetches a set of 16 single-precision (32-bit) memory locations pointed by base address "mv" and 32-bit integer index vector "index" with scale "scale" to L1 or L2 level of cache depending on the value of "hint". The "hint" parameter may be 1 (_MM_HINT_T0) for prefetching to L1 cache, or 2 (_MM_HINT_T1) for prefetching to L2 cache. -The "conv" parameter specifies the granularity used by compilers to better encode the instruction. It should be the same as the "conv" parameter specified for the subsequent gather intrinsic. + AVX512F + Swizzle + + + + + Load contiguous active single-precision (32-bit) floating-point elements from "a" (those with their respective bit set in mask "k"), and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). +m := 0 FOR j := 0 to 15 - addr := MEM[mv + index[j] * scale] i := j*32 - CASE hint OF - _MM_HINT_T0: PrefetchL1WithT0Hint(addr[i+31:i]) - _MM_HINT_T1: PrefetchL2WithT1Hint(addr[i+31:i]) - ESAC + IF k[j] + dst[i+31:i] := a[m+31:m] + m := m + 32 + ELSE + dst[i+31:i] := src[i+31:i] + FI ENDFOR dst[MAX:512] := 0 - - +
immintrin.h
 - + Floating Point - AVX512PF/KNCNI + AVX512F Load - - - - - - - Prefetches a set of 16 single-precision (32-bit) memory locations pointed by base address "mv" and 32-bit integer index vector "index" with scale "scale" to L1 or L2 level of cache depending on the value of "hint". Gathered elements are merged in cache using writemask "k" (elements are brought into cache only when their corresponding mask bits are set). The "hint" parameter may be 1 (_MM_HINT_T0) for prefetching to L1 cache, or 2 (_MM_HINT_T1) for prefetching to L2 cache. -The "conv" parameter specifies the granularity used by compilers to better encode the instruction. It should be the same as the "conv" parameter specified for the subsequent gather intrinsic. + Swizzle + + + + + Load contiguous active single-precision (32-bit) floating-point elements from unaligned memory at "mem_addr" (those with their respective bit set in mask "k"), and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). +m := 0 FOR j := 0 to 15 - addr := MEM[mv + index[j] * scale] i := j*32 - IF k[j] THEN - CASE hint OF - _MM_HINT_T0: PrefetchL1WithT0Hint(addr[i+31:i]) - _MM_HINT_T1: PrefetchL2WithT1Hint(addr[i+31:i]) - ESAC + IF k[j] + dst[i+31:i] := MEM[mem_addr+m+31:mem_addr+m] + m := m + 32 + ELSE + dst[i+31:i] := src[i+31:i] FI ENDFOR dst[MAX:512] := 0 - - +
immintrin.h
 - + Floating Point - AVX512F/KNCNI - Store - - - - - - - Down-converts 16 packed single-precision (32-bit) floating-point elements in "v1" and stores them in memory locations starting at location "mv" at packed 32-bit integer indices stored in "index" scaled by "scale" using "conv". + AVX512F + Swizzle + + + + Load contiguous active single-precision (32-bit) floating-point elements from "a" (those with their respective bit set in mask "k"), and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). +m := 0 FOR j := 0 to 15 - addr := MEM[mv + index[j] * scale] i := j*32 - CASE conv OF - _MM_DOWNCONV_PS_NONE: - n := j*32 - addr[i+31:i] := v1[n+31:n] - _MM_DOWNCONV_PS_FLOAT16: - i := j*16 - addr[i+15:i] := Float32ToFloat16(v1[n+31:n]) - _MM_DOWNCONV_PS_UINT8: - i := j*8 - addr[i+7:i] := Float32ToUInt8(v1[n+31:n]) - _MM_DOWNCONV_PS_SINT8: - i := j*8 - addr[i+7:i] := Float32ToSInt8(v1[n+31:n]) - _MM_DOWNCONV_PS_UINT16: - i := j*8 - addr[i+15:i] := Float32ToUInt16(v1[n+31:n]) - _MM_DOWNCONV_PS_SINT16: - i := j*8 - addr[i+15:i] := Float32ToSInt16(v1[n+31:n]) - ESAC + IF k[j] + dst[i+31:i] := a[m+31:m] + m := m + 32 + ELSE + dst[i+31:i] := 0 + FI ENDFOR +dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - AVX512F/KNCNI - Store - - - - - - - - Down-converts 16 packed single-precision (32-bit) floating-point elements in "v1" according to "conv" and stores them in memory locations starting at location "mv" at packed 32-bit integer indices stored in "index" scaled by "scale" using writemask "k" (elements are written only when the corresponding mask bit is not set). + AVX512F + Load + Swizzle + + + + Load contiguous active single-precision (32-bit) floating-point elements from unaligned memory at "mem_addr" (those with their respective bit set in mask "k"), and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). +m := 0 FOR j := 0 to 15 i := j*32 IF k[j] - addr := MEM[mv + index[j] * scale] - CASE conv OF - _MM_DOWNCONV_PS_NONE: - n := j*32 - addr[i+31:i] := v1[n+31:n] - _MM_DOWNCONV_PS_FLOAT16: - i := j*16 - addr[i+15:i] := Float32ToFloat16(v1[n+31:n]) - _MM_DOWNCONV_PS_UINT8: - i := j*8 - addr[i+7:i] := Float32ToUInt8(v1[n+31:n]) - _MM_DOWNCONV_PS_SINT8: - i := j*8 - addr[i+7:i] := Float32ToSInt8(v1[n+31:n]) - _MM_DOWNCONV_PS_UINT16: - i := j*8 - addr[i+15:i] := Float32ToUInt16(v1[n+31:n]) - _MM_DOWNCONV_PS_SINT16: - i := j*8 - addr[i+15:i] := Float32ToSInt16(v1[n+31:n]) - ESAC + dst[i+31:i] := MEM[mem_addr+m+31:mem_addr+m] + m := m + 32 + ELSE + dst[i+31:i] := 0 FI ENDFOR +dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - AVX512F/KNCNI - Store - - - - - - - Down-converts 8 packed double-precision (64-bit) floating-point elements in "v1" and stores them in memory locations starting at location "mv" at packed 32-bit integer indices stored in "index" scaled by "scale" using "conv". + AVX512F + Swizzle + + + + Extract 128 bits (composed of 4 packed single-precision (32-bit) floating-point elements) from "a", selected with "imm8", and store the result in "dst". -FOR j := 0 to 7 - addr := MEM[mv + index[j] * scale] - i := j*64 - CASE conv OF - _MM_DOWNCONV_PD_NONE: addr[i+63:i] := v1[i+63:i] - ESAC -ENDFOR +CASE imm8[1:0] OF +0: dst[127:0] := a[127:0] +1: dst[127:0] := a[255:128] +2: dst[127:0] := a[383:256] +3: dst[127:0] := a[511:384] +ESAC +dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point - AVX512F/KNCNI - Store - - - - - - - - Down-converts 8 packed double-precision (64-bit) floating-point elements in "v1" and stores them in memory locations starting at location "mv" at packed 32-bit integer indices stored in "index" scaled by "scale" using "conv". Only those elements whose corresponding mask bit is set in writemask "k" are written to memory. - -FOR j := 0 to 7 + AVX512F + Swizzle + + + + + + Extract 128 bits (composed of 4 packed single-precision (32-bit) floating-point elements) from "a", selected with "imm8", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +CASE imm8[1:0] OF +0: tmp[127:0] := a[127:0] +1: tmp[127:0] := a[255:128] +2: tmp[127:0] := a[383:256] +3: tmp[127:0] := a[511:384] +ESAC +FOR j := 0 to 3 + i := j*32 IF k[j] - addr := MEM[mv + index[j] * scale] - i := j*64 - CASE conv OF - _MM_DOWNCONV_PD_NONE: addr[i+63:i] := v1[i+63:i] - ESAC + dst[i+31:i] := tmp[i+31:i] + ELSE + dst[i+31:i] := src[i+31:i] FI ENDFOR +dst[MAX:128] := 0 - -
immintrin.h
- - - Integer - AVX512F/KNCNI - Store - - - - - - - Down-converts 8 packed 64-bit integer elements in "v1" and stores them in memory locations starting at location "mv" at packed 32-bit integer indices stored in "index" scaled by "scale" using "conv". - -FOR j := 0 to 7 - addr := MEM[mv + index[j] * scale] - i := j*64 - CASE conv OF - _MM_DOWNCONV_EPI64_NONE: addr[i+63:i] := v1[i+63:i] - ESAC -ENDFOR - - +
immintrin.h
 - - Integer - AVX512F/KNCNI - Store - - - - - - - - Down-converts 8 packed 64-bit integer elements in "v1" and stores them in memory locations starting at location "mv" at packed 32-bit integer indices stored in "index" scaled by "scale" using "conv". Only those elements whose corresponding mask bit is set in writemask "k" are written to memory. + + Floating Point + AVX512F + Swizzle + + + + + Extract 128 bits (composed of 4 packed single-precision (32-bit) floating-point elements) from "a", selected with "imm8", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 +CASE imm8[1:0] OF +0: tmp[127:0] := a[127:0] +1: tmp[127:0] := a[255:128] +2: tmp[127:0] := a[383:256] +3: tmp[127:0] := a[511:384] +ESAC +FOR j := 0 to 3 + i := j*32 IF k[j] - addr := MEM[mv + index[j] * scale] - i := j*64 - CASE conv OF - _MM_DOWNCONV_EPI64_NONE: addr[i+63:i] := v1[i+63:i] - ESAC + dst[i+31:i] := tmp[i+31:i] + ELSE + dst[i+31:i] := 0 FI ENDFOR +dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point - AVX512PF/KNCNI - Store - - - - - - Prefetches a set of 16 single-precision (32-bit) memory locations pointed by base address "mv" and 32-bit integer index vector "index" with scale "scale" to L1 or L2 level of cache depending on the value of "hint", with a request for exclusive ownership. The "hint" parameter may be one of the following: _MM_HINT_T0 = 1 for prefetching to L1 cache, _MM_HINT_T1 = 2 for prefetching to L2 cache, _MM_HINT_T2 = 3 for prefetching to L2 cache non-temporal, _MM_HINT_NTA = 0 for prefetching to L1 cache non-temporal. The "conv" parameter specifies the granularity used by compilers to better encode the instruction. It should be the same as the "conv" parameter specified for the subsequent scatter intrinsic. + AVX512F + Swizzle + + + + Extract 256 bits (composed of 4 packed double-precision (64-bit) floating-point elements) from "a", selected with "imm8", and store the result in "dst". -cachev := 0 -FOR j := 0 to 15 - i := j*32 - addr := MEM[mv + index[j] * scale] - CASE hint OF - _MM_HINT_T0: PrefetchL1WithT0Hint(addr[i+31:i]) - _MM_HINT_T1: PrefetchL2WithT1Hint(addr[i+31:i]) - _MM_HINT_T2: PrefetchL2WithT1HintNonTemporal(addr[i+31:i]) - _MM_HINT_NTA: PrefetchL1WithT0HintNonTemporal(addr[i+31:i]) - ESAC -ENDFOR +CASE imm8[0] OF +0: dst[255:0] := a[255:0] +1: dst[255:0] := a[511:256] +ESAC +dst[MAX:256] := 0 - - +
immintrin.h
 - + Floating Point - AVX512PF/KNCNI - Store - - - - - - - Prefetches a set of 16 single-precision (32-bit) memory locations pointed by base address "mv" and 32-bit integer index vector "index" with scale "scale" to L1 or L2 level of cache depending on the value of "hint". The "hint" parameter may be 1 (_MM_HINT_T0) for prefetching to L1 cache, or 2 (_MM_HINT_T1) for prefetching to L2 cache. -The "conv" parameter specifies the granularity used by compilers to better encode the instruction. It should be the same as the "conv" parameter specified for the subsequent gather intrinsic. Only those elements whose corresponding mask bit in "k" is set are loaded into cache. - -cachev := 0 -FOR j := 0 to 15 - i := j*32 + AVX512F + Swizzle + + + + + + Extract 256 bits (composed of 4 packed double-precision (64-bit) floating-point elements) from "a", selected with "imm8", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +CASE imm8[0] OF +0: tmp[255:0] := a[255:0] +1: tmp[255:0] := a[511:256] +ESAC +FOR j := 0 to 3 + i := j*64 IF k[j] - addr := MEM[mv + index[j] * scale] - CASE hint OF - _MM_HINT_T0: PrefetchL1WithT0Hint(addr[i+31:i]) - _MM_HINT_T1: PrefetchL2WithT1Hint(addr[i+31:i]) - ESAC - FI -ENDFOR - - - -
immintrin.h
- - - Integer - KNCNI - Load - - - - - Loads the high-64-byte-aligned portion of the byte/word/doubleword stream starting at element-aligned address mt-64, up-converted depending on the value of "conv", and expanded into packed 32-bit integers in "dst". The initial values of "dst" are copied from "src". Only those converted doublewords that occur at or after the first 64-byte-aligned address following (mt-64) are loaded. Elements in the resulting vector that do not map to those doublewords are taken from "src". "hint" indicates to the processor whether the loaded data is non-temporal. - -UPCONVERT(address, offset, convertTo) { - CASE conv OF - _MM_UPCONV_EPI32_NONE: RETURN MEM[addr + 4*offset] - _MM_UPCONV_EPI32_UINT8: RETURN UInt8ToInt32(MEM[addr + offset]) - _MM_UPCONV_EPI32_SINT8: RETURN SInt8ToInt32(MEM[addr + offset]) - _MM_UPCONV_EPI32_UINT16: RETURN UInt16ToInt32(MEM[addr + 2*offset]) - _MM_UPCONV_EPI32_SINT16: RETURN SInt16ToInt32(MEM[addr + 2*offset]) - ESAC -} - -UPCONVERTSIZE(convertTo) { - CASE conv OF - _MM_UPCONV_EPI32_NONE: RETURN 4 - _MM_UPCONV_EPI32_UINT8: RETURN 1 - _MM_UPCONV_EPI32_SINT8: RETURN 1 - _MM_UPCONV_EPI32_UINT16: RETURN 2 - _MM_UPCONV_EPI32_SINT16: RETURN 2 - ESAC -} - -dst[511:0] := src[511:0] -loadOffset := 0 -foundNext64BytesBoundary := false -upSize := UPCONVERTSIZE(conv) -addr = mt-64 -FOR j := 0 to 15 - IF foundNext64BytesBoundary == false - IF (addr + (loadOffset + 1)*upSize % 64) == 0 - foundNext64BytesBoundary := true - FI + dst[i+63:i] := tmp[i+63:i] ELSE - i := j*32 - dst[i+31:i] := UPCONVERT(addr, loadOffset, conv) + dst[i+63:i] := src[i+63:i] FI - loadOffset := loadOffset + 1 ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - - Integer - KNCNI - Load - - - - - - Loads the high-64-byte-aligned portion of the byte/word/doubleword stream starting at element-aligned address mt-64, up-converted depending on the value of "conv", and expanded into packed 32-bit integers in "dst". The initial values of "dst" are copied from "src". Only those converted doublewords that occur at or after the first 64-byte-aligned address following (mt-64) are loaded. Elements in the resulting vector that do not map to those doublewords are taken from "src". "hint" indicates to the processor whether the loaded data is non-temporal. Elements are copied to "dst" according to element selector "k" (elements are skipped when the corresponding mask bit is not set). + + Floating Point + AVX512F + Swizzle + + + + + Extract 256 bits (composed of 4 packed double-precision (64-bit) floating-point elements) from "a", selected with "imm8", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -UPCONVERT(address, offset, convertTo) { - CASE conv OF - _MM_UPCONV_EPI32_NONE: RETURN MEM[addr + 4*offset] - _MM_UPCONV_EPI32_UINT8: RETURN UInt8ToInt32(MEM[addr + offset]) - _MM_UPCONV_EPI32_SINT8: RETURN SInt8ToInt32(MEM[addr + offset]) - _MM_UPCONV_EPI32_UINT16: RETURN UInt16ToInt32(MEM[addr + 2*offset]) - _MM_UPCONV_EPI32_SINT16: RETURN SInt16ToInt32(MEM[addr + 2*offset]) - ESAC -} - -UPCONVERTSIZE(convertTo) { - CASE conv OF - _MM_UPCONV_EPI32_NONE: RETURN 4 - _MM_UPCONV_EPI32_UINT8: RETURN 1 - _MM_UPCONV_EPI32_SINT8: RETURN 1 - _MM_UPCONV_EPI32_UINT16: RETURN 2 - _MM_UPCONV_EPI32_SINT16: RETURN 2 - ESAC -} - -dst[511:0] := src[511:0] -loadOffset := 0 -foundNext64BytesBoundary := false -upSize := UPCONVERTSIZE(conv) -addr = mt-64 -FOR j := 0 to 15 +CASE imm8[0] OF +0: tmp[255:0] := a[255:0] +1: tmp[255:0] := a[511:256] +ESAC +FOR j := 0 to 3 + i := j*64 IF k[j] - IF foundNext64BytesBoundary == false - IF (addr + (loadOffset + 1)*upSize % 64) == 0 - foundNext64BytesBoundary := true - FI - ELSE - i := j*32 - dst[i+31:i] := UPCONVERT(addr, loadOffset, conv) - FI - loadOffset := loadOffset + 1 + dst[i+63:i] := tmp[i+63:i] + ELSE + dst[i+63:i] := 0 FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - KNCNI - Load - - - - - Loads the low-64-byte-aligned portion of the byte/word/doubleword stream starting at element-aligned address mt, up-converted depending on the value of "conv", and expanded into packed 32-bit integers in "dst". The initial values of "dst" are copied from "src". Only those converted doublewords that occur before first 64-byte-aligned address following "mt" are loaded. Elements in the resulting vector that do not map to those doublewords are taken from "src". "hint" indicates to the processor whether the loaded data is non-temporal. + AVX512F + Swizzle + + + + Extract 128 bits (composed of 4 packed 32-bit integers) from "a", selected with "imm8", and store the result in "dst". -UPCONVERT(address, offset, convertTo) { - CASE conv OF - _MM_UPCONV_EPI32_NONE: RETURN MEM[addr + 4*offset] - _MM_UPCONV_EPI32_UINT8: RETURN UInt8ToInt32(MEM[addr + offset]) - _MM_UPCONV_EPI32_SINT8: RETURN SInt8ToInt32(MEM[addr + offset]) - _MM_UPCONV_EPI32_UINT16: RETURN UInt16ToInt32(MEM[addr + 2*offset]) - _MM_UPCONV_EPI32_SINT16: RETURN SInt16ToInt32(MEM[addr + 2*offset]) - ESAC -} - -UPCONVERTSIZE(convertTo) { - CASE conv OF - _MM_UPCONV_EPI32_NONE: RETURN 4 - _MM_UPCONV_EPI32_UINT8: RETURN 1 - _MM_UPCONV_EPI32_SINT8: RETURN 1 - _MM_UPCONV_EPI32_UINT16: RETURN 2 - _MM_UPCONV_EPI32_SINT16: RETURN 2 - ESAC -} - -dst[511:0] := src[511:0] -loadOffset := 0 -upSize := UPCONVERTSIZE(conv) -addr = mt -FOR j := 0 to 15 - i := j*32 - dst[i+31:i] := UPCONVERT(addr, loadOffset, conv) - loadOffset := loadOffset + 1 - IF (mt + loadOffset * upSize) % 64 == 0 - break - FI -ENDFOR -dst[MAX:512] := 0 +CASE imm8[1:0] OF +0: dst[127:0] := a[127:0] +1: dst[127:0] := a[255:128] +2: dst[127:0] := a[383:256] +3: dst[127:0] := a[511:384] +ESAC +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - KNCNI - Load - - - - - - Loads the low-64-byte-aligned portion of the byte/word/doubleword stream starting at element-aligned address mt, up-converted depending on the value of "conv", and expanded into packed 32-bit integers in "dst". The initial values of "dst" are copied from "src". Only those converted doublewords that occur before first 64-byte-aligned address following "mt" are loaded. Elements in the resulting vector that do not map to those doublewords are taken from "src". "hint" indicates to the processor whether the loaded data is non-temporal. Elements are copied to "dst" according to element selector "k" (elements are skipped when the corresponding mask bit is not set). - -UPCONVERT(address, offset, convertTo) { - CASE conv OF - _MM_UPCONV_EPI32_NONE: RETURN MEM[addr + 4*offset] - _MM_UPCONV_EPI32_UINT8: RETURN UInt8ToInt32(MEM[addr + offset]) - _MM_UPCONV_EPI32_SINT8: RETURN SInt8ToInt32(MEM[addr + offset]) - _MM_UPCONV_EPI32_UINT16: RETURN UInt16ToInt32(MEM[addr + 2*offset]) - _MM_UPCONV_EPI32_SINT16: RETURN SInt16ToInt32(MEM[addr + 2*offset]) - ESAC -} - -UPCONVERTSIZE(convertTo) { - CASE conv OF - _MM_UPCONV_EPI32_NONE: RETURN 4 - _MM_UPCONV_EPI32_UINT8: RETURN 1 - _MM_UPCONV_EPI32_SINT8: RETURN 1 - _MM_UPCONV_EPI32_UINT16: RETURN 2 - _MM_UPCONV_EPI32_SINT16: RETURN 2 - ESAC -} - -dst[511:0] := src[511:0] -loadOffset := 0 -upSize := UPCONVERTSIZE(conv) -addr = mt -FOR j := 0 to 15 + AVX512F + Swizzle + + + + + + Extract 128 bits (composed of 4 packed 32-bit integers) from "a", selected with "imm8", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +CASE imm8[1:0] OF +0: tmp[127:0] := a[127:0] +1: tmp[127:0] := a[255:128] +2: tmp[127:0] := a[383:256] +3: tmp[127:0] := a[511:384] +ESAC +FOR j := 0 to 3 + i := j*32 IF k[j] - i := j*32 - dst[i+31:i] := UPCONVERT(addr, loadOffset, conv) - loadOffset := loadOffset + 1 - IF (mt + loadOffset * upSize) % 64 == 0 - break - FI + dst[i+31:i] := tmp[i+31:i] + ELSE + dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - KNCNI - Load - - - - - Loads the high-64-byte-aligned portion of the quadword stream starting at element-aligned address mt-64, up-converted depending on the value of "conv", and expanded into packed 64-bit integers in "dst". The initial values of "dst" are copied from "src". Only those converted quadwords that occur at or after the first 64-byte-aligned address following (mt-64) are loaded. Elements in the resulting vector that do not map to those quadwords are taken from "src". "hint" indicates to the processor whether the loaded data is non-temporal. + AVX512F + Swizzle + + + + + Extract 128 bits (composed of 4 packed 32-bit integers) from "a", selected with "imm8", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -UPCONVERT(address, offset, convertTo) { - CASE conv OF - _MM_UPCONV_EPI64_NONE: RETURN MEM[addr + 8*offset] - ESAC -} - -UPCONVERTSIZE(convertTo) { - CASE conv OF - _MM_UPCONV_EPI64_NONE: RETURN 8 - ESAC -} - -dst[511:0] := src[511:0] -loadOffset := 0 -foundNext64BytesBoundary := false -upSize := UPCONVERTSIZE(conv) -addr = mt-64 -FOR j := 0 to 7 - IF foundNext64BytesBoundary == false - IF (addr + (loadOffset + 1)*upSize) == 0 - foundNext64BytesBoundary := true - FI +CASE imm8[1:0] OF +0: tmp[127:0] := a[127:0] +1: tmp[127:0] := a[255:128] +2: tmp[127:0] := a[383:256] +3: tmp[127:0] := a[511:384] +ESAC +FOR j := 0 to 3 + i := j*32 + IF k[j] + dst[i+31:i] := tmp[i+31:i] ELSE - i := j*64 - dst[i+63:i] := UPCONVERT(addr, loadOffset, conv) + dst[i+31:i] := 0 FI - loadOffset := loadOffset + 1 ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - KNCNI - Load - - - - - - Loads the high-64-byte-aligned portion of the quadword stream starting at element-aligned address mt-64, up-converted depending on the value of "conv", and expanded into packed 64-bit integers in "dst". The initial values of "dst" are copied from "src". Only those converted quadwords that occur at or after the first 64-byte-aligned address following (mt-64) are loaded. Elements in the resulting vector that do not map to those quadwords are taken from "src". "hint" indicates to the processor whether the loaded data is non-temporal. Elements are copied to "dst" according to element selector "k" (elements are skipped when the corresponding mask bit is not set). + AVX512F + Swizzle + + + + Extract 256 bits (composed of 4 packed 64-bit integers) from "a", selected with "imm8", and store the result in "dst". -UPCONVERT(address, offset, convertTo) { - CASE conv OF - _MM_UPCONV_EPI64_NONE: RETURN MEM[addr + 8*offset] - ESAC -} - -UPCONVERTSIZE(convertTo) { - CASE conv OF - _MM_UPCONV_EPI64_NONE: RETURN 8 - ESAC -} - -dst[511:0] := src[511:0] -loadOffset := 0 -foundNext64BytesBoundary := false -upSize := UPCONVERTSIZE(conv) -addr = mt-64 -FOR j := 0 to 7 - IF k[j] - IF foundNext64BytesBoundary == false - IF (addr + (loadOffset + 1)*upSize) == 0 - foundNext64BytesBoundary := true - FI - ELSE - i := j*64 - dst[i+63:i] := UPCONVERT(addr, loadOffset, conv) - FI - loadOffset := loadOffset + 1 - FI -ENDFOR -dst[MAX:512] := 0 +CASE imm8[0] OF +0: dst[255:0] := a[255:0] +1: dst[255:0] := a[511:256] +ESAC +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - KNCNI - Load - - - - - Loads the low-64-byte-aligned portion of the quadword stream starting at element-aligned address mt, up-converted depending on the value of "conv", and expanded into packed 64-bit integers in "dst". The initial values of "dst" are copied from "src". Only those converted quad that occur before first 64-byte-aligned address following "mt" are loaded. Elements in the resulting vector that do not map to those quadwords are taken from "src". "hint" indicates to the processor whether the loaded data is non-temporal. - -UPCONVERT(address, offset, convertTo) { - CASE conv OF - _MM_UPCONV_EPI64_NONE: RETURN MEM[addr + 8*offset] - ESAC -} - -UPCONVERTSIZE(convertTo) { - CASE conv OF - _MM_UPCONV_EPI64_NONE: RETURN 8 - ESAC -} - -dst[511:0] := src[511:0] -loadOffset := 0 -upSize := UPCONVERTSIZE(conv) -addr = mt -FOR j := 0 to 7 + AVX512F + Swizzle + + + + + + Extract 256 bits (composed of 4 packed 64-bit integers) from "a", selected with "imm8", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +CASE imm8[0] OF +0: tmp[255:0] := a[255:0] +1: tmp[255:0] := a[511:256] +ESAC +FOR j := 0 to 3 i := j*64 - dst[i+63:i] := UPCONVERT(addr, loadOffset, conv) - loadOffset := loadOffset + 1 - IF (addr + loadOffset*upSize % 64) == 0 - BREAK + IF k[j] + dst[i+63:i] := tmp[i+63:i] + ELSE + dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - KNCNI - Load - - - - - - Loads the low-64-byte-aligned portion of the quadword stream starting at element-aligned address mt, up-converted depending on the value of "conv", and expanded into packed 64-bit integers in "dst". The initial values of "dst" are copied from "src". Only those converted quadwords that occur before first 64-byte-aligned address following "mt" are loaded. Elements in the resulting vector that do not map to those quadwords are taken from "src". "hint" indicates to the processor whether the loaded data is non-temporal. Elements are copied to "dst" according to element selector "k" (elements are skipped when the corresponding mask bit is not set). + AVX512F + Swizzle + + + + + Extract 256 bits (composed of 4 packed 64-bit integers) from "a", selected with "imm8", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -UPCONVERT(address, offset, convertTo) { - CASE conv OF - _MM_UPCONV_EPI64_NONE: RETURN MEM[addr + 8*offset] - ESAC -} - -UPCONVERTSIZE(convertTo) { - CASE conv OF - _MM_UPCONV_EPI64_NONE: RETURN 8 - ESAC -} - -dst[511:0] := src[511:0] -loadOffset := 0 -upSize := UPCONVERTSIZE(conv) -addr = mt -FOR j := 0 to 7 +CASE imm8[0] OF +0: tmp[255:0] := a[255:0] +1: tmp[255:0] := a[511:256] +ESAC +FOR j := 0 to 3 + i := j*64 IF k[j] - i := j*64 - dst[i+63:i] := UPCONVERT(addr, loadOffset, conv) - loadOffset := loadOffset + 1 - IF (addr + loadOffset*upSize % 64) == 0 - BREAK - FI + dst[i+63:i] := tmp[i+63:i] + ELSE + dst[i+63:i] := 0 FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Floating Point - KNCNI - Load - - - - - Loads the high-64-byte-aligned portion of the byte/word/doubleword stream starting at element-aligned address mt-64, up-converted depending on the value of "conv", and expanded into packed single-precision (32-bit) floating-point elements in "dst". The initial values of "dst" are copied from "src". Only those converted quadwords that occur at or after the first 64-byte-aligned address following (mt-64) are loaded. Elements in the resulting vector that do not map to those quadwords are taken from "src". "hint" indicates to the processor whether the loaded data is non-temporal. - -UPCONVERT(address, offset, convertTo) { - CASE conv OF - _MM_UPCONV_PS_NONE: RETURN MEM[addr + 4*offset] - _MM_UPCONV_PS_FLOAT16: RETURN Float16ToFloat32(MEM[addr + 4*offset]) - _MM_UPCONV_PS_UINT8: RETURN UInt8ToFloat32(MEM[addr + offset]) - _MM_UPCONV_PS_SINT8: RETURN SInt8ToFloat32(MEM[addr + offset]) - _MM_UPCONV_PS_UINT16: RETURN UInt16ToFloat32(MEM[addr + 2*offset]) - _MM_UPCONV_PS_SINT16: RETURN SInt16ToFloat32(MEM[addr + 2*offset]) - ESAC + AVX512F + Miscellaneous + + + + + + Fix up packed double-precision (64-bit) floating-point elements in "a" and "b" using packed 64-bit integers in "c", and store the results in "dst". "imm8" is used to set the required flags reporting. + enum TOKEN_TYPE { + QNAN_TOKEN := 0, \ + SNAN_TOKEN := 1, \ + ZERO_VALUE_TOKEN := 2, \ + ONE_VALUE_TOKEN := 3, \ + NEG_INF_TOKEN := 4, \ + POS_INF_TOKEN := 5, \ + NEG_VALUE_TOKEN := 6, \ + POS_VALUE_TOKEN := 7 } - -UPCONVERTSIZE(convertTo) { - CASE conv OF - _MM_UPCONV_PS_NONE: RETURN 4 - _MM_UPCONV_PS_FLOAT16: RETURN 2 - _MM_UPCONV_PS_UINT8: RETURN 1 - _MM_UPCONV_PS_SINT8: RETURN 1 - _MM_UPCONV_PS_UINT16: RETURN 2 - _MM_UPCONV_PS_SINT16: RETURN 2 +DEFINE FIXUPIMMPD(src1[63:0], src2[63:0], src3[63:0], imm8[7:0]) { + tsrc[63:0] := ((src2[62:52] == 0) AND (MXCSR.DAZ == 1)) ? 0.0 : src2[63:0] + CASE(tsrc[63:0]) OF + QNAN_TOKEN:j := 0 + SNAN_TOKEN:j := 1 + ZERO_VALUE_TOKEN: j := 2 + ONE_VALUE_TOKEN: j := 3 + NEG_INF_TOKEN: j := 4 + POS_INF_TOKEN: j := 5 + NEG_VALUE_TOKEN: j := 6 + POS_VALUE_TOKEN: j := 7 + ESAC + + token_response[3:0] := src3[3+4*j:4*j] + + CASE(token_response[3:0]) OF + 0 : dest[63:0] := src1[63:0] + 1 : dest[63:0] := tsrc[63:0] + 2 : dest[63:0] := QNaN(tsrc[63:0]) + 3 : dest[63:0] := QNAN_Indefinite + 4 : dest[63:0] := -INF + 5 : dest[63:0] := +INF + 6 : dest[63:0] := tsrc.sign? -INF : +INF + 7 : dest[63:0] := -0 + 8 : dest[63:0] := +0 + 9 : dest[63:0] := -1 + 10: dest[63:0] := +1 + 11: dest[63:0] := 1/2 + 12: dest[63:0] := 90.0 + 13: dest[63:0] := PI/2 + 14: dest[63:0] := MAX_FLOAT + 15: dest[63:0] := -MAX_FLOAT ESAC + + CASE(tsrc[31:0]) OF + ZERO_VALUE_TOKEN: + IF (imm8[0]) #ZE; FI + ZERO_VALUE_TOKEN: + IF (imm8[1]) #IE; FI + ONE_VALUE_TOKEN: + IF (imm8[2]) #ZE; FI + ONE_VALUE_TOKEN: + IF (imm8[3]) #IE; FI + SNAN_TOKEN: + IF (imm8[4]) #IE; FI + NEG_INF_TOKEN: + IF (imm8[5]) #IE; FI + NEG_VALUE_TOKEN: + IF (imm8[6]) #IE; FI + POS_INF_TOKEN: + IF (imm8[7]) #IE; FI + ESAC + RETURN dest[63:0] } - -dst[511:0] := src[511:0] -loadOffset := 0 -foundNext64BytesBoundary := false -upSize := UPCONVERTSIZE(conv) -addr = mt-64 -FOR j := 0 to 15 - IF foundNext64BytesBoundary == false - IF (addr + (loadOffset + 1)*upSize % 64) == 0 - foundNext64BytesBoundary := true - FI - ELSE - i := j*32 - dst[i+31:i] := UPCONVERT(addr, loadOffset, conv) - FI - loadOffset := loadOffset + 1 +FOR j := 0 to 7 + i := j*64 + dst[i+63:i] := FIXUPIMMPD(a[i+63:i], b[i+63:i], c[i+63:i], imm8[7:0]) ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - KNCNI - Load - - - - - - Loads the high-64-byte-aligned portion of the byte/word/doubleword stream starting at element-aligned address mt-64, up-converted depending on the value of "conv", and expanded into packed single-precision (32-bit) floating-point elements in "dst". The initial values of "dst" are copied from "src". Only those converted quadwords that occur at or after the first 64-byte-aligned address following (mt-64) are loaded. Elements in the resulting vector that do not map to those quadwords are taken from "src". "hint" indicates to the processor whether the loaded data is non-temporal. Elements are copied to "dst" according to element selector "k" (elements are skipped when the corresponding mask bit is not set). - -UPCONVERT(address, offset, convertTo) { - CASE conv OF - _MM_UPCONV_PS_NONE: RETURN MEM[addr + 4*offset] - _MM_UPCONV_PS_FLOAT16: RETURN Float16ToFloat32(MEM[addr + 4*offset]) - _MM_UPCONV_PS_UINT8: RETURN UInt8ToFloat32(MEM[addr + offset]) - _MM_UPCONV_PS_SINT8: RETURN SInt8ToFloat32(MEM[addr + offset]) - _MM_UPCONV_PS_UINT16: RETURN UInt16ToFloat32(MEM[addr + 2*offset]) - _MM_UPCONV_PS_SINT16: RETURN SInt16ToFloat32(MEM[addr + 2*offset]) - ESAC + AVX512F + Miscellaneous + + + + + + + Fix up packed double-precision (64-bit) floating-point elements in "a" and "b" using packed 64-bit integers in "c", and store the results in "dst". "imm8" is used to set the required flags reporting. + [sae_note] + enum TOKEN_TYPE { + QNAN_TOKEN := 0, \ + SNAN_TOKEN := 1, \ + ZERO_VALUE_TOKEN := 2, \ + ONE_VALUE_TOKEN := 3, \ + NEG_INF_TOKEN := 4, \ + POS_INF_TOKEN := 5, \ + NEG_VALUE_TOKEN := 6, \ + POS_VALUE_TOKEN := 7 } - -UPCONVERTSIZE(convertTo) { - CASE conv OF - _MM_UPCONV_PS_NONE: RETURN 4 - _MM_UPCONV_PS_FLOAT16: RETURN 2 - _MM_UPCONV_PS_UINT8: RETURN 1 - _MM_UPCONV_PS_SINT8: RETURN 1 - _MM_UPCONV_PS_UINT16: RETURN 2 - _MM_UPCONV_PS_SINT16: RETURN 2 +DEFINE FIXUPIMMPD(src1[63:0], src2[63:0], src3[63:0], imm8[7:0]) { + tsrc[63:0] := ((src2[62:52] == 0) AND (MXCSR.DAZ == 1)) ? 0.0 : src2[63:0] + CASE(tsrc[63:0]) OF + QNAN_TOKEN:j := 0 + SNAN_TOKEN:j := 1 + ZERO_VALUE_TOKEN: j := 2 + ONE_VALUE_TOKEN: j := 3 + NEG_INF_TOKEN: j := 4 + POS_INF_TOKEN: j := 5 + NEG_VALUE_TOKEN: j := 6 + POS_VALUE_TOKEN: j := 7 + ESAC + + token_response[3:0] := src3[3+4*j:4*j] + + CASE(token_response[3:0]) OF + 0 : dest[63:0] := src1[63:0] + 1 : dest[63:0] := tsrc[63:0] + 2 : dest[63:0] := QNaN(tsrc[63:0]) + 3 : dest[63:0] := QNAN_Indefinite + 4 : dest[63:0] := -INF + 5 : dest[63:0] := +INF + 6 : dest[63:0] := tsrc.sign? -INF : +INF + 7 : dest[63:0] := -0 + 8 : dest[63:0] := +0 + 9 : dest[63:0] := -1 + 10: dest[63:0] := +1 + 11: dest[63:0] := 1/2 + 12: dest[63:0] := 90.0 + 13: dest[63:0] := PI/2 + 14: dest[63:0] := MAX_FLOAT + 15: dest[63:0] := -MAX_FLOAT + ESAC + + CASE(tsrc[31:0]) OF + ZERO_VALUE_TOKEN: + IF (imm8[0]) #ZE; FI + ZERO_VALUE_TOKEN: + IF (imm8[1]) #IE; FI + ONE_VALUE_TOKEN: + IF (imm8[2]) #ZE; FI + ONE_VALUE_TOKEN: + IF (imm8[3]) #IE; FI + SNAN_TOKEN: + IF (imm8[4]) #IE; FI + NEG_INF_TOKEN: + IF (imm8[5]) #IE; FI + NEG_VALUE_TOKEN: + IF (imm8[6]) #IE; FI + POS_INF_TOKEN: + IF (imm8[7]) #IE; FI ESAC + RETURN dest[63:0] } - -dst[511:0] := src[511:0] -loadOffset := 0 -foundNext64BytesBoundary := false -upSize := UPCONVERTSIZE(conv) -addr = mt-64 -FOR j := 0 to 15 - IF k[j] - IF foundNext64BytesBoundary == false - IF (addr + (loadOffset + 1)*upSize % 64) == 0 - foundNext64BytesBoundary := true - FI - ELSE - i := j*32 - dst[i+31:i] := UPCONVERT(addr, loadOffset, conv) - FI - loadOffset := loadOffset + 1 - FI +FOR j := 0 to 7 + i := j*64 + dst[i+63:i] := FIXUPIMMPD(a[i+63:i], b[i+63:i], c[i+63:i], imm8[7:0]) ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - KNCNI - Load - - - - - Loads the low-64-byte-aligned portion of the byte/word/doubleword stream starting at element-aligned address mt, up-converted depending on the value of "conv", and expanded into packed single-precision (32-bit) floating-point elements in "dst". The initial values of "dst" are copied from "src". Only those converted doublewords that occur before first 64-byte-aligned address following "mt" are loaded. Elements in the resulting vector that do not map to those doublewords are taken from "src". "hint" indicates to the processor whether the loaded data is non-temporal. - -UPCONVERT(address, offset, convertTo) { - CASE conv OF - _MM_UPCONV_PS_NONE: RETURN MEM[addr + 4*offset] - _MM_UPCONV_PS_FLOAT16: RETURN Float16ToFloat32(MEM[addr + 4*offset]) - _MM_UPCONV_PS_UINT8: RETURN UInt8ToFloat32(MEM[addr + offset]) - _MM_UPCONV_PS_SINT8: RETURN SInt8ToFloat32(MEM[addr + offset]) - _MM_UPCONV_PS_UINT16: RETURN UInt16ToFloat32(MEM[addr + 2*offset]) - _MM_UPCONV_PS_SINT16: RETURN SInt16ToFloat32(MEM[addr + 2*offset]) - ESAC + AVX512F + Miscellaneous + + + + + + + Fix up packed double-precision (64-bit) floating-point elements in "a" and "b" using packed 64-bit integers in "c", and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). "imm8" is used to set the required flags reporting. + enum TOKEN_TYPE { + QNAN_TOKEN := 0, \ + SNAN_TOKEN := 1, \ + ZERO_VALUE_TOKEN := 2, \ + ONE_VALUE_TOKEN := 3, \ + NEG_INF_TOKEN := 4, \ + POS_INF_TOKEN := 5, \ + NEG_VALUE_TOKEN := 6, \ + POS_VALUE_TOKEN := 7 } - -UPCONVERTSIZE(convertTo) { - CASE conv OF - _MM_UPCONV_PS_NONE: RETURN 4 - _MM_UPCONV_PS_FLOAT16: RETURN 2 - _MM_UPCONV_PS_UINT8: RETURN 1 - _MM_UPCONV_PS_SINT8: RETURN 1 - _MM_UPCONV_PS_UINT16: RETURN 2 - _MM_UPCONV_PS_SINT16: RETURN 2 +DEFINE FIXUPIMMPD(src1[63:0], src2[63:0], src3[63:0], imm8[7:0]) { + tsrc[63:0] := ((src2[62:52] == 0) AND (MXCSR.DAZ == 1)) ? 0.0 : src2[63:0] + CASE(tsrc[63:0]) OF + QNAN_TOKEN:j := 0 + SNAN_TOKEN:j := 1 + ZERO_VALUE_TOKEN: j := 2 + ONE_VALUE_TOKEN: j := 3 + NEG_INF_TOKEN: j := 4 + POS_INF_TOKEN: j := 5 + NEG_VALUE_TOKEN: j := 6 + POS_VALUE_TOKEN: j := 7 + ESAC + + token_response[3:0] := src3[3+4*j:4*j] + + CASE(token_response[3:0]) OF + 0 : dest[63:0] := src1[63:0] + 1 : dest[63:0] := tsrc[63:0] + 2 : dest[63:0] := QNaN(tsrc[63:0]) + 3 : dest[63:0] := QNAN_Indefinite + 4 : dest[63:0] := -INF + 5 : dest[63:0] := +INF + 6 : dest[63:0] := tsrc.sign? -INF : +INF + 7 : dest[63:0] := -0 + 8 : dest[63:0] := +0 + 9 : dest[63:0] := -1 + 10: dest[63:0] := +1 + 11: dest[63:0] := 1/2 + 12: dest[63:0] := 90.0 + 13: dest[63:0] := PI/2 + 14: dest[63:0] := MAX_FLOAT + 15: dest[63:0] := -MAX_FLOAT + ESAC + + CASE(tsrc[31:0]) OF + ZERO_VALUE_TOKEN: + IF (imm8[0]) #ZE; FI + ZERO_VALUE_TOKEN: + IF (imm8[1]) #IE; FI + ONE_VALUE_TOKEN: + IF (imm8[2]) #ZE; FI + ONE_VALUE_TOKEN: + IF (imm8[3]) #IE; FI + SNAN_TOKEN: + IF (imm8[4]) #IE; FI + NEG_INF_TOKEN: + IF (imm8[5]) #IE; FI + NEG_VALUE_TOKEN: + IF (imm8[6]) #IE; FI + POS_INF_TOKEN: + IF (imm8[7]) #IE; FI ESAC + RETURN dest[63:0] } - -dst[511:0] := src[511:0] -loadOffset := 0 -upSize := UPCONVERTSIZE(conv) -addr = MEM[mt] -FOR j := 0 to 15 - i := j*32 - dst[i+31:i] := UPCONVERT(addr, loadOffset, conv) - loadOffset := loadOffset + 1 - IF (mt + loadOffset * upSize) % 64 == 0 - break +FOR j := 0 to 7 + i := j*64 + IF k[j] + dst[i+63:i] := FIXUPIMMPD(a[i+63:i], b[i+63:i], c[i+63:i], imm8[7:0]) + ELSE + dst[i+63:i] := a[i+63:i] FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - KNCNI - Load - - - - - - Loads the low-64-byte-aligned portion of the byte/word/doubleword stream starting at element-aligned address mt, up-converted depending on the value of "conv", and expanded into packed single-precision (32-bit) floating-point elements in "dst". The initial values of "dst" are copied from "src". Only those converted doublewords that occur before first 64-byte-aligned address following "mt" are loaded. Elements in the resulting vector that do not map to those doublewords are taken from "src". "hint" indicates to the processor whether the loaded data is non-temporal. Elements are copied to "dst" according to element selector "k" (elements are skipped when the corresponding mask bit is not set). - -UPCONVERT(address, offset, convertTo) { - CASE conv OF - _MM_UPCONV_PS_NONE: RETURN MEM[addr + 4*offset] - _MM_UPCONV_PS_FLOAT16: RETURN Float16ToFloat32(MEM[addr + 4*offset]) - _MM_UPCONV_PS_UINT8: RETURN UInt8ToFloat32(MEM[addr + offset]) - _MM_UPCONV_PS_SINT8: RETURN SInt8ToFloat32(MEM[addr + offset]) - _MM_UPCONV_PS_UINT16: RETURN UInt16ToFloat32(MEM[addr + 2*offset]) - _MM_UPCONV_PS_SINT16: RETURN SInt16ToFloat32(MEM[addr + 2*offset]) - ESAC + AVX512F + Miscellaneous + + + + + + + + Fix up packed double-precision (64-bit) floating-point elements in "a" and "b" using packed 64-bit integers in "c", and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). "imm8" is used to set the required flags reporting. + [sae_note] + enum TOKEN_TYPE { + QNAN_TOKEN := 0, \ + SNAN_TOKEN := 1, \ + ZERO_VALUE_TOKEN := 2, \ + ONE_VALUE_TOKEN := 3, \ + NEG_INF_TOKEN := 4, \ + POS_INF_TOKEN := 5, \ + NEG_VALUE_TOKEN := 6, \ + POS_VALUE_TOKEN := 7 } - -UPCONVERTSIZE(convertTo) { - CASE conv OF - _MM_UPCONV_PS_NONE: RETURN 4 - _MM_UPCONV_PS_FLOAT16: RETURN 2 - _MM_UPCONV_PS_UINT8: RETURN 1 - _MM_UPCONV_PS_SINT8: RETURN 1 - _MM_UPCONV_PS_UINT16: RETURN 2 - _MM_UPCONV_PS_SINT16: RETURN 2 +DEFINE FIXUPIMMPD(src1[63:0], src2[63:0], src3[63:0], imm8[7:0]) { + tsrc[63:0] := ((src2[62:52] == 0) AND (MXCSR.DAZ == 1)) ? 0.0 : src2[63:0] + CASE(tsrc[63:0]) OF + QNAN_TOKEN:j := 0 + SNAN_TOKEN:j := 1 + ZERO_VALUE_TOKEN: j := 2 + ONE_VALUE_TOKEN: j := 3 + NEG_INF_TOKEN: j := 4 + POS_INF_TOKEN: j := 5 + NEG_VALUE_TOKEN: j := 6 + POS_VALUE_TOKEN: j := 7 + ESAC + + token_response[3:0] := src3[3+4*j:4*j] + + CASE(token_response[3:0]) OF + 0 : dest[63:0] := src1[63:0] + 1 : dest[63:0] := tsrc[63:0] + 2 : dest[63:0] := QNaN(tsrc[63:0]) + 3 : dest[63:0] := QNAN_Indefinite + 4 : dest[63:0] := -INF + 5 : dest[63:0] := +INF + 6 : dest[63:0] := tsrc.sign? -INF : +INF + 7 : dest[63:0] := -0 + 8 : dest[63:0] := +0 + 9 : dest[63:0] := -1 + 10: dest[63:0] := +1 + 11: dest[63:0] := 1/2 + 12: dest[63:0] := 90.0 + 13: dest[63:0] := PI/2 + 14: dest[63:0] := MAX_FLOAT + 15: dest[63:0] := -MAX_FLOAT ESAC + + CASE(tsrc[31:0]) OF + ZERO_VALUE_TOKEN: + IF (imm8[0]) #ZE; FI + ZERO_VALUE_TOKEN: + IF (imm8[1]) #IE; FI + ONE_VALUE_TOKEN: + IF (imm8[2]) #ZE; FI + ONE_VALUE_TOKEN: + IF (imm8[3]) #IE; FI + SNAN_TOKEN: + IF (imm8[4]) #IE; FI + NEG_INF_TOKEN: + IF (imm8[5]) #IE; FI + NEG_VALUE_TOKEN: + IF (imm8[6]) #IE; FI + POS_INF_TOKEN: + IF (imm8[7]) #IE; FI + ESAC + RETURN dest[63:0] } - -dst[511:0] := src[511:0] -loadOffset := 0 -upSize := UPCONVERTSIZE(conv) -addr = MEM[mt] -FOR j := 0 to 15 +FOR j := 0 to 7 + i := j*64 IF k[j] - i := j*32 - dst[i+31:i] := UPCONVERT(addr, loadOffset, conv) - loadOffset := loadOffset + 1 - IF (mt + loadOffset * upSize) % 64 == 0 - break - FI + dst[i+63:i] := FIXUPIMMPD(a[i+63:i], b[i+63:i], c[i+63:i], imm8[7:0]) + ELSE + dst[i+63:i] := a[i+63:i] FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - KNCNI - Load - - - - - Loads the high-64-byte-aligned portion of the quadword stream starting at element-aligned address mt-64, up-converted depending on the value of "conv", and expanded into packed double-precision (64-bit) floating-point values in "dst". The initial values of "dst" are copied from "src". Only those converted quadwords that occur at or after the first 64-byte-aligned address following (mt-64) are loaded. Elements in the resulting vector that do not map to those quadwords are taken from "src". "hint" indicates to the processor whether the loaded data is non-temporal. - -UPCONVERT(address, offset, convertTo) { - CASE conv OF - _MM_UPCONV_PD_NONE: RETURN MEM[addr + 8*offset] - ESAC + AVX512F + Miscellaneous + + + + + + + Fix up packed double-precision (64-bit) floating-point elements in "a" and "b" using packed 64-bit integers in "c", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). "imm8" is used to set the required flags reporting. + enum TOKEN_TYPE { + QNAN_TOKEN := 0, \ + SNAN_TOKEN := 1, \ + ZERO_VALUE_TOKEN := 2, \ + ONE_VALUE_TOKEN := 3, \ + NEG_INF_TOKEN := 4, \ + POS_INF_TOKEN := 5, \ + NEG_VALUE_TOKEN := 6, \ + POS_VALUE_TOKEN := 7 } - -UPCONVERTSIZE(convertTo) { - CASE conv OF - _MM_UPCONV_PD_NONE: RETURN 8 +DEFINE FIXUPIMMPD(src1[63:0], src2[63:0], src3[63:0], imm8[7:0]) { + tsrc[63:0] := ((src2[62:52] == 0) AND (MXCSR.DAZ == 1)) ? 0.0 : src2[63:0] + CASE(tsrc[63:0]) OF + QNAN_TOKEN:j := 0 + SNAN_TOKEN:j := 1 + ZERO_VALUE_TOKEN: j := 2 + ONE_VALUE_TOKEN: j := 3 + NEG_INF_TOKEN: j := 4 + POS_INF_TOKEN: j := 5 + NEG_VALUE_TOKEN: j := 6 + POS_VALUE_TOKEN: j := 7 + ESAC + + token_response[3:0] := src3[3+4*j:4*j] + + CASE(token_response[3:0]) OF + 0 : dest[63:0] := src1[63:0] + 1 : dest[63:0] := tsrc[63:0] + 2 : dest[63:0] := QNaN(tsrc[63:0]) + 3 : dest[63:0] := QNAN_Indefinite + 4 : dest[63:0] := -INF + 5 : dest[63:0] := +INF + 6 : dest[63:0] := tsrc.sign? -INF : +INF + 7 : dest[63:0] := -0 + 8 : dest[63:0] := +0 + 9 : dest[63:0] := -1 + 10: dest[63:0] := +1 + 11: dest[63:0] := 1/2 + 12: dest[63:0] := 90.0 + 13: dest[63:0] := PI/2 + 14: dest[63:0] := MAX_FLOAT + 15: dest[63:0] := -MAX_FLOAT + ESAC + + CASE(tsrc[31:0]) OF + ZERO_VALUE_TOKEN: + IF (imm8[0]) #ZE; FI + ZERO_VALUE_TOKEN: + IF (imm8[1]) #IE; FI + ONE_VALUE_TOKEN: + IF (imm8[2]) #ZE; FI + ONE_VALUE_TOKEN: + IF (imm8[3]) #IE; FI + SNAN_TOKEN: + IF (imm8[4]) #IE; FI + NEG_INF_TOKEN: + IF (imm8[5]) #IE; FI + NEG_VALUE_TOKEN: + IF (imm8[6]) #IE; FI + POS_INF_TOKEN: + IF (imm8[7]) #IE; FI ESAC + RETURN dest[63:0] } - -dst[511:0] := src[511:0] -loadOffset := 0 -foundNext64BytesBoundary := false -upSize := UPCONVERTSIZE(conv) -addr = mt-64 FOR j := 0 to 7 - IF foundNext64BytesBoundary == false - IF (addr + (loadOffset + 1)*upSize) % 64 == 0 - foundNext64BytesBoundary := true - FI + i := j*64 + IF k[j] + dst[i+63:i] := FIXUPIMMPD(a[i+63:i], b[i+63:i], c[i+63:i], imm8[7:0]) ELSE - i := j*64 - dst[i+63:i] := UPCONVERT(addr, loadOffset, conv) + dst[i+63:i] := 0 FI - loadOffset := loadOffset + 1 ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - KNCNI - Load - - - - - - Loads the high-64-byte-aligned portion of the quadword stream starting at element-aligned address mt-64, up-converted depending on the value of "conv", and expanded into packed double-precision (64-bit) floating-point values in "dst". The initial values of "dst" are copied from "src". Only those converted quadwords that occur at or after the first 64-byte-aligned address following (mt-64) are loaded. Elements in the resulting vector that do not map to those quadwords are taken from "src". "hint" indicates to the processor whether the loaded data is non-temporal. Elements are copied to "dst" according to element selector "k" (elements are skipped when the corresponding mask bit is not set). - -UPCONVERT(address, offset, convertTo) { - CASE conv OF - _MM_UPCONV_PD_NONE: RETURN MEM[addr + 8*offset] - ESAC + AVX512F + Miscellaneous + + + + + + + + Fix up packed double-precision (64-bit) floating-point elements in "a" and "b" using packed 64-bit integers in "c", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). "imm8" is used to set the required flags reporting. + [sae_note] + enum TOKEN_TYPE { + QNAN_TOKEN := 0, \ + SNAN_TOKEN := 1, \ + ZERO_VALUE_TOKEN := 2, \ + ONE_VALUE_TOKEN := 3, \ + NEG_INF_TOKEN := 4, \ + POS_INF_TOKEN := 5, \ + NEG_VALUE_TOKEN := 6, \ + POS_VALUE_TOKEN := 7 } - -UPCONVERTSIZE(convertTo) { - CASE conv OF - _MM_UPCONV_PD_NONE: RETURN 8 +DEFINE FIXUPIMMPD(src1[63:0], src2[63:0], src3[63:0], imm8[7:0]) { + tsrc[63:0] := ((src2[62:52] == 0) AND (MXCSR.DAZ == 1)) ? 0.0 : src2[63:0] + CASE(tsrc[63:0]) OF + QNAN_TOKEN:j := 0 + SNAN_TOKEN:j := 1 + ZERO_VALUE_TOKEN: j := 2 + ONE_VALUE_TOKEN: j := 3 + NEG_INF_TOKEN: j := 4 + POS_INF_TOKEN: j := 5 + NEG_VALUE_TOKEN: j := 6 + POS_VALUE_TOKEN: j := 7 + ESAC + + token_response[3:0] := src3[3+4*j:4*j] + + CASE(token_response[3:0]) OF + 0 : dest[63:0] := src1[63:0] + 1 : dest[63:0] := tsrc[63:0] + 2 : dest[63:0] := QNaN(tsrc[63:0]) + 3 : dest[63:0] := QNAN_Indefinite + 4 : dest[63:0] := -INF + 5 : dest[63:0] := +INF + 6 : dest[63:0] := tsrc.sign? -INF : +INF + 7 : dest[63:0] := -0 + 8 : dest[63:0] := +0 + 9 : dest[63:0] := -1 + 10: dest[63:0] := +1 + 11: dest[63:0] := 1/2 + 12: dest[63:0] := 90.0 + 13: dest[63:0] := PI/2 + 14: dest[63:0] := MAX_FLOAT + 15: dest[63:0] := -MAX_FLOAT + ESAC + + CASE(tsrc[31:0]) OF + ZERO_VALUE_TOKEN: + IF (imm8[0]) #ZE; FI + ZERO_VALUE_TOKEN: + IF (imm8[1]) #IE; FI + ONE_VALUE_TOKEN: + IF (imm8[2]) #ZE; FI + ONE_VALUE_TOKEN: + IF (imm8[3]) #IE; FI + SNAN_TOKEN: + IF (imm8[4]) #IE; FI + NEG_INF_TOKEN: + IF (imm8[5]) #IE; FI + NEG_VALUE_TOKEN: + IF (imm8[6]) #IE; FI + POS_INF_TOKEN: + IF (imm8[7]) #IE; FI ESAC + RETURN dest[63:0] } - -dst[511:0] := src[511:0] -loadOffset := 0 -foundNext64BytesBoundary := false -upSize := UPCONVERTSIZE(conv) -addr = mt-64 FOR j := 0 to 7 + i := j*64 IF k[j] - IF foundNext64BytesBoundary == false - IF (addr + (loadOffset + 1)*upSize) % 64 == 0 - foundNext64BytesBoundary := true - FI - ELSE - i := j*64 - dst[i+63:i] := UPCONVERT(addr, loadOffset, conv) - FI - loadOffset := loadOffset + 1 + dst[i+63:i] := FIXUPIMMPD(a[i+63:i], b[i+63:i], c[i+63:i], imm8[7:0]) + ELSE + dst[i+63:i] := 0 FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - KNCNI - Load - - - - - Loads the low-64-byte-aligned portion of the quadword stream starting at element-aligned address mt, up-converted depending on the value of "conv", and expanded into packed double-precision (64-bit) floating-point elements in "dst". The initial values of "dst" are copied from "src". Only those converted quad that occur before first 64-byte-aligned address following "mt" are loaded. Elements in the resulting vector that do not map to those quadwords are taken from "src". "hint" indicates to the processor whether the loaded data is non-temporal. - -UPCONVERT(address, offset, convertTo) { - CASE conv OF - _MM_UPCONV_PD_NONE: RETURN MEM[addr + 8*offset] - ESAC + AVX512F + Miscellaneous + + + + + + Fix up packed single-precision (32-bit) floating-point elements in "a" and "b" using packed 32-bit integers in "c", and store the results in "dst". "imm8" is used to set the required flags reporting. + enum TOKEN_TYPE { + QNAN_TOKEN := 0, \ + SNAN_TOKEN := 1, \ + ZERO_VALUE_TOKEN := 2, \ + ONE_VALUE_TOKEN := 3, \ + NEG_INF_TOKEN := 4, \ + POS_INF_TOKEN := 5, \ + NEG_VALUE_TOKEN := 6, \ + POS_VALUE_TOKEN := 7 } - -UPCONVERTSIZE(convertTo) { - CASE conv OF - _MM_UPCONV_PD_NONE: RETURN 8 +DEFINE FIXUPIMMPD(src1[31:0], src2[31:0], src3[31:0], imm8[7:0]) { + tsrc[31:0] := ((src2[30:23] == 0) AND (MXCSR.DAZ == 1)) ? 0.0 : src2[31:0] + CASE(tsrc[31:0]) OF + QNAN_TOKEN:j := 0 + SNAN_TOKEN:j := 1 + ZERO_VALUE_TOKEN: j := 2 + ONE_VALUE_TOKEN: j := 3 + NEG_INF_TOKEN: j := 4 + POS_INF_TOKEN: j := 5 + NEG_VALUE_TOKEN: j := 6 + POS_VALUE_TOKEN: j := 7 + ESAC + + token_response[3:0] := src3[3+4*j:4*j] + + CASE(token_response[3:0]) OF + 0 : dest[31:0] := src1[31:0] + 1 : dest[31:0] := tsrc[31:0] + 2 : dest[31:0] := QNaN(tsrc[31:0]) + 3 : dest[31:0] := QNAN_Indefinite + 4 : dest[31:0] := -INF + 5 : dest[31:0] := +INF + 6 : dest[31:0] := tsrc.sign? -INF : +INF + 7 : dest[31:0] := -0 + 8 : dest[31:0] := +0 + 9 : dest[31:0] := -1 + 10: dest[31:0] := +1 + 11: dest[31:0] := 1/2 + 12: dest[31:0] := 90.0 + 13: dest[31:0] := PI/2 + 14: dest[31:0] := MAX_FLOAT + 15: dest[31:0] := -MAX_FLOAT + ESAC + + CASE(tsrc[31:0]) OF + ZERO_VALUE_TOKEN: + IF (imm8[0]) #ZE; FI + ZERO_VALUE_TOKEN: + IF (imm8[1]) #IE; FI + ONE_VALUE_TOKEN: + IF (imm8[2]) #ZE; FI + ONE_VALUE_TOKEN: + IF (imm8[3]) #IE; FI + SNAN_TOKEN: + IF (imm8[4]) #IE; FI + NEG_INF_TOKEN: + IF (imm8[5]) #IE; FI + NEG_VALUE_TOKEN: + IF (imm8[6]) #IE; FI + POS_INF_TOKEN: + IF (imm8[7]) #IE; FI ESAC + RETURN dest[31:0] } - -dst[511:0] := src[511:0] -loadOffset := 0 -upSize := UPCONVERTSIZE(conv) -addr = mt -FOR j := 0 to 7 - i := j*64 - dst[i+63:i] := UPCONVERT(addr, loadOffset, conv) - loadOffset := loadOffset + 1 - IF (mt + loadOffset * upSize) % 64 == 0 - break - FI +FOR j := 0 to 15 + i := j*32 + dst[i+31:i] := FIXUPIMMPD(a[i+31:i], b[i+31:i], c[i+31:i], imm8[7:0]) ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - KNCNI - Load - - - - - - Loads the low-64-byte-aligned portion of the quadword stream starting at element-aligned address mt, up-converted depending on the value of "conv", and expanded into packed double-precision (64-bit) floating-point elements in "dst". The initial values of "dst" are copied from "src". Only those converted quad that occur before first 64-byte-aligned address following "mt" are loaded. Elements in the resulting vector that do not map to those quadwords are taken from "src". "hint" indicates to the processor whether the loaded data is non-temporal. Elements are copied to "dst" according to element selector "k" (elemenst are skipped when the corresponding mask bit is not set). - -UPCONVERT(address, offset, convertTo) { - CASE conv OF - _MM_UPCONV_PD_NONE: RETURN MEM[addr + 8*offset] - ESAC + AVX512F + Miscellaneous + + + + + + + Fix up packed single-precision (32-bit) floating-point elements in "a" and "b" using packed 32-bit integers in "c", and store the results in "dst". "imm8" is used to set the required flags reporting. + [sae_note] + enum TOKEN_TYPE { + QNAN_TOKEN := 0, \ + SNAN_TOKEN := 1, \ + ZERO_VALUE_TOKEN := 2, \ + ONE_VALUE_TOKEN := 3, \ + NEG_INF_TOKEN := 4, \ + POS_INF_TOKEN := 5, \ + NEG_VALUE_TOKEN := 6, \ + POS_VALUE_TOKEN := 7 } - -UPCONVERTSIZE(convertTo) { - CASE conv OF - _MM_UPCONV_PD_NONE: RETURN 8 +DEFINE FIXUPIMMPD(src1[31:0], src2[31:0], src3[31:0], imm8[7:0]) { + tsrc[31:0] := ((src2[30:23] == 0) AND (MXCSR.DAZ == 1)) ? 0.0 : src2[31:0] + CASE(tsrc[31:0]) OF + QNAN_TOKEN:j := 0 + SNAN_TOKEN:j := 1 + ZERO_VALUE_TOKEN: j := 2 + ONE_VALUE_TOKEN: j := 3 + NEG_INF_TOKEN: j := 4 + POS_INF_TOKEN: j := 5 + NEG_VALUE_TOKEN: j := 6 + POS_VALUE_TOKEN: j := 7 + ESAC + + token_response[3:0] := src3[3+4*j:4*j] + + CASE(token_response[3:0]) OF + 0 : dest[31:0] := src1[31:0] + 1 : dest[31:0] := tsrc[31:0] + 2 : dest[31:0] := QNaN(tsrc[31:0]) + 3 : dest[31:0] := QNAN_Indefinite + 4 : dest[31:0] := -INF + 5 : dest[31:0] := +INF + 6 : dest[31:0] := tsrc.sign? -INF : +INF + 7 : dest[31:0] := -0 + 8 : dest[31:0] := +0 + 9 : dest[31:0] := -1 + 10: dest[31:0] := +1 + 11: dest[31:0] := 1/2 + 12: dest[31:0] := 90.0 + 13: dest[31:0] := PI/2 + 14: dest[31:0] := MAX_FLOAT + 15: dest[31:0] := -MAX_FLOAT + ESAC + + CASE(tsrc[31:0]) OF + ZERO_VALUE_TOKEN: + IF (imm8[0]) #ZE; FI + ZERO_VALUE_TOKEN: + IF (imm8[1]) #IE; FI + ONE_VALUE_TOKEN: + IF (imm8[2]) #ZE; FI + ONE_VALUE_TOKEN: + IF (imm8[3]) #IE; FI + SNAN_TOKEN: + IF (imm8[4]) #IE; FI + NEG_INF_TOKEN: + IF (imm8[5]) #IE; FI + NEG_VALUE_TOKEN: + IF (imm8[6]) #IE; FI + POS_INF_TOKEN: + IF (imm8[7]) #IE; FI ESAC + RETURN dest[31:0] } - -dst[511:0] := src[511:0] -loadOffset := 0 -upSize := UPCONVERTSIZE(conv) -addr = mt -FOR j := 0 to 7 - IF k[j] - i := j*64 - dst[i+63:i] := UPCONVERT(addr, loadOffset, conv) - loadOffset := loadOffset + 1 - IF (mt + loadOffset * upSize) % 64 == 0 - break - FI - FI +FOR j := 0 to 15 + i := j*32 + dst[i+31:i] := FIXUPIMMPD(a[i+31:i], b[i+31:i], c[i+31:i], imm8[7:0]) ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - KNCNI - Store - - - - - Down-converts and stores packed 32-bit integer elements of "v1" into a byte/word/doubleword stream according to "conv" at a logically mapped starting address (mt-64), storing the high-64-byte elements of that stream (those elemetns of the stream that map at or after the first 64-byte-aligned address following (m5-64)). "hint" indicates to the processor whether the data is non-temporal. - -DOWNCONVERT(element, convertTo) { - CASE converTo OF - _MM_DOWNCONV_EPI32_NONE: RETURN element[i+31:i] - _MM_DOWNCONV_EPI32_UINT8: RETURN UInt32ToUInt8(element[i+31:i]) - _MM_DOWNCONV_EPI32_SINT8: RETURN SInt32ToSInt8(element[i+31:i]) - _MM_DOWNCONV_EPI32_UINT16: RETURN UInt32ToUInt16(element[i+31:i]) - _MM_DOWNCONV_EPI32_SINT16: RETURN SInt32ToSInt16(element[i+31:i]) + + Floating Point + AVX512F + Miscellaneous + + + + + + + Fix up packed single-precision (32-bit) floating-point elements in "a" and "b" using packed 32-bit integers in "c", and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). "imm8" is used to set the required flags reporting. + enum TOKEN_TYPE { + QNAN_TOKEN := 0, \ + SNAN_TOKEN := 1, \ + ZERO_VALUE_TOKEN := 2, \ + ONE_VALUE_TOKEN := 3, \ + NEG_INF_TOKEN := 4, \ + POS_INF_TOKEN := 5, \ + NEG_VALUE_TOKEN := 6, \ + POS_VALUE_TOKEN := 7 +} +DEFINE FIXUPIMMPD(src1[31:0], src2[31:0], src3[31:0], imm8[7:0]) { + tsrc[31:0] := ((src2[30:23] == 0) AND (MXCSR.DAZ == 1)) ? 0.0 : src2[31:0] + CASE(tsrc[31:0]) OF + QNAN_TOKEN:j := 0 + SNAN_TOKEN:j := 1 + ZERO_VALUE_TOKEN: j := 2 + ONE_VALUE_TOKEN: j := 3 + NEG_INF_TOKEN: j := 4 + POS_INF_TOKEN: j := 5 + NEG_VALUE_TOKEN: j := 6 + POS_VALUE_TOKEN: j := 7 ESAC -} - -DOWNCONVERTSIZE(convertTo) { - CASE converTo OF - _MM_DOWNCONV_EPI32_NONE: RETURN 4 - _MM_DOWNCONV_EPI32_UINT8: RETURN 1 - _MM_DOWNCONV_EPI32_SINT8: RETURN 1 - _MM_DOWNCONV_EPI32_UINT16: RETURN 2 - _MM_DOWNCONV_EPI32_SINT16: RETURN 2 + + token_response[3:0] := src3[3+4*j:4*j] + + CASE(token_response[3:0]) OF + 0 : dest[31:0] := src1[31:0] + 1 : dest[31:0] := tsrc[31:0] + 2 : dest[31:0] := QNaN(tsrc[31:0]) + 3 : dest[31:0] := QNAN_Indefinite + 4 : dest[31:0] := -INF + 5 : dest[31:0] := +INF + 6 : dest[31:0] := tsrc.sign? -INF : +INF + 7 : dest[31:0] := -0 + 8 : dest[31:0] := +0 + 9 : dest[31:0] := -1 + 10: dest[31:0] := +1 + 11: dest[31:0] := 1/2 + 12: dest[31:0] := 90.0 + 13: dest[31:0] := PI/2 + 14: dest[31:0] := MAX_FLOAT + 15: dest[31:0] := -MAX_FLOAT + ESAC + + CASE(tsrc[31:0]) OF + ZERO_VALUE_TOKEN: + IF (imm8[0]) #ZE; FI + ZERO_VALUE_TOKEN: + IF (imm8[1]) #IE; FI + ONE_VALUE_TOKEN: + IF (imm8[2]) #ZE; FI + ONE_VALUE_TOKEN: + IF (imm8[3]) #IE; FI + SNAN_TOKEN: + IF (imm8[4]) #IE; FI + NEG_INF_TOKEN: + IF (imm8[5]) #IE; FI + NEG_VALUE_TOKEN: + IF (imm8[6]) #IE; FI + POS_INF_TOKEN: + IF (imm8[7]) #IE; FI ESAC + RETURN dest[31:0] } - -storeOffset := 0 -foundNext64BytesBoundary := false -downSize := DOWNCONVERTSIZE(conv) -addr = mt-64 FOR j := 0 to 15 - IF foundNext64BytesBoundary == false - IF ((addr + (storeOffset + 1)*downSize) % 64) == 0 - foundNext64BytesBoundary = true - FI + i := j*32 + IF k[j] + dst[i+31:i] := FIXUPIMMPD(a[i+31:i], b[i+31:i], c[i+31:i], imm8[7:0]) ELSE - i := j*32 - tmp := DOWNCONVERT(v1[i+31:i], conv) - storeAddr := addr + storeOffset * downSize - CASE downSize OF - 4: MEM[storeAddr] := tmp[31:0] - 2: MEM[storeAddr] := tmp[15:0] - 1: MEM[storeAddr] := tmp[7:0] - ESAC + dst[i+31:i] := a[i+31:i] FI - storeOffset := storeOffset + 1 ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - KNCNI - Store - - - - - - Down-converts and stores packed 32-bit integer elements of "v1" into a byte/word/doubleword stream according to "conv" at a logically mapped starting address (mt-64), storing the high-64-byte elements of that stream (those elemetns of the stream that map at or after the first 64-byte-aligned address following (m5-64)). "hint" indicates to the processor whether the data is non-temporal. Elements are stored to memory according to element selector "k" (elements are skipped when the corresonding mask bit is not set). - -DOWNCONVERT(element, convertTo) { - CASE converTo OF - _MM_DOWNCONV_EPI32_NONE: RETURN element[i+31:i] - _MM_DOWNCONV_EPI32_UINT8: RETURN UInt32ToUInt8(element[i+31:i]) - _MM_DOWNCONV_EPI32_SINT8: RETURN SInt32ToSInt8(element[i+31:i]) - _MM_DOWNCONV_EPI32_UINT16: RETURN UInt32ToUInt16(element[i+31:i]) - _MM_DOWNCONV_EPI32_SINT16: RETURN SInt32ToSInt16(element[i+31:i]) - ESAC + + Floating Point + AVX512F + Miscellaneous + + + + + + + + Fix up packed single-precision (32-bit) floating-point elements in "a" and "b" using packed 32-bit integers in "c", and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). "imm8" is used to set the required flags reporting. + [sae_note] + enum TOKEN_TYPE { + QNAN_TOKEN := 0, \ + SNAN_TOKEN := 1, \ + ZERO_VALUE_TOKEN := 2, \ + ONE_VALUE_TOKEN := 3, \ + NEG_INF_TOKEN := 4, \ + POS_INF_TOKEN := 5, \ + NEG_VALUE_TOKEN := 6, \ + POS_VALUE_TOKEN := 7 } - -DOWNCONVERTSIZE(convertTo) { - CASE converTo OF - _MM_DOWNCONV_EPI32_NONE: RETURN 4 - _MM_DOWNCONV_EPI32_UINT8: RETURN 1 - _MM_DOWNCONV_EPI32_SINT8: RETURN 1 - _MM_DOWNCONV_EPI32_UINT16: RETURN 2 - _MM_DOWNCONV_EPI32_SINT16: RETURN 2 +DEFINE FIXUPIMMPD(src1[31:0], src2[31:0], src3[31:0], imm8[7:0]) { + tsrc[31:0] := ((src2[30:23] == 0) AND (MXCSR.DAZ == 1)) ? 0.0 : src2[31:0] + CASE(tsrc[31:0]) OF + QNAN_TOKEN:j := 0 + SNAN_TOKEN:j := 1 + ZERO_VALUE_TOKEN: j := 2 + ONE_VALUE_TOKEN: j := 3 + NEG_INF_TOKEN: j := 4 + POS_INF_TOKEN: j := 5 + NEG_VALUE_TOKEN: j := 6 + POS_VALUE_TOKEN: j := 7 + ESAC + + token_response[3:0] := src3[3+4*j:4*j] + + CASE(token_response[3:0]) OF + 0 : dest[31:0] := src1[31:0] + 1 : dest[31:0] := tsrc[31:0] + 2 : dest[31:0] := QNaN(tsrc[31:0]) + 3 : dest[31:0] := QNAN_Indefinite + 4 : dest[31:0] := -INF + 5 : dest[31:0] := +INF + 6 : dest[31:0] := tsrc.sign? -INF : +INF + 7 : dest[31:0] := -0 + 8 : dest[31:0] := +0 + 9 : dest[31:0] := -1 + 10: dest[31:0] := +1 + 11: dest[31:0] := 1/2 + 12: dest[31:0] := 90.0 + 13: dest[31:0] := PI/2 + 14: dest[31:0] := MAX_FLOAT + 15: dest[31:0] := -MAX_FLOAT + ESAC + + CASE(tsrc[31:0]) OF + ZERO_VALUE_TOKEN: + IF (imm8[0]) #ZE; FI + ZERO_VALUE_TOKEN: + IF (imm8[1]) #IE; FI + ONE_VALUE_TOKEN: + IF (imm8[2]) #ZE; FI + ONE_VALUE_TOKEN: + IF (imm8[3]) #IE; FI + SNAN_TOKEN: + IF (imm8[4]) #IE; FI + NEG_INF_TOKEN: + IF (imm8[5]) #IE; FI + NEG_VALUE_TOKEN: + IF (imm8[6]) #IE; FI + POS_INF_TOKEN: + IF (imm8[7]) #IE; FI ESAC + RETURN dest[31:0] } - -storeOffset := 0 -foundNext64BytesBoundary := false -downSize := DOWNCONVERTSIZE(conv) -addr = mt-64 FOR j := 0 to 15 + i := j*32 IF k[j] - IF foundNext64BytesBoundary == false - IF ((addr + (storeOffset + 1)*downSize) % 64) == 0 - foundNext64BytesBoundary = true - FI - ELSE - i := j*32 - tmp := DOWNCONVERT(v1[i+31:i], conv) - storeAddr := addr + storeOffset * downSize - CASE downSize OF - 4: MEM[storeAddr] := tmp[31:0] - 2: MEM[storeAddr] := tmp[15:0] - 1: MEM[storeAddr] := tmp[7:0] - ESAC - FI - storeOffset := storeOffset + 1 + dst[i+31:i] := FIXUPIMMPD(a[i+31:i], b[i+31:i], c[i+31:i], imm8[7:0]) + ELSE + dst[i+31:i] := a[i+31:i] FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - KNCNI - Store - - - - - Down-converts and stores packed 32-bit integer elements of "v1" into a byte/word/doubleword stream according to "conv" at a logically mapped starting address "mt", storing the low-64-byte elements of that stream (those elements of the stream that map before the first 64-byte-aligned address follwing "mt"). "hint" indicates to the processor whether the data is non-temporal. - -DOWNCONVERT(element, convertTo) { - CASE converTo OF - _MM_DOWNCONV_EPI32_NONE: RETURN element[i+31:i] - _MM_DOWNCONV_EPI32_UINT8: RETURN UInt32ToUInt8(element[i+31:i]) - _MM_DOWNCONV_EPI32_SINT8: RETURN SInt32ToSInt8(element[i+31:i]) - _MM_DOWNCONV_EPI32_UINT16: RETURN UInt32ToUInt16(element[i+31:i]) - _MM_DOWNCONV_EPI32_SINT16: RETURN SInt32ToSInt16(element[i+31:i]) - ESAC + + Floating Point + AVX512F + Miscellaneous + + + + + + + Fix up packed single-precision (32-bit) floating-point elements in "a" and "b" using packed 32-bit integers in "c", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). "imm8" is used to set the required flags reporting. + enum TOKEN_TYPE { + QNAN_TOKEN := 0, \ + SNAN_TOKEN := 1, \ + ZERO_VALUE_TOKEN := 2, \ + ONE_VALUE_TOKEN := 3, \ + NEG_INF_TOKEN := 4, \ + POS_INF_TOKEN := 5, \ + NEG_VALUE_TOKEN := 6, \ + POS_VALUE_TOKEN := 7 } - -DOWNCONVERTSIZE(convertTo) { - CASE converTo OF - _MM_DOWNCONV_EPI32_NONE: RETURN 4 - _MM_DOWNCONV_EPI32_UINT8: RETURN 1 - _MM_DOWNCONV_EPI32_SINT8: RETURN 1 - _MM_DOWNCONV_EPI32_UINT16: RETURN 2 - _MM_DOWNCONV_EPI32_SINT16: RETURN 2 +DEFINE FIXUPIMMPD(src1[31:0], src2[31:0], src3[31:0], imm8[7:0]) { + tsrc[31:0] := ((src2[30:23] == 0) AND (MXCSR.DAZ == 1)) ? 0.0 : src2[31:0] + CASE(tsrc[31:0]) OF + QNAN_TOKEN:j := 0 + SNAN_TOKEN:j := 1 + ZERO_VALUE_TOKEN: j := 2 + ONE_VALUE_TOKEN: j := 3 + NEG_INF_TOKEN: j := 4 + POS_INF_TOKEN: j := 5 + NEG_VALUE_TOKEN: j := 6 + POS_VALUE_TOKEN: j := 7 + ESAC + + token_response[3:0] := src3[3+4*j:4*j] + + CASE(token_response[3:0]) OF + 0 : dest[31:0] := src1[31:0] + 1 : dest[31:0] := tsrc[31:0] + 2 : dest[31:0] := QNaN(tsrc[31:0]) + 3 : dest[31:0] := QNAN_Indefinite + 4 : dest[31:0] := -INF + 5 : dest[31:0] := +INF + 6 : dest[31:0] := tsrc.sign? -INF : +INF + 7 : dest[31:0] := -0 + 8 : dest[31:0] := +0 + 9 : dest[31:0] := -1 + 10: dest[31:0] := +1 + 11: dest[31:0] := 1/2 + 12: dest[31:0] := 90.0 + 13: dest[31:0] := PI/2 + 14: dest[31:0] := MAX_FLOAT + 15: dest[31:0] := -MAX_FLOAT + ESAC + + CASE(tsrc[31:0]) OF + ZERO_VALUE_TOKEN: + IF (imm8[0]) #ZE; FI + ZERO_VALUE_TOKEN: + IF (imm8[1]) #IE; FI + ONE_VALUE_TOKEN: + IF (imm8[2]) #ZE; FI + ONE_VALUE_TOKEN: + IF (imm8[3]) #IE; FI + SNAN_TOKEN: + IF (imm8[4]) #IE; FI + NEG_INF_TOKEN: + IF (imm8[5]) #IE; FI + NEG_VALUE_TOKEN: + IF (imm8[6]) #IE; FI + POS_INF_TOKEN: + IF (imm8[7]) #IE; FI ESAC + RETURN dest[31:0] } - -storeOffset := 0 -downSize := DOWNCONVERTSIZE(conv) -addr = mt FOR j := 0 to 15 i := j*32 - tmp := DOWNCONVERT(v1[i+31:i], conv) - storeAddr := addr + storeOffset * downSize - CASE downSize OF - 4: MEM[storeAddr] := tmp[31:0] - 2: MEM[storeAddr] := tmp[15:0] - 1: MEM[storeAddr] := tmp[7:0] - ESAC - storeOffset := storeOffset + 1 - IF ((addr + storeOffset * downSize) % 64) == 0 - BREAK + IF k[j] + dst[i+31:i] := FIXUPIMMPD(a[i+31:i], b[i+31:i], c[i+31:i], imm8[7:0]) + ELSE + dst[i+31:i] := 0 FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - KNCNI - Store - - - - - - Down-converts and stores packed 32-bit integer elements of "v1" into a byte/word/doubleword stream according to "conv" at a logically mapped starting address "mt", storing the low-64-byte elements of that stream (those elements of the stream that map before the first 64-byte-aligned address follwing "mt"). "hint" indicates to the processor whether the data is non-temporal. Elements are written to memory according to element selector "k" (elements are skipped when the corresponding mask bit is not set). - -DOWNCONVERT(element, convertTo) { - CASE converTo OF - _MM_DOWNCONV_EPI32_NONE: RETURN element[i+31:i] - _MM_DOWNCONV_EPI32_UINT8: RETURN UInt32ToUInt8(element[i+31:i]) - _MM_DOWNCONV_EPI32_SINT8: RETURN SInt32ToSInt8(element[i+31:i]) - _MM_DOWNCONV_EPI32_UINT16: RETURN UInt32ToUInt16(element[i+31:i]) - _MM_DOWNCONV_EPI32_SINT16: RETURN SInt32ToSInt16(element[i+31:i]) - ESAC + + Floating Point + AVX512F + Miscellaneous + + + + + + + + Fix up packed single-precision (32-bit) floating-point elements in "a" and "b" using packed 32-bit integers in "c", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). "imm8" is used to set the required flags reporting. + [sae_note] + enum TOKEN_TYPE { + QNAN_TOKEN := 0, \ + SNAN_TOKEN := 1, \ + ZERO_VALUE_TOKEN := 2, \ + ONE_VALUE_TOKEN := 3, \ + NEG_INF_TOKEN := 4, \ + POS_INF_TOKEN := 5, \ + NEG_VALUE_TOKEN := 6, \ + POS_VALUE_TOKEN := 7 } - -DOWNCONVERTSIZE(convertTo) { - CASE converTo OF - _MM_DOWNCONV_EPI32_NONE: RETURN 4 - _MM_DOWNCONV_EPI32_UINT8: RETURN 1 - _MM_DOWNCONV_EPI32_SINT8: RETURN 1 - _MM_DOWNCONV_EPI32_UINT16: RETURN 2 - _MM_DOWNCONV_EPI32_SINT16: RETURN 2 +DEFINE FIXUPIMMPD(src1[31:0], src2[31:0], src3[31:0], imm8[7:0]) { + tsrc[31:0] := ((src2[30:23] == 0) AND (MXCSR.DAZ == 1)) ? 0.0 : src2[31:0] + CASE(tsrc[31:0]) OF + QNAN_TOKEN:j := 0 + SNAN_TOKEN:j := 1 + ZERO_VALUE_TOKEN: j := 2 + ONE_VALUE_TOKEN: j := 3 + NEG_INF_TOKEN: j := 4 + POS_INF_TOKEN: j := 5 + NEG_VALUE_TOKEN: j := 6 + POS_VALUE_TOKEN: j := 7 + ESAC + + token_response[3:0] := src3[3+4*j:4*j] + + CASE(token_response[3:0]) OF + 0 : dest[31:0] := src1[31:0] + 1 : dest[31:0] := tsrc[31:0] + 2 : dest[31:0] := QNaN(tsrc[31:0]) + 3 : dest[31:0] := QNAN_Indefinite + 4 : dest[31:0] := -INF + 5 : dest[31:0] := +INF + 6 : dest[31:0] := tsrc.sign? -INF : +INF + 7 : dest[31:0] := -0 + 8 : dest[31:0] := +0 + 9 : dest[31:0] := -1 + 10: dest[31:0] := +1 + 11: dest[31:0] := 1/2 + 12: dest[31:0] := 90.0 + 13: dest[31:0] := PI/2 + 14: dest[31:0] := MAX_FLOAT + 15: dest[31:0] := -MAX_FLOAT ESAC + + CASE(tsrc[31:0]) OF + ZERO_VALUE_TOKEN: + IF (imm8[0]) #ZE; FI + ZERO_VALUE_TOKEN: + IF (imm8[1]) #IE; FI + ONE_VALUE_TOKEN: + IF (imm8[2]) #ZE; FI + ONE_VALUE_TOKEN: + IF (imm8[3]) #IE; FI + SNAN_TOKEN: + IF (imm8[4]) #IE; FI + NEG_INF_TOKEN: + IF (imm8[5]) #IE; FI + NEG_VALUE_TOKEN: + IF (imm8[6]) #IE; FI + POS_INF_TOKEN: + IF (imm8[7]) #IE; FI + ESAC + RETURN dest[31:0] } - -storeOffset := 0 -downSize := DOWNCONVERTSIZE(conv) -addr = mt FOR j := 0 to 15 + i := j*32 IF k[j] - i := j*32 - tmp := DOWNCONVERT(v1[i+31:i], conv) - storeAddr := addr + storeOffset * downSize - CASE downSize OF - 4: MEM[storeAddr] := tmp[31:0] - 2: MEM[storeAddr] := tmp[15:0] - 1: MEM[storeAddr] := tmp[7:0] - ESAC - storeOffset := storeOffset + 1 - IF ((addr + storeOffset * downSize) % 64) == 0 - BREAK - FI + dst[i+31:i] := FIXUPIMMPD(a[i+31:i], b[i+31:i], c[i+31:i], imm8[7:0]) + ELSE + dst[i+31:i] := 0 FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - KNCNI - Store - - - - - Down-converts and stores packed 64-bit integer elements of "v1" into a quadword stream according to "conv" at a logically mapped starting address (mt-64), storing the high-64-byte elements of that stream (those elemetns of the stream that map at or after the first 64-byte-aligned address following (m5-64)). "hint" indicates to the processor whether the data is non-temporal. - -DOWNCONVERT(element, convertTo) { - CASE converTo OF - _MM_UPCONV_EPI64_NONE: RETURN element[i+63:i] - ESAC + + Floating Point + AVX512F + Miscellaneous + + + + + + + Fix up the lower double-precision (64-bit) floating-point elements in "a" and "b" using the lower 64-bit integer in "c", store the result in the lower element of "dst", and copy the upper element from "a" to the upper element of "dst". "imm8" is used to set the required flags reporting. + [sae_note] + enum TOKEN_TYPE { + QNAN_TOKEN := 0, \ + SNAN_TOKEN := 1, \ + ZERO_VALUE_TOKEN := 2, \ + ONE_VALUE_TOKEN := 3, \ + NEG_INF_TOKEN := 4, \ + POS_INF_TOKEN := 5, \ + NEG_VALUE_TOKEN := 6, \ + POS_VALUE_TOKEN := 7 } - -DOWNCONVERTSIZE(convertTo) { - CASE converTo OF - _MM_UPCONV_EPI64_NONE: RETURN 8 +DEFINE FIXUPIMMPD(src1[63:0], src2[63:0], src3[63:0], imm8[7:0]) { + tsrc[63:0] := ((src2[62:52] == 0) AND (MXCSR.DAZ == 1)) ? 0.0 : src2[63:0] + CASE(tsrc[63:0]) OF + QNAN_TOKEN:j := 0 + SNAN_TOKEN:j := 1 + ZERO_VALUE_TOKEN: j := 2 + ONE_VALUE_TOKEN: j := 3 + NEG_INF_TOKEN: j := 4 + POS_INF_TOKEN: j := 5 + NEG_VALUE_TOKEN: j := 6 + POS_VALUE_TOKEN: j := 7 + ESAC + + token_response[3:0] := src3[3+4*j:4*j] + + CASE(token_response[3:0]) OF + 0 : dest[63:0] := src1[63:0] + 1 : dest[63:0] := tsrc[63:0] + 2 : dest[63:0] := QNaN(tsrc[63:0]) + 3 : dest[63:0] := QNAN_Indefinite + 4 : dest[63:0] := -INF + 5 : dest[63:0] := +INF + 6 : dest[63:0] := tsrc.sign? -INF : +INF + 7 : dest[63:0] := -0 + 8 : dest[63:0] := +0 + 9 : dest[63:0] := -1 + 10: dest[63:0] := +1 + 11: dest[63:0] := 1/2 + 12: dest[63:0] := 90.0 + 13: dest[63:0] := PI/2 + 14: dest[63:0] := MAX_FLOAT + 15: dest[63:0] := -MAX_FLOAT + ESAC + + CASE(tsrc[31:0]) OF + ZERO_VALUE_TOKEN: + IF (imm8[0]) #ZE; FI + ZERO_VALUE_TOKEN: + IF (imm8[1]) #IE; FI + ONE_VALUE_TOKEN: + IF (imm8[2]) #ZE; FI + ONE_VALUE_TOKEN: + IF (imm8[3]) #IE; FI + SNAN_TOKEN: + IF (imm8[4]) #IE; FI + NEG_INF_TOKEN: + IF (imm8[5]) #IE; FI + NEG_VALUE_TOKEN: + IF (imm8[6]) #IE; FI + POS_INF_TOKEN: + IF (imm8[7]) #IE; FI ESAC + RETURN dest[63:0] } - -storeOffset := 0 -foundNext64BytesBoundary := false -downSize := DOWNCONVERTSIZE(conv) -addr = mt-64 -FOR j := 0 to 7 - IF foundNext64BytesBoundary == false - IF ((addr + (storeOffset + 1)*downSize) % 64) == 0 - foundNext64BytesBoundary = true - FI - ELSE - i := j*64 - tmp := DOWNCONVERT(v1[i+63:i], conv) - storeAddr := addr + storeOffset * downSize - CASE downSize OF - 8: MEM[storeAddr] := tmp[63:0] - ESAC - FI - storeOffset := storeOffset + 1 -ENDFOR -dst[MAX:512] := 0 +dst[63:0] := FIXUPIMMPD(a[63:0], b[63:0], c[63:0], imm8[7:0]) +dst[127:64] := a[127:64] +dst[MAX:128] := 0 - +
immintrin.h
 - - Integer - KNCNI - Store - - - - - - Down-converts and stores packed 64-bit integer elements of "v1" into a quadword stream according to "conv" at a logically mapped starting address (mt-64), storing the high-64-byte elements of that stream (those elemetns of the stream that map at or after the first 64-byte-aligned address following (mt-64)). "hint" indicates to the processor whether the data is non-temporal. Elements are stored to memory according to element selector "k" (elements are skipped when the corresonding mask bit is not set). - -DOWNCONVERT(element, convertTo) { - CASE converTo OF - _MM_UPCONV_EPI64_NONE: RETURN element[i+63:i] - ESAC + + Floating Point + AVX512F + Miscellaneous + + + + + + Fix up the lower double-precision (64-bit) floating-point elements in "a" and "b" using the lower 64-bit integer in "c", store the result in the lower element of "dst", and copy the upper element from "a" to the upper element of "dst". "imm8" is used to set the required flags reporting. + enum TOKEN_TYPE { + QNAN_TOKEN := 0, \ + SNAN_TOKEN := 1, \ + ZERO_VALUE_TOKEN := 2, \ + ONE_VALUE_TOKEN := 3, \ + NEG_INF_TOKEN := 4, \ + POS_INF_TOKEN := 5, \ + NEG_VALUE_TOKEN := 6, \ + POS_VALUE_TOKEN := 7 } - -DOWNCONVERTSIZE(convertTo) { - CASE converTo OF - _MM_UPCONV_EPI64_NONE: RETURN 8 +DEFINE FIXUPIMMPD(src1[63:0], src2[63:0], src3[63:0], imm8[7:0]) { + tsrc[63:0] := ((src2[62:52] == 0) AND (MXCSR.DAZ == 1)) ? 0.0 : src2[63:0] + CASE(tsrc[63:0]) OF + QNAN_TOKEN:j := 0 + SNAN_TOKEN:j := 1 + ZERO_VALUE_TOKEN: j := 2 + ONE_VALUE_TOKEN: j := 3 + NEG_INF_TOKEN: j := 4 + POS_INF_TOKEN: j := 5 + NEG_VALUE_TOKEN: j := 6 + POS_VALUE_TOKEN: j := 7 + ESAC + + token_response[3:0] := src3[3+4*j:4*j] + + CASE(token_response[3:0]) OF + 0 : dest[63:0] := src1[63:0] + 1 : dest[63:0] := tsrc[63:0] + 2 : dest[63:0] := QNaN(tsrc[63:0]) + 3 : dest[63:0] := QNAN_Indefinite + 4 : dest[63:0] := -INF + 5 : dest[63:0] := +INF + 6 : dest[63:0] := tsrc.sign? -INF : +INF + 7 : dest[63:0] := -0 + 8 : dest[63:0] := +0 + 9 : dest[63:0] := -1 + 10: dest[63:0] := +1 + 11: dest[63:0] := 1/2 + 12: dest[63:0] := 90.0 + 13: dest[63:0] := PI/2 + 14: dest[63:0] := MAX_FLOAT + 15: dest[63:0] := -MAX_FLOAT + ESAC + + CASE(tsrc[31:0]) OF + ZERO_VALUE_TOKEN: + IF (imm8[0]) #ZE; FI + ZERO_VALUE_TOKEN: + IF (imm8[1]) #IE; FI + ONE_VALUE_TOKEN: + IF (imm8[2]) #ZE; FI + ONE_VALUE_TOKEN: + IF (imm8[3]) #IE; FI + SNAN_TOKEN: + IF (imm8[4]) #IE; FI + NEG_INF_TOKEN: + IF (imm8[5]) #IE; FI + NEG_VALUE_TOKEN: + IF (imm8[6]) #IE; FI + POS_INF_TOKEN: + IF (imm8[7]) #IE; FI ESAC + RETURN dest[63:0] } - -storeOffset := 0 -foundNext64BytesBoundary := false -downSize := DOWNCONVERTSIZE(conv) -addr = mt-64 -FOR j := 0 to 7 - IF k[j] - IF foundNext64BytesBoundary == false - IF ((addr + (storeOffset + 1)*downSize) % 64) == 0 - foundNext64BytesBoundary = true - FI - ELSE - i := j*64 - tmp := DOWNCONVERT(v1[i+63:i], conv) - storeAddr := addr + storeOffset * downSize - CASE downSize OF - 8: MEM[storeAddr] := tmp[63:0] - ESAC - FI - storeOffset := storeOffset + 1 - FI -ENDFOR -dst[MAX:512] := 0 +dst[63:0] := FIXUPIMMPD(a[63:0], b[63:0], c[63:0], imm8[7:0]) +dst[127:64] := a[127:64] +dst[MAX:128] := 0 - +
immintrin.h
 - - Integer - KNCNI - Store - - - - - Down-converts and stores packed 64-bit integer elements of "v1" into a quadword stream according to "conv" at a logically mapped starting address "mt", storing the low-64-byte elements of that stream (those elements of the stream that map before the first 64-byte-aligned address follwing "mt"). "hint" indicates to the processor whether the data is non-temporal. - -DOWNCONVERT(element, convertTo) { - CASE converTo OF - _MM_UPCONV_EPI64_NONE: RETURN element[i+63:i] - ESAC + + Floating Point + AVX512F + Miscellaneous + + + + + + + + Fix up the lower double-precision (64-bit) floating-point elements in "a" and "b" using the lower 64-bit integer in "c", store the result in the lower element of "dst" using writemask "k" (the element is copied from "a" when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". "imm8" is used to set the required flags reporting. + [sae_note] + enum TOKEN_TYPE { + QNAN_TOKEN := 0, \ + SNAN_TOKEN := 1, \ + ZERO_VALUE_TOKEN := 2, \ + ONE_VALUE_TOKEN := 3, \ + NEG_INF_TOKEN := 4, \ + POS_INF_TOKEN := 5, \ + NEG_VALUE_TOKEN := 6, \ + POS_VALUE_TOKEN := 7 } - -DOWNCONVERTSIZE(convertTo) { - CASE converTo OF - _MM_UPCONV_EPI64_NONE: RETURN 8 +DEFINE FIXUPIMMPD(src1[63:0], src2[63:0], src3[63:0], imm8[7:0]) { + tsrc[63:0] := ((src2[62:52] == 0) AND (MXCSR.DAZ == 1)) ? 0.0 : src2[63:0] + CASE(tsrc[63:0]) OF + QNAN_TOKEN:j := 0 + SNAN_TOKEN:j := 1 + ZERO_VALUE_TOKEN: j := 2 + ONE_VALUE_TOKEN: j := 3 + NEG_INF_TOKEN: j := 4 + POS_INF_TOKEN: j := 5 + NEG_VALUE_TOKEN: j := 6 + POS_VALUE_TOKEN: j := 7 ESAC -} - -storeOffset := 0 -downSize := DOWNCONVERTSIZE(conv) -addr = mt -FOR j := 0 to 7 - i := j*63 - tmp := DOWNCONVERT(v1[i+63:i], conv) - storeAddr := addr + storeOffset * downSize - CASE downSize OF - 8: MEM[storeAddr] := tmp[63:0] + + token_response[3:0] := src3[3+4*j:4*j] + + CASE(token_response[3:0]) OF + 0 : dest[63:0] := src1[63:0] + 1 : dest[63:0] := tsrc[63:0] + 2 : dest[63:0] := QNaN(tsrc[63:0]) + 3 : dest[63:0] := QNAN_Indefinite + 4 : dest[63:0] := -INF + 5 : dest[63:0] := +INF + 6 : dest[63:0] := tsrc.sign? -INF : +INF + 7 : dest[63:0] := -0 + 8 : dest[63:0] := +0 + 9 : dest[63:0] := -1 + 10: dest[63:0] := +1 + 11: dest[63:0] := 1/2 + 12: dest[63:0] := 90.0 + 13: dest[63:0] := PI/2 + 14: dest[63:0] := MAX_FLOAT + 15: dest[63:0] := -MAX_FLOAT ESAC - storeOffset := storeOffset + 1 - IF ((addr + storeOffset * downSize) % 64) == 0 - BREAK - FI -ENDFOR -dst[MAX:512] := 0 + + CASE(tsrc[31:0]) OF + ZERO_VALUE_TOKEN: + IF (imm8[0]) #ZE; FI + ZERO_VALUE_TOKEN: + IF (imm8[1]) #IE; FI + ONE_VALUE_TOKEN: + IF (imm8[2]) #ZE; FI + ONE_VALUE_TOKEN: + IF (imm8[3]) #IE; FI + SNAN_TOKEN: + IF (imm8[4]) #IE; FI + NEG_INF_TOKEN: + IF (imm8[5]) #IE; FI + NEG_VALUE_TOKEN: + IF (imm8[6]) #IE; FI + POS_INF_TOKEN: + IF (imm8[7]) #IE; FI + ESAC + RETURN dest[63:0] +} +IF k[0] + dst[63:0] := FIXUPIMMPD(a[63:0], b[63:0], c[63:0], imm8[7:0]) +ELSE + dst[63:0] := a[63:0] +FI +dst[127:64] := a[127:64] +dst[MAX:128] := 0 - +
immintrin.h
 - - Integer - KNCNI - Store - - - - - - Down-converts and stores packed 64-bit integer elements of "v1" into a quadword stream according to "conv" at a logically mapped starting address "mt", storing the low-64-byte elements of that stream (those elements of the stream that map before the first 64-byte-aligned address follwing "mt"). "hint" indicates to the processor whether the data is non-temporal. Elements are stored to memory according to element selector "k" (elements are skipped whent he corresponding mask bit is not set). - -DOWNCONVERT(element, convertTo) { - CASE converTo OF - _MM_UPCONV_EPI64_NONE: RETURN element[i+63:i] - ESAC + + Floating Point + AVX512F + Miscellaneous + + + + + + + Fix up the lower double-precision (64-bit) floating-point elements in "a" and "b" using the lower 64-bit integer in "c", store the result in the lower element of "dst" using writemask "k" (the element is copied from "a" when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". "imm8" is used to set the required flags reporting. + enum TOKEN_TYPE { + QNAN_TOKEN := 0, \ + SNAN_TOKEN := 1, \ + ZERO_VALUE_TOKEN := 2, \ + ONE_VALUE_TOKEN := 3, \ + NEG_INF_TOKEN := 4, \ + POS_INF_TOKEN := 5, \ + NEG_VALUE_TOKEN := 6, \ + POS_VALUE_TOKEN := 7 } - -DOWNCONVERTSIZE(convertTo) { - CASE converTo OF - _MM_UPCONV_EPI64_NONE: RETURN 8 +DEFINE FIXUPIMMPD(src1[63:0], src2[63:0], src3[63:0], imm8[7:0]) { + tsrc[63:0] := ((src2[62:52] == 0) AND (MXCSR.DAZ == 1)) ? 0.0 : src2[63:0] + CASE(tsrc[63:0]) OF + QNAN_TOKEN:j := 0 + SNAN_TOKEN:j := 1 + ZERO_VALUE_TOKEN: j := 2 + ONE_VALUE_TOKEN: j := 3 + NEG_INF_TOKEN: j := 4 + POS_INF_TOKEN: j := 5 + NEG_VALUE_TOKEN: j := 6 + POS_VALUE_TOKEN: j := 7 + ESAC + + token_response[3:0] := src3[3+4*j:4*j] + + CASE(token_response[3:0]) OF + 0 : dest[63:0] := src1[63:0] + 1 : dest[63:0] := tsrc[63:0] + 2 : dest[63:0] := QNaN(tsrc[63:0]) + 3 : dest[63:0] := QNAN_Indefinite + 4 : dest[63:0] := -INF + 5 : dest[63:0] := +INF + 6 : dest[63:0] := tsrc.sign? -INF : +INF + 7 : dest[63:0] := -0 + 8 : dest[63:0] := +0 + 9 : dest[63:0] := -1 + 10: dest[63:0] := +1 + 11: dest[63:0] := 1/2 + 12: dest[63:0] := 90.0 + 13: dest[63:0] := PI/2 + 14: dest[63:0] := MAX_FLOAT + 15: dest[63:0] := -MAX_FLOAT ESAC + + CASE(tsrc[31:0]) OF + ZERO_VALUE_TOKEN: + IF (imm8[0]) #ZE; FI + ZERO_VALUE_TOKEN: + IF (imm8[1]) #IE; FI + ONE_VALUE_TOKEN: + IF (imm8[2]) #ZE; FI + ONE_VALUE_TOKEN: + IF (imm8[3]) #IE; FI + SNAN_TOKEN: + IF (imm8[4]) #IE; FI + NEG_INF_TOKEN: + IF (imm8[5]) #IE; FI + NEG_VALUE_TOKEN: + IF (imm8[6]) #IE; FI + POS_INF_TOKEN: + IF (imm8[7]) #IE; FI + ESAC + RETURN dest[63:0] } - -storeOffset := 0 -downSize := DOWNCONVERTSIZE(conv) -addr = mt -FOR j := 0 to 7 - IF k[j] - i := j*63 - tmp := DOWNCONVERT(v1[i+63:i], conv) - storeAddr := addr + storeOffset * downSize - CASE downSize OF - 8: MEM[storeAddr] := tmp[63:0] - ESAC - storeOffset := storeOffset + 1 - IF ((addr + storeOffset * downSize) % 64) == 0 - BREAK - FI - FI -ENDFOR -dst[MAX:512] := 0 +IF k[0] + dst[63:0] := FIXUPIMMPD(a[63:0], b[63:0], c[63:0], imm8[7:0]) +ELSE + dst[63:0] := a[63:0] +FI +dst[127:64] := a[127:64] +dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point - KNCNI - Store - - - - - Down-converts and stores packed single-precision (32-bit) floating-point elements of "v1" into a byte/word/doubleword stream according to "conv" at a logically mapped starting address (mt-64), storing the high-64-byte elements of that stream (those elemetns of the stream that map at or after the first 64-byte-aligned address following (m5-64)). "hint" indicates to the processor whether the data is non-temporal. - -DOWNCONVERT(element, convertTo) { - CASE converTo OF - _MM_UPCONV_PS_NONE: RETURN element[i+31:i] - _MM_UPCONV_PS_FLOAT16: RETURN Float32ToFloat16(element[i+31:i]) - _MM_UPCONV_PS_UINT8: RETURN UInt32ToUInt8(element[i+31:i]) - _MM_UPCONV_PS_SINT8: RETURN SInt32ToSInt8(element[i+31:i]) - _MM_UPCONV_PS_UINT16: RETURN UInt32ToUInt16(element[i+31:i]) - _MM_UPCONV_PS_SINT16: RETURN SInt32ToSInt16(element[i+31:i]) - ESAC + AVX512F + Miscellaneous + + + + + + + + Fix up the lower double-precision (64-bit) floating-point elements in "a" and "b" using the lower 64-bit integer in "c", store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". "imm8" is used to set the required flags reporting. + [sae_note] + enum TOKEN_TYPE { + QNAN_TOKEN := 0, \ + SNAN_TOKEN := 1, \ + ZERO_VALUE_TOKEN := 2, \ + ONE_VALUE_TOKEN := 3, \ + NEG_INF_TOKEN := 4, \ + POS_INF_TOKEN := 5, \ + NEG_VALUE_TOKEN := 6, \ + POS_VALUE_TOKEN := 7 } - -DOWNCONVERTSIZE(convertTo) { - CASE converTo OF - _MM_UPCONV_PS_NONE: RETURN 4 - _MM_UPCONV_PS_FLOAT16: RETURN 2 - _MM_UPCONV_PS_UINT8: RETURN 1 - _MM_UPCONV_PS_SINT8: RETURN 1 - _MM_UPCONV_PS_UINT16: RETURN 2 - _MM_UPCONV_PS_SINT16: RETURN 2 +DEFINE FIXUPIMMPD(src1[63:0], src2[63:0], src3[63:0], imm8[7:0]) { + tsrc[63:0] := ((src2[62:52] == 0) AND (MXCSR.DAZ == 1)) ? 0.0 : src2[63:0] + CASE(tsrc[63:0]) OF + QNAN_TOKEN:j := 0 + SNAN_TOKEN:j := 1 + ZERO_VALUE_TOKEN: j := 2 + ONE_VALUE_TOKEN: j := 3 + NEG_INF_TOKEN: j := 4 + POS_INF_TOKEN: j := 5 + NEG_VALUE_TOKEN: j := 6 + POS_VALUE_TOKEN: j := 7 + ESAC + + token_response[3:0] := src3[3+4*j:4*j] + + CASE(token_response[3:0]) OF + 0 : dest[63:0] := src1[63:0] + 1 : dest[63:0] := tsrc[63:0] + 2 : dest[63:0] := QNaN(tsrc[63:0]) + 3 : dest[63:0] := QNAN_Indefinite + 4 : dest[63:0] := -INF + 5 : dest[63:0] := +INF + 6 : dest[63:0] := tsrc.sign? -INF : +INF + 7 : dest[63:0] := -0 + 8 : dest[63:0] := +0 + 9 : dest[63:0] := -1 + 10: dest[63:0] := +1 + 11: dest[63:0] := 1/2 + 12: dest[63:0] := 90.0 + 13: dest[63:0] := PI/2 + 14: dest[63:0] := MAX_FLOAT + 15: dest[63:0] := -MAX_FLOAT + ESAC + + CASE(tsrc[31:0]) OF + ZERO_VALUE_TOKEN: + IF (imm8[0]) #ZE; FI + ZERO_VALUE_TOKEN: + IF (imm8[1]) #IE; FI + ONE_VALUE_TOKEN: + IF (imm8[2]) #ZE; FI + ONE_VALUE_TOKEN: + IF (imm8[3]) #IE; FI + SNAN_TOKEN: + IF (imm8[4]) #IE; FI + NEG_INF_TOKEN: + IF (imm8[5]) #IE; FI + NEG_VALUE_TOKEN: + IF (imm8[6]) #IE; FI + POS_INF_TOKEN: + IF (imm8[7]) #IE; FI ESAC + RETURN dest[63:0] } - -storeOffset := 0 -foundNext64BytesBoundary := false -downSize := DOWNCONVERTSIZE(conv) -addr = mt-64 -FOR j := 0 to 15 - IF foundNext64BytesBoundary == false - IF ((addr + (storeOffset + 1)*downSize) % 64) == 0 - foundNext64BytesBoundary = true - FI - ELSE - i := j*32 - tmp := DOWNCONVERT(v1[i+31:i], conv) - storeAddr := addr + storeOffset * downSize - CASE downSize OF - 4: MEM[storeAddr] := tmp[31:0] - 2: MEM[storeAddr] := tmp[15:0] - 1: MEM[storeAddr] := tmp[7:0] - ESAC - FI - storeOffset := storeOffset + 1 -ENDFOR -dst[MAX:512] := 0 +IF k[0] + dst[63:0] := FIXUPIMMPD(a[63:0], b[63:0], c[63:0], imm8[7:0]) +ELSE + dst[63:0] := 0 +FI +dst[127:64] := a[127:64] +dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point - KNCNI - Store - - - - - - Down-converts and stores packed single-precision (32-bit) floating-point elements of "v1" into a byte/word/doubleword stream according to "conv" at a logically mapped starting address (mt-64), storing the high-64-byte elements of that stream (those elemetns of the stream that map at or after the first 64-byte-aligned address following (m5-64)). "hint" indicates to the processor whether the data is non-temporal. Elements are stored to memory according to element selector "k" (elements are skipped when the corresponding mask bit is not set). - -DOWNCONVERT(element, convertTo) { - CASE converTo OF - _MM_UPCONV_PS_NONE: RETURN element[i+31:i] - _MM_UPCONV_PS_FLOAT16: RETURN Float32ToFloat16(element[i+31:i]) - _MM_UPCONV_PS_UINT8: RETURN UInt32ToUInt8(element[i+31:i]) - _MM_UPCONV_PS_SINT8: RETURN SInt32ToSInt8(element[i+31:i]) - _MM_UPCONV_PS_UINT16: RETURN UInt32ToUInt16(element[i+31:i]) - _MM_UPCONV_PS_SINT16: RETURN SInt32ToSInt16(element[i+31:i]) - ESAC + AVX512F + Miscellaneous + + + + + + + Fix up the lower double-precision (64-bit) floating-point elements in "a" and "b" using the lower 64-bit integer in "c", store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". "imm8" is used to set the required flags reporting. + enum TOKEN_TYPE { + QNAN_TOKEN := 0, \ + SNAN_TOKEN := 1, \ + ZERO_VALUE_TOKEN := 2, \ + ONE_VALUE_TOKEN := 3, \ + NEG_INF_TOKEN := 4, \ + POS_INF_TOKEN := 5, \ + NEG_VALUE_TOKEN := 6, \ + POS_VALUE_TOKEN := 7 } - -DOWNCONVERTSIZE(convertTo) { - CASE converTo OF - _MM_UPCONV_PS_NONE: RETURN 4 - _MM_UPCONV_PS_FLOAT16: RETURN 2 - _MM_UPCONV_PS_UINT8: RETURN 1 - _MM_UPCONV_PS_SINT8: RETURN 1 - _MM_UPCONV_PS_UINT16: RETURN 2 - _MM_UPCONV_PS_SINT16: RETURN 2 +DEFINE FIXUPIMMPD(src1[63:0], src2[63:0], src3[63:0], imm8[7:0]) { + tsrc[63:0] := ((src2[62:52] == 0) AND (MXCSR.DAZ == 1)) ? 0.0 : src2[63:0] + CASE(tsrc[63:0]) OF + QNAN_TOKEN:j := 0 + SNAN_TOKEN:j := 1 + ZERO_VALUE_TOKEN: j := 2 + ONE_VALUE_TOKEN: j := 3 + NEG_INF_TOKEN: j := 4 + POS_INF_TOKEN: j := 5 + NEG_VALUE_TOKEN: j := 6 + POS_VALUE_TOKEN: j := 7 + ESAC + + token_response[3:0] := src3[3+4*j:4*j] + + CASE(token_response[3:0]) OF + 0 : dest[63:0] := src1[63:0] + 1 : dest[63:0] := tsrc[63:0] + 2 : dest[63:0] := QNaN(tsrc[63:0]) + 3 : dest[63:0] := QNAN_Indefinite + 4 : dest[63:0] := -INF + 5 : dest[63:0] := +INF + 6 : dest[63:0] := tsrc.sign? -INF : +INF + 7 : dest[63:0] := -0 + 8 : dest[63:0] := +0 + 9 : dest[63:0] := -1 + 10: dest[63:0] := +1 + 11: dest[63:0] := 1/2 + 12: dest[63:0] := 90.0 + 13: dest[63:0] := PI/2 + 14: dest[63:0] := MAX_FLOAT + 15: dest[63:0] := -MAX_FLOAT + ESAC + + CASE(tsrc[31:0]) OF + ZERO_VALUE_TOKEN: + IF (imm8[0]) #ZE; FI + ZERO_VALUE_TOKEN: + IF (imm8[1]) #IE; FI + ONE_VALUE_TOKEN: + IF (imm8[2]) #ZE; FI + ONE_VALUE_TOKEN: + IF (imm8[3]) #IE; FI + SNAN_TOKEN: + IF (imm8[4]) #IE; FI + NEG_INF_TOKEN: + IF (imm8[5]) #IE; FI + NEG_VALUE_TOKEN: + IF (imm8[6]) #IE; FI + POS_INF_TOKEN: + IF (imm8[7]) #IE; FI ESAC + RETURN dest[63:0] } - -storeOffset := 0 -foundNext64BytesBoundary := false -downSize := DOWNCONVERTSIZE(conv) -addr = mt-64 -FOR j := 0 to 15 - IF k[j] - IF foundNext64BytesBoundary == false - IF ((addr + (storeOffset + 1)*downSize) % 64) == 0 - foundNext64BytesBoundary = true - FI - ELSE - i := j*32 - tmp := DOWNCONVERT(v1[i+31:i], conv) - storeAddr := addr + storeOffset * downSize - CASE downSize OF - 4: MEM[storeAddr] := tmp[31:0] - 2: MEM[storeAddr] := tmp[15:0] - 1: MEM[storeAddr] := tmp[7:0] - ESAC - FI - storeOffset := storeOffset + 1 - FI -ENDFOR -dst[MAX:512] := 0 +IF k[0] + dst[63:0] := FIXUPIMMPD(a[63:0], b[63:0], c[63:0], imm8[7:0]) +ELSE + dst[63:0] := 0 +FI +dst[127:64] := a[127:64] +dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point - KNCNI - Store - - - - - Down-converts and stores packed single-precision (32-bit) floating-point elements of "v1" into a byte/word/doubleword stream according to "conv" at a logically mapped starting address "mt", storing the low-64-byte elements of that stream (those elements of the stream that map before the first 64-byte-aligned address follwing "mt"). "hint" indicates to the processor whether the data is non-temporal. - -DOWNCONVERT(element, convertTo) { - CASE converTo OF - _MM_UPCONV_PS_NONE: RETURN element[i+31:i] - _MM_UPCONV_PS_FLOAT16: RETURN Float32ToFloat16(element[i+31:i]) - _MM_UPCONV_PS_UINT8: RETURN UInt32ToUInt8(element[i+31:i]) - _MM_UPCONV_PS_SINT8: RETURN SInt32ToSInt8(element[i+31:i]) - _MM_UPCONV_PS_UINT16: RETURN UInt32ToUInt16(element[i+31:i]) - _MM_UPCONV_PS_SINT16: RETURN SInt32ToSInt16(element[i+31:i]) - ESAC + AVX512F + Miscellaneous + + + + + + + Fix up the lower single-precision (32-bit) floating-point elements in "a" and "b" using the lower 32-bit integer in "c", store the result in the lower element of "dst", and copy the upper 3 packed elements from "a" to the upper elements of "dst". "imm8" is used to set the required flags reporting. + [sae_note] + enum TOKEN_TYPE { + QNAN_TOKEN := 0, \ + SNAN_TOKEN := 1, \ + ZERO_VALUE_TOKEN := 2, \ + ONE_VALUE_TOKEN := 3, \ + NEG_INF_TOKEN := 4, \ + POS_INF_TOKEN := 5, \ + NEG_VALUE_TOKEN := 6, \ + POS_VALUE_TOKEN := 7 } - -DOWNCONVERTSIZE(convertTo) { - CASE converTo OF - _MM_UPCONV_PS_NONE: RETURN 4 - _MM_UPCONV_PS_FLOAT16: RETURN 2 - _MM_UPCONV_PS_UINT8: RETURN 1 - _MM_UPCONV_PS_SINT8: RETURN 1 - _MM_UPCONV_PS_UINT16: RETURN 2 - _MM_UPCONV_PS_SINT16: RETURN 2 +DEFINE FIXUPIMMPD(src1[31:0], src2[31:0], src3[31:0], imm8[7:0]) { + tsrc[31:0] := ((src2[30:23] == 0) AND (MXCSR.DAZ == 1)) ? 0.0 : src2[31:0] + CASE(tsrc[31:0]) OF + QNAN_TOKEN:j := 0 + SNAN_TOKEN:j := 1 + ZERO_VALUE_TOKEN: j := 2 + ONE_VALUE_TOKEN: j := 3 + NEG_INF_TOKEN: j := 4 + POS_INF_TOKEN: j := 5 + NEG_VALUE_TOKEN: j := 6 + POS_VALUE_TOKEN: j := 7 ESAC -} - -storeOffset := 0 -downSize := DOWNCONVERTSIZE(conv) -addr = mt -FOR j := 0 to 15 - i := j*32 - tmp := DOWNCONVERT(v1[i+31:i], conv) - storeAddr := addr + storeOffset * downSize - CASE downSize OF - 4: MEM[storeAddr] := tmp[31:0] - 2: MEM[storeAddr] := tmp[15:0] - 1: MEM[storeAddr] := tmp[7:0] + + token_response[3:0] := src3[3+4*j:4*j] + + CASE(token_response[3:0]) OF + 0 : dest[31:0] := src1[31:0] + 1 : dest[31:0] := tsrc[31:0] + 2 : dest[31:0] := QNaN(tsrc[31:0]) + 3 : dest[31:0] := QNAN_Indefinite + 4 : dest[31:0] := -INF + 5 : dest[31:0] := +INF + 6 : dest[31:0] := tsrc.sign? -INF : +INF + 7 : dest[31:0] := -0 + 8 : dest[31:0] := +0 + 9 : dest[31:0] := -1 + 10: dest[31:0] := +1 + 11: dest[31:0] := 1/2 + 12: dest[31:0] := 90.0 + 13: dest[31:0] := PI/2 + 14: dest[31:0] := MAX_FLOAT + 15: dest[31:0] := -MAX_FLOAT ESAC - storeOffset := storeOffset + 1 - IF ((addr + storeOffset * downSize) % 64) == 0 - BREAK - FI -ENDFOR -dst[MAX:512] := 0 + + CASE(tsrc[31:0]) OF + ZERO_VALUE_TOKEN: + IF (imm8[0]) #ZE; FI + ZERO_VALUE_TOKEN: + IF (imm8[1]) #IE; FI + ONE_VALUE_TOKEN: + IF (imm8[2]) #ZE; FI + ONE_VALUE_TOKEN: + IF (imm8[3]) #IE; FI + SNAN_TOKEN: + IF (imm8[4]) #IE; FI + NEG_INF_TOKEN: + IF (imm8[5]) #IE; FI + NEG_VALUE_TOKEN: + IF (imm8[6]) #IE; FI + POS_INF_TOKEN: + IF (imm8[7]) #IE; FI + ESAC + RETURN dest[31:0] +} +dst[31:0] := FIXUPIMMPD(a[31:0], b[31:0], c[31:0], imm8[7:0]) +dst[127:32] := a[127:32] +dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point - KNCNI - Store - - - - - - Down-converts and stores packed single-precision (32-bit) floating-point elements of "v1" into a byte/word/doubleword stream according to "conv" at a logically mapped starting address "mt", storing the low-64-byte elements of that stream (those elements of the stream that map before the first 64-byte-aligned address follwing "mt"). "hint" indicates to the processor whether the data is non-temporal. Elements are stored to memory according to element selector "k" (elements are skipped when the corresponding mask bit is not set). - -DOWNCONVERT(element, convertTo) { - CASE converTo OF - _MM_UPCONV_PS_NONE: RETURN element[i+31:i] - _MM_UPCONV_PS_FLOAT16: RETURN Float32ToFloat16(element[i+31:i]) - _MM_UPCONV_PS_UINT8: RETURN UInt32ToUInt8(element[i+31:i]) - _MM_UPCONV_PS_SINT8: RETURN SInt32ToSInt8(element[i+31:i]) - _MM_UPCONV_PS_UINT16: RETURN UInt32ToUInt16(element[i+31:i]) - _MM_UPCONV_PS_SINT16: RETURN SInt32ToSInt16(element[i+31:i]) - ESAC + AVX512F + Miscellaneous + + + + + + Fix up the lower single-precision (32-bit) floating-point elements in "a" and "b" using the lower 32-bit integer in "c", store the result in the lower element of "dst", and copy the upper 3 packed elements from "a" to the upper elements of "dst". "imm8" is used to set the required flags reporting. + enum TOKEN_TYPE { + QNAN_TOKEN := 0, \ + SNAN_TOKEN := 1, \ + ZERO_VALUE_TOKEN := 2, \ + ONE_VALUE_TOKEN := 3, \ + NEG_INF_TOKEN := 4, \ + POS_INF_TOKEN := 5, \ + NEG_VALUE_TOKEN := 6, \ + POS_VALUE_TOKEN := 7 } - -DOWNCONVERTSIZE(convertTo) { - CASE converTo OF - _MM_UPCONV_PS_NONE: RETURN 4 - _MM_UPCONV_PS_FLOAT16: RETURN 2 - _MM_UPCONV_PS_UINT8: RETURN 1 - _MM_UPCONV_PS_SINT8: RETURN 1 - _MM_UPCONV_PS_UINT16: RETURN 2 - _MM_UPCONV_PS_SINT16: RETURN 2 +DEFINE FIXUPIMMPD(src1[31:0], src2[31:0], src3[31:0], imm8[7:0]) { + tsrc[31:0] := ((src2[30:23] == 0) AND (MXCSR.DAZ == 1)) ? 0.0 : src2[31:0] + CASE(tsrc[31:0]) OF + QNAN_TOKEN:j := 0 + SNAN_TOKEN:j := 1 + ZERO_VALUE_TOKEN: j := 2 + ONE_VALUE_TOKEN: j := 3 + NEG_INF_TOKEN: j := 4 + POS_INF_TOKEN: j := 5 + NEG_VALUE_TOKEN: j := 6 + POS_VALUE_TOKEN: j := 7 + ESAC + + token_response[3:0] := src3[3+4*j:4*j] + + CASE(token_response[3:0]) OF + 0 : dest[31:0] := src1[31:0] + 1 : dest[31:0] := tsrc[31:0] + 2 : dest[31:0] := QNaN(tsrc[31:0]) + 3 : dest[31:0] := QNAN_Indefinite + 4 : dest[31:0] := -INF + 5 : dest[31:0] := +INF + 6 : dest[31:0] := tsrc.sign? -INF : +INF + 7 : dest[31:0] := -0 + 8 : dest[31:0] := +0 + 9 : dest[31:0] := -1 + 10: dest[31:0] := +1 + 11: dest[31:0] := 1/2 + 12: dest[31:0] := 90.0 + 13: dest[31:0] := PI/2 + 14: dest[31:0] := MAX_FLOAT + 15: dest[31:0] := -MAX_FLOAT ESAC + + CASE(tsrc[31:0]) OF + ZERO_VALUE_TOKEN: + IF (imm8[0]) #ZE; FI + ZERO_VALUE_TOKEN: + IF (imm8[1]) #IE; FI + ONE_VALUE_TOKEN: + IF (imm8[2]) #ZE; FI + ONE_VALUE_TOKEN: + IF (imm8[3]) #IE; FI + SNAN_TOKEN: + IF (imm8[4]) #IE; FI + NEG_INF_TOKEN: + IF (imm8[5]) #IE; FI + NEG_VALUE_TOKEN: + IF (imm8[6]) #IE; FI + POS_INF_TOKEN: + IF (imm8[7]) #IE; FI + ESAC + RETURN dest[31:0] } - -storeOffset := 0 -downSize := DOWNCONVERTSIZE(conv) -addr = mt -FOR j := 0 to 15 - IF k[j] - i := j*32 - tmp := DOWNCONVERT(v1[i+31:i], conv) - storeAddr := addr + storeOffset * downSize - CASE downSize OF - 4: MEM[storeAddr] := tmp[31:0] - 2: MEM[storeAddr] := tmp[15:0] - 1: MEM[storeAddr] := tmp[7:0] - ESAC - storeOffset := storeOffset + 1 - IF ((addr + storeOffset * downSize) % 64) == 0 - BREAK - FI - FI -ENDFOR -dst[MAX:512] := 0 +dst[31:0] := FIXUPIMMPD(a[31:0], b[31:0], c[31:0], imm8[7:0]) +dst[127:32] := a[127:32] +dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point - KNCNI - Store - - - - - Down-converts and stores packed double-precision (64-bit) floating-point elements of "v1" into a quadword stream according to "conv" at a logically mapped starting address (mt-64), storing the high-64-byte elements of that stream (those elemetns of the stream that map at or after the first 64-byte-aligned address following (m5-64)). "hint" indicates to the processor whether the data is non-temporal. - -DOWNCONVERT(element, convertTo) { - CASE converTo OF - _MM_UPCONV_PD_NONE: RETURN element[i+63:i] - ESAC + AVX512F + Miscellaneous + + + + + + + + Fix up the lower single-precision (32-bit) floating-point elements in "a" and "b" using the lower 32-bit integer in "c", store the result in the lower element of "dst" using writemask "k" (the element is copied from "a" when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". "imm8" is used to set the required flags reporting. + [sae_note] + enum TOKEN_TYPE { + QNAN_TOKEN := 0, \ + SNAN_TOKEN := 1, \ + ZERO_VALUE_TOKEN := 2, \ + ONE_VALUE_TOKEN := 3, \ + NEG_INF_TOKEN := 4, \ + POS_INF_TOKEN := 5, \ + NEG_VALUE_TOKEN := 6, \ + POS_VALUE_TOKEN := 7 } - -DOWNCONVERTSIZE(convertTo) { - CASE converTo OF - _MM_UPCONV_PD_NONE: RETURN 8 +DEFINE FIXUPIMMPD(src1[31:0], src2[31:0], src3[31:0], imm8[7:0]) { + tsrc[31:0] := ((src2[30:23] == 0) AND (MXCSR.DAZ == 1)) ? 0.0 : src2[31:0] + CASE(tsrc[31:0]) OF + QNAN_TOKEN:j := 0 + SNAN_TOKEN:j := 1 + ZERO_VALUE_TOKEN: j := 2 + ONE_VALUE_TOKEN: j := 3 + NEG_INF_TOKEN: j := 4 + POS_INF_TOKEN: j := 5 + NEG_VALUE_TOKEN: j := 6 + POS_VALUE_TOKEN: j := 7 + ESAC + + token_response[3:0] := src3[3+4*j:4*j] + + CASE(token_response[3:0]) OF + 0 : dest[31:0] := src1[31:0] + 1 : dest[31:0] := tsrc[31:0] + 2 : dest[31:0] := QNaN(tsrc[31:0]) + 3 : dest[31:0] := QNAN_Indefinite + 4 : dest[31:0] := -INF + 5 : dest[31:0] := +INF + 6 : dest[31:0] := tsrc.sign? -INF : +INF + 7 : dest[31:0] := -0 + 8 : dest[31:0] := +0 + 9 : dest[31:0] := -1 + 10: dest[31:0] := +1 + 11: dest[31:0] := 1/2 + 12: dest[31:0] := 90.0 + 13: dest[31:0] := PI/2 + 14: dest[31:0] := MAX_FLOAT + 15: dest[31:0] := -MAX_FLOAT + ESAC + + CASE(tsrc[31:0]) OF + ZERO_VALUE_TOKEN: + IF (imm8[0]) #ZE; FI + ZERO_VALUE_TOKEN: + IF (imm8[1]) #IE; FI + ONE_VALUE_TOKEN: + IF (imm8[2]) #ZE; FI + ONE_VALUE_TOKEN: + IF (imm8[3]) #IE; FI + SNAN_TOKEN: + IF (imm8[4]) #IE; FI + NEG_INF_TOKEN: + IF (imm8[5]) #IE; FI + NEG_VALUE_TOKEN: + IF (imm8[6]) #IE; FI + POS_INF_TOKEN: + IF (imm8[7]) #IE; FI ESAC + RETURN dest[31:0] } - -storeOffset := 0 -foundNext64BytesBoundary := false -downSize := DOWNCONVERTSIZE(conv) -addr = mt-64 -FOR j := 0 to 7 - IF foundNext64BytesBoundary == false - IF ((addr + (storeOffset + 1)*downSize) % 64) == 0 - foundNext64BytesBoundary = true - FI - ELSE - i := j*64 - tmp := DOWNCONVERT(v1[i+63:i], conv) - storeAddr := addr + storeOffset * downSize - CASE downSize OF - 8: MEM[storeAddr] := tmp[63:0] - ESAC - FI - storeOffset := storeOffset + 1 -ENDFOR -dst[MAX:512] := 0 +IF k[0] + dst[31:0] := FIXUPIMMPD(a[31:0], b[31:0], c[31:0], imm8[7:0]) +ELSE + dst[31:0] := a[31:0] +FI +dst[127:32] := a[127:32] +dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point - KNCNI - Store - - - - - - Down-converts and stores packed double-precision (64-bit) floating-point elements of "v1" into a quadword stream according to "conv" at a logically mapped starting address (mt-64), storing the high-64-byte elements of that stream (those elemetns of the stream that map at or after the first 64-byte-aligned address following (m5-64)). "hint" indicates to the processor whether the data is non-temporal. Elements are stored to memory according to element selector "k" (elements are skipped when the corresponding mask bit is not set). - -DOWNCONVERT(element, convertTo) { - CASE converTo OF - _MM_UPCONV_PD_NONE: RETURN element[i+63:i] - ESAC + AVX512F + Miscellaneous + + + + + + + Fix up the lower single-precision (32-bit) floating-point elements in "a" and "b" using the lower 32-bit integer in "c", store the result in the lower element of "dst" using writemask "k" (the element is copied from "a" when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". "imm8" is used to set the required flags reporting. + enum TOKEN_TYPE { + QNAN_TOKEN := 0, \ + SNAN_TOKEN := 1, \ + ZERO_VALUE_TOKEN := 2, \ + ONE_VALUE_TOKEN := 3, \ + NEG_INF_TOKEN := 4, \ + POS_INF_TOKEN := 5, \ + NEG_VALUE_TOKEN := 6, \ + POS_VALUE_TOKEN := 7 } - -DOWNCONVERTSIZE(convertTo) { - CASE converTo OF - _MM_UPCONV_PD_NONE: RETURN 8 +DEFINE FIXUPIMMPD(src1[31:0], src2[31:0], src3[31:0], imm8[7:0]) { + tsrc[31:0] := ((src2[30:23] == 0) AND (MXCSR.DAZ == 1)) ? 0.0 : src2[31:0] + CASE(tsrc[31:0]) OF + QNAN_TOKEN:j := 0 + SNAN_TOKEN:j := 1 + ZERO_VALUE_TOKEN: j := 2 + ONE_VALUE_TOKEN: j := 3 + NEG_INF_TOKEN: j := 4 + POS_INF_TOKEN: j := 5 + NEG_VALUE_TOKEN: j := 6 + POS_VALUE_TOKEN: j := 7 + ESAC + + token_response[3:0] := src3[3+4*j:4*j] + + CASE(token_response[3:0]) OF + 0 : dest[31:0] := src1[31:0] + 1 : dest[31:0] := tsrc[31:0] + 2 : dest[31:0] := QNaN(tsrc[31:0]) + 3 : dest[31:0] := QNAN_Indefinite + 4 : dest[31:0] := -INF + 5 : dest[31:0] := +INF + 6 : dest[31:0] := tsrc.sign? -INF : +INF + 7 : dest[31:0] := -0 + 8 : dest[31:0] := +0 + 9 : dest[31:0] := -1 + 10: dest[31:0] := +1 + 11: dest[31:0] := 1/2 + 12: dest[31:0] := 90.0 + 13: dest[31:0] := PI/2 + 14: dest[31:0] := MAX_FLOAT + 15: dest[31:0] := -MAX_FLOAT + ESAC + + CASE(tsrc[31:0]) OF + ZERO_VALUE_TOKEN: + IF (imm8[0]) #ZE; FI + ZERO_VALUE_TOKEN: + IF (imm8[1]) #IE; FI + ONE_VALUE_TOKEN: + IF (imm8[2]) #ZE; FI + ONE_VALUE_TOKEN: + IF (imm8[3]) #IE; FI + SNAN_TOKEN: + IF (imm8[4]) #IE; FI + NEG_INF_TOKEN: + IF (imm8[5]) #IE; FI + NEG_VALUE_TOKEN: + IF (imm8[6]) #IE; FI + POS_INF_TOKEN: + IF (imm8[7]) #IE; FI ESAC + RETURN dest[31:0] } - -storeOffset := 0 -foundNext64BytesBoundary := false -downSize := DOWNCONVERTSIZE(conv) -addr = mt-64 -FOR j := 0 to 7 - IF k[j] - IF foundNext64BytesBoundary == false - IF ((addr + (storeOffset + 1)*downSize) % 64) == 0 - foundNext64BytesBoundary = true - FI - ELSE - i := j*64 - tmp := DOWNCONVERT(v1[i+63:i], conv) - storeAddr := addr + storeOffset * downSize - CASE downSize OF - 8: MEM[storeAddr] := tmp[63:0] - ESAC - FI - storeOffset := storeOffset + 1 - FI -ENDFOR -dst[MAX:512] := 0 +IF k[0] + dst[31:0] := FIXUPIMMPD(a[31:0], b[31:0], c[31:0], imm8[7:0]) +ELSE + dst[31:0] := a[31:0] +FI +dst[127:32] := a[127:32] +dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point - KNCNI - Store - - - - - Down-converts and stores packed double-precision (64-bit) floating-point elements of "v1" into a quadword stream according to "conv" at a logically mapped starting address "mt", storing the low-64-byte elements of that stream (those elements of the stream that map before the first 64-byte-aligned address follwing "mt"). "hint" indicates to the processor whether the data is non-temporal. - -DOWNCONVERT(element, convertTo) { - CASE converTo OF - _MM_UPCONV_PD_NONE: RETURN element[i+63:i] - ESAC + AVX512F + Miscellaneous + + + + + + + + Fix up the lower single-precision (32-bit) floating-point elements in "a" and "b" using the lower 32-bit integer in "c", store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". "imm8" is used to set the required flags reporting. + [sae_note] + enum TOKEN_TYPE { + QNAN_TOKEN := 0, \ + SNAN_TOKEN := 1, \ + ZERO_VALUE_TOKEN := 2, \ + ONE_VALUE_TOKEN := 3, \ + NEG_INF_TOKEN := 4, \ + POS_INF_TOKEN := 5, \ + NEG_VALUE_TOKEN := 6, \ + POS_VALUE_TOKEN := 7 } - -DOWNCONVERTSIZE(convertTo) { - CASE converTo OF - _MM_UPCONV_PD_NONE: RETURN 8 +DEFINE FIXUPIMMPD(src1[31:0], src2[31:0], src3[31:0], imm8[7:0]) { + tsrc[31:0] := ((src2[30:23] == 0) AND (MXCSR.DAZ == 1)) ? 0.0 : src2[31:0] + CASE(tsrc[31:0]) OF + QNAN_TOKEN:j := 0 + SNAN_TOKEN:j := 1 + ZERO_VALUE_TOKEN: j := 2 + ONE_VALUE_TOKEN: j := 3 + NEG_INF_TOKEN: j := 4 + POS_INF_TOKEN: j := 5 + NEG_VALUE_TOKEN: j := 6 + POS_VALUE_TOKEN: j := 7 ESAC -} - -storeOffset := 0 -downSize := DOWNCONVERTSIZE(conv) -addr = mt -FOR j := 0 to 7 - i := j*63 - tmp := DOWNCONVERT(v1[i+63:i], conv) - storeAddr := addr + storeOffset * downSize - CASE downSize OF - 8: MEM[storeAddr] := tmp[63:0] + + token_response[3:0] := src3[3+4*j:4*j] + + CASE(token_response[3:0]) OF + 0 : dest[31:0] := src1[31:0] + 1 : dest[31:0] := tsrc[31:0] + 2 : dest[31:0] := QNaN(tsrc[31:0]) + 3 : dest[31:0] := QNAN_Indefinite + 4 : dest[31:0] := -INF + 5 : dest[31:0] := +INF + 6 : dest[31:0] := tsrc.sign? -INF : +INF + 7 : dest[31:0] := -0 + 8 : dest[31:0] := +0 + 9 : dest[31:0] := -1 + 10: dest[31:0] := +1 + 11: dest[31:0] := 1/2 + 12: dest[31:0] := 90.0 + 13: dest[31:0] := PI/2 + 14: dest[31:0] := MAX_FLOAT + 15: dest[31:0] := -MAX_FLOAT ESAC - storeOffset := storeOffset + 1 - IF ((addr + storeOffset * downSize) % 64) == 0 - BREAK - FI -ENDFOR -dst[MAX:512] := 0 + + CASE(tsrc[31:0]) OF + ZERO_VALUE_TOKEN: + IF (imm8[0]) #ZE; FI + ZERO_VALUE_TOKEN: + IF (imm8[1]) #IE; FI + ONE_VALUE_TOKEN: + IF (imm8[2]) #ZE; FI + ONE_VALUE_TOKEN: + IF (imm8[3]) #IE; FI + SNAN_TOKEN: + IF (imm8[4]) #IE; FI + NEG_INF_TOKEN: + IF (imm8[5]) #IE; FI + NEG_VALUE_TOKEN: + IF (imm8[6]) #IE; FI + POS_INF_TOKEN: + IF (imm8[7]) #IE; FI + ESAC + RETURN dest[31:0] +} +IF k[0] + dst[31:0] := FIXUPIMMPD(a[31:0], b[31:0], c[31:0], imm8[7:0]) +ELSE + dst[31:0] := 0 +FI +dst[127:32] := a[127:32] +dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point - KNCNI - Store - - - - - - Down-converts and stores packed double-precision (64-bit) floating-point elements of "v1" into a quadword stream according to "conv" at a logically mapped starting address "mt", storing the low-64-byte elements of that stream (those elements of the stream that map before the first 64-byte-aligned address follwing "mt"). "hint" indicates to the processor whether the data is non-temporal. Elements are stored to memory according to element selector "k" (elements are skipped when the corresponding mask bit is not set). - -DOWNCONVERT(element, convertTo) { - CASE converTo OF - _MM_UPCONV_PD_NONE: RETURN element[i+63:i] - ESAC + AVX512F + Miscellaneous + + + + + + + Fix up the lower single-precision (32-bit) floating-point elements in "a" and "b" using the lower 32-bit integer in "c", store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". "imm8" is used to set the required flags reporting. + enum TOKEN_TYPE { + QNAN_TOKEN := 0, \ + SNAN_TOKEN := 1, \ + ZERO_VALUE_TOKEN := 2, \ + ONE_VALUE_TOKEN := 3, \ + NEG_INF_TOKEN := 4, \ + POS_INF_TOKEN := 5, \ + NEG_VALUE_TOKEN := 6, \ + POS_VALUE_TOKEN := 7 } - -DOWNCONVERTSIZE(convertTo) { - CASE converTo OF - _MM_UPCONV_PD_NONE: RETURN 8 +DEFINE FIXUPIMMPD(src1[31:0], src2[31:0], src3[31:0], imm8[7:0]) { + tsrc[31:0] := ((src2[30:23] == 0) AND (MXCSR.DAZ == 1)) ? 0.0 : src2[31:0] + CASE(tsrc[31:0]) OF + QNAN_TOKEN:j := 0 + SNAN_TOKEN:j := 1 + ZERO_VALUE_TOKEN: j := 2 + ONE_VALUE_TOKEN: j := 3 + NEG_INF_TOKEN: j := 4 + POS_INF_TOKEN: j := 5 + NEG_VALUE_TOKEN: j := 6 + POS_VALUE_TOKEN: j := 7 + ESAC + + token_response[3:0] := src3[3+4*j:4*j] + + CASE(token_response[3:0]) OF + 0 : dest[31:0] := src1[31:0] + 1 : dest[31:0] := tsrc[31:0] + 2 : dest[31:0] := QNaN(tsrc[31:0]) + 3 : dest[31:0] := QNAN_Indefinite + 4 : dest[31:0] := -INF + 5 : dest[31:0] := +INF + 6 : dest[31:0] := tsrc.sign? -INF : +INF + 7 : dest[31:0] := -0 + 8 : dest[31:0] := +0 + 9 : dest[31:0] := -1 + 10: dest[31:0] := +1 + 11: dest[31:0] := 1/2 + 12: dest[31:0] := 90.0 + 13: dest[31:0] := PI/2 + 14: dest[31:0] := MAX_FLOAT + 15: dest[31:0] := -MAX_FLOAT ESAC + + CASE(tsrc[31:0]) OF + ZERO_VALUE_TOKEN: + IF (imm8[0]) #ZE; FI + ZERO_VALUE_TOKEN: + IF (imm8[1]) #IE; FI + ONE_VALUE_TOKEN: + IF (imm8[2]) #ZE; FI + ONE_VALUE_TOKEN: + IF (imm8[3]) #IE; FI + SNAN_TOKEN: + IF (imm8[4]) #IE; FI + NEG_INF_TOKEN: + IF (imm8[5]) #IE; FI + NEG_VALUE_TOKEN: + IF (imm8[6]) #IE; FI + POS_INF_TOKEN: + IF (imm8[7]) #IE; FI + ESAC + RETURN dest[31:0] } - -storeOffset := 0 -downSize := DOWNCONVERTSIZE(conv) -addr = mt -FOR j := 0 to 7 - IF k[j] - i := j*63 - tmp := DOWNCONVERT(v1[i+63:i], conv) - storeAddr := addr + storeOffset * downSize - CASE downSize OF - 8: MEM[storeAddr] := tmp[63:0] - ESAC - storeOffset := storeOffset + 1 - IF ((addr + storeOffset * downSize) % 64) == 0 - BREAK - FI - FI -ENDFOR -dst[MAX:512] := 0 +IF k[0] + dst[31:0] := FIXUPIMMPD(a[31:0], b[31:0], c[31:0], imm8[7:0]) +ELSE + dst[31:0] := 0 +FI +dst[127:32] := a[127:32] +dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point - AVX512F/KNCNI - Convert - - Performs an element-by-element conversion of packed double-precision (64-bit) floating-point elements in "v2" to single-precision (32-bit) floating-point elements and stores them in "dst". The elements are stored in the lower half of the results vector, while the remaining upper half locations are set to 0. + AVX512F + Arithmetic + + + + + + Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", add the intermediate result to packed elements in "c", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). FOR j := 0 to 7 i := j*64 - k := j*32 - dst[k+31:k] := Float64ToFloat32(v2[i+63:i]) + IF k[j] + dst[i+63:i] := (a[i+63:i] * b[i+63:i]) + c[i+63:i] + ELSE + dst[i+63:i] := 0 + FI ENDFOR dst[MAX:512] := 0 - + + +
immintrin.h
 - + Floating Point - AVX512F/KNCNI - Convert - - - - Performs an element-by-element conversion of packed double-precision (64-bit) floating-point elements in "v2" to single-precision (32-bit) floating-point elements and stores them in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). The elements are stored in the lower half of the results vector, while the remaining upper half locations are set to 0. + AVX512F + Arithmetic + + + + + + + Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", add the intermediate result to packed elements in "c", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + [round_note] FOR j := 0 to 7 i := j*64 - l := j*32 IF k[j] - dst[l+31:l] := Float64ToFloat32(v2[i+63:i]) + dst[i+63:i] := (a[i+63:i] * b[i+63:i]) + c[i+63:i] ELSE - dst[l+31:l] := src[l+31:l] + dst[i+63:i] := 0 FI ENDFOR dst[MAX:512] := 0 - -
immintrin.h
- - - Integer - AVX512F/KNCNI - Load - - - - Loads 8 64-bit integer elements from memory starting at location "mv" at packed 32-bit integer indices stored in the lower half of "index" scaled by "scale" and stores them in "dst". - -FOR j := 0 to 7 - i := j*64 - addr := MEM[mv + index[j] * scale] - dst[i+63:i] := addr[i+63:i] -ENDFOR -dst[MAX:512] := 0 - - + + +
immintrin.h
 - - Integer - AVX512F/KNCNI - Load - - - - - - Loads 8 64-bit integer elements from memory starting at location "mv" at packed 32-bit integer indices stored in the lower half of "index" scaled by "scale" and stores them in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Floating Point + AVX512F + Arithmetic + + + + + + Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", add the intermediate result to packed elements in "c", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := j*64 +FOR j := 0 to 15 + i := j*32 IF k[j] - addr := MEM[mv + index[j] * scale] - dst[i+63:i] := addr[i+63:i] + dst[i+31:i] := (a[i+31:i] * b[i+31:i]) + c[i+31:i] ELSE - dst[i+63:i] := src[i+63:i] + dst[i+31:i] := 0 FI ENDFOR dst[MAX:512] := 0 - + + +
immintrin.h
 - + Floating Point - AVX512F/KNCNI - Load - - - - Loads 8 double-precision (64-bit) floating-point elements stored at memory locations starting at location "mv" at packed 32-bit integer indices stored in the lower half of "index" scaled by "scale" them in "dst". + AVX512F + Arithmetic + + + + + + + Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", add the intermediate result to packed elements in "c", and store the results in "a" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + [round_note] -FOR j := 0 to 7 - i := j*64 - addr := MEM[mv + index[j] * scale] - dst[i+63:i] := addr[i+63:i] +FOR j := 0 to 15 + i := j*32 + IF k[j] + dst[i+31:i] := (a[i+31:i] * b[i+31:i]) + c[i+31:i] + ELSE + dst[i+31:i] := 0 + FI ENDFOR dst[MAX:512] := 0 - + + +
immintrin.h
 - + Floating Point - AVX512F/KNCNI - Load - - - - - - Loads 8 double-precision (64-bit) floating-point elements from memory starting at location "mv" at packed 32-bit integer indices stored in the lower half of "index" scaled by "scale" into "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + AVX512F + Arithmetic + + + + + + Multiply the lower double-precision (64-bit) floating-point elements in "a" and "b", and add the intermediate result to the lower element in "c". Store the result in the lower element of "dst", and copy the upper element from "a" to the upper element of "dst". + [round_note] -FOR j := 0 to 7 - i := j*64 - IF k[j] - addr := MEM[mv + index[j] * scale] - dst[i+63:i] := addr[i+63:i] - ELSE - dst[i+63:i] := src[i+63:i] - FI -ENDFOR -dst[MAX:512] := 0 +dst[63:0] := (a[63:0] * b[63:0]) + c[63:0] +dst[127:64] := a[127:64] +dst[MAX:128] := 0 - + + +
immintrin.h
 - + Floating Point - AVX512PF/KNCNI - Load - - - - - Prefetches 16 single-precision (32-bit) floating-point elements in memory starting at location "mv" at packed 32-bit integer indices stored in "index" scaled by "scale". The "hint" parameter may be 1 (_MM_HINT_T0) for prefetching to L1 cache, or 2 (_MM_HINT_T1) for prefetching to L2 cache. + AVX512F + Arithmetic + + + + + + + Multiply the lower double-precision (64-bit) floating-point elements in "a" and "b", and add the intermediate result to the lower element in "c". Store the result in the lower element of "dst" using writemask "k" (the element is copied from "c" when mask bit 0 is not set), and copy the upper element from "c" to the upper element of "dst". + [round_note] -cachev := 0 -FOR j := 0 to 15 - i := j*32 - addr := MEM[mv + index[j] * scale] - cachev[i+31:i] := addr[i+63:i] -ENDFOR +IF k[0] + dst[63:0] := (a[63:0] * b[63:0]) + c[63:0] +ELSE + dst[63:0] := c[63:0] +FI +dst[127:64] := c[127:64] +dst[MAX:128] := 0 - - + + +
immintrin.h
 - + Floating Point - AVX512F/KNCNI - Store - - - - - Stores 8 packed double-precision (64-bit) floating-point elements in "v1" and to memory locations starting at location "mv" at packed 32-bit integer indices stored in "index" scaled by "scale". + AVX512F + Arithmetic + + + + + + Multiply the lower double-precision (64-bit) floating-point elements in "a" and "b", and add the intermediate result to the lower element in "c". Store the result in the lower element of "dst" using writemask "k" (the element is copied from "c" when mask bit 0 is not set), and copy the upper element from "c" to the upper element of "dst". -FOR j := 0 to 7 - addr := MEM[mv + index[j] * scale] - i := j*64 - addr[i+63:i] := v1[k+63:j] -ENDFOR +IF k[0] + dst[63:0] := (a[63:0] * b[63:0]) + c[63:0] +ELSE + dst[63:0] := c[63:0] +FI +dst[127:64] := c[127:64] +dst[MAX:128] := 0 - + + +
immintrin.h
 - + Floating Point - AVX512F/KNCNI - Store - - - - - - Stores 8 packed double-precision (64-bit) floating-point elements in "v1" to memory locations starting at location "mv" at packed 32-bit integer indices stored in "index" scaled by "scale". Only those elements whose corresponding mask bit is set in writemask "k" are written to memory. + AVX512F + Arithmetic + + + + + + + Multiply the lower double-precision (64-bit) floating-point elements in "a" and "b", and add the intermediate result to the lower element in "c". Store the result in the lower element of "dst" using writemask "k" (the element is copied from "a" when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". + [round_note] -FOR j := 0 to 7 - IF k[j] - addr := MEM[mv + index[j] * scale] - i := j*64 - addr[i+63:i] := v1[k+63:j] - FI -ENDFOR +IF k[0] + dst[63:0] := (a[63:0] * b[63:0]) + c[63:0] +ELSE + dst[63:0] := a[63:0] +FI +dst[127:64] := a[127:64] +dst[MAX:128] := 0 - + + +
immintrin.h
 - - Integer - KNCNI - Store - - - - - Stores 8 packed 64-bit integer elements located in "v1" and stores them in memory locations starting at location "mv" at packed 32-bit integer indices stored in "index" scaled by "scale". + + Floating Point + AVX512F + Arithmetic + + + + + + Multiply the lower double-precision (64-bit) floating-point elements in "a" and "b", and add the intermediate result to the lower element in "c". Store the result in the lower element of "dst" using writemask "k" (the element is copied from "a" when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". -FOR j := 0 to 7 - addr := MEM[mv + index[j] * scale] - i := j*64 - addr[i+63:i] := v1[k+63:j] -ENDFOR +IF k[0] + dst[63:0] := (a[63:0] * b[63:0]) + c[63:0] +ELSE + dst[63:0] := a[63:0] +FI +dst[127:64] := a[127:64] +dst[MAX:128] := 0 - + + +
immintrin.h
 - - Integer - KNCNI - Store - - - - - - Stores 8 packed 64-bit integer elements located in "v1" and stores them in memory locations starting at location "mv" at packed 32-bit integer indices stored in "index" scaled by "scale" using writemask "k" (elements whose corresponding mask bit is not set are not written to memory). + + Floating Point + AVX512F + Arithmetic + + + + + + + Multiply the lower double-precision (64-bit) floating-point elements in "a" and "b", and add the intermediate result to the lower element in "c". Store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". + [round_note] -FOR j := 0 to 7 - IF k[j] - addr := MEM[mv + index[j] * scale] - addr[i+63:i] := v1[i+63:i] - FI -ENDFOR +IF k[0] + dst[63:0] := (a[63:0] * b[63:0]) + c[63:0] +ELSE + dst[63:0] := 0 +FI +dst[127:64] := a[127:64] +dst[MAX:128] := 0 - + + +
immintrin.h
 - + Floating Point - AVX512PF/KNCNI - Store - - - - - Prefetches 16 single-precision (32-bit) floating-point elements in memory starting at location "mv" at packed 32-bit integer indices stored in "index" scaled by "scale". The "hint" parameter may be 1 (_MM_HINT_T0) for prefetching to L1 cache, or 2 (_MM_HINT_T1) for prefetching to L2 cache. + AVX512F + Arithmetic + + + + + + Multiply the lower double-precision (64-bit) floating-point elements in "a" and "b", and add the intermediate result to the lower element in "c". Store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". -FOR j := 0 to 15 - i := j*32 - addr := MEM[mv + index[j] * scale] - CASE hint OF - _MM_HINT_T0: PrefetchL1WithT0Hint(addr[i+31:i]) - _MM_HINT_T1: PrefetchL2WithT1Hint(addr[i+31:i]) - _MM_HINT_T2: PrefetchL2WithT1HintNonTemporal(addr[i+31:i]) - _MM_HINT_NTA: PrefetchL1WithT0HintNonTemporal(addr[i+31:i]) - ESAC -ENDFOR +IF k[0] + dst[63:0] := (a[63:0] * b[63:0]) + c[63:0] +ELSE + dst[63:0] := 0 +FI +dst[127:64] := a[127:64] +dst[MAX:128] := 0 - - + + +
immintrin.h
 - + Floating Point - AVX512PF/KNCNI - Store - - - - - - Prefetches 16 single-precision (32-bit) floating-point elements in memory starting at location "mv" at packed 32-bit integer indices stored in "index" scaled by "scale". The "hint" parameter may be 1 (_MM_HINT_T0) for prefetching to L1 cache, or 2 (_MM_HINT_T1) for prefetching to L2 cache. Only those elements whose corresponding mask bit in "k" is set are loaded into the desired cache. + AVX512F + Arithmetic + + + + + + + Multiply the lower single-precision (32-bit) floating-point elements in "a" and "b", and add the intermediate result to the lower element in "c". Store the result in the lower element of "dst" using writemask "k" (the element is copied from "c" when mask bit 0 is not set), and copy the upper 3 packed elements from "c" to the upper elements of "dst". + [round_note] -FOR j := 0 to 15 - i := j*32 - IF k[j] - addr := MEM[mv + index[j] * scale] - CASE hint OF - _MM_HINT_T0: PrefetchL1WithT0Hint(addr[i+31:i]) - _MM_HINT_T1: PrefetchL2WithT1Hint(addr[i+31:i]) - _MM_HINT_T2: PrefetchL2WithT1HintNonTemporal(addr[i+31:i]) - _MM_HINT_NTA: PrefetchL1WithT0HintNonTemporal(addr[i+31:i]) - FI -ENDFOR +IF k[0] + dst[31:0] := (a[31:0] * b[31:0]) + c[31:0] +ELSE + dst[31:0] := c[31:0] +FI +dst[127:32] := c[127:32] +dst[MAX:128] := 0 - - + + +
immintrin.h
 - - Integer - KNCNI - Load - - - Loads the high-64-byte-aligned portion of the byte/word/doubleword stream starting at element-aligned address mt-64 and expands them into packed 32-bit integers in "dst". The initial values of "dst" are copied from "src". Only those converted doublewords that occur at or after the first 64-byte-aligned address following (mt-64) are loaded. Elements in the resulting vector that do not map to those doublewords are taken from "src". + + Floating Point + AVX512F + Arithmetic + + + + + + Multiply the lower single-precision (32-bit) floating-point elements in "a" and "b", and add the intermediate result to the lower element in "c". Store the result in the lower element of "dst" using writemask "k" (the element is copied from "c" when mask bit 0 is not set), and copy the upper 3 packed elements from "c" to the upper elements of "dst". -dst[511:0] := src[511:0] -loadOffset := 0 -foundNext64BytesBoundary := false -addr = mt-64 -FOR j := 0 to 15 - IF foundNext64BytesBoundary == false - IF (addr + (loadOffset + 1)*4 % 64) == 0 - foundNext64BytesBoundary := true - FI - ELSE - i := j*32 - tmp := MEM[addr + loadOffset*4] - dst[i+31:i] := tmp[i+31:i] - FI - loadOffset := loadOffset + 1 -ENDFOR -dst[MAX:512] := 0 +IF k[0] + dst[31:0] := (a[31:0] * b[31:0]) + c[31:0] +ELSE + dst[31:0] := c[31:0] +FI +dst[127:32] := c[127:32] +dst[MAX:128] := 0 - + + +
immintrin.h
 - - Integer - KNCNI - Load - - - - Loads the high-64-byte-aligned portion of the byte/word/doubleword stream starting at element-aligned address mt-64 and expands them into packed 32-bit integers in "dst". The initial values of "dst" are copied from "src". Only those converted doublewords that occur at or after the first 64-byte-aligned address following (mt-64) are loaded. Elements in the resulting vector that do not map to those doublewords are taken from "src". Elements are loaded from memory according to element selector "k" (elements are skipped when the corresponding mask bit is not set). + + Floating Point + AVX512F + Arithmetic + + + + + + Multiply the lower single-precision (32-bit) floating-point elements in "a" and "b", and add the intermediate result to the lower element in "c". Store the result in the lower element of "dst", and copy the upper 3 packed elements from "a" to the upper elements of "dst". + [round_note] -dst[511:0] := src[511:0] -loadOffset := 0 -foundNext64BytesBoundary := false -addr = mt-64 -FOR j := 0 to 15 - IF k[j] - IF foundNext64BytesBoundary == false - IF (addr + (loadOffset + 1)*4 % 64) == 0 - foundNext64BytesBoundary := true - FI - ELSE - i := j*32 - tmp := MEM[addr + loadOffset*4] - dst[i+31:i] := tmp[i+31:i] - FI - loadOffset := loadOffset + 1 - FI -ENDFOR -dst[MAX:512] := 0 +dst[31:0] := (a[31:0] * b[31:0]) + c[31:0] +dst[127:32] := a[127:32] +dst[MAX:128] := 0 - + + +
immintrin.h
 - - Integer - KNCNI - Load - - - Loads the low-64-byte-aligned portion of the byte/word/doubleword stream starting at element-aligned address mt and expanded into packed 32-bit integers in "dst". The initial values of "dst" are copied from "src". Only those converted doublewords that occur before first 64-byte-aligned address following "mt" are loaded. Elements in the resulting vector that do not map to those doublewords are taken from "src". + + Floating Point + AVX512F + Arithmetic + + + + + + + Multiply the lower single-precision (32-bit) floating-point elements in "a" and "b", and add the intermediate result to the lower element in "c". Store the result in the lower element of "dst" using writemask "k" (the element is copied from "a" when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". + [round_note] -dst[511:0] := src[511:0] -loadOffset := 0 -addr = mt -FOR j := 0 to 15 - i := j*32 - tmp := MEM[addr + loadOffset*4] - dst[i+31:i] := tmp[i+31:i] - loadOffset := loadOffset + 1 - IF (mt + loadOffset * 4) % 64 == 0 - break - FI -ENDFOR -dst[MAX:512] := 0 +IF k[0] + dst[31:0] := (a[31:0] * b[31:0]) + c[31:0] +ELSE + dst[31:0] := a[31:0] +FI +dst[127:32] := a[127:32] +dst[MAX:128] := 0 - + + +
immintrin.h
 - - Integer - KNCNI - Load - - - - Loads the low-64-byte-aligned portion of the byte/word/doubleword stream starting at element-aligned address mt and expands them into packed 32-bit integers in "dst". The initial values of "dst" are copied from "src". Only those converted doublewords that occur before first 64-byte-aligned address following "mt" are loaded. Elements in the resulting vector that do not map to those doublewords are taken from "src". Elements are loaded from memory according to element selector "k" (elements are skipped when the corresponding mask bit is not set). + + Floating Point + AVX512F + Arithmetic + + + + + + Multiply the lower single-precision (32-bit) floating-point elements in "a" and "b", and add the intermediate result to the lower element in "c". Store the result in the lower element of "dst" using writemask "k" (the element is copied from "a" when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". -dst[511:0] := src[511:0] -loadOffset := 0 -addr = mt -FOR j := 0 to 15 - i := j*32 - IF k[j] - tmp := MEM[addr + loadOffset*4] - dst[i+31:i] := tmp[i+31:i] - loadOffset := loadOffset + 1 - IF (mt + loadOffset * 4) % 64 == 0 - break - FI - FI -ENDFOR -dst[MAX:512] := 0 +IF k[0] + dst[31:0] := (a[31:0] * b[31:0]) + c[31:0] +ELSE + dst[31:0] := a[31:0] +FI +dst[127:32] := a[127:32] +dst[MAX:128] := 0 - + + +
immintrin.h
 - - Integer - KNCNI - Load - - - Loads the high-64-byte-aligned portion of the quadword stream starting at element-aligned address mt-64 and expands them into packed 64-bit integers in "dst". The initial values of "dst" are copied from "src". Only those converted quadwords that occur at or after the first 64-byte-aligned address following (mt-64) are loaded. Elements in the resulting vector that do not map to those quadwords are taken from "src". + + Floating Point + AVX512F + Arithmetic + + + + + + + Multiply the lower single-precision (32-bit) floating-point elements in "a" and "b", and add the intermediate result to the lower element in "c". Store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". + [round_note] -dst[511:0] := src[511:0] -loadOffset := 0 -foundNext64BytesBoundary := false -addr = mt-64 -FOR j := 0 to 7 - IF foundNext64BytesBoundary == false - IF (addr + (loadOffset + 1)*8) == 0 - foundNext64BytesBoundary := true - FI - ELSE - i := j*64 - tmp := MEM[addr + loadOffset*8] - dst[i+63:i] := tmp[i+63:i] - FI - loadOffset := loadOffset + 1 -ENDFOR -dst[MAX:512] := 0 +IF k[0] + dst[31:0] := (a[31:0] * b[31:0]) + c[31:0] +ELSE + dst[31:0] := 0 +FI +dst[127:32] := a[127:32] +dst[MAX:128] := 0 - + + +
immintrin.h
 - - Integer - KNCNI - Load - - - - Loads the high-64-byte-aligned portion of the quadword stream starting at element-aligned address mt-64 and expands them into packed 64-bit integers in "dst". The initial values of "dst" are copied from "src". Only those converted quadwords that occur at or after the first 64-byte-aligned address following (mt-64) are loaded. Elements in the resulting vector that do not map to those quadwords are taken from "src". Elements are loaded from memory according to element selector "k" (elements are skipped when the corresponding mask bit is not set). + + Floating Point + AVX512F + Arithmetic + + + + + + Multiply the lower single-precision (32-bit) floating-point elements in "a" and "b", and add the intermediate result to the lower element in "c". Store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". -dst[511:0] := src[511:0] -loadOffset := 0 -foundNext64BytesBoundary := false -addr = mt-64 -FOR j := 0 to 7 - IF k[j] - IF foundNext64BytesBoundary == false - IF (addr + (loadOffset + 1)*8) == 0 - foundNext64BytesBoundary := true - FI - ELSE - i := j*64 - tmp := MEM[addr + loadOffset*8] - dst[i+63:i] := tmp[i+63:i] - FI - loadOffset := loadOffset + 1 - FI -ENDFOR -dst[MAX:512] := 0 +IF k[0] + dst[31:0] := (a[31:0] * b[31:0]) + c[31:0] +ELSE + dst[31:0] := 0 +FI +dst[127:32] := a[127:32] +dst[MAX:128] := 0 - + + +
immintrin.h
 - - Integer - KNCNI - Load - - - Loads the low-64-byte-aligned portion of the quadword stream starting at element-aligned address mt and expands them into packed 64-bit integers in "dst". The initial values of "dst" are copied from "src". Only those converted quad that occur before first 64-byte-aligned address following "mt" are loaded. Elements in the resulting vector that do not map to those quadwords are taken from "src". + + Floating Point + AVX512F + Arithmetic + + + + + Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", alternatively add and subtract packed elements in "c" to/from the intermediate result, and store the results in "dst". -dst[511:0] := src[511:0] -loadOffset := 0 -addr = mt FOR j := 0 to 7 i := j*64 - tmp := MEM[addr + loadOffset*8] - dst[i+63:i] := tmp[i+63:i] - loadOffset := loadOffset + 1 - IF (addr + loadOffset*8 % 64) == 0 - break + IF ((j & 1) == 0) + dst[i+63:i] := (a[i+63:i] * b[i+63:i]) - c[i+63:i] + ELSE + dst[i+63:i] := (a[i+63:i] * b[i+63:i]) + c[i+63:i] FI ENDFOR dst[MAX:512] := 0 - + + +
immintrin.h
 - - Integer - KNCNI - Load - - - - Loads the low-64-byte-aligned portion of the quadword stream starting at element-aligned address mt and expands them into packed 64-bit integers in "dst". The initial values of "dst" are copied from "src". Only those converted quad that occur before first 64-byte-aligned address following "mt" are loaded. Elements in the resulting vector that do not map to those quadwords are taken from "src". Elements are loaded from memory according to element selector "k" (elements are skipped when the corresponding mask bit is not set). + + Floating Point + AVX512F + Arithmetic + + + + + + Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", alternatively add and subtract packed elements in "c" to/from the intermediate result, and store the results in "dst". + [round_note] -dst[511:0] := src[511:0] -loadOffset := 0 -addr = mt FOR j := 0 to 7 i := j*64 - IF k[j] - tmp := MEM[addr + loadOffset*8] - dst[i+63:i] := tmp[i+63:i] - loadOffset := loadOffset + 1 - IF (addr + loadOffset*8 % 64) == 0 - break - FI + IF ((j & 1) == 0) + dst[i+63:i] := (a[i+63:i] * b[i+63:i]) - c[i+63:i] + ELSE + dst[i+63:i] := (a[i+63:i] * b[i+63:i]) + c[i+63:i] FI ENDFOR dst[MAX:512] := 0 - + + +
immintrin.h
 - + Floating Point - KNCNI - Load - - - Loads the high-64-byte-aligned portion of the byte/word/doubleword stream starting at element-aligned address mt-64 and expands them into packed single-precision (32-bit) floating-point elements in "dst". The initial values of "dst" are copied from "src". Only those converted quadwords that occur at or after the first 64-byte-aligned address following (mt-64) are loaded. Elements in the resulting vector that do not map to those quadwords are taken from "src". + AVX512F + Arithmetic + + + + + + Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", alternatively add and subtract packed elements in "c" to/from the intermediate result, and store the results in "dst" using writemask "k" (elements are copied from "c" when the corresponding mask bit is not set). -dst[511:0] := src[511:0] -loadOffset := 0 -foundNext64BytesBoundary := false -addr = mt-64 -FOR j := 0 to 15 - IF foundNext64BytesBoundary == false - IF (addr + (loadOffset + 1)*4 % 64) == 0 - foundNext64BytesBoundary := true +FOR j := 0 to 7 + i := j*64 + IF k[j] + IF ((j & 1) == 0) + dst[i+63:i] := (a[i+63:i] * b[i+63:i]) - c[i+63:i] + ELSE + dst[i+63:i] := (a[i+63:i] * b[i+63:i]) + c[i+63:i] FI ELSE - i := j*32 - tmp := MEM[addr + loadOffset*4] - dst[i+31:i] := tmp[i+31:i] + dst[i+63:i] := c[i+63:i] FI - loadOffset := loadOffset + 1 ENDFOR dst[MAX:512] := 0 - + + +
immintrin.h
 - + Floating Point - KNCNI - Load - - - - Loads the high-64-byte-aligned portion of the doubleword stream starting at element-aligned address mt-64 and expands them into packed single-precision (32-bit) floating-point elements in "dst". The initial values of "dst" are copied from "src". Only those converted quadwords that occur at or after the first 64-byte-aligned address following (mt-64) are loaded. Elements in the resulting vector that do not map to those quadwords are taken from "src". Elements are loaded from memory according to element selector "k" (elements are skipped when the corresponding mask bit is not set). + AVX512F + Arithmetic + + + + + + + Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", alternatively add and subtract packed elements in "c" to/from the intermediate result, and store the results in "dst" using writemask "k" (elements are copied from "c" when the corresponding mask bit is not set). [round_note] -dst[511:0] := src[511:0] -loadOffset := 0 -foundNext64BytesBoundary := false -addr = mt-64 -FOR j := 0 to 15 +FOR j := 0 to 7 + i := j*64 IF k[j] - IF foundNext64BytesBoundary == false - IF (addr + (loadOffset + 1)*4 % 64) == 0 - foundNext64BytesBoundary := true - FI + IF ((j & 1) == 0) + dst[i+63:i] := (a[i+63:i] * b[i+63:i]) - c[i+63:i] ELSE - i := j*32 - tmp := MEM[addr + loadOffset*4] - dst[i+31:i] := tmp[i+31:i] + dst[i+63:i] := (a[i+63:i] * b[i+63:i]) + c[i+63:i] FI - loadOffset := loadOffset + 1 + ELSE + dst[i+63:i] := c[i+63:i] FI ENDFOR dst[MAX:512] := 0 - + + +
immintrin.h
 - + Floating Point - KNCNI - Load - - - Loads the low-64-byte-aligned portion of the doubleword stream starting at element-aligned address mt and expanded into packed single-precision (32-bit) floating-point elements in "dst". The initial values of "dst" are copied from "src". Only those converted doublewords that occur before first 64-byte-aligned address following "mt" are loaded. Elements in the resulting vector that do not map to those doublewords are taken from "src". + AVX512F + Arithmetic + + + + + + Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", alternatively add and subtract packed elements in "c" to/from the intermediate result, and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). -dst[511:0] := src[511:0] -loadOffset := 0 -addr = mt -FOR j := 0 to 15 - i := j*32 - tmp := MEM[addr + loadOffset*4] - dst[i+31:i] := tmp[i+31:i] - loadOffset := loadOffset + 1 - IF (mt + loadOffset * 4) % 64 == 0 - BREAK +FOR j := 0 to 7 + i := j*64 + IF k[j] + IF ((j & 1) == 0) + dst[i+63:i] := (a[i+63:i] * b[i+63:i]) - c[i+63:i] + ELSE + dst[i+63:i] := (a[i+63:i] * b[i+63:i]) + c[i+63:i] + FI + ELSE + dst[i+63:i] := a[i+63:i] FI ENDFOR dst[MAX:512] := 0 - + + +
immintrin.h
 - + Floating Point - KNCNI - Load - - - - Loads the low-64-byte-aligned portion of the doubleword stream starting at element-aligned address mt and expanded into packed single-precision (32-bit) floating-point elements in "dst". The initial values of "dst" are copied from "src". Only those converted doublewords that occur before first 64-byte-aligned address following "mt" are loaded. Elements in the resulting vector that do not map to those doublewords are taken from "src". Elements are loaded from memory according to element selector "k" (elements are skipped when the corresponding mask bit is not set). + AVX512F + Arithmetic + + + + + + + Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", alternatively add and subtract packed elements in "c" to/from the intermediate result, and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). [round_note] -dst[511:0] := src[511:0] -loadOffset := 0 -addr = mt -FOR j := 0 to 15 - i := j*32 +FOR j := 0 to 7 + i := j*64 IF k[j] - tmp := MEM[addr + loadOffset*4] - dst[i+31:i] := tmp[i+31:i] - loadOffset := loadOffset + 1 - IF (mt + loadOffset * 4) % 64 == 0 - break + IF ((j & 1) == 0) + dst[i+63:i] := (a[i+63:i] * b[i+63:i]) - c[i+63:i] + ELSE + dst[i+63:i] := (a[i+63:i] * b[i+63:i]) + c[i+63:i] FI + ELSE + dst[i+63:i] := a[i+63:i] FI ENDFOR dst[MAX:512] := 0 - + + +
immintrin.h
 - + Floating Point - KNCNI - Load - - - Loads the high-64-byte-aligned portion of the quadword stream starting at element-aligned address mt-64 and expands them into packed double-precision (64-bit) floating-point values in "dst". The initial values of "dst" are copied from "src". Only those converted quadwords that occur at or after the first 64-byte-aligned address following (mt-64) are loaded. Elements in the resulting vector that do not map to those quadwords are taken from "src". + AVX512F + Arithmetic + + + + + + Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", alternatively add and subtract packed elements in "c" to/from the intermediate result, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -dst[511:0] := src[511:0] -loadOffset := 0 -foundNext64BytesBoundary := false -addr = mt-64 FOR j := 0 to 7 - IF foundNext64BytesBoundary == false - IF (addr + (loadOffset + 1)*8) % 64 == 0 - foundNext64BytesBoundary := true + i := j*64 + IF k[j] + IF ((j & 1) == 0) + dst[i+63:i] := (a[i+63:i] * b[i+63:i]) - c[i+63:i] + ELSE + dst[i+63:i] := (a[i+63:i] * b[i+63:i]) + c[i+63:i] FI ELSE - i := j*64 - tmp := MEM[addr + loadOffset*8] - dst[i+63:i] := tmp[i+63:i] + dst[i+63:i] := 0 FI - loadOffset := loadOffset + 1 ENDFOR dst[MAX:512] := 0 - + + +
immintrin.h
 - + Floating Point - KNCNI - Load - - - - Loads the high-64-byte-aligned portion of the quadword stream starting at element-aligned address mt-64 and expands them into packed double-precision (64-bit) floating-point values in "dst". The initial values of "dst" are copied from "src". Only those converted quadwords that occur at or after the first 64-byte-aligned address following (mt-64) are loaded. Elements in the resulting vector that do not map to those quadwords are taken from "src". Elements are loaded from memory according to element selector "k" (elements are skipped when the corresponding mask bit is not set). + AVX512F + Arithmetic + + + + + + + Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", alternatively add and subtract packed elements in "c" to/from the intermediate result, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + [round_note] -dst[511:0] := src[511:0] -loadOffset := 0 -foundNext64BytesBoundary := false -addr = mt-64 FOR j := 0 to 7 + i := j*64 IF k[j] - IF foundNext64BytesBoundary == false - IF (addr + (loadOffset + 1)*8) % 64 == 0 - foundNext64BytesBoundary := true - FI + IF ((j & 1) == 0) + dst[i+63:i] := (a[i+63:i] * b[i+63:i]) - c[i+63:i] ELSE - i := j*64 - tmp := MEM[addr + loadOffset*8] - dst[i+63:i] := tmp[i+63:i] + dst[i+63:i] := (a[i+63:i] * b[i+63:i]) + c[i+63:i] FI - loadOffset := loadOffset + 1 + ELSE + dst[i+63:i] := 0 FI ENDFOR dst[MAX:512] := 0 - + + +
immintrin.h
 - + Floating Point - KNCNI - Load - - - Loads the low-64-byte-aligned portion of the quadword stream starting at element-aligned address mt and expands them into packed double-precision (64-bit) floating-point elements in "dst". The initial values of "dst" are copied from "src". Only those converted quad that occur before first 64-byte-aligned address following "mt" are loaded. Elements in the resulting vector that do not map to those quadwords are taken from "src". + AVX512F + Arithmetic + + + + + Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", alternatively add and subtract packed elements in "c" to/from the intermediate result, and store the results in "dst". -dst[511:0] := src[511:0] -loadOffset := 0 -addr = mt -FOR j := 0 to 7 - i := j*64 - tmp := MEM[addr + loadOffset*8] - dst[i+63:i] := tmp[i+63:i] - loadOffset := loadOffset + 1 - IF ((addr + 8*loadOffset) % 64) == 0 - BREAK +FOR j := 0 to 15 + i := j*32 + IF ((j & 1) == 0) + dst[i+31:i] := (a[i+31:i] * b[i+31:i]) - c[i+31:i] + ELSE + dst[i+31:i] := (a[i+31:i] * b[i+31:i]) + c[i+31:i] FI ENDFOR dst[MAX:512] := 0 - + + +
immintrin.h
 - + Floating Point - KNCNI - Load - - - - Loads the low-64-byte-aligned portion of the quadword stream starting at element-aligned address mt and expands them into packed double-precision (64-bit) floating-point values in "dst". The initial values of "dst" are copied from "src". Only those converted quad that occur before first 64-byte-aligned address following "mt" are loaded. Elements in the resulting vector that do not map to those quadwords are taken from "src". Elements are loaded from memory according to element selector "k" (elements are skipped when the corresponding mask bit is not set). + AVX512F + Arithmetic + + + + + + Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", alternatively add and subtract packed elements in "c" to/from the intermediate result, and store the results in "dst". + [round_note] -dst[511:0] := src[511:0] -loadOffset := 0 -addr = mt -FOR j := 0 to 7 - i := j*64 - IF k[j] - tmp := MEM[addr + loadOffset*8] - dst[i+63:i] := tmp[i+63:i] - loadOffset := loadOffset + 1 - IF ((addr + 8*loadOffset) % 64) == 0 - BREAK - FI +FOR j := 0 to 15 + i := j*32 + IF ((j & 1) == 0) + dst[i+31:i] := (a[i+31:i] * b[i+31:i]) - c[i+31:i] + ELSE + dst[i+31:i] := (a[i+31:i] * b[i+31:i]) + c[i+31:i] FI ENDFOR dst[MAX:512] := 0 - + + +
immintrin.h
 - - Integer - KNCNI - Store - - - Stores packed 32-bit integer elements of "v1" into a doubleword stream at a logically mapped starting address (mt-64), storing the high-64-byte elements of that stream (those elements of the stream that map at or after the first 64-byte-aligned address following (m5-64)). + + Floating Point + AVX512F + Arithmetic + + + + + + Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", alternatively add and subtract packed elements in "c" to/from the intermediate result, and store the results in "dst" using writemask "k" (elements are copied from "c" when the corresponding mask bit is not set). -storeOffset := 0 -foundNext64BytesBoundary := false -addr = mt-64 FOR j := 0 to 15 - IF foundNext64BytesBoundary == false - IF ((addr + (storeOffset + 1)*4) % 64) == 0 - foundNext64BytesBoundary = true + i := j*32 + IF k[j] + IF ((j & 1) == 0) + dst[i+31:i] := (a[i+31:i] * b[i+31:i]) - c[i+31:i] + ELSE + dst[i+31:i] := (a[i+31:i] * b[i+31:i]) + c[i+31:i] FI ELSE - i := j*32 - MEM[addr + storeOffset*4] := v1[i+31:i] + dst[i+31:i] := c[i+31:i] FI - storeOffset := storeOffset + 1 ENDFOR dst[MAX:512] := 0 - + + +
immintrin.h
 - - Integer - KNCNI - Store - - - - Stores packed 32-bit integer elements of "v1" into a doubleword stream at a logically mapped starting address (mt-64), storing the high-64-byte elements of that stream (those elements of the stream that map at or after the first 64-byte-aligned address following (m5-64)). Elements are loaded from memory according to element selector "k" (elements are skipped when the corresponding mask bit is not set). + + Floating Point + AVX512F + Arithmetic + + + + + + + Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", alternatively add and subtract packed elements in "c" to/from the intermediate result, and store the results in "dst" using writemask "k" (elements are copied from "c" when the corresponding mask bit is not set). [round_note] -storeOffset := 0 -foundNext64BytesBoundary := false -addr = mt-64 FOR j := 0 to 15 + i := j*32 IF k[j] - IF foundNext64BytesBoundary == false - IF ((addr + (storeOffset + 1)*4) % 64) == 0 - foundNext64BytesBoundary = true - FI + IF ((j & 1) == 0) + dst[i+31:i] := (a[i+31:i] * b[i+31:i]) - c[i+31:i] ELSE - i := j*32 - MEM[addr + storeOffset*4] := v1[i+31:i] + dst[i+31:i] := (a[i+31:i] * b[i+31:i]) + c[i+31:i] FI - storeOffset := storeOffset + 1 + ELSE + dst[i+31:i] := c[i+31:i] FI ENDFOR dst[MAX:512] := 0 - + + +
immintrin.h
 - - Integer - KNCNI - Store - - - Stores packed 32-bit integer elements of "v1" into a doubleword stream at a logically mapped starting address "mt", storing the low-64-byte elements of that stream (those elements of the stream that map before the first 64-byte-aligned address follwing "mt"). + + Floating Point + AVX512F + Arithmetic + + + + + + Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", alternatively add and subtract packed elements in "c" to/from the intermediate result, and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). -storeOffset := 0 -addr = mt FOR j := 0 to 15 i := j*32 - MEM[addr + storeOffset*4] := v1[i+31:i] - storeOffset := storeOffset + 1 - IF ((addr + storeOffset*4) % 64) == 0 - BREAK + IF k[j] + IF ((j & 1) == 0) + dst[i+31:i] := (a[i+31:i] * b[i+31:i]) - c[i+31:i] + ELSE + dst[i+31:i] := (a[i+31:i] * b[i+31:i]) + c[i+31:i] + FI + ELSE + dst[i+31:i] := a[i+31:i] FI ENDFOR dst[MAX:512] := 0 - + + +
immintrin.h
 - - Integer - KNCNI - Store - - - - Stores packed 32-bit integer elements of "v1" into a doubleword stream at a logically mapped starting address "mt", storing the low-64-byte elements of that stream (those elements of the stream that map before the first 64-byte-aligned address follwing "mt"). Elements are loaded from memory according to element selector "k" (elements are skipped when the corresponding mask bit is not set). + + Floating Point + AVX512F + Arithmetic + + + + + + + Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", alternatively add and subtract packed elements in "c" to/from the intermediate result, and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). [round_note] -storeOffset := 0 -addr = mt FOR j := 0 to 15 + i := j*32 IF k[j] - i := j*32 - MEM[addr + storeOffset*4] := v1[i+31:i] - storeOffset := storeOffset + 1 - IF ((addr + storeOffset*4) % 64) == 0 - BREAK + IF ((j & 1) == 0) + dst[i+31:i] := (a[i+31:i] * b[i+31:i]) - c[i+31:i] + ELSE + dst[i+31:i] := (a[i+31:i] * b[i+31:i]) + c[i+31:i] FI + ELSE + dst[i+31:i] := a[i+31:i] FI ENDFOR dst[MAX:512] := 0 - + + +
immintrin.h
 - - Integer - KNCNI - Store - - - Stores packed 64-bit integer elements of "v1" into a quadword stream at a logically mapped starting address (mt-64), storing the high-64-byte elements of that stream (those elemetns of the stream that map at or after the first 64-byte-aligned address following (m5-64)). + + Floating Point + AVX512F + Arithmetic + + + + + + Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", alternatively add and subtract packed elements in "c" to/from the intermediate result, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -storeOffset := 0 -foundNext64BytesBoundary := false -addr = mt-64 -FOR j := 0 to 7 - IF foundNext64BytesBoundary == false - IF ((addr + (storeOffset + 1)*8) % 64) == 0 - foundNext64BytesBoundary = true +FOR j := 0 to 15 + i := j*32 + IF k[j] + IF ((j & 1) == 0) + dst[i+31:i] := (a[i+31:i] * b[i+31:i]) - c[i+31:i] + ELSE + dst[i+31:i] := (a[i+31:i] * b[i+31:i]) + c[i+31:i] FI ELSE - i := j*64 - MEM[addr + storeOffset*8] := v1[i+63:i] + dst[i+31:i] := 0 FI - storeOffset := storeOffset + 1 ENDFOR dst[MAX:512] := 0 - + + +
immintrin.h
 - - Integer - KNCNI - Store - - - - Stores packed 64-bit integer elements of "v1" into a quadword stream at a logically mapped starting address (mt-64), storing the high-64-byte elements of that stream (those elemetns of the stream that map at or after the first 64-byte-aligned address following (m5-64)). Elements are loaded from memory according to element selector "k" (elements are skipped when the corresponding mask bit is not set). + + Floating Point + AVX512F + Arithmetic + + + + + + + Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", alternatively add and subtract packed elements in "c" to/from the intermediate result, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + [round_note] -storeOffset := 0 -foundNext64BytesBoundary := false -addr = mt-64 -FOR j := 0 to 7 +FOR j := 0 to 15 + i := j*32 IF k[j] - IF foundNext64BytesBoundary == false - IF ((addr + (storeOffset + 1)*8) % 64) == 0 - foundNext64BytesBoundary = true - FI + IF ((j & 1) == 0) + dst[i+31:i] := (a[i+31:i] * b[i+31:i]) - c[i+31:i] ELSE - i := j*64 - MEM[addr + storeOffset*8] := v1[i+63:i] + dst[i+31:i] := (a[i+31:i] * b[i+31:i]) + c[i+31:i] FI - storeOffset := storeOffset + 1 + ELSE + dst[i+31:i] := 0 FI ENDFOR dst[MAX:512] := 0 - + + +
immintrin.h
 - - Integer - KNCNI - Store - - - Stores packed 64-bit integer elements of "v1" into a quadword stream at a logically mapped starting address "mt", storing the low-64-byte elements of that stream (those elements of the stream that map before the first 64-byte-aligned address follwing "mt"). + + Floating Point + AVX512F + Arithmetic + + + + + + Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", subtract packed elements in "c" from the intermediate result, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -storeOffset := 0 -addr = mt FOR j := 0 to 7 i := j*64 - MEM[addr + storeOffset*8] := v1[i+63:i] - storeOffset := storeOffset + 1 - IF ((addr + storeOffset*8) % 64) == 0 - BREAK + IF k[j] + dst[i+63:i] := (a[i+63:i] * b[i+63:i]) - c[i+63:i] + ELSE + dst[i+63:i] := 0 FI ENDFOR dst[MAX:512] := 0 - + + +
immintrin.h
 - - Integer - KNCNI - Store - - - - Stores packed 64-bit integer elements of "v1" into a quadword stream at a logically mapped starting address "mt", storing the low-64-byte elements of that stream (those elements of the stream that map before the first 64-byte-aligned address follwing "mt"). Elements are loaded from memory according to element selector "k" (elements are skipped when the corresponding mask bit is not set). + + Floating Point + AVX512F + Arithmetic + + + + + + + Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", subtract packed elements in "c" from the intermediate result, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + [round_note] -storeOffset := 0 -addr = mt FOR j := 0 to 7 + i := j*64 IF k[j] - i := j*64 - MEM[addr + storeOffset*8] := v1[i+63:i] - storeOffset := storeOffset + 1 - IF ((addr + storeOffset*8) % 64) == 0 - BREAK - FI + dst[i+63:i] := (a[i+63:i] * b[i+63:i]) - c[i+63:i] + ELSE + dst[i+63:i] := 0 FI ENDFOR dst[MAX:512] := 0 - + + +
immintrin.h
 - + Floating Point - KNCNI - Store - - - Stores packed single-precision (32-bit) floating-point elements of "v1" into a doubleword stream at a logically mapped starting address (mt-64), storing the high-64-byte elements of that stream (those elemetns of the stream that map at or after the first 64-byte-aligned address following (m5-64)). + AVX512F + Arithmetic + + + + + + Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", subtract packed elements in "c" from the intermediate result, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -storeOffset := 0 -foundNext64BytesBoundary := false -addr = mt-64 FOR j := 0 to 15 - IF foundNext64BytesBoundary == false - IF ((addr + (storeOffset + 1)*4) % 64) == 0 - foundNext64BytesBoundary = true - FI + i := j*32 + IF k[j] + dst[i+31:i] := (a[i+31:i] * b[i+31:i]) - c[i+31:i] ELSE - i := j*32 - MEM[addr + storeOffset*4] := v1[i+31:i] + dst[i+31:i] := 0 FI - storeOffset := storeOffset + 1 ENDFOR dst[MAX:512] := 0 - + + +
immintrin.h
 - + Floating Point - KNCNI - Store - - - - Stores packed single-precision (32-bit) floating-point elements of "v1" into a doubleword stream at a logically mapped starting address (mt-64), storing the high-64-byte elements of that stream (those elemetns of the stream that map at or after the first 64-byte-aligned address following (m5-64)). Elements are loaded from memory according to element selector "k" (elements are skipped when the corresponding mask bit is not set). + AVX512F + Arithmetic + + + + + + + Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", subtract packed elements in "c" from the intermediate result, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + [round_note] -storeOffset := 0 -foundNext64BytesBoundary := false -addr = mt-64 FOR j := 0 to 15 + i := j*32 IF k[j] - IF foundNext64BytesBoundary == false - IF ((addr + (storeOffset + 1)*4) % 64) == 0 - foundNext64BytesBoundary = true - FI - ELSE - i := j*32 - MEM[addr + storeOffset*4] := v1[i+31:i] - FI - storeOffset := storeOffset + 1 + dst[i+31:i] := (a[i+31:i] * b[i+31:i]) - c[i+31:i] + ELSE + dst[i+31:i] := 0 FI ENDFOR dst[MAX:512] := 0 - + + +
immintrin.h
 - + Floating Point - KNCNI - Store - - - Stores packed single-precision (32-bit) floating-point elements of "v1" into a doubleword stream at a logically mapped starting address "mt", storing the low-64-byte elements of that stream (those elements of the stream that map before the first 64-byte-aligned address follwing "mt"). + AVX512F + Arithmetic + + + + + + Multiply the lower double-precision (64-bit) floating-point elements in "a" and "b", and subtract the lower element in "c" from the intermediate result. Store the result in the lower element of "dst", and copy the upper element from "a" to the upper element of "dst". + [round_note] -storeOffset := 0 -addr = mt -FOR j := 0 to 15 - i := j*32 - MEM[addr + storeOffset*4] := v1[i+31:i] - storeOffset := storeOffset + 1 - IF ((addr + storeOffset*4) % 64) == 0 - BREAK - FI -ENDFOR -dst[MAX:512] := 0 +dst[63:0] := (a[63:0] * b[63:0]) - c[63:0] +dst[127:64] := a[127:64] +dst[MAX:128] := 0 - + + +
immintrin.h
 - + Floating Point - KNCNI - Store - - - - Stores packed single-precision (32-bit) floating-point elements of "v1" into a doubleword stream at a logically mapped starting address "mt", storing the low-64-byte elements of that stream (those elements of the stream that map before the first 64-byte-aligned address follwing "mt"). Elements are loaded from memory according to element selector "k" (elements are skipped when the corresponding mask bit is not set). + AVX512F + Arithmetic + + + + + + + Multiply the lower double-precision (64-bit) floating-point elements in "a" and "b", and subtract the lower element in "c" from the intermediate result. Store the result in the lower element of "dst" using writemask "k" (the element is copied from "c" when mask bit 0 is not set), and copy the upper element from "c" to the upper element of "dst". + [round_note] -storeOffset := 0 -addr = mt -FOR j := 0 to 15 - IF k[j] - i := j*32 - MEM[addr + storeOffset*4] := v1[i+31:i] - storeOffset := storeOffset + 1 - IF ((addr + storeOffset*4) % 64) == 0 - BREAK - FI - FI -ENDFOR -dst[MAX:512] := 0 +IF k[0] + dst[63:0] := (a[63:0] * b[63:0]) - c[63:0] +ELSE + dst[63:0] := c[63:0] +FI +dst[127:64] := c[127:64] +dst[MAX:128] := 0 - + + +
immintrin.h
 - + Floating Point - KNCNI - Store - - - Stores packed double-precision (64-bit) floating-point elements of "v1" into a quadword stream at a logically mapped starting address (mt-64), storing the high-64-byte elements of that stream (those elemetns of the stream that map at or after the first 64-byte-aligned address following (m5-64)). + AVX512F + Arithmetic + + + + + + Multiply the lower double-precision (64-bit) floating-point elements in "a" and "b", and subtract the lower element in "c" from the intermediate result. Store the result in the lower element of "dst" using writemask "k" (the element is copied from "c" when mask bit 0 is not set), and copy the upper element from "c" to the upper element of "dst". -storeOffset := 0 -foundNext64BytesBoundary := false -addr = mt-64 -FOR j := 0 to 7 - IF foundNext64BytesBoundary == false - IF ((addr + (storeOffset + 1)*8) % 64) == 0 - foundNext64BytesBoundary = true - FI - ELSE - i := j*64 - MEM[addr + storeOffset*4] := v1[i+63:i] - FI - storeOffset := storeOffset + 1 -ENDFOR -dst[MAX:512] := 0 +IF k[0] + dst[63:0] := (a[63:0] * b[63:0]) - c[63:0] +ELSE + dst[63:0] := c[63:0] +FI +dst[127:64] := c[127:64] +dst[MAX:128] := 0 - + + +
immintrin.h
 - + Floating Point - KNCNI - Store - - - - Stores packed double-precision (64-bit) floating-point elements of "v1" into a quadword stream at a logically mapped starting address (mt-64), storing the high-64-byte elements of that stream (those elemetns of the stream that map at or after the first 64-byte-aligned address following (m5-64)). Elements are loaded from memory according to element selector "k" (elements are skipped when the corresponding mask bit is not set). + AVX512F + Arithmetic + + + + + + + Multiply the lower double-precision (64-bit) floating-point elements in "a" and "b", and subtract the lower element in "c" from the intermediate result. Store the result in the lower element of "dst" using writemask "k" (the element is copied from "a" when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". + [round_note] -storeOffset := 0 -foundNext64BytesBoundary := false -addr = mt-64 -FOR j := 0 to 7 - IF k[j] - IF foundNext64BytesBoundary == false - IF ((addr + (storeOffset + 1)*8) % 64) == 0 - foundNext64BytesBoundary = true - FI - ELSE - i := j*64 - MEM[addr + storeOffset*4] := v1[i+63:i] - FI - storeOffset := storeOffset + 1 - FI -ENDFOR -dst[MAX:512] := 0 +IF k[0] + dst[63:0] := (a[63:0] * b[63:0]) - c[63:0] +ELSE + dst[63:0] := a[63:0] +FI +dst[127:64] := a[127:64] +dst[MAX:128] := 0 - + + +
immintrin.h
 - + Floating Point - KNCNI - Store - - - Stores packed double-precision (64-bit) floating-point elements of "v1" into a quadword stream at a logically mapped starting address "mt", storing the low-64-byte elements of that stream (those elements of the stream that map before the first 64-byte-aligned address follwing "mt"). + AVX512F + Arithmetic + + + + + + Multiply the lower double-precision (64-bit) floating-point elements in "a" and "b", and subtract the lower element in "c" from the intermediate result. Store the result in the lower element of "dst" using writemask "k" (the element is copied from "a" when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". -storeOffset := 0 -addr = mt -FOR j := 0 to 7 - i := j*64 - MEM[addr + storeOffset*8] := v1[i+63:i] - storeOffset := storeOffset + 1 - IF ((addr + storeOffset*8) % 64) == 0 - BREAK - FI -ENDFOR -dst[MAX:512] := 0 +IF k[0] + dst[63:0] := (a[63:0] * b[63:0]) - c[63:0] +ELSE + dst[63:0] := a[63:0] +FI +dst[127:64] := a[127:64] +dst[MAX:128] := 0 - + + +
immintrin.h
 - + Floating Point - KNCNI - Store - - - - Stores packed double-precision (64-bit) floating-point elements of "v1" into a quadword stream at a logically mapped starting address "mt", storing the low-64-byte elements of that stream (those elements of the stream that map before the first 64-byte-aligned address follwing "mt"). Elements are loaded from memory according to element selector "k" (elements are skipped when the corresponding mask bit is not set). + AVX512F + Arithmetic + + + + + + + Multiply the lower double-precision (64-bit) floating-point elements in "a" and "b", and subtract the lower element in "c" from the intermediate result. Store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". + [round_note] -storeOffset := 0 -addr = mt -FOR j := 0 to 7 - IF k[j] - i := j*64 - MEM[addr + storeOffset*8] := v1[i+63:i] - storeOffset := storeOffset + 1 - IF ((addr + storeOffset*8) % 64) == 0 - BREAK - FI - FI -ENDFOR -dst[MAX:512] := 0 +IF k[0] + dst[63:0] := (a[63:0] * b[63:0]) - c[63:0] +ELSE + dst[63:0] := 0 +FI +dst[127:64] := a[127:64] +dst[MAX:128] := 0 - + + +
immintrin.h
 - - KNCNI - Bit Manipulation - - Counts the number of set bits in 32-bit unsigned integer "r1", returning the results in "dst". + + Floating Point + AVX512F + Arithmetic + + + + + + Multiply the lower double-precision (64-bit) floating-point elements in "a" and "b", and subtract the lower element in "c" from the intermediate result. Store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". + +IF k[0] + dst[63:0] := (a[63:0] * b[63:0]) - c[63:0] +ELSE + dst[63:0] := 0 +FI +dst[127:64] := a[127:64] +dst[MAX:128] := 0 + + + + +
immintrin.h
+ + + Floating Point + AVX512F + Arithmetic + + + + + + Multiply the lower single-precision (32-bit) floating-point elements in "a" and "b", and subtract the lower element in "c" from the intermediate result. Store the result in the lower element of "dst", and copy the upper 3 packed elements from "a" to the upper elements of "dst". + [round_note] -dst[31:0] := PopCount(r1[31:0]) +dst[31:0] := (a[31:0] * b[31:0]) - c[31:0] +dst[127:32] := a[127:32] +dst[MAX:128] := 0 - + + +
immintrin.h
 - - KNCNI - Bit Manipulation - - Counts the number of set bits in 64-bit unsigned integer "r1", returning the results in "dst". + + Floating Point + AVX512F + Arithmetic + + + + + + + Multiply the lower single-precision (32-bit) floating-point elements in "a" and "b", and subtract the lower element in "c" from the intermediate result. Store the result in the lower element of "dst" using writemask "k" (the element is copied from "c" when mask bit 0 is not set), and copy the upper 3 packed elements from "c" to the upper elements of "dst". + [round_note] -dst[63:0] := PopCount(r1[63:0]) +IF k[0] + dst[31:0] := (a[31:0] * b[31:0]) - c[31:0] +ELSE + dst[31:0] := c[31:0] +FI +dst[127:32] := c[127:32] +dst[MAX:128] := 0 - + + +
immintrin.h
 - - Mask - KNCNI - Mask - - - Inserts the low byte of mask "k2" into the high byte of "dst", and copies the low byte of "k1" to the low byte of "dst". + + Floating Point + AVX512F + Arithmetic + + + + + + Multiply the lower single-precision (32-bit) floating-point elements in "a" and "b", and subtract the lower element in "c" from the intermediate result. Store the result in the lower element of "dst" using writemask "k" (the element is copied from "c" when mask bit 0 is not set), and copy the upper 3 packed elements from "c" to the upper elements of "dst". -dst[7:0] := k1[7:0] -dst[15:8] := k2[7:0] +IF k[0] + dst[31:0] := (a[31:0] * b[31:0]) - c[31:0] +ELSE + dst[31:0] := c[31:0] +FI +dst[127:32] := c[127:32] +dst[MAX:128] := 0 - + + +
immintrin.h
 - + Floating Point - Integer - KNCNI - Convert - - - Performs an element-by-element conversion of elements in packed double-precision (64-bit) floating-point vector "v2" to 32-bit integer elements, storing them in the lower half of "dst". The elements in the upper half of "dst" are set to 0. - [round_note] - + AVX512F + Arithmetic + + + + + + + Multiply the lower single-precision (32-bit) floating-point elements in "a" and "b", and subtract the lower element in "c" from the intermediate result. Store the result in the lower element of "dst" using writemask "k" (the element is copied from "a" when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". + [round_note] + +IF k[0] + dst[31:0] := (a[31:0] * b[31:0]) - c[31:0] +ELSE + dst[31:0] := a[31:0] +FI +dst[127:32] := a[127:32] +dst[MAX:128] := 0 + + + + +
immintrin.h
+ + + Floating Point + AVX512F + Arithmetic + + + + + + Multiply the lower single-precision (32-bit) floating-point elements in "a" and "b", and subtract the lower element in "c" from the intermediate result. Store the result in the lower element of "dst" using writemask "k" (the element is copied from "a" when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". + +IF k[0] + dst[31:0] := (a[31:0] * b[31:0]) - c[31:0] +ELSE + dst[31:0] := a[31:0] +FI +dst[127:32] := a[127:32] +dst[MAX:128] := 0 + + + + +
immintrin.h
+ + + Floating Point + AVX512F + Arithmetic + + + + + + + Multiply the lower single-precision (32-bit) floating-point elements in "a" and "b", and subtract the lower element in "c" from the intermediate result. Store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". + [round_note] + +IF k[0] + dst[31:0] := (a[31:0] * b[31:0]) - c[31:0] +ELSE + dst[31:0] := 0 +FI +dst[127:32] := a[127:32] +dst[MAX:128] := 0 + + + + +
immintrin.h
+ + + Floating Point + AVX512F + Arithmetic + + + + + + Multiply the lower single-precision (32-bit) floating-point elements in "a" and "b", and subtract the lower element in "c" from the intermediate result. Store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". + +IF k[0] + dst[31:0] := (a[31:0] * b[31:0]) - c[31:0] +ELSE + dst[31:0] := 0 +FI +dst[127:32] := a[127:32] +dst[MAX:128] := 0 + + + + +
immintrin.h
+ + + Floating Point + AVX512F + Arithmetic + + + + + Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", alternatively subtract and add packed elements in "c" from/to the intermediate result, and store the results in "dst". FOR j := 0 to 7 i := j*64 - k := j*32 - dst[k+31:k] := Float64ToInt32(v2[i+63:i]) + IF ((j & 1) == 0) + dst[i+63:i] := (a[i+63:i] * b[i+63:i]) + c[i+63:i] + ELSE + dst[i+63:i] := (a[i+63:i] * b[i+63:i]) - c[i+63:i] + FI ENDFOR dst[MAX:512] := 0 - + + +
immintrin.h
 - + Floating Point - Integer - KNCNI - Convert - - - - - Performs an element-by-element conversion of elements in packed double-precision (64-bit) floating-point vector "v2" to 32-bit integer elements, storing them in the lower half of "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). The elements in the upper half of "dst" are set to 0. - [round_note] - + AVX512F + Arithmetic + + + + + + Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", alternatively subtract and add packed elements in "c" from/to the intermediate result, and store the results in "dst". + [round_note] FOR j := 0 to 7 i := j*64 - l := j*32 - IF k[j] - dst[l+31:l] := Float64ToInt32(v2[i+63:i]) + IF ((j & 1) == 0) + dst[i+63:i] := (a[i+63:i] * b[i+63:i]) + c[i+63:i] ELSE - dst[l+31:l] := src[l+31:l] + dst[i+63:i] := (a[i+63:i] * b[i+63:i]) - c[i+63:i] FI ENDFOR dst[MAX:512] := 0 - + + +
immintrin.h
 - + Floating Point - Integer - KNCNI - Convert - - - - Performs element-by-element conversion of packed 32-bit integer elements in "v2" to packed single-precision (32-bit) floating-point elements and performing an optional exponent adjust using "expadj", storing the results in "dst". - [round_note] - + AVX512F + Arithmetic + + + + + + Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", alternatively subtract and add packed elements in "c" from/to the intermediate result, and store the results in "dst" using writemask "k" (elements are copied from "c" when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*32 - dst[i+31:i] := Int32ToFloat32(v2[i+31:i]) - CASE expadj OF - _MM_EXPADJ_NONE: dst[i+31:i] = dst[i+31:i] * 2**0 - _MM_EXPADJ_4: dst[i+31:i] = dst[i+31:i] * 2**4 - _MM_EXPADJ_5: dst[i+31:i] = dst[i+31:i] * 2**5 - _MM_EXPADJ_8: dst[i+31:i] = dst[i+31:i] * 2**8 - _MM_EXPADJ_16: dst[i+31:i] = dst[i+31:i] * 2**16 - _MM_EXPADJ_24: dst[i+31:i] = dst[i+31:i] * 2**24 - _MM_EXPADJ_31: dst[i+31:i] = dst[i+31:i] * 2**31 - _MM_EXPADJ_32: dst[i+31:i] = dst[i+31:i] * 2**32 - ESAC +FOR j := 0 to 7 + i := j*64 + IF k[j] + IF ((j & 1) == 0) + dst[i+63:i] := (a[i+63:i] * b[i+63:i]) + c[i+63:i] + ELSE + dst[i+63:i] := (a[i+63:i] * b[i+63:i]) - c[i+63:i] + FI + ELSE + dst[i+63:i] := c[i+63:i] + FI ENDFOR dst[MAX:512] := 0 - + + +
immintrin.h
 - - + Floating Point - AVX512F/KNCNI + AVX512F Arithmetic - - Finds the absolute value of each packed single-precision (32-bit) floating-point element in "v2", storing the results in "dst". + + + + + + + Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", alternatively subtract and add packed elements in "c" from/to the intermediate result, and store the results in "dst" using writemask "k" (elements are copied from "c" when the corresponding mask bit is not set). [round_note] -FOR j := 0 to 15 - i := j*32 - dst[i+31:i] := ABS(v2[i+31:i]) +FOR j := 0 to 7 + i := j*64 + IF k[j] + IF ((j & 1) == 0) + dst[i+63:i] := (a[i+63:i] * b[i+63:i]) + c[i+63:i] + ELSE + dst[i+63:i] := (a[i+63:i] * b[i+63:i]) - c[i+63:i] + FI + ELSE + dst[i+63:i] := c[i+63:i] + FI ENDFOR dst[MAX:512] := 0 - + + +
immintrin.h
 - - + Floating Point - AVX512F/KNCNI + AVX512F Arithmetic - - - - Finds the absolute value of each packed single-precision (32-bit) floating-point element in "v2", storing the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + + + + + Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", alternatively subtract and add packed elements in "c" from/to the intermediate result, and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*32 +FOR j := 0 to 7 + i := j*64 IF k[j] - dst[i+31:i] := ABS(v2[i+31:i]) + IF ((j & 1) == 0) + dst[i+63:i] := (a[i+63:i] * b[i+63:i]) + c[i+63:i] + ELSE + dst[i+63:i] := (a[i+63:i] * b[i+63:i]) - c[i+63:i] + FI ELSE - dst[i+31:i] := src[i+31:i] + dst[i+63:i] := a[i+63:i] FI ENDFOR dst[MAX:512] := 0 - + + +
immintrin.h
 - - + Floating Point - AVX512F/KNCNI + AVX512F Arithmetic - - Finds the absolute value of each packed double-precision (64-bit) floating-point element in "v2", storing the results in "dst". + + + + + + + Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", alternatively subtract and add packed elements in "c" from/to the intermediate result, and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). [round_note] FOR j := 0 to 7 i := j*64 - dst[i+63:i] := ABS(v2[i+63:i]) + IF k[j] + IF ((j & 1) == 0) + dst[i+63:i] := (a[i+63:i] * b[i+63:i]) + c[i+63:i] + ELSE + dst[i+63:i] := (a[i+63:i] * b[i+63:i]) - c[i+63:i] + FI + ELSE + dst[i+63:i] := a[i+63:i] + FI ENDFOR dst[MAX:512] := 0 - + + +
immintrin.h
 - - + Floating Point - AVX512F/KNCNI + AVX512F Arithmetic - - - - Finds the absolute value of each packed double-precision (64-bit) floating-point element in "v2", storing the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + + + + + Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", alternatively subtract and add packed elements in "c" from/to the intermediate result, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). FOR j := 0 to 7 i := j*64 IF k[j] - dst[i+63:i] := ABS(v2[i+63:i]) + IF ((j & 1) == 0) + dst[i+63:i] := (a[i+63:i] * b[i+63:i]) + c[i+63:i] + ELSE + dst[i+63:i] := (a[i+63:i] * b[i+63:i]) - c[i+63:i] + FI ELSE - dst[i+63:i] := src[i+63:i] + dst[i+63:i] := 0 FI ENDFOR dst[MAX:512] := 0 - + + +
immintrin.h
 - + Floating Point - KNCNI - Elementary Math Functions - - Compute the base-2 logarithm of packed single-precision (32-bit) floating-point elements in "a" with absolute error of 2^(-23) and store the results in "dst". + AVX512F + Arithmetic + + + + + + + Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", alternatively subtract and add packed elements in "c" from/to the intermediate result, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + [round_note] -FOR j := 0 to 15 - i := j*32 - dst[i+31:i] := Log2ae23(a[i+31:i]) +FOR j := 0 to 7 + i := j*64 + IF k[j] + IF ((j & 1) == 0) + dst[i+63:i] := (a[i+63:i] * b[i+63:i]) + c[i+63:i] + ELSE + dst[i+63:i] := (a[i+63:i] * b[i+63:i]) - c[i+63:i] + FI + ELSE + dst[i+63:i] := 0 + FI ENDFOR dst[MAX:512] := 0 - + + +
immintrin.h
 - + Floating Point - KNCNI - Elementary Math Functions - - - - Compute the base-2 logarithm of packed single-precision (32-bit) floating-point elements in "a" with absolute error of 2^(-23) and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + AVX512F + Arithmetic + + + + + Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", alternatively subtract and add packed elements in "c" from/to the intermediate result, and store the results in "dst". FOR j := 0 to 15 i := j*32 - IF k[j] - dst[i+31:i] := Log2ae23(a[i+31:i]) + IF ((j & 1) == 0) + dst[i+31:i] := (a[i+31:i] * b[i+31:i]) + c[i+31:i] ELSE - dst[i+31:i] := src[i+31:i] + dst[i+31:i] := (a[i+31:i] * b[i+31:i]) - c[i+31:i] FI ENDFOR dst[MAX:512] := 0 - + + +
immintrin.h
 - - Integer - KNCNI + + Floating Point + AVX512F Arithmetic - - - - Multiply packed 32-bit integer elements in "a" and "b", add the intermediate result to packed elements in "c" and store the results in "dst". + + + + + + Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", alternatively subtract and add packed elements in "c" from/to the intermediate result, and store the results in "dst". + [round_note] FOR j := 0 to 15 i := j*32 - dst[i+31:i] := (a[i+31:i] * b[i+31:i]) + c[i+31:i] + IF ((j & 1) == 0) + dst[i+31:i] := (a[i+31:i] * b[i+31:i]) + c[i+31:i] + ELSE + dst[i+31:i] := (a[i+31:i] * b[i+31:i]) - c[i+31:i] + FI ENDFOR dst[MAX:512] := 0 - + + +
immintrin.h
 - - Integer - KNCNI + + Floating Point + AVX512F Arithmetic - - - - - Multiply packed 32-bit integer elements in "a" and "b", add the intermediate result to packed elements in "c" and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). + + + + + + Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", alternatively subtract and add packed elements in "c" from/to the intermediate result, and store the results in "dst" using writemask "k" (elements are copied from "c" when the corresponding mask bit is not set). FOR j := 0 to 15 i := j*32 IF k[j] - dst[i+31:i] := (a[i+31:i] * b[i+31:i]) + c[i+31:i] + IF ((j & 1) == 0) + dst[i+31:i] := (a[i+31:i] * b[i+31:i]) + c[i+31:i] + ELSE + dst[i+31:i] := (a[i+31:i] * b[i+31:i]) - c[i+31:i] + FI ELSE - dst[i+31:i] := a[i+31:i] + dst[i+31:i] := c[i+31:i] FI ENDFOR dst[MAX:512] := 0 - + + +
immintrin.h
 - - Integer - KNCNI + + Floating Point + AVX512F Arithmetic - - - - - Multiply packed 32-bit integer elements in "a" and "b", add the intermediate result to packed elements in "c" and store the results in "dst" using writemask "k" (elements are copied from "c" when the corresponding mask bit is not set). + + + + + + + Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", alternatively subtract and add packed elements in "c" from/to the intermediate result, and store the results in "dst" using writemask "k" (elements are copied from "c" when the corresponding mask bit is not set). [round_note] FOR j := 0 to 15 i := j*32 IF k[j] - dst[i+31:i] := (a[i+31:i] * b[i+31:i]) + c[i+31:i] + IF ((j & 1) == 0) + dst[i+31:i] := (a[i+31:i] * b[i+31:i]) + c[i+31:i] + ELSE + dst[i+31:i] := (a[i+31:i] * b[i+31:i]) - c[i+31:i] + FI ELSE dst[i+31:i] := c[i+31:i] FI ENDFOR dst[MAX:512] := 0 - + + +
immintrin.h
 - - Integer - KNCNI + + Floating Point + AVX512F Arithmetic - - - Multiply packed 32-bit integer elements in each 4-element set of "a" and by element 1 of the corresponding 4-element set from "b", add the intermediate result to element 0 of the corresponding 4-element set from "b", and store the results in "dst". + + + + + + Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", alternatively subtract and add packed elements in "c" from/to the intermediate result, and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). FOR j := 0 to 15 i := j*32 - base := (j & ~0x3) * 32 - scale[31:0] := b[base+63:base+32] - bias[31:0] := b[base+31:base] - dst[i+31:i] := (a[i+31:i] * scale[31:0]) + bias[31:0] + IF k[j] + IF ((j & 1) == 0) + dst[i+31:i] := (a[i+31:i] * b[i+31:i]) + c[i+31:i] + ELSE + dst[i+31:i] := (a[i+31:i] * b[i+31:i]) - c[i+31:i] + FI + ELSE + dst[i+31:i] := a[i+31:i] + FI ENDFOR dst[MAX:512] := 0 - + + +
immintrin.h
 - - Integer - KNCNI + + Floating Point + AVX512F Arithmetic - - - - - Multiply packed 32-bit integer elements in each 4-element set of "a" and by element 1 of the corresponding 4-element set from "b", add the intermediate result to element 0 of the corresponding 4-element set from "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + + + + + + Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", alternatively subtract and add packed elements in "c" from/to the intermediate result, and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). + [round_note] FOR j := 0 to 15 i := j*32 IF k[j] - base := (j & ~0x3) * 32 - scale[31:0] := b[base+63:base+32] - bias[31:0] := b[base+31:base] - dst[i+31:i] := (a[i+31:i] * scale[31:0]) + bias[31:0] + IF ((j & 1) == 0) + dst[i+31:i] := (a[i+31:i] * b[i+31:i]) + c[i+31:i] + ELSE + dst[i+31:i] := (a[i+31:i] * b[i+31:i]) - c[i+31:i] + FI ELSE - dst[i+31:i] := src[i+31:i] + dst[i+31:i] := a[i+31:i] FI ENDFOR dst[MAX:512] := 0 - + + +
immintrin.h
 - + Floating Point - KNCNI + AVX512F Arithmetic - - - - Multiply packed single-precision (32-bit) floating-point elements in each 4-element set of "a" and by element 1 of the corresponding 4-element set from "b", add the intermediate result to element 0 of the corresponding 4-element set from "b", and store the results in "dst". - [round_note] - + + + + + + Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", alternatively subtract and add packed elements in "c" from/to the intermediate result, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). FOR j := 0 to 15 i := j*32 - base := (j & ~0x3) * 32 - scale[31:0] := b[base+63:base+32] - bias[31:0] := b[base+31:base] - dst[i+31:i] := (a[i+31:i] * scale[31:0]) + bias[31:0] + IF k[j] + IF ((j & 1) == 0) + dst[i+31:i] := (a[i+31:i] * b[i+31:i]) + c[i+31:i] + ELSE + dst[i+31:i] := (a[i+31:i] * b[i+31:i]) - c[i+31:i] + FI + ELSE + dst[i+31:i] := 0 + FI ENDFOR dst[MAX:512] := 0 - + + +
immintrin.h
 - + Floating Point - KNCNI + AVX512F Arithmetic - - - - - - Multiply packed single-precision (32-bit) floating-point elements in each 4-element set of "a" and by element 1 of the corresponding 4-element set from "b", add the intermediate result to element 0 of the corresponding 4-element set from "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - [round_note] - + + + + + + + Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", alternatively subtract and add packed elements in "c" from/to the intermediate result, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + [round_note] FOR j := 0 to 15 i := j*32 IF k[j] - base := (j & ~0x3) * 32 - scale[31:0] := b[base+63:base+32] - bias[31:0] := b[base+31:base] - dst[i+31:i] := (a[i+31:i] * scale[31:0]) + bias[31:0] + IF ((j & 1) == 0) + dst[i+31:i] := (a[i+31:i] * b[i+31:i]) + c[i+31:i] + ELSE + dst[i+31:i] := (a[i+31:i] * b[i+31:i]) - c[i+31:i] + FI ELSE - dst[i+31:i] := src[i+31:i] + dst[i+31:i] := 0 FI ENDFOR dst[MAX:512] := 0 - + + +
immintrin.h
 - + Floating Point - KNCNI - Special Math Functions - - - Determines the maximum of the absolute elements of each pair of corresponding elements of packed single-precision (32-bit) floating-point elements in "a" and "b", storing the results in "dst". + AVX512F + Arithmetic + + + + + + Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", add the negated intermediate result to packed elements in "c", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*32 - dst[i+31:i] := FpMax(Abs(a[i+31:i]), Abs(b[i+31:i])) -ENDFOR +FOR j := 0 to 7 + i := j*64 + IF k[j] + dst[i+63:i] := -(a[i+63:i] * b[i+63:i]) + c[i+63:i] + ELSE + dst[i+63:i] := 0 + FI +ENDFOR dst[MAX:512] := 0 - + + +
immintrin.h
 - + Floating Point - KNCNI - Special Math Functions - - - - - Determines the maximum of the absolute elements of each pair of corresponding elements of packed single-precision (32-bit) floating-point elements in "a" and "b", storing the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + AVX512F + Arithmetic + + + + + + + Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", add the negated intermediate result to packed elements in "c", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). [round_note] -FOR j := 0 to 15 - i := j*32 +FOR j := 0 to 7 + i := j*64 IF k[j] - dst[i+31:i] := FpMax(Abs(a[i+31:i]), Abs(b[i+31:i])) + dst[i+63:i] := -(a[i+63:i] * b[i+63:i]) + c[i+63:i] ELSE - dst[i+31:i] := src[i+31:i] + dst[i+63:i] := 0 FI -ENDFOR +ENDFOR dst[MAX:512] := 0 - + + +
immintrin.h
 - + Floating Point - KNCNI - Special Math Functions - - - Determines the maximum of each pair of corresponding elements in packed single-precision (32-bit) floating-point elements in "a" and "b", storing the results in "dst". + AVX512F + Arithmetic + + + + + + Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", add the negated intermediate result to packed elements in "c", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). FOR j := 0 to 15 i := j*32 - dst[i+31:i] := FpMax(a[i+31:i], b[i+31:i]) -ENDFOR + IF k[j] + dst[i+31:i] := -(a[i+31:i] * b[i+31:i]) + c[i+31:i] + ELSE + dst[i+31:i] := 0 + FI +ENDFOR dst[MAX:512] := 0 - + + +
immintrin.h
 - + Floating Point - KNCNI - Special Math Functions - - - - - Determines the maximum of each pair of corresponding elements of packed single-precision (32-bit) floating-point elements in "a" and "b", storing the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + AVX512F + Arithmetic + + + + + + + Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", add the negated intermediate result to packed elements in "c", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). [round_note] FOR j := 0 to 15 i := j*32 IF k[j] - dst[i+31:i] := FpMax(a[i+31:i], b[i+31:i]) + dst[i+31:i] := -(a[i+31:i] * b[i+31:i]) + c[i+31:i] ELSE - dst[i+31:i] := src[i+31:i] + dst[i+31:i] := 0 FI -ENDFOR +ENDFOR dst[MAX:512] := 0 - + + +
immintrin.h
 - + Floating Point - KNCNI - Special Math Functions - - - Determines the maximum of the absolute elements of each pair of corresponding elements of packed single-precision (32-bit) floating-point elements in "a" and "b", storing the results in "dst". + AVX512F + Arithmetic + + + + + + Multiply the lower double-precision (64-bit) floating-point elements in "a" and "b", and add the negated intermediate result to the lower element in "c". Store the result in the lower element of "dst", and copy the upper element from "a" to the upper element of "dst". + [round_note] -FOR j := 0 to 15 - i := j*32 - dst[i+31:i] := FpMax(Abs(a[i+31:i]), Abs(b[i+31:i])) -ENDFOR -dst[MAX:512] := 0 +dst[63:0] := -(a[63:0] * b[63:0]) + c[63:0] +dst[127:64] := a[127:64] +dst[MAX:128] := 0 - + + +
immintrin.h
 - + Floating Point - KNCNI - Special Math Functions - - - - - Determines the maximum of the absolute elements of each pair of corresponding elements of packed single-precision (32-bit) floating-point elements in "a" and "b", storing the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + AVX512F + Arithmetic + + + + + + + Multiply the lower double-precision (64-bit) floating-point elements in "a" and "b", and add the negated intermediate result to the lower element in "c". Store the result in the lower element of "dst" using writemask "k" (the element is copied from "c" when mask bit 0 is not set), and copy the upper element from "c" to the upper element of "dst". + [round_note] -FOR j := 0 to 15 - i := j*32 - IF k[j] - dst[i+31:i] := FpMax(Abs(a[i+31:i]), Abs(b[i+31:i])) - ELSE - dst[i+31:i] := src[i+31:i] - FI -ENDFOR -dst[MAX:512] := 0 +IF k[0] + dst[63:0] := -(a[63:0] * b[63:0]) + c[63:0] +ELSE + dst[63:0] := c[63:0] +FI +dst[127:64] := c[127:64] +dst[MAX:128] := 0 - + + +
immintrin.h
 - + Floating Point - KNCNI - Special Math Functions - - - Determines the maximum of each pair of corresponding elements in packed double-precision (64-bit) floating-point elements in "a" and "b", storing the results in "dst". + AVX512F + Arithmetic + + + + + + Multiply the lower double-precision (64-bit) floating-point elements in "a" and "b", and add the negated intermediate result to the lower element in "c". Store the result in the lower element of "dst" using writemask "k" (the element is copied from "c" when mask bit 0 is not set), and copy the upper element from "c" to the upper element of "dst". + +IF k[0] + dst[63:0] := -(a[63:0] * b[63:0]) + c[63:0] +ELSE + dst[63:0] := c[63:0] +FI +dst[127:64] := c[127:64] +dst[MAX:128] := 0 + + + + +
immintrin.h
+ + + Floating Point + AVX512F + Arithmetic + + + + + + + Multiply the lower double-precision (64-bit) floating-point elements in "a" and "b", and add the negated intermediate result to the lower element in "c". Store the result in the lower element of "dst" using writemask "k" (the element is copied from "a" when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". + [round_note] + +IF k[0] + dst[63:0] := -(a[63:0] * b[63:0]) + c[63:0] +ELSE + dst[63:0] := a[63:0] +FI +dst[127:64] := a[127:64] +dst[MAX:128] := 0 + + + + +
immintrin.h
+ + + Floating Point + AVX512F + Arithmetic + + + + + + Multiply the lower double-precision (64-bit) floating-point elements in "a" and "b", and add the negated intermediate result to the lower element in "c". Store the result in the lower element of "dst" using writemask "k" (the element is copied from "a" when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". + +IF k[0] + dst[63:0] := -(a[63:0] * b[63:0]) + c[63:0] +ELSE + dst[63:0] := a[63:0] +FI +dst[127:64] := a[127:64] +dst[MAX:128] := 0 + + + + +
immintrin.h
+ + + Floating Point + AVX512F + Arithmetic + + + + + + + Multiply the lower double-precision (64-bit) floating-point elements in "a" and "b", and add the negated intermediate result to the lower element in "c". Store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". + [round_note] + +IF k[0] + dst[63:0] := -(a[63:0] * b[63:0]) + c[63:0] +ELSE + dst[63:0] := 0 +FI +dst[127:64] := a[127:64] +dst[MAX:128] := 0 + + + + +
immintrin.h
+ + + Floating Point + AVX512F + Arithmetic + + + + + + Multiply the lower double-precision (64-bit) floating-point elements in "a" and "b", and add the negated intermediate result to the lower element in "c". Store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". + +IF k[0] + dst[63:0] := -(a[63:0] * b[63:0]) + c[63:0] +ELSE + dst[63:0] := 0 +FI +dst[127:64] := a[127:64] +dst[MAX:128] := 0 + + + + +
immintrin.h
+ + + Floating Point + AVX512F + Arithmetic + + + + + + Multiply the lower single-precision (32-bit) floating-point elements in "a" and "b", and add the negated intermediate result to the lower element in "c". Store the result in the lower element of "dst", and copy the upper 3 packed elements from "a" to the upper elements of "dst". + [round_note] + +dst[31:0] := -(a[31:0] * b[31:0]) + c[31:0] +dst[127:32] := a[127:32] +dst[MAX:128] := 0 + + + + +
immintrin.h
+ + + Floating Point + AVX512F + Arithmetic + + + + + + + Multiply the lower single-precision (32-bit) floating-point elements in "a" and "b", and add the negated intermediate result to the lower element in "c". Store the result in the lower element of "dst" using writemask "k" (the element is copied from "c" when mask bit 0 is not set), and copy the upper 3 packed elements from "c" to the upper elements of "dst". + [round_note] + +IF k[0] + dst[31:0] := -(a[31:0] * b[31:0]) + c[31:0] +ELSE + dst[31:0] := c[31:0] +FI +dst[127:32] := c[127:32] +dst[MAX:128] := 0 + + + + +
immintrin.h
+ + + Floating Point + AVX512F + Arithmetic + + + + + + Multiply the lower single-precision (32-bit) floating-point elements in "a" and "b", and add the negated intermediate result to the lower element in "c". Store the result in the lower element of "dst" using writemask "k" (the element is copied from "c" when mask bit 0 is not set), and copy the upper 3 packed elements from "c" to the upper elements of "dst". + +IF k[0] + dst[31:0] := -(a[31:0] * b[31:0]) + c[31:0] +ELSE + dst[31:0] := c[31:0] +FI +dst[127:32] := c[127:32] +dst[MAX:128] := 0 + + + + +
immintrin.h
+ + + Floating Point + AVX512F + Arithmetic + + + + + + + Multiply the lower single-precision (32-bit) floating-point elements in "a" and "b", and add the negated intermediate result to the lower element in "c". Store the result in the lower element of "dst" using writemask "k" (the element is copied from "a" when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". + [round_note] + +IF k[0] + dst[31:0] := -(a[31:0] * b[31:0]) + c[31:0] +ELSE + dst[31:0] := a[31:0] +FI +dst[127:32] := a[127:32] +dst[MAX:128] := 0 + + + + +
immintrin.h
+ + + Floating Point + AVX512F + Arithmetic + + + + + + Multiply the lower single-precision (32-bit) floating-point elements in "a" and "b", and add the negated intermediate result to the lower element in "c". Store the result in the lower element of "dst" using writemask "k" (the element is copied from "a" when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". + +IF k[0] + dst[31:0] := -(a[31:0] * b[31:0]) + c[31:0] +ELSE + dst[31:0] := a[31:0] +FI +dst[127:32] := a[127:32] +dst[MAX:128] := 0 + + + + +
immintrin.h
+ + + Floating Point + AVX512F + Arithmetic + + + + + + + Multiply the lower single-precision (32-bit) floating-point elements in "a" and "b", and add the negated intermediate result to the lower element in "c". Store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". + [round_note] + +IF k[0] + dst[31:0] := -(a[31:0] * b[31:0]) + c[31:0] +ELSE + dst[31:0] := 0 +FI +dst[127:32] := a[127:32] +dst[MAX:128] := 0 + + + + +
immintrin.h
+ + + Floating Point + AVX512F + Arithmetic + + + + + + Multiply the lower single-precision (32-bit) floating-point elements in "a" and "b", and add the negated intermediate result to the lower element in "c". Store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". + +IF k[0] + dst[31:0] := -(a[31:0] * b[31:0]) + c[31:0] +ELSE + dst[31:0] := 0 +FI +dst[127:32] := a[127:32] +dst[MAX:128] := 0 + + + + +
immintrin.h
+ + + Floating Point + AVX512F + Arithmetic + + + + + + Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", subtract packed elements in "c" from the negated intermediate result, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). FOR j := 0 to 7 i := j*64 - dst[i+63:i] := FpMax(a[i+63:i], b[i+63:i]) -ENDFOR + IF k[j] + dst[i+63:i] := -(a[i+63:i] * b[i+63:i]) - c[i+63:i] + ELSE + dst[i+63:i] := 0 + FI +ENDFOR dst[MAX:512] := 0 - + + +
immintrin.h
 - + Floating Point - KNCNI - Special Math Functions - - - - - Determines the maximum of each pair of corresponding elements of packed double-precision (64-bit) floating-point elements in "a" and "b", storing the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + AVX512F + Arithmetic + + + + + + + Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", subtract packed elements in "c" from the negated intermediate result, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). [round_note] FOR j := 0 to 7 i := j*64 IF k[j] - dst[i+63:i] := FpMax(a[i+63:i], b[i+63:i]) + dst[i+63:i] := -(a[i+63:i] * b[i+63:i]) - c[i+63:i] ELSE - dst[i+63:i] := src[i+63:i] + dst[i+63:i] := 0 FI -ENDFOR +ENDFOR dst[MAX:512] := 0 - + + +
immintrin.h
 - + Floating Point - KNCNI - Special Math Functions - - - Determines the minimum of each pair of corresponding elements in packed single-precision (32-bit) floating-point elements in "a" and "b", storing the results in "dst". + AVX512F + Arithmetic + + + + + + Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", subtract packed elements in "c" from the negated intermediate result, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). FOR j := 0 to 15 i := j*32 - dst[i+31:i] := FpMin(a[i+31:i], b[i+31:i]) -ENDFOR + IF k[j] + dst[i+31:i] := -(a[i+31:i] * b[i+31:i]) - c[i+31:i] + ELSE + dst[i+31:i] := 0 + FI +ENDFOR dst[MAX:512] := 0 - + + +
immintrin.h
 - + Floating Point - KNCNI - Special Math Functions - - - - - Determines the maximum of each pair of corresponding elements of packed single-precision (32-bit) floating-point elements in "a" and "b", storing the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + AVX512F + Arithmetic + + + + + + + Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", subtract packed elements in "c" from the negated intermediate result, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + [round_note] FOR j := 0 to 15 i := j*32 IF k[j] - dst[i+31:i] := FpMin(a[i+31:i], b[i+31:i]) + dst[i+31:i] := -(a[i+31:i] * b[i+31:i]) - c[i+31:i] ELSE - dst[i+31:i] := src[i+31:i] + dst[i+31:i] := 0 FI -ENDFOR +ENDFOR dst[MAX:512] := 0 - + + +
immintrin.h
 - + Floating Point - KNCNI - Special Math Functions - - - Determines the minimum of each pair of corresponding elements in packed double-precision (64-bit) floating-point elements in "a" and "b", storing the results in "dst". + AVX512F + Arithmetic + + + + + + Multiply the lower double-precision (64-bit) floating-point elements in "a" and "b", and subtract the lower element in "c" from the negated intermediate result. Store the result in the lower element of "dst", and copy the upper element from "a" to the upper element of "dst". + [round_note] -FOR j := 0 to 7 - i := j*64 - dst[i+63:i] := FpMin(a[i+63:i], b[i+63:i]) -ENDFOR -dst[MAX:512] := 0 +dst[63:0] := -(a[63:0] * b[63:0]) - c[63:0] +dst[127:64] := a[127:64] +dst[MAX:128] := 0 - + + +
immintrin.h
 - + Floating Point - KNCNI - Special Math Functions - - - - - Determines the maximum of each pair of corresponding elements of packed double-precision (64-bit) floating-point elements in "a" and "b", storing the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + AVX512F + Arithmetic + + + + + + + Multiply the lower double-precision (64-bit) floating-point elements in "a" and "b", and subtract the lower element in "c" from the negated intermediate result. Store the result in the lower element of "dst" using writemask "k" (the element is copied from "c" when mask bit 0 is not set), and copy the upper element from "c" to the upper element of "dst". + [round_note] -FOR j := 0 to 7 - i := j*64 - IF k[j] - dst[i+63:i] := FpMin(a[i+63:i], b[i+63:i]) - ELSE - dst[i+63:i] := src[i+63:i] - FI -ENDFOR -dst[MAX:512] := 0 +IF k[0] + dst[63:0] := -(a[63:0] * b[63:0]) - c[63:0] +ELSE + dst[63:0] := c[63:0] +FI +dst[127:64] := c[127:64] +dst[MAX:128] := 0 - + + +
immintrin.h
 - - Integer - KNCNI + + Floating Point + AVX512F Arithmetic - - - Performs element-by-element multiplication between packed 32-bit integer elements in "a" and "b" and stores the high 32 bits of each result into "dst". + + + + + + Multiply the lower double-precision (64-bit) floating-point elements in "a" and "b", and subtract the lower element in "c" from the negated intermediate result. Store the result in the lower element of "dst" using writemask "k" (the element is copied from "c" when mask bit 0 is not set), and copy the upper element from "c" to the upper element of "dst". -FOR j := 0 to 15 - i := j*32 - dst[i+31:i] := (a[i+31:i] * b[i+31:i]) >> 32 -ENDFOR -dst[MAX:512] := 0 +IF k[0] + dst[63:0] := -(a[63:0] * b[63:0]) - c[63:0] +ELSE + dst[63:0] := c[63:0] +FI +dst[127:64] := c[127:64] +dst[MAX:128] := 0 - + + +
immintrin.h
 - - Integer - KNCNI + + Floating Point + AVX512F Arithmetic - - - - - Performs element-by-element multiplication between packed 32-bit integer elements in "a" and "b" and stores the high 32 bits of each result into "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + + + + + + Multiply the lower double-precision (64-bit) floating-point elements in "a" and "b", and subtract the lower element in "c" from the negated intermediate result. Store the result in the lower element of "dst" using writemask "k" (the element is copied from "c" when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". + [round_note] -FOR j := 0 to 15 - i := j*32 - IF k[j] - dst[i+31:i] := (a[i+31:i] * b[i+31:i]) >> 32 - ELSE - dst[i+31:i] := src[i+31:i] - FI -ENDFOR -dst[MAX:512] := 0 +IF k[0] + dst[63:0] := -(a[63:0] * b[63:0]) - c[63:0] +ELSE + dst[63:0] := a[63:0] +FI +dst[127:64] := a[127:64] +dst[MAX:128] := 0 - + + +
immintrin.h
 - - Integer - KNCNI + + Floating Point + AVX512F Arithmetic - - - Performs element-by-element multiplication between packed unsigned 32-bit integer elements in "a" and "b" and stores the high 32 bits of each result into "dst". + + + + + + Multiply the lower double-precision (64-bit) floating-point elements in "a" and "b", and subtract the lower element in "c" from the negated intermediate result. Store the result in the lower element of "dst" using writemask "k" (the element is copied from "c" when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". -FOR j := 0 to 15 - i := j*32 - dst[i+31:i] := (a[i+31:i] * b[i+31:i]) >> 32 -ENDFOR -dst[MAX:512] := 0 +IF k[0] + dst[63:0] := -(a[63:0] * b[63:0]) - c[63:0] +ELSE + dst[63:0] := a[63:0] +FI +dst[127:64] := a[127:64] +dst[MAX:128] := 0 - + + +
immintrin.h
 - - Integer - KNCNI + + Floating Point + AVX512F Arithmetic - - - - - Performs element-by-element multiplication between packed unsigned 32-bit integer elements in "a" and "b" and stores the high 32 bits of each result into "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + + + + + + Multiply the lower double-precision (64-bit) floating-point elements in "a" and "b", and subtract the lower element in "c" from the negated intermediate result. Store the result in "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". + [round_note] -FOR j := 0 to 15 - i := j*32 - IF k[j] - dst[i+31:i] := (a[i+31:i] * b[i+31:i]) >> 32 - ELSE - dst[i+31:i] := src[i+31:i] - FI -ENDFOR -dst[MAX:512] := 0 +IF k[0] + dst[63:0] := -(a[63:0] * b[63:0]) - c[63:0] +ELSE + dst[63:0] := 0 +FI +dst[127:64] := a[127:64] +dst[MAX:128] := 0 - + + +
immintrin.h
 - - Integer - KNCNI - Swizzle - - - Permutes 128-bit blocks of the packed 32-bit integer vector "a" using constant "imm8". The results are stored in "dst". + + Floating Point + AVX512F + Arithmetic + + + + + + Multiply the lower double-precision (64-bit) floating-point elements in "a" and "b", and subtract the lower element in "c" from the negated intermediate result. Store the result in "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". -SELECT4(src, control) { - CASE control[1:0] OF - 0: tmp[127:0] := src[127:0] - 1: tmp[127:0] := src[255:128] - 2: tmp[127:0] := src[383:256] - 3: tmp[127:0] := src[511:384] - ESAC - RETURN tmp[127:0] -} - -FOR j := 0 to 3 - i := j*128 - n := j*2 - dst[i+127:i] := SELECT4(a[511:0], imm8[n+1:n]) -ENDFOR -dst[MAX:512] := 0 +IF k[0] + dst[63:0] := -(a[63:0] * b[63:0]) - c[63:0] +ELSE + dst[63:0] := 0 +FI +dst[127:64] := a[127:64] +dst[MAX:128] := 0 - + + +
immintrin.h
 - - Integer - KNCNI - Swizzle - - - - - Permutes 128-bit blocks of the packed 32-bit integer vector "a" using constant "imm8". The results are stored in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Floating Point + AVX512F + Arithmetic + + + + + + Multiply the lower single-precision (32-bit) floating-point elements in "a" and "b", subtract the lower element in "c" from the negated intermediate result, store the result in the lower element of "dst", and copy the upper 3 packed elements from "a" to the upper elements of "dst". + [round_note] -SELECT4(src, control) { - CASE control[1:0] OF - 0: tmp[127:0] := src[127:0] - 1: tmp[127:0] := src[255:128] - 2: tmp[127:0] := src[383:256] - 3: tmp[127:0] := src[511:384] - ESAC - RETURN tmp[127:0] -} - -tmp[511:0] := 0 -FOR j := 0 to 4 - i := j*128 - n := j*2 - tmp[i+127:i] := SELECT4(a[511:0], imm8[n+1:n]) -ENDFOR -FOR j := 0 to 15 - IF k[j] - dst[i+31:i] := tmp[i+31:i] - ELSE - dst[i+31:i] := src[i+31:i] - FI -ENDFOR -dst[MAX:512] := 0 +dst[31:0] := -(a[31:0] * b[31:0]) - c[31:0] +dst[127:32] := a[127:32] +dst[MAX:128] := 0 - + + +
immintrin.h
 - + Floating Point - KNCNI - Elementary Math Functions - - Approximates the reciprocals of packed single-precision (32-bit) floating-point elements in "a" to 23 bits of precision, storing the results in "dst". + AVX512F + Arithmetic + + + + + + + Multiply the lower single-precision (32-bit) floating-point elements in "a" and "b", subtract the lower element in "c" from the negated intermediate result. Store the result in the lower element of "dst" using writemask "k" (the element is copied from "c" when mask bit 0 is not set), and copy the upper 3 packed elements from "c" to the upper elements of "dst". + [round_note] -FOR j := 0 to 15 - i := j*32 - dst[i+31:i] := APPROXIMATE(1.0/a[i+31:i]) -ENDFOR -dst[MAX:512] := 0 +IF k[0] + dst[31:0] := -(a[31:0] * b[31:0]) - c[31:0] +ELSE + dst[31:0] := c[31:0] +FI +dst[127:32] := c[127:32] +dst[MAX:128] := 0 - + + +
immintrin.h
 - + Floating Point - KNCNI - Elementary Math Functions - - - - Approximates the reciprocals of packed single-precision (32-bit) floating-point elements in "a" to 23 bits of precision, storing the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + AVX512F + Arithmetic + + + + + + Multiply the lower single-precision (32-bit) floating-point elements in "a" and "b", and subtract the lower element in "c" from the negated intermediate result. Store the result in the lower element of "dst" using writemask "k" (the element is copied from "c" when mask bit 0 is not set), and copy the upper 3 packed elements from "c" to the upper elements of "dst". -FOR j := 0 to 15 - i := j*32 - IF k[j] - dst[i+31:i] := APPROXIMATE(1.0/a[i+31:i]) - ELSE - dst[i+31:i] := src[i+31:i] - FI -ENDFOR -dst[MAX:512] := 0 +IF k[0] + dst[31:0] := -(a[31:0] * b[31:0]) - c[31:0] +ELSE + dst[31:0] := c[31:0] +FI +dst[127:32] := c[127:32] +dst[MAX:128] := 0 - + + +
immintrin.h
 - + Floating Point - KNCNI - Convert - - - - Round the packed single-precision (32-bit) floating-point elements in "a" to the nearest integer value using "expadj" and in the direction of "rounding", and store the results as packed single-precision floating-point elements in "dst". - [round_note] - + AVX512F + Arithmetic + + + + + + + Multiply the lower single-precision (32-bit) floating-point elements in "a" and "b", and subtract the lower element in "c" from the negated intermediate result. Store the result in the lower element of "dst" using writemask "k" (the element is copied from "c" when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". + [round_note] -FOR j := 0 to 15 - i := j*32 - dst[i+31:i] := ROUND(a[i+31:i]) - CASE expadj OF - _MM_EXPADJ_NONE: dst[i+31:i] = dst[i+31:i] * 2**0 - _MM_EXPADJ_4: dst[i+31:i] = dst[i+31:i] * 2**4 - _MM_EXPADJ_5: dst[i+31:i] = dst[i+31:i] * 2**5 - _MM_EXPADJ_8: dst[i+31:i] = dst[i+31:i] * 2**8 - _MM_EXPADJ_16: dst[i+31:i] = dst[i+31:i] * 2**16 - _MM_EXPADJ_24: dst[i+31:i] = dst[i+31:i] * 2**24 - _MM_EXPADJ_31: dst[i+31:i] = dst[i+31:i] * 2**31 - _MM_EXPADJ_32: dst[i+31:i] = dst[i+31:i] * 2**32 - ESAC -ENDFOR -dst[MAX:512] := 0 +IF k[0] + dst[31:0] := -(a[31:0] * b[31:0]) - c[31:0] +ELSE + dst[31:0] := a[31:0] +FI +dst[127:32] := a[127:32] +dst[MAX:128] := 0 - + + +
immintrin.h
 - + Floating Point - KNCNI - Convert - - - - - - Round the packed single-precision (32-bit) floating-point elements in "a" to the nearest integer value using "expadj" and in the direction of "rounding", and store the results as packed single-precision floating-point elements in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - [round_note] - + AVX512F + Arithmetic + + + + + + Multiply the lower single-precision (32-bit) floating-point elements in "a" and "b", and subtract the lower element in "c" from the negated intermediate result. Store the result in the lower element of "dst" using writemask "k" (the element is copied from "c" when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". -FOR j := 0 to 15 - i := j*32 - IF k[j] - dst[i+31:i] := ROUND(a[i+31:i]) - CASE expadj OF - _MM_EXPADJ_NONE: dst[i+31:i] = dst[i+31:i] * 2**0 - _MM_EXPADJ_4: dst[i+31:i] = dst[i+31:i] * 2**4 - _MM_EXPADJ_5: dst[i+31:i] = dst[i+31:i] * 2**5 - _MM_EXPADJ_8: dst[i+31:i] = dst[i+31:i] * 2**8 - _MM_EXPADJ_16: dst[i+31:i] = dst[i+31:i] * 2**16 - _MM_EXPADJ_24: dst[i+31:i] = dst[i+31:i] * 2**24 - _MM_EXPADJ_31: dst[i+31:i] = dst[i+31:i] * 2**31 - _MM_EXPADJ_32: dst[i+31:i] = dst[i+31:i] * 2**32 - ESAC - ELSE - dst[i+31:i] := src[i+31:i] - FI -ENDFOR -dst[MAX:512] := 0 +IF k[0] + dst[31:0] := -(a[31:0] * b[31:0]) - c[31:0] +ELSE + dst[31:0] := a[31:0] +FI +dst[127:32] := a[127:32] +dst[MAX:128] := 0 - + + +
immintrin.h
 - + Floating Point - KNCNI - Miscellaneous - - - - Performs element-by-element rounding of packed single-precision (32-bit) floating-point elements in "a" using "expadj" and in the direction of "rounding" and stores results in "dst". - [round_note] - + AVX512F + Arithmetic + + + + + + + Multiply the lower single-precision (32-bit) floating-point elements in "a" and "b", and subtract the lower element in "c" from the negated intermediate result. Store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". + [round_note] -FOR j := 0 to 15 - i := j*32 - dst[i+31:i] := ROUND(a[i+31:i]) - CASE expadj OF - _MM_EXPADJ_NONE: dst[i+31:i] = dst[i+31:i] * 2**0 - _MM_EXPADJ_4: dst[i+31:i] = dst[i+31:i] * 2**4 - _MM_EXPADJ_5: dst[i+31:i] = dst[i+31:i] * 2**5 - _MM_EXPADJ_8: dst[i+31:i] = dst[i+31:i] * 2**8 - _MM_EXPADJ_16: dst[i+31:i] = dst[i+31:i] * 2**16 - _MM_EXPADJ_24: dst[i+31:i] = dst[i+31:i] * 2**24 - _MM_EXPADJ_31: dst[i+31:i] = dst[i+31:i] * 2**31 - _MM_EXPADJ_32: dst[i+31:i] = dst[i+31:i] * 2**32 - ESAC -ENDFOR -dst[MAX:512] := 0 +IF k[0] + dst[31:0] := -(a[31:0] * b[31:0]) - c[31:0] +ELSE + dst[31:0] := 0 +FI +dst[127:32] := a[127:32] +dst[MAX:128] := 0 - + + +
immintrin.h
 - + Floating Point - KNCNI - Miscellaneous - - - - - - Performs element-by-element rounding of packed single-precision (32-bit) floating-point elements in "a" using "expadj" and in the direction of "rounding" and stores results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - [round_note] - + AVX512F + Arithmetic + + + + + + Multiply the lower single-precision (32-bit) floating-point elements in "a" and "b", and subtract the lower element in "c" from the negated intermediate result. Store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". -FOR j := 0 to 15 - i := j*32 - IF k[j] - dst[i+31:i] := ROUND(a[i+31:i]) - CASE expadj OF - _MM_EXPADJ_NONE: dst[i+31:i] = dst[i+31:i] * 2**0 - _MM_EXPADJ_4: dst[i+31:i] = dst[i+31:i] * 2**4 - _MM_EXPADJ_5: dst[i+31:i] = dst[i+31:i] * 2**5 - _MM_EXPADJ_8: dst[i+31:i] = dst[i+31:i] * 2**8 - _MM_EXPADJ_16: dst[i+31:i] = dst[i+31:i] * 2**16 - _MM_EXPADJ_24: dst[i+31:i] = dst[i+31:i] * 2**24 - _MM_EXPADJ_31: dst[i+31:i] = dst[i+31:i] * 2**31 - _MM_EXPADJ_32: dst[i+31:i] = dst[i+31:i] * 2**32 - ESAC - ELSE - dst[i+31:i] := src[i+31:i] - FI -ENDFOR -dst[MAX:512] := 0 +IF k[0] + dst[31:0] := -(a[31:0] * b[31:0]) - c[31:0] +ELSE + dst[31:0] := 0 +FI +dst[127:32] := a[127:32] +dst[MAX:128] := 0 - + + +
immintrin.h
 - + Floating Point - KNCNI - Miscellaneous - - - - Performs element-by-element rounding of packed double-precision (64-bit) floating-point elements in "a" using "expadj" and in the direction of "rounding" and stores results in "dst". - [round_note] - + AVX512F + Load + + + + + Gather double-precision (64-bit) floating-point elements from memory using 32-bit indices. 64-bit elements are loaded from addresses starting at "base_addr" and offset by each 32-bit element in "vindex" (each index is scaled by the factor in "scale"). Gathered elements are merged into "dst". "scale" should be 1, 2, 4 or 8. FOR j := 0 to 7 i := j*64 - dst[i+63:i] := ROUND(a[i+63:i]) - CASE expadj OF - _MM_EXPADJ_NONE: dst[i+31:i] = dst[i+31:i] * 2**0 - _MM_EXPADJ_4: dst[i+31:i] = dst[i+31:i] * 2**4 - _MM_EXPADJ_5: dst[i+31:i] = dst[i+31:i] * 2**5 - _MM_EXPADJ_8: dst[i+31:i] = dst[i+31:i] * 2**8 - _MM_EXPADJ_16: dst[i+31:i] = dst[i+31:i] * 2**16 - _MM_EXPADJ_24: dst[i+31:i] = dst[i+31:i] * 2**24 - _MM_EXPADJ_31: dst[i+31:i] = dst[i+31:i] * 2**31 - _MM_EXPADJ_32: dst[i+31:i] = dst[i+31:i] * 2**32 - ESAC + m := j*32 + addr := base_addr + SignExtend64(vindex[m+31:m]) * ZeroExtend64(scale) * 8 + dst[i+63:i] := MEM[addr+63:addr] ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - KNCNI - Miscellaneous - - - - - - Performs element-by-element rounding of packed double-precision (64-bit) floating-point elements in "a" using "expadj" and in the direction of "rounding" and stores results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - [round_note] - + AVX512F + Load + + + + + + + Gather double-precision (64-bit) floating-point elements from memory using 32-bit indices. 64-bit elements are loaded from addresses starting at "base_addr" and offset by each 32-bit element in "vindex" (each index is scaled by the factor in "scale"). Gathered elements are merged into "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). "scale" should be 1, 2, 4 or 8. FOR j := 0 to 7 i := j*64 + m := j*32 IF k[j] - dst[i+63:i] := ROUND(a[i+63:i]) - CASE expadj OF - _MM_EXPADJ_NONE: dst[i+31:i] = dst[i+31:i] * 2**0 - _MM_EXPADJ_4: dst[i+31:i] = dst[i+31:i] * 2**4 - _MM_EXPADJ_5: dst[i+31:i] = dst[i+31:i] * 2**5 - _MM_EXPADJ_8: dst[i+31:i] = dst[i+31:i] * 2**8 - _MM_EXPADJ_16: dst[i+31:i] = dst[i+31:i] * 2**16 - _MM_EXPADJ_24: dst[i+31:i] = dst[i+31:i] * 2**24 - _MM_EXPADJ_31: dst[i+31:i] = dst[i+31:i] * 2**31 - _MM_EXPADJ_32: dst[i+31:i] = dst[i+31:i] * 2**32 - ESAC + addr := base_addr + SignExtend64(vindex[m+31:m]) * ZeroExtend64(scale) * 8 + dst[i+63:i] := MEM[addr+63:addr] ELSE dst[i+63:i] := src[i+63:i] FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - KNCNI - Elementary Math Functions - - Calculates the reciprocal square root of packed single-precision (32-bit) floating-point elements in "a" to 23 bits of accuracy and stores the result in "dst". + AVX512F + Load + + + + + Gather double-precision (64-bit) floating-point elements from memory using 64-bit indices. 64-bit elements are loaded from addresses starting at "base_addr" and offset by each 64-bit element in "vindex" (each index is scaled by the factor in "scale"). Gathered elements are merged into "dst". "scale" should be 1, 2, 4 or 8. -FOR j := 0 to 15 - i := j*32 - dst[i+31:i] := Sqrt(1.0 / a[i+31:i]) +FOR j := 0 to 7 + i := j*64 + m := j*64 + addr := base_addr + vindex[m+63:m] * ZeroExtend64(scale) * 8 + dst[i+63:i] := MEM[addr+63:addr] ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - KNCNI - Elementary Math Functions - - - - Calculates the reciprocal square root of packed single-precision (32-bit) floating-point elements in "a" to 23 bits of accuracy and stores the result in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + AVX512F + Load + + + + + + + Gather double-precision (64-bit) floating-point elements from memory using 64-bit indices. 64-bit elements are loaded from addresses starting at "base_addr" and offset by each 64-bit element in "vindex" (each index is scaled by the factor in "scale"). Gathered elements are merged into "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). "scale" should be 1, 2, 4 or 8. -FOR j := 0 to 15 - i := j*32 +FOR j := 0 to 7 + i := j*64 + m := j*64 IF k[j] - dst[i+31:i] := Sqrt(1.0 / a[i+31:i]) + addr := base_addr + vindex[m+63:m] * ZeroExtend64(scale) * 8 + dst[i+63:i] := MEM[addr+63:addr] ELSE - dst[i+31:i] := src[i+31:i] + dst[i+63:i] := src[i+63:i] FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - KNCNI - Arithmetic - - - Scales each single-precision (32-bit) floating-point element in "a" by multiplying it by 2**exponent, where the exponent is the corresponding 32-bit integer element in "b", storing results in "dst". + AVX512F + Load + + + + + Gather single-precision (32-bit) floating-point elements from memory using 64-bit indices. 32-bit elements are loaded from addresses starting at "base_addr" and offset by each 64-bit element in "vindex" (each index is scaled by the factor in "scale"). Gathered elements are merged into "dst". "scale" should be 1, 2, 4 or 8. -FOR j := 0 to 15 +FOR j := 0 to 7 i := j*32 - dst[i+31:i] := a[i+31:i] * Pow(2, b[i+31:i]) + m := j*64 + addr := base_addr + vindex[m+63:m] * ZeroExtend64(scale) * 8 + dst[i+31:i] := MEM[addr+31:addr] ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Floating Point - KNCNI - Arithmetic - - - - - Scales each single-precision (32-bit) floating-point element in "a" by multiplying it by 2**exponent, where the exponenet is the corresponding 32-bit integer element in "b", storing results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + AVX512F + Load + + + + + + + Gather single-precision (32-bit) floating-point elements from memory using 64-bit indices. 32-bit elements are loaded from addresses starting at "base_addr" and offset by each 64-bit element in "vindex" (each index is scaled by the factor in "scale"). Gathered elements are merged into "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). "scale" should be 1, 2, 4 or 8. -FOR j := 0 to 15 +FOR j := 0 to 7 i := j*32 + m := j*64 IF k[j] - dst[i+31:i] := a[i+31:i] * Pow(2, b[i+31:i]) + addr := base_addr + vindex[m+63:m] * ZeroExtend64(scale) * 8 + dst[i+31:i] := MEM[addr+31:addr] ELSE dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Floating Point - KNCNI - Arithmetic - - - - Scales each single-precision (32-bit) floating-point element in "a" by multiplying it by 2**exponent, where the exponenet is the corresponding 32-bit integer element in "b", storing results in "dst". Intermediate elements are rounded using "rounding". - [round_note] - - -FOR j := 0 to 15 - i := j*32 - dst[i+31:i] := a[i+31:i] * Pow(2, b[i+31:i]) + AVX512F + Miscellaneous + + + + Convert the exponent of each packed double-precision (64-bit) floating-point element in "a" to a double-precision (64-bit) floating-point number representing the integer exponent, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). This intrinsic essentially calculates "floor(log2(x))" for each element. + FOR j := 0 to 7 + i := j*64 + IF k[j] + dst[i+63:i] := ConvertExpFP64(a[i+63:i]) + ELSE + dst[i+63:i] := 0 + FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - KNCNI - Arithmetic - - - - - - Scales each single-precision (32-bit) floating-point element in "a" by multiplying it by 2**exp, where the exp is the corresponding 32-bit integer element in "b", storing results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). Results are rounded using constant "rounding". - [round_note] - - -FOR j := 0 to 15 - i := j*32 + AVX512F + Miscellaneous + + + + + Convert the exponent of each packed double-precision (64-bit) floating-point element in "a" to a double-precision (64-bit) floating-point number representing the integer exponent, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). This intrinsic essentially calculates "floor(log2(x))" for each element. + [sae_note] + FOR j := 0 to 7 + i := j*64 IF k[j] - dst[i+31:i] := a[i+31:i] * Pow(2, b[i+31:i]) + dst[i+63:i] := ConvertExpFP64(a[i+63:i]) ELSE - dst[i+31:i] := src[i+31:i] + dst[i+63:i] := 0 FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point AVX512F - Trigonometry - - Compute the inverse cosine of packed double-precision (64-bit) floating-point elements in "a" expressed in radians, and store the results in "dst". - -FOR j := 0 to 7 - i := j*64 - dst[i+63:i] := ACOS(a[i+63:i]) + Miscellaneous + + + + Convert the exponent of each packed single-precision (32-bit) floating-point element in "a" to a single-precision (32-bit) floating-point number representing the integer exponent, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). This intrinsic essentially calculates "floor(log2(x))" for each element. + FOR j := 0 to 15 + i := j*32 + IF k[j] + dst[i+31:i] := ConvertExpFP32(a[i+31:i]) + ELSE + dst[i+31:i] := 0 + FI ENDFOR dst[MAX:512] := 0 +
immintrin.h
 - + Floating Point AVX512F - Trigonometry - - - - Compute the inverse cosine of packed double-precision (64-bit) floating-point elements in "a" expressed in radians, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -FOR j := 0 to 7 - i := j*64 + Miscellaneous + + + + + Convert the exponent of each packed single-precision (32-bit) floating-point element in "a" to a single-precision (32-bit) floating-point number representing the integer exponent, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). This intrinsic essentially calculates "floor(log2(x))" for each element. + [sae_note] + FOR j := 0 to 15 + i := j*32 IF k[j] - dst[i+63:i] := ACOS(a[i+63:i]) + dst[i+31:i] := ConvertExpFP32(a[i+31:i]) ELSE - dst[i+63:i] := src[i+63:i] + dst[i+31:i] := 0 FI ENDFOR dst[MAX:512] := 0 +
immintrin.h
 - + Floating Point AVX512F - Trigonometry - - Compute the inverse cosine of packed single-precision (32-bit) floating-point elements in "a" expressed in radians, and store the results in "dst". - -FOR j := 0 to 15 - i := j*32 - dst[i+31:i] := ACOS(a[i+31:i]) -ENDFOR -dst[MAX:512] := 0 + Miscellaneous + + + + + Convert the exponent of the lower double-precision (64-bit) floating-point element in "b" to a double-precision (64-bit) floating-point number representing the integer exponent, store the result in the lower element of "dst", and copy the upper element from "a" to the upper element of "dst". This intrinsic essentially calculates "floor(log2(x))" for the lower element. + [sae_note] + dst[63:0] := ConvertExpFP64(b[63:0]) +dst[127:64] := a[127:64] +dst[MAX:128] := 0 +
immintrin.h
 - + Floating Point AVX512F - Trigonometry - - - - Compute the inverse cosine of packed single-precision (32-bit) floating-point elements in "a" expressed in radians, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -FOR j := 0 to 15 - i := j*32 - IF k[j] - dst[i+31:i] := ACOS(a[i+31:i]) - ELSE - dst[i+31:i] := src[i+31:i] - FI -ENDFOR -dst[MAX:512] := 0 + Miscellaneous + + + + Convert the exponent of the lower double-precision (64-bit) floating-point element in "b" to a double-precision (64-bit) floating-point number representing the integer exponent, store the result in the lower element of "dst", and copy the upper element from "a" to the upper element of "dst". This intrinsic essentially calculates "floor(log2(x))" for the lower element. + dst[63:0] := ConvertExpFP64(b[63:0]) +dst[127:64] := a[127:64] +dst[MAX:128] := 0 +
immintrin.h
 - + Floating Point AVX512F - Trigonometry - - Compute the inverse hyperbolic cosine of packed double-precision (64-bit) floating-point elements in "a" expressed in radians, and store the results in "dst". - -FOR j := 0 to 7 - i := j*64 - dst[i+63:i] := ACOSH(a[i+63:i]) -ENDFOR -dst[MAX:512] := 0 + Miscellaneous + + + + + + + Convert the exponent of the lower double-precision (64-bit) floating-point element in "b" to a double-precision (64-bit) floating-point number representing the integer exponent, store the result in the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". This intrinsic essentially calculates "floor(log2(x))" for the lower element. + [sae_note] + IF k[0] + dst[63:0] := ConvertExpFP64(b[63:0]) +ELSE + dst[63:0] := src[63:0] +FI +dst[127:64] := a[127:64] +dst[MAX:128] := 0 +
immintrin.h
 - + Floating Point AVX512F - Trigonometry - - - - Compute the inverse hyperbolic cosine of packed double-precision (64-bit) floating-point elements in "a" expressed in radians, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -FOR j := 0 to 7 - i := j*64 - IF k[j] - dst[i+63:i] := ACOSH(a[i+63:i]) - ELSE - dst[i+63:i] := src[i+63:i] - FI -ENDFOR -dst[MAX:512] := 0 + Miscellaneous + + + + + + Convert the exponent of the lower double-precision (64-bit) floating-point element in "b" to a double-precision (64-bit) floating-point number representing the integer exponent, store the result in the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". This intrinsic essentially calculates "floor(log2(x))" for the lower element. + IF k[0] + dst[63:0] := ConvertExpFP64(b[63:0]) +ELSE + dst[63:0] := src[63:0] +FI +dst[127:64] := a[127:64] +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512F + Miscellaneous + + + + + + Convert the exponent of the lower double-precision (64-bit) floating-point element in "b" to a double-precision (64-bit) floating-point number representing the integer exponent, store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". This intrinsic essentially calculates "floor(log2(x))" for the lower element. + [sae_note] + IF k[0] + dst[63:0] := ConvertExpFP64(b[63:0]) +ELSE + dst[63:0] := 0 +FI +dst[127:64] := a[127:64] +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512F + Miscellaneous + + + + + Convert the exponent of the lower double-precision (64-bit) floating-point element in "b" to a double-precision (64-bit) floating-point number representing the integer exponent, store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". This intrinsic essentially calculates "floor(log2(x))" for the lower element. + IF k[0] + dst[63:0] := ConvertExpFP64(b[63:0]) +ELSE + dst[63:0] := 0 +FI +dst[127:64] := a[127:64] +dst[MAX:128] := 0 +
immintrin.h
 - + Floating Point AVX512F - Trigonometry - - Compute the inverse hyperbolic cosine of packed single-precision (32-bit) floating-point elements in "a" expressed in radians, and store the results in "dst". - -FOR j := 0 to 15 - i := j*32 - dst[i+31:i] := ACOSH(a[i+31:i]) -ENDFOR -dst[MAX:512] := 0 + Miscellaneous + + + + + Convert the exponent of the lower single-precision (32-bit) floating-point element in "b" to a single-precision (32-bit) floating-point number representing the integer exponent, store the result in the lower element of "dst", and copy the upper 3 packed elements from "a" to the upper elements of "dst". This intrinsic essentially calculates "floor(log2(x))" for the lower element. + [sae_note] + dst[31:0] := ConvertExpFP32(b[31:0]) +dst[127:32] := a[127:32] +dst[MAX:128] := 0 +
immintrin.h
 - + Floating Point AVX512F - Trigonometry - - - - Compute the inverse hyperbolic cosine of packed single-precision (32-bit) floating-point elements in "a" expressed in radians, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -FOR j := 0 to 15 - i := j*32 - IF k[j] - dst[i+31:i] := ACOSH(a[i+31:i]) - ELSE - dst[i+31:i] := src[i+31:i] - FI -ENDFOR -dst[MAX:512] := 0 + Miscellaneous + + + + Convert the exponent of the lower single-precision (32-bit) floating-point element in "b" to a single-precision (32-bit) floating-point number representing the integer exponent, store the result in the lower element of "dst", and copy the upper 3 packed elements from "a" to the upper elements of "dst". This intrinsic essentially calculates "floor(log2(x))" for the lower element. + dst[31:0] := ConvertExpFP32(b[31:0]) +dst[127:32] := a[127:32] +dst[MAX:128] := 0 +
immintrin.h
 - + Floating Point AVX512F - Trigonometry - - Compute the inverse sine of packed double-precision (64-bit) floating-point elements in "a" expressed in radians, and store the results in "dst". - -FOR j := 0 to 7 - i := j*64 - dst[i+63:i] := ASIN(a[i+63:i]) -ENDFOR -dst[MAX:512] := 0 + Miscellaneous + + + + + + + Convert the exponent of the lower single-precision (32-bit) floating-point element in "b" to a single-precision (32-bit) floating-point number representing the integer exponent, store the result in the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". This intrinsic essentially calculates "floor(log2(x))" for the lower element. + [sae_note] + IF k[0] + dst[31:0] := ConvertExpFP32(b[31:0]) +ELSE + dst[31:0] := src[31:0] +FI +dst[127:32] := a[127:32] +dst[MAX:128] := 0 +
immintrin.h
 - + Floating Point AVX512F - Trigonometry - - - - Compute the inverse sine of packed double-precision (64-bit) floating-point elements in "a" expressed in radians, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -FOR j := 0 to 7 - i := j*64 - IF k[j] - dst[i+63:i] := ASIN(a[i+63:i]) - ELSE - dst[i+63:i] := src[i+63:i] - FI -ENDFOR -dst[MAX:512] := 0 + Miscellaneous + + + + + + Convert the exponent of the lower single-precision (32-bit) floating-point element in "b" to a single-precision (32-bit) floating-point number representing the integer exponent, store the result in the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". This intrinsic essentially calculates "floor(log2(x))" for the lower element. + IF k[0] + dst[31:0] := ConvertExpFP32(b[31:0]) +ELSE + dst[31:0] := src[31:0] +FI +dst[127:32] := a[127:32] +dst[MAX:128] := 0 +
immintrin.h
 - + Floating Point AVX512F - Trigonometry - - Compute the inverse sine of packed single-precision (32-bit) floating-point elements in "a" expressed in radians, and store the results in "dst". - -FOR j := 0 to 15 - i := j*32 - dst[i+31:i] := ASIN(a[i+31:i]) -ENDFOR -dst[MAX:512] := 0 + Miscellaneous + + + + + + Convert the exponent of the lower single-precision (32-bit) floating-point element in "b" to a single-precision (32-bit) floating-point number representing the integer exponent, store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". This intrinsic essentially calculates "floor(log2(x))" for the lower element. + [sae_note] + IF k[0] + dst[31:0] := ConvertExpFP32(b[31:0]) +ELSE + dst[31:0] := 0 +FI +dst[127:32] := a[127:32] +dst[MAX:128] := 0 +
immintrin.h
 - + Floating Point AVX512F - Trigonometry - - - - Compute the inverse sine of packed single-precision (32-bit) floating-point elements in "a" expressed in radians, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -FOR j := 0 to 15 - i := j*32 - IF k[j] - dst[i+31:i] := ASIN(a[i+31:i]) - ELSE - dst[i+31:i] := src[i+31:i] - FI -ENDFOR -dst[MAX:512] := 0 + Miscellaneous + + + + + Convert the exponent of the lower single-precision (32-bit) floating-point element in "b" to a single-precision (32-bit) floating-point number representing the integer exponent, store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". This intrinsic essentially calculates "floor(log2(x))" for the lower element. + IF k[0] + dst[31:0] := ConvertExpFP32(b[31:0]) +ELSE + dst[31:0] := 0 +FI +dst[127:32] := a[127:32] +dst[MAX:128] := 0 +
immintrin.h
 - + Floating Point AVX512F - Trigonometry - - Compute the inverse hyperbolic sine of packed double-precision (64-bit) floating-point elements in "a" expressed in radians, and store the results in "dst". - -FOR j := 0 to 7 + Miscellaneous + + + + + + Normalize the mantissas of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). This intrinsic essentially calculates "±(2^k)*|x.significand|", where "k" depends on the interval range defined by "interv" and the sign depends on "sc" and the source sign. + [getmant_note] + FOR j := 0 to 7 i := j*64 - dst[i+63:i] := ASINH(a[i+63:i]) + IF k[j] + dst[i+63:i] := GetNormalizedMantissa(a[i+63:i], sc, interv) + ELSE + dst[i+63:i] := 0 + FI ENDFOR dst[MAX:512] := 0 +
immintrin.h
 - + Floating Point AVX512F - Trigonometry - - - - Compute the inverse hyperbolic sine of packed double-precision (64-bit) floating-point elements in "a" expressed in radians, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -FOR j := 0 to 7 + Miscellaneous + + + + + + + Normalize the mantissas of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). This intrinsic essentially calculates "±(2^k)*|x.significand|", where "k" depends on the interval range defined by "interv" and the sign depends on "sc" and the source sign. + [getmant_note][sae_note] + FOR j := 0 to 7 i := j*64 IF k[j] - dst[i+63:i] := ASINH(a[i+63:i]) + dst[i+63:i] := GetNormalizedMantissa(a[i+63:i], sc, interv) ELSE - dst[i+63:i] := src[i+63:i] + dst[i+63:i] := 0 FI ENDFOR dst[MAX:512] := 0 +
immintrin.h
 - + Floating Point AVX512F - Trigonometry - - Compute the inverse hyperbolic sine of packed single-precision (32-bit) floating-point elements in "a" expressed in radians, and store the results in "dst". - -FOR j := 0 to 15 + Miscellaneous + + + + + + Normalize the mantissas of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). This intrinsic essentially calculates "±(2^k)*|x.significand|", where "k" depends on the interval range defined by "interv" and the sign depends on "sc" and the source sign. + [getmant_note] + FOR j := 0 to 15 i := j*32 - dst[i+31:i] := ASINH(a[i+31:i]) + IF k[j] + dst[i+31:i] := GetNormalizedMantissa(a[i+31:i], sc, interv) + ELSE + dst[i+31:i] := 0 + FI ENDFOR dst[MAX:512] := 0 +
immintrin.h
 - + Floating Point AVX512F - Trigonometry - - - - Compute the inverse hyperbolic sine of packed single-precision (32-bit) floating-point elements in "a" expressed in radians, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -FOR j := 0 to 15 + Miscellaneous + + + + + + + Normalize the mantissas of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). This intrinsic essentially calculates "±(2^k)*|x.significand|", where "k" depends on the interval range defined by "interv" and the sign depends on "sc" and the source sign. + [getmant_note][sae_note] + FOR j := 0 to 15 i := j*32 IF k[j] - dst[i+31:i] := ASINH(a[i+31:i]) + dst[i+31:i] := GetNormalizedMantissa(a[i+31:i], sc, interv) ELSE - dst[i+31:i] := src[i+31:i] + dst[i+31:i] := 0 FI ENDFOR dst[MAX:512] := 0 +
immintrin.h
 - + Floating Point AVX512F - Trigonometry - - - Compute the inverse tangent of packed double-precision (64-bit) floating-point elements in "a" divided by packed elements in "b", and store the results in "dst" expressed in radians. - -FOR j := 0 to 7 - i := j*64 - dst[i+63:i] := ATAN(a[i+63:i] / b[i+63:i]) -ENDFOR -dst[MAX:512] := 0 + Miscellaneous + + + + + + + Normalize the mantissas of the lower double-precision (64-bit) floating-point element in "b", store the result in the lower element of "dst", and copy the upper element from "a" to the upper element of "dst". This intrinsic essentially calculates "±(2^k)*|x.significand|", where "k" depends on the interval range defined by "interv" and the sign depends on "sc" and the source sign. + [getmant_note][sae_note] + dst[63:0] := GetNormalizedMantissa(b[63:0], sc, interv) +dst[127:64] := a[127:64] +dst[MAX:128] := 0 +
immintrin.h
 - + Floating Point AVX512F - Trigonometry - - - - - Compute the inverse tangent of packed double-precision (64-bit) floating-point elements in "a" divided by packed elements in "b", and store the results in "dst" expressed in radians using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -FOR j := 0 to - i := j*64 - IF k[j] - dst[i+63:i] := ATAN(a[i+63:i] / b[i+63:i]) - ELSE - dst[i+63:i] := src[i+63:i] - FI -ENDFOR -dst[MAX:512] := 0 + Miscellaneous + + + + + + Normalize the mantissas of the lower double-precision (64-bit) floating-point element in "b", store the result in the lower element of "dst", and copy the upper element from "a" to the upper element of "dst". This intrinsic essentially calculates "±(2^k)*|x.significand|", where "k" depends on the interval range defined by "interv" and the sign depends on "sc" and the source sign. + [getmant_note] + dst[63:0] := GetNormalizedMantissa(b[63:0], sc, interv) +dst[127:64] := a[127:64] +dst[MAX:128] := 0 +
immintrin.h
 - + Floating Point AVX512F - Trigonometry - - - Compute the inverse tangent of packed single-precision (32-bit) floating-point elements in "a" divided by packed elements in "b", and store the results in "dst" expressed in radians. - -FOR j := 0 to 15 - i := j*32 - dst[i+31:i] := ATAN(a[i+31:i] / b[i+31:i]) -ENDFOR -dst[MAX:512] := 0 + Miscellaneous + + + + + + + + + Normalize the mantissas of the lower double-precision (64-bit) floating-point element in "b", store the result in the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". This intrinsic essentially calculates "±(2^k)*|x.significand|", where "k" depends on the interval range defined by "interv" and the sign depends on "sc" and the source sign. + [getmant_note][sae_note] + IF k[0] + dst[63:0] := GetNormalizedMantissa(b[63:0], sc, interv) +ELSE + dst[63:0] := src[63:0] +FI +dst[127:64] := a[127:64] +dst[MAX:128] := 0 +
immintrin.h
 - + Floating Point AVX512F - Trigonometry - - - - - Compute the inverse tangent of packed single-precision (32-bit) floating-point elements in "a" divided by packed elements in "b", and store the results in "dst" expressed in radians using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -FOR j := 0 to 15 - i := j*32 - IF k[j] - dst[i+31:i] := ATAN(a[i+31:i] / b[i+31:i]) - ELSE - dst[i+31:i] := src[i+31:i] - FI -ENDFOR -dst[MAX:512] := 0 + Miscellaneous + + + + + + + + Normalize the mantissas of the lower double-precision (64-bit) floating-point element in "b", store the result in the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". This intrinsic essentially calculates "±(2^k)*|x.significand|", where "k" depends on the interval range defined by "interv" and the sign depends on "sc" and the source sign. + [getmant_note] + IF k[0] + dst[63:0] := GetNormalizedMantissa(b[63:0], sc, interv) +ELSE + dst[63:0] := src[63:0] +FI +dst[127:64] := a[127:64] +dst[MAX:128] := 0 +
immintrin.h
 - + Floating Point AVX512F - Trigonometry - - Compute the inverse tangent of packed double-precision (64-bit) floating-point elements in "a" and store the results in "dst" expressed in radians. - -FOR j := 0 to 7 - i := j*64 - dst[i+63:i] := ATAN(a[i+63:i]) -ENDFOR -dst[MAX:512] := 0 + Miscellaneous + + + + + + + + Normalize the mantissas of the lower double-precision (64-bit) floating-point element in "b", store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". This intrinsic essentially calculates "±(2^k)*|x.significand|", where "k" depends on the interval range defined by "interv" and the sign depends on "sc" and the source sign. + [getmant_note][sae_note] + IF k[0] + dst[63:0] := GetNormalizedMantissa(b[63:0], sc, interv) +ELSE + dst[63:0] := 0 +FI +dst[127:64] := a[127:64] +dst[MAX:128] := 0 +
immintrin.h
 - + Floating Point AVX512F - Trigonometry - - - - Compute the inverse tangent of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst" expressed in radians using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -FOR j := 0 to 7 - i := j*64 - IF k[j] - dst[i+63:i] := ATAN(a[i+63:i]) - ELSE - dst[i+63:i] := src[i+63:i] - FI -ENDFOR -dst[MAX:512] := 0 + Miscellaneous + + + + + + + Normalize the mantissas of the lower double-precision (64-bit) floating-point element in "b", store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". This intrinsic essentially calculates "±(2^k)*|x.significand|", where "k" depends on the interval range defined by "interv" and the sign depends on "sc" and the source sign. + [getmant_note] + IF k[0] + dst[63:0] := GetNormalizedMantissa(b[63:0], sc, interv) +ELSE + dst[63:0] := 0 +FI +dst[127:64] := a[127:64] +dst[MAX:128] := 0 +
immintrin.h
 - + Floating Point AVX512F - Trigonometry - - Compute the inverse tangent of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst" expressed in radians. - -FOR j := 0 to 15 - i := j*32 - dst[i+31:i] := ATAN(a[i+31:i]) -ENDFOR -dst[MAX:512] := 0 + Miscellaneous + + + + + + + Normalize the mantissas of the lower single-precision (32-bit) floating-point element in "b", store the result in the lower element of "dst", and copy the upper 3 packed elements from "a" to the upper elements of "dst". This intrinsic essentially calculates "±(2^k)*|x.significand|", where "k" depends on the interval range defined by "interv" and the sign depends on "sc" and the source sign. + [getmant_note][sae_note] + dst[31:0] := GetNormalizedMantissa(b[31:0], sc, interv) +dst[127:32] := a[127:32] +dst[MAX:128] := 0 +
immintrin.h
 - + Floating Point AVX512F - Trigonometry - - - - Compute the inverse tangent of packed single-precision (32-bit) floating-point elements in "a" expressed in radians, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -FOR j := 0 to 15 - i := j*32 - IF k[j] - dst[i+31:i] := ATAN(a[i+31:i]) - ELSE - dst[i+31:i] := src[i+31:i] - FI -ENDFOR -dst[MAX:512] := 0 + Miscellaneous + + + + + + Normalize the mantissas of the lower single-precision (32-bit) floating-point element in "b", store the result in the lower element of "dst", and copy the upper 3 packed elements from "a" to the upper elements of "dst". This intrinsic essentially calculates "±(2^k)*|x.significand|", where "k" depends on the interval range defined by "interv" and the sign depends on "sc" and the source sign. + [getmant_note] + dst[31:0] := GetNormalizedMantissa(b[31:0], sc, interv) +dst[127:32] := a[127:32] +dst[MAX:128] := 0 +
immintrin.h
 - + Floating Point AVX512F - Trigonometry - - Compute the inverse hyperbolic tangent of packed double-precision (64-bit) floating-point elements in "a" and store the results in "dst" expressed in radians. - -FOR j := 0 to 7 - i := j*64 - dst[i+63:i] := ATANH(a[i+63:i]) -ENDFOR -dst[MAX:512] := 0 + Miscellaneous + + + + + + + + + Normalize the mantissas of the lower single-precision (32-bit) floating-point element in "b", store the result in the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". This intrinsic essentially calculates "±(2^k)*|x.significand|", where "k" depends on the interval range defined by "interv" and the sign depends on "sc" and the source sign. + [getmant_note][sae_note] + IF k[0] + dst[31:0] := GetNormalizedMantissa(b[31:0], sc, interv) +ELSE + dst[31:0] := src[31:0] +FI +dst[127:32] := a[127:32] +dst[MAX:128] := 0 +
immintrin.h
 - + Floating Point AVX512F - Trigonometry - - - - Compute the inverse hyperbolic tangent of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst" expressed in radians using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -FOR j := 0 to 7 - i := j*64 - IF k[j] - dst[i+63:i] := ATANH(a[i+63:i]) - ELSE - dst[i+63:i] := src[i+63:i] - FI -ENDFOR -dst[MAX:512] := 0 + Miscellaneous + + + + + + + + Normalize the mantissas of the lower single-precision (32-bit) floating-point element in "b", store the result in the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". This intrinsic essentially calculates "±(2^k)*|x.significand|", where "k" depends on the interval range defined by "interv" and the sign depends on "sc" and the source sign. + [getmant_note] + IF k[0] + dst[31:0] := GetNormalizedMantissa(b[31:0], sc, interv) +ELSE + dst[31:0] := src[31:0] +FI +dst[127:32] := a[127:32] +dst[MAX:128] := 0 +
immintrin.h
 - + Floating Point AVX512F - Trigonometry - - Compute the inverse hyperblic tangent of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst" expressed in radians. - -FOR j := 0 to 15 - i := j*32 - dst[i+31:i] := ATANH(a[i+31:i]) -ENDFOR -dst[MAX:512] := 0 + Miscellaneous + + + + + + + + Normalize the mantissas of the lower single-precision (32-bit) floating-point element in "b", store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". This intrinsic essentially calculates "±(2^k)*|x.significand|", where "k" depends on the interval range defined by "interv" and the sign depends on "sc" and the source sign. + [getmant_note][sae_note] + IF k[0] + dst[31:0] := GetNormalizedMantissa(b[31:0], sc, interv) +ELSE + dst[31:0] := 0 +FI +dst[127:32] := a[127:32] +dst[MAX:128] := 0 +
immintrin.h
 - + Floating Point AVX512F - Trigonometry - - - - Compute the inverse hyperbolic tangent of packed single-precision (32-bit) floating-point elements in "a" expressed in radians, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -FOR j := 0 to 15 - i := j*32 - IF k[j] - dst[i+31:i] := ATANH(a[i+31:i]) - ELSE - dst[i+31:i] := src[i+31:i] - FI -ENDFOR -dst[MAX:512] := 0 + Miscellaneous + + + + + + + Normalize the mantissas of the lower single-precision (32-bit) floating-point element in "b", store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". This intrinsic essentially calculates "±(2^k)*|x.significand|", where "k" depends on the interval range defined by "interv" and the sign depends on "sc" and the source sign. + [getmant_note] + IF k[0] + dst[31:0] := GetNormalizedMantissa(b[31:0], sc, interv) +ELSE + dst[31:0] := 0 +FI +dst[127:32] := a[127:32] +dst[MAX:128] := 0 +
immintrin.h
 - + Floating Point AVX512F - Elementary Math Functions - - Compute the cube root of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst". + Swizzle + + + + + Copy "a" to "dst", then insert 128 bits (composed of 4 packed single-precision (32-bit) floating-point elements) from "b" into "dst" at the location specified by "imm8". -FOR j := 0 to 7 - i := j*64 - dst[i+63:i] := CubeRoot(a[i+63:i]) -ENDFOR +dst[511:0] := a[511:0] +CASE (imm8[1:0]) OF +0: dst[127:0] := b[127:0] +1: dst[255:128] := b[127:0] +2: dst[383:256] := b[127:0] +3: dst[511:384] := b[127:0] +ESAC dst[MAX:512] := 0 +
immintrin.h
 - + Floating Point AVX512F - Elementary Math Functions - - - - Compute the cube root of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Swizzle + + + + + + + Copy "a" to "tmp", then insert 128 bits (composed of 4 packed single-precision (32-bit) floating-point elements) from "b" into "tmp" at the location specified by "imm8". Store "tmp" to "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := j*64 +tmp[511:0] := a[511:0] +CASE (imm8[1:0]) OF +0: tmp[127:0] := b[127:0] +1: tmp[255:128] := b[127:0] +2: tmp[383:256] := b[127:0] +3: tmp[511:384] := b[127:0] +ESAC +FOR j := 0 to 15 + i := j*32 IF k[j] - dst[i+63:i] := CubeRoot(a[i+63:i]) + dst[i+31:i] := tmp[i+31:i] ELSE - dst[i+63:i] := src[i+63:i] + dst[i+31:i] := src[i+31:i] FI ENDFOR dst[MAX:512] := 0 +
immintrin.h
 - + Floating Point AVX512F - Elementary Math Functions - - Compute the cube root of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst". + Swizzle + + + + + + Copy "a" to "tmp", then insert 128 bits (composed of 4 packed single-precision (32-bit) floating-point elements) from "b" into "tmp" at the location specified by "imm8". Store "tmp" to "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). +tmp[511:0] := a[511:0] +CASE (imm8[1:0]) OF +0: tmp[127:0] := b[127:0] +1: tmp[255:128] := b[127:0] +2: tmp[383:256] := b[127:0] +3: tmp[511:384] := b[127:0] +ESAC FOR j := 0 to 15 i := j*32 - dst[i+31:i] := CubeRoot(a[i+31:i]) + IF k[j] + dst[i+31:i] := tmp[i+31:i] + ELSE + dst[i+31:i] := 0 + FI ENDFOR dst[MAX:512] := 0 +
immintrin.h
 - + Floating Point AVX512F - Elementary Math Functions - - - - Compute the cube root of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Swizzle + + + + + Copy "a" to "dst", then insert 256 bits (composed of 4 packed double-precision (64-bit) floating-point elements) from "b" into "dst" at the location specified by "imm8". -FOR j := 0 to 15 - i := j*32 - IF k[j] - dst[i+31:i] := CubeRoot(a[i+31:i]) - ELSE - dst[i+31:i] := src[i+31:i] - FI -ENDFOR +dst[511:0] := a[511:0] +CASE (imm8[0]) OF +0: dst[255:0] := b[255:0] +1: dst[511:256] := b[255:0] +ESAC dst[MAX:512] := 0 +
immintrin.h
 - + Floating Point AVX512F - Probability/Statistics - - Compute the cumulative distribution function of packed double-precision (64-bit) floating-point elements in "a" using the normal distribution, and store the results in "dst". + Swizzle + + + + + + + Copy "a" to "tmp", then insert 256 bits (composed of 4 packed double-precision (64-bit) floating-point elements) from "b" into "tmp" at the location specified by "imm8". Store "tmp" to "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). +tmp[511:0] := a[511:0] +CASE (imm8[0]) OF +0: tmp[255:0] := b[255:0] +1: tmp[511:256] := b[255:0] +ESAC FOR j := 0 to 7 i := j*64 - dst[i+63:i] := CDFNormal(a[i+63:i]) + IF k[j] + dst[i+63:i] := tmp[i+63:i] + ELSE + dst[i+63:i] := src[i+63:i] + FI ENDFOR dst[MAX:512] := 0 +
immintrin.h
 - + Floating Point AVX512F - Probability/Statistics - - - - Compute the cumulative distribution function of packed double-precision (64-bit) floating-point elements in "a" using the normal distribution, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Swizzle + + + + + + Copy "a" to "tmp", then insert 256 bits (composed of 4 packed double-precision (64-bit) floating-point elements) from "b" into "tmp" at the location specified by "imm8". Store "tmp" to "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). +tmp[511:0] := a[511:0] +CASE (imm8[0]) OF +0: tmp[255:0] := b[255:0] +1: tmp[511:256] := b[255:0] +ESAC FOR j := 0 to 7 i := j*64 IF k[j] - dst[i+63:i] := CDFNormal(a[i+63:i]) + dst[i+63:i] := tmp[i+63:i] ELSE - dst[i+63:i] := src[i+63:i] + dst[i+63:i] := 0 FI ENDFOR dst[MAX:512] := 0 +
immintrin.h
 - - Floating Point + + Integer AVX512F - Probability/Statistics - - Compute the cumulative distribution function of packed single-precision (32-bit) floating-point elements in "a" using the normal distribution, and store the results in "dst". + Swizzle + + + + + Copy "a" to "dst", then insert 128 bits (composed of 4 packed 32-bit integers) from "b" into "dst" at the location specified by "imm8". + +dst[511:0] := a[511:0] +CASE (imm8[1:0]) OF +0: dst[127:0] := b[127:0] +1: dst[255:128] := b[127:0] +2: dst[383:256] := b[127:0] +3: dst[511:384] := b[127:0] +ESAC +dst[MAX:512] := 0 + + +
immintrin.h
+ + + Integer + AVX512F + Swizzle + + + + + + + Copy "a" to "tmp", then insert 128 bits (composed of 4 packed 32-bit integers) from "b" into "tmp" at the location specified by "imm8". Store "tmp" to "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). +tmp[511:0] := a[511:0] +CASE (imm8[1:0]) OF +0: tmp[127:0] := b[127:0] +1: tmp[255:128] := b[127:0] +2: tmp[383:256] := b[127:0] +3: tmp[511:384] := b[127:0] +ESAC FOR j := 0 to 15 i := j*32 - dst[i+31:i] := CDFNormal(a[i+31:i]) + IF k[j] + dst[i+31:i] := tmp[i+31:i] + ELSE + dst[i+31:i] := src[i+31:i] + FI ENDFOR dst[MAX:512] := 0 +
immintrin.h
 - - Floating Point + + Integer AVX512F - Probability/Statistics - - - - Compute the cumulative distribution function of packed single-precision (32-bit) floating-point elements in "a" using the normal distribution, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Swizzle + + + + + + Copy "a" to "tmp", then insert 128 bits (composed of 4 packed 32-bit integers) from "b" into "tmp" at the location specified by "imm8". Store "tmp" to "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). +tmp[511:0] := a[511:0] +CASE (imm8[1:0]) OF +0: tmp[127:0] := b[127:0] +1: tmp[255:128] := b[127:0] +2: tmp[383:256] := b[127:0] +3: tmp[511:384] := b[127:0] +ESAC FOR j := 0 to 15 i := j*32 IF k[j] - dst[i+31:i] := CDFNormal(a[i+31:i]) + dst[i+31:i] := tmp[i+31:i] ELSE - dst[i+31:i] := src[i+31:i] + dst[i+31:i] := 0 FI ENDFOR dst[MAX:512] := 0 +
immintrin.h
 - - Floating Point + + Integer AVX512F - Probability/Statistics - - Compute the inverse cumulative distribution function of packed double-precision (64-bit) floating-point elements in "a" using the normal distribution, and store the results in "dst". + Swizzle + + + + + Copy "a" to "dst", then insert 256 bits (composed of 4 packed 64-bit integers) from "b" into "dst" at the location specified by "imm8". -FOR j := 0 to 7 - i := j*64 - dst[i+63:i] := InverseCDFNormal(a[i+63:i]) -ENDFOR +dst[511:0] := a[511:0] +CASE (imm8[0]) OF +0: dst[255:0] := b[255:0] +1: dst[511:256] := b[255:0] +ESAC dst[MAX:512] := 0 +
immintrin.h
 - - Floating Point + + Integer AVX512F - Probability/Statistics - - - - Compute the inverse cumulative distribution function of packed double-precision (64-bit) floating-point elements in "a" using the normal distribution, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Swizzle + + + + + + + Copy "a" to "tmp", then insert 256 bits (composed of 4 packed 64-bit integers) from "b" into "tmp" at the location specified by "imm8". Store "tmp" to "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). +tmp[511:0] := a[511:0] +CASE (imm8[0]) OF +0: tmp[255:0] := b[255:0] +1: tmp[511:256] := b[255:0] +ESAC FOR j := 0 to 7 i := j*64 IF k[j] - dst[i+63:i] := InverseCDFNormal(a[i+63:i]) + dst[i+63:i] := tmp[i+63:i] ELSE dst[i+63:i] := src[i+63:i] FI ENDFOR dst[MAX:512] := 0 +
immintrin.h
 - - Floating Point - AVX512F - Probability/Statistics - - Compute the inverse cumulative distribution function of packed single-precision (32-bit) floating-point elements in "a" using the normal distribution, and store the results in "dst". - -FOR j := 0 to 15 - i := j*32 - dst[i+31:i] := InverseCDFNormal(a[i+31:i]) -ENDFOR -dst[MAX:512] := 0 - -
immintrin.h
- - - Floating Point + + Integer AVX512F - Probability/Statistics - - - - Compute the inverse cumulative distribution function of packed single-precision (32-bit) floating-point elements in "a" using the normal distribution, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Swizzle + + + + + + Copy "a" to "tmp", then insert 256 bits (composed of 4 packed 64-bit integers) from "b" into "tmp" at the location specified by "imm8". Store "tmp" to "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*32 +tmp[511:0] := a[511:0] +CASE (imm8[0]) OF +0: tmp[255:0] := b[255:0] +1: tmp[511:256] := b[255:0] +ESAC +FOR j := 0 to 7 + i := j*64 IF k[j] - dst[i+31:i] := InverseCDFNormal(a[i+31:i]) + dst[i+63:i] := tmp[i+63:i] ELSE - dst[i+31:i] := src[i+31:i] + dst[i+63:i] := 0 FI ENDFOR dst[MAX:512] := 0 +
immintrin.h
 - + Floating Point AVX512F Special Math Functions - - Round the packed double-precision (64-bit) floating-point elements in "a" up to an integer value, and store the results as packed double-precision floating-point elements in "dst". + + + + + + Compare packed double-precision (64-bit) floating-point elements in "a" and "b", and store packed maximum values in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). FOR j := 0 to 7 i := j*64 - dst[i+63:i] := CEIL(a[i+63:i]) + IF k[j] + dst[i+63:i] := MAX(a[i+63:i], b[i+63:i]) + ELSE + dst[i+63:i] := src[i+63:i] + FI ENDFOR dst[MAX:512] := 0 +
immintrin.h
 - + Floating Point AVX512F Special Math Functions - - - - Round the packed double-precision (64-bit) floating-point elements in "a" up to an integer value, and store the results as packed double-precision floating-point elements in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + + + + + + Compare packed double-precision (64-bit) floating-point elements in "a" and "b", and store packed maximum values in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). [sae_note] FOR j := 0 to 7 i := j*64 IF k[j] - dst[i+63:i] := CEIL(a[i+63:i]) + dst[i+63:i] := MAX(a[i+63:i], b[i+63:i]) ELSE dst[i+63:i] := src[i+63:i] FI ENDFOR dst[MAX:512] := 0 +
immintrin.h
 - + Floating Point AVX512F Special Math Functions - - Round the packed single-precision (32-bit) floating-point elements in "a" up to an integer value, and store the results as packed single-precision floating-point elements in "dst". + + + + + Compare packed double-precision (64-bit) floating-point elements in "a" and "b", and store packed maximum values in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*32 - dst[i+31:i] := CEIL(a[i+31:i]) +FOR j := 0 to 7 + i := j*64 + IF k[j] + dst[i+63:i] := MAX(a[i+63:i], b[i+63:i]) + ELSE + dst[i+63:i] := 0 + FI ENDFOR dst[MAX:512] := 0 +
immintrin.h
 - + Floating Point AVX512F Special Math Functions - - - - Round the packed single-precision (32-bit) floating-point elements in "a" up to an integer value, and store the results as packed single-precision floating-point elements in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + + + + + Compare packed double-precision (64-bit) floating-point elements in "a" and "b", and store packed maximum values in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). [sae_note] -FOR j := 0 to 15 - i := j*32 +FOR j := 0 to 7 + i := j*64 IF k[j] - dst[i+31:i] := CEIL(a[i+31:i]) + dst[i+63:i] := MAX(a[i+63:i], b[i+63:i]) ELSE - dst[i+31:i] := src[i+31:i] + dst[i+63:i] := 0 FI ENDFOR dst[MAX:512] := 0 +
immintrin.h
 - + Floating Point AVX512F - Trigonometry - - Compute the cosine of packed double-precision (64-bit) floating-point elements in "a" expressed in radians, and store the results in "dst". + Special Math Functions + + + + Compare packed double-precision (64-bit) floating-point elements in "a" and "b", and store packed maximum values in "dst". FOR j := 0 to 7 i := j*64 - dst[i+63:i] := COS(a[i+63:i]) + dst[i+63:i] := MAX(a[i+63:i], b[i+63:i]) ENDFOR dst[MAX:512] := 0 +
immintrin.h
 - + Floating Point AVX512F - Trigonometry - - - - Compute the cosine of packed double-precision (64-bit) floating-point elements in "a" expressed in radians, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Special Math Functions + + + + + Compare packed double-precision (64-bit) floating-point elements in "a" and "b", and store packed maximum values in "dst". [sae_note] FOR j := 0 to 7 i := j*64 - IF k[j] - dst[i+63:i] := COS(a[i+63:i]) - ELSE - dst[i+63:i] := src[i+63:i] - FI + dst[i+63:i] := MAX(a[i+63:i], b[i+63:i]) ENDFOR dst[MAX:512] := 0 +
immintrin.h
 - + Floating Point AVX512F - Trigonometry - - Compute the cosine of packed single-precision (32-bit) floating-point elements in "a" expressed in radians, and store the results in "dst". + Special Math Functions + + + + + + Compare packed single-precision (32-bit) floating-point elements in "a" and "b", and store packed maximum values in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). FOR j := 0 to 15 i := j*32 - dst[i+31:i] := COS(a[i+31:i]) + IF k[j] + dst[i+31:i] := MAX(a[i+31:i], b[i+31:i]) + ELSE + dst[i+31:i] := src[i+31:i] + FI ENDFOR dst[MAX:512] := 0 +
immintrin.h
 - + Floating Point AVX512F - Trigonometry - - - - Compute the cosine of packed single-precision (32-bit) floating-point elements in "a" expressed in radians, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Special Math Functions + + + + + + + Compare packed single-precision (32-bit) floating-point elements in "a" and "b", and store packed maximum values in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). [sae_note] FOR j := 0 to 15 i := j*32 IF k[j] - dst[i+31:i] := COS(a[i+31:i]) + dst[i+31:i] := MAX(a[i+31:i], b[i+31:i]) ELSE dst[i+31:i] := src[i+31:i] FI ENDFOR dst[MAX:512] := 0 +
immintrin.h
 - - Floating Point - AVX512F - Trigonometry - - Compute the cosine of packed double-precision (64-bit) floating-point elements in "a" expressed in degrees, and store the results in "dst". - -FOR j := 0 to 7 - i := j*64 - dst[i+63:i] := COSD(a[i+63:i]) -ENDFOR -dst[MAX:512] := 0 - -
immintrin.h
- - + Floating Point AVX512F - Trigonometry - - - - Compute the cosine of packed double-precision (64-bit) floating-point elements in "a" expressed in degrees, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Special Math Functions + + + + + Compare packed single-precision (32-bit) floating-point elements in "a" and "b", and store packed maximum values in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := j*64 +FOR j := 0 to 15 + i := j*32 IF k[j] - dst[i+63:i] := COSD(a[i+63:i]) + dst[i+31:i] := MAX(a[i+31:i], b[i+31:i]) ELSE - dst[i+63:i] := src[i+63:i] + dst[i+31:i] := 0 FI ENDFOR dst[MAX:512] := 0 +
immintrin.h
 - + Floating Point AVX512F - Trigonometry - - Compute the cosine of packed single-precision (32-bit) floating-point elements in "a" expressed in degrees, and store the results in "dst". + Special Math Functions + + + + + + Compare packed single-precision (32-bit) floating-point elements in "a" and "b", and store packed maximum values in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). [sae_note] FOR j := 0 to 15 i := j*32 - dst[i+31:i] := COSD(a[i+31:i]) + IF k[j] + dst[i+31:i] := MAX(a[i+31:i], b[i+31:i]) + ELSE + dst[i+31:i] := 0 + FI ENDFOR dst[MAX:512] := 0 +
immintrin.h
 - + Floating Point AVX512F - Trigonometry - - - - Compute the cosine of packed single-precision (32-bit) floating-point elements in "a" expressed in degrees, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Special Math Functions + + + + Compare packed single-precision (32-bit) floating-point elements in "a" and "b", and store packed maximum values in "dst". FOR j := 0 to 15 i := j*32 - IF k[j] - dst[i+31:i] := COSD(a[i+31:i]) - ELSE - dst[i+31:i] := src[i+31:i] - FI + dst[i+31:i] := MAX(a[i+31:i], b[i+31:i]) ENDFOR dst[MAX:512] := 0 +
immintrin.h
 - + Floating Point AVX512F - Trigonometry - - Compute the hyperbolic cosine of packed double-precision (64-bit) floating-point elements in "a" expressed in radians, and store the results in "dst". + Special Math Functions + + + + + Compare packed single-precision (32-bit) floating-point elements in "a" and "b", and store packed maximum values in "dst". [sae_note] -FOR j := 0 to 7 - i := j*64 - dst[i+63:i] := COSH(a[i+63:i]) +FOR j := 0 to 15 + i := j*32 + dst[i+31:i] := MAX(a[i+31:i], b[i+31:i]) ENDFOR dst[MAX:512] := 0 +
immintrin.h
 - + Floating Point AVX512F - Trigonometry - - - - Compute the hyperbolic cosine of packed double-precision (64-bit) floating-point elements in "a" expressed in radians, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Special Math Functions + + + + + + + Compare the lower double-precision (64-bit) floating-point elements in "a" and "b", store the maximum value in the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". [sae_note] -FOR j := 0 to 7 - i := j*64 - IF k[j] - dst[i+63:i] := COSH(a[i+63:i]) - ELSE - dst[i+63:i] := src[i+63:i] - FI -ENDFOR -dst[MAX:512] := 0 +IF k[0] + dst[63:0] := MAX(a[63:0], b[63:0]) +ELSE + dst[63:0] := src[63:0] +FI +dst[127:64] := a[127:64] +dst[MAX:128] := 0 +
immintrin.h
 - + Floating Point AVX512F - Trigonometry - - Compute the hyperbolic cosine of packed single-precision (32-bit) floating-point elements in "a" expressed in radians, and store the results in "dst". + Special Math Functions + + + + + + Compare the lower double-precision (64-bit) floating-point elements in "a" and "b", store the maximum value in the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". -FOR j := 0 to 15 - i := j*32 - dst[i+31:i] := COSH(a[i+31:i]) -ENDFOR -dst[MAX:512] := 0 +IF k[0] + dst[63:0] := MAX(a[63:0], b[63:0]) +ELSE + dst[63:0] := src[63:0] +FI +dst[127:64] := a[127:64] +dst[MAX:128] := 0 +
immintrin.h
 - + Floating Point AVX512F - Trigonometry - - - - Compute the hyperbolic cosine of packed single-precision (32-bit) floating-point elements in "a" expressed in radians, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Special Math Functions + + + + + + Compare the lower double-precision (64-bit) floating-point elements in "a" and "b", store the maximum value in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". [sae_note] -FOR j := 0 to 15 - i := j*32 - IF k[j] - dst[i+31:i] := COSH(a[i+31:i]) - ELSE - dst[i+31:i] := src[i+31:i] - FI -ENDFOR -dst[MAX:512] := 0 +IF k[0] + dst[63:0] := MAX(a[63:0], b[63:0]) +ELSE + dst[63:0] := 0 +FI +dst[127:64] := a[127:64] +dst[MAX:128] := 0 +
immintrin.h
 - + Floating Point AVX512F - Probability/Statistics - - Compute the error function of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst". + Special Math Functions + + + + + Compare the lower double-precision (64-bit) floating-point elements in "a" and "b", store the maximum value in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". -FOR j := 0 to 7 - i := j*64 - dst[i+63:i] := ERF(a[i+63:i]) -ENDFOR -dst[MAX:512] := 0 +IF k[0] + dst[63:0] := MAX(a[63:0], b[63:0]) +ELSE + dst[63:0] := 0 +FI +dst[127:64] := a[127:64] +dst[MAX:128] := 0 +
immintrin.h
 - + Floating Point AVX512F - Probability/Statistics - - - - Compute the error function of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Special Math Functions + + + + + Compare the lower double-precision (64-bit) floating-point elements in "a" and "b", store the maximum value in the lower element of "dst", and copy the upper element from "a" to the upper element of "dst". [sae_note] -FOR j := 0 to 7 - i := j*64 - IF k[j] - dst[i+63:i] := ERF(a[i+63:i]) - ELSE - dst[i+63:i] := src[i+63:i] - FI -ENDFOR -dst[MAX:512] := 0 +dst[63:0] := MAX(a[63:0], b[63:0]) +dst[127:64] := a[127:64] +dst[MAX:128] := 0 +
immintrin.h
 - + Floating Point AVX512F - Probability/Statistics - - Compute the complementary error function of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst". + Special Math Functions + + + + + + + Compare the lower single-precision (32-bit) floating-point elements in "a" and "b", store the maximum value in the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". [sae_note] -FOR j := 0 to 7 - i := j*64 - dst[i+63:i] := 1.0 - ERF(a[i+63:i]) -ENDFOR -dst[MAX:512] := 0 +IF k[0] + dst[31:0] := MAX(a[31:0], b[31:0]) +ELSE + dst[31:0] := src[31:0] +FI +dst[127:32] := a[127:32] +dst[MAX:128] := 0 +
immintrin.h
 - + Floating Point AVX512F - Probability/Statistics - - - - Compute the complementary error function of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Special Math Functions + + + + + + Compare the lower single-precision (32-bit) floating-point elements in "a" and "b", store the maximum value in the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". -FOR j := 0 to 7 - i := j*64 - IF k[j] - dst[i+63:i] := 1.0 - ERF(a[i+63:i]) - ELSE - dst[i+63:i] := src[i+63:i] - FI -ENDFOR -dst[MAX:512] := 0 +IF k[0] + dst[31:0] := MAX(a[31:0], b[31:0]) +ELSE + dst[31:0] := src[31:0] +FI +dst[127:32] := a[127:32] +dst[MAX:128] := 0 +
immintrin.h
 - + Floating Point AVX512F - Probability/Statistics - - Compute the error function of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst". + Special Math Functions + + + + + + Compare the lower single-precision (32-bit) floating-point elements in "a" and "b", store the maximum value in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". [sae_note] -FOR j := 0 to 15 - i := j*32 - dst[i+31:i] := ERF(a[i+31:i]) -ENDFOR -dst[MAX:512] := 0 +IF k[0] + dst[31:0] := MAX(a[31:0], b[31:0]) +ELSE + dst[31:0] := 0 +FI +dst[127:32] := a[127:32] +dst[MAX:128] := 0 +
immintrin.h
 - + Floating Point AVX512F - Probability/Statistics - - - - Compute the error function of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Special Math Functions + + + + + Compare the lower single-precision (32-bit) floating-point elements in "a" and "b", store the maximum value in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". -FOR j := 0 to 15 - i := j*32 - IF k[j] - dst[i+31:i] := ERF(a[i+31:i]) - ELSE - dst[i+31:i] := src[i+31:i] - FI -ENDFOR -dst[MAX:512] := 0 +IF k[0] + dst[31:0] := MAX(a[31:0], b[31:0]) +ELSE + dst[31:0] := 0 +FI +dst[127:32] := a[127:32] +dst[MAX:128] := 0 +
immintrin.h
 - + Floating Point AVX512F - Probability/Statistics - - Compute the complementary error function of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst". + Special Math Functions + + + + + Compare the lower single-precision (32-bit) floating-point elements in "a" and "b", store the maximum value in the lower element of "dst", and copy the upper 3 packed elements from "a" to the upper elements of "dst". [sae_note] -FOR j := 0 to 15 - i := j*32 - dst[i+31:i] := 1.0 - ERF(a[i+31:i]) -ENDFOR -dst[MAX:512] := 0 +dst[31:0] := MAX(a[31:0], b[31:0]) +dst[127:32] := a[127:32] +dst[MAX:128] := 0 +
immintrin.h
 - + Floating Point AVX512F - Probability/Statistics - - - - Compute the complementary error function of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Special Math Functions + + + + + + Compare packed double-precision (64-bit) floating-point elements in "a" and "b", and store packed minimum values in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*32 +FOR j := 0 to 7 + i := j*64 IF k[j] - dst[i+31:i] := 1.0 - ERF(a[i+31:i]) + dst[i+63:i] := MIN(a[i+63:i], b[i+63:i]) ELSE - dst[i+31:i] := src[i+31:i] + dst[i+63:i] := src[i+63:i] FI ENDFOR dst[MAX:512] := 0 +
immintrin.h
 - + Floating Point AVX512F - Probability/Statistics - - Compute the inverse error function of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst". + Special Math Functions + + + + + + + Compare packed double-precision (64-bit) floating-point elements in "a" and "b", and store packed minimum values in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). [sae_note] FOR j := 0 to 7 i := j*64 - dst[i+63:i] := 1.0 / ERF(a[i+63:i]) + IF k[j] + dst[i+63:i] := MIN(a[i+63:i], b[i+63:i]) + ELSE + dst[i+63:i] := src[i+63:i] + FI ENDFOR dst[MAX:512] := 0 +
immintrin.h
 - + Floating Point AVX512F - Probability/Statistics - - - - Compute the inverse error function of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Special Math Functions + + + + + Compare packed double-precision (64-bit) floating-point elements in "a" and "b", and store packed minimum values in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). FOR j := 0 to 7 i := j*64 IF k[j] - dst[i+63:i] := 1.0 / ERF(a[i+63:i]) + dst[i+63:i] := MIN(a[i+63:i], b[i+63:i]) ELSE - dst[i+63:i] := src[i+63:i] + dst[i+63:i] := 0 FI ENDFOR dst[MAX:512] := 0 +
immintrin.h
 - - Floating Point - AVX512F - Probability/Statistics - - Compute the inverse error function of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst". - -FOR j := 0 to 15 - i := j*32 - dst[i+31:i] := 1.0 / ERF(a[i+31:i]) -ENDFOR -dst[MAX:512] := 0 - -
immintrin.h
- - + Floating Point AVX512F - Probability/Statistics - - - - Compute the inverse error function of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Special Math Functions + + + + + + Compare packed double-precision (64-bit) floating-point elements in "a" and "b", and store packed minimum values in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). [sae_note] -FOR j := 0 to 15 - i := j*32 +FOR j := 0 to 7 + i := j*64 IF k[j] - dst[i+31:i] := 1.0 / ERF(a[i+31:i]) + dst[i+63:i] := MIN(a[i+63:i], b[i+63:i]) ELSE - dst[i+31:i] := src[i+31:i] + dst[i+63:i] := 0 FI ENDFOR dst[MAX:512] := 0 +
immintrin.h
 - + Floating Point AVX512F - Probability/Statistics - - Compute the inverse complementary error function of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst". + Special Math Functions + + + + Compare packed double-precision (64-bit) floating-point elements in "a" and "b", and store packed minimum values in "dst". FOR j := 0 to 7 i := j*64 - dst[i+63:i] := 1.0 / (1.0 - ERF(a[i+63:i])) + dst[i+63:i] := MIN(a[i+63:i], b[i+63:i]) ENDFOR dst[MAX:512] := 0 +
immintrin.h
 - + Floating Point AVX512F - Probability/Statistics - - - - Compute the inverse complementary error function of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Special Math Functions + + + + + Compare packed double-precision (64-bit) floating-point elements in "a" and "b", and store packed minimum values in "dst". [sae_note] FOR j := 0 to 7 i := j*64 - IF k[j] - dst[i+63:i] := 1.0 / (1.0 - ERF(a[i+63:i])) - ELSE - dst[i+63:i] := src[i+63:i] - FI + dst[i+63:i] := MIN(a[i+63:i], b[i+63:i]) ENDFOR dst[MAX:512] := 0 +
immintrin.h
 - + Floating Point AVX512F - Probability/Statistics - - Compute the inverse complementary error function of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst". + Special Math Functions + + + + + + Compare packed single-precision (32-bit) floating-point elements in "a" and "b", and store packed minimum values in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). FOR j := 0 to 15 i := j*32 - dst[i+31:i] := 1.0 / (1.0 - ERF(a[i+31:i])) + IF k[j] + dst[i+31:i] := MIN(a[i+31:i], b[i+31:i]) + ELSE + dst[i+31:i] := src[i+31:i] + FI ENDFOR dst[MAX:512] := 0 +
immintrin.h
 - + Floating Point AVX512F - Probability/Statistics - - - - Compute the inverse complementary error function of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Special Math Functions + + + + + + + Compare packed single-precision (32-bit) floating-point elements in "a" and "b", and store packed minimum values in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). [sae_note] FOR j := 0 to 15 i := j*32 IF k[j] - dst[i+31:i] := 1.0 / (1.0 - ERF(a[i+31:i])) + dst[i+31:i] := MIN(a[i+31:i], b[i+31:i]) ELSE dst[i+31:i] := src[i+31:i] FI ENDFOR dst[MAX:512] := 0 +
immintrin.h
 - + Floating Point AVX512F - Elementary Math Functions - - Compute the exponential value of 10 raised to the power of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst". + Special Math Functions + + + + + Compare packed single-precision (32-bit) floating-point elements in "a" and "b", and store packed minimum values in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := j*64 - dst[i+63:i] := 10^(a[i+63:i]) +FOR j := 0 to 15 + i := j*32 + IF k[j] + dst[i+31:i] := MIN(a[i+31:i], b[i+31:i]) + ELSE + dst[i+31:i] := 0 + FI ENDFOR dst[MAX:512] := 0 +
immintrin.h
 - + Floating Point AVX512F - Elementary Math Functions - - - - Compute the exponential value of 10 raised to the power of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Special Math Functions + + + + + + Compare packed single-precision (32-bit) floating-point elements in "a" and "b", and store packed minimum values in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). [sae_note] -FOR j := 0 to 7 - i := j*64 +FOR j := 0 to 15 + i := j*32 IF k[j] - dst[i+63:i] := 10^(a[i+63:i]) + dst[i+31:i] := MIN(a[i+31:i], b[i+31:i]) ELSE - dst[i+63:i] := src[i+63:i] + dst[i+31:i] := 0 FI ENDFOR dst[MAX:512] := 0 +
immintrin.h
 - + Floating Point AVX512F - Elementary Math Functions - - Compute the exponential value of 10 raised to the power of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst". + Special Math Functions + + + + Compare packed single-precision (32-bit) floating-point elements in "a" and "b", and store packed minimum values in "dst". FOR j := 0 to 15 i := j*32 - dst[i+31:i] := 10^(a[i+31:i]) + dst[i+31:i] := MIN(a[i+31:i], b[i+31:i]) ENDFOR dst[MAX:512] := 0 +
immintrin.h
 - + Floating Point AVX512F - Elementary Math Functions - - - - Compute the exponential value of 10 raised to the power of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Special Math Functions + + + + + Compare packed single-precision (32-bit) floating-point elements in "a" and "b", and store packed minimum values in "dst". [sae_note] FOR j := 0 to 15 i := j*32 - IF k[j] - dst[i+31:i] := 10^(a[i+31:i]) - ELSE - dst[i+31:i] := src[i+31:i] - FI + dst[i+31:i] := MIN(a[i+31:i], b[i+31:i]) ENDFOR dst[MAX:512] := 0 +
immintrin.h
 - + Floating Point AVX512F - Elementary Math Functions - - Compute the exponential value of 2 raised to the power of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst". + Special Math Functions + + + + + + + Compare the lower double-precision (64-bit) floating-point elements in "a" and "b", store the minimum value in the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". [sae_note] -FOR j := 0 to 7 - i := j*64 - dst[i+63:i] := 2^(a[i+63:i]) -ENDFOR -dst[MAX:512] := 0 +IF k[0] + dst[63:0] := MIN(a[63:0], b[63:0]) +ELSE + dst[63:0] := src[63:0] +FI +dst[127:64] := a[127:64] +dst[MAX:128] := 0 +
immintrin.h
 - + Floating Point AVX512F - Elementary Math Functions - - - - Compute the exponential value of 2 raised to the power of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Special Math Functions + + + + + + Compare the lower double-precision (64-bit) floating-point elements in "a" and "b", store the minimum value in the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". -FOR j := 0 to 7 - i := j*64 - IF k[j] - dst[i+63:i] := 2^(a[i+63:i]) - ELSE - dst[i+63:i] := src[i+63:i] - FI -ENDFOR -dst[MAX:512] := 0 +IF k[0] + dst[63:0] := MIN(a[63:0], b[63:0]) +ELSE + dst[63:0] := src[63:0] +FI +dst[127:64] := a[127:64] +dst[MAX:128] := 0 +
immintrin.h
 - + Floating Point AVX512F - Elementary Math Functions - - Compute the exponential value of 2 raised to the power of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst". + Special Math Functions + + + + + + Compare the lower double-precision (64-bit) floating-point elements in "a" and "b", store the minimum value in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". [sae_note] -FOR j := 0 to 15 - i := j*32 - dst[i+31:i] := 2^(a[i+31:i]) -ENDFOR -dst[MAX:512] := 0 +IF k[0] + dst[63:0] := MIN(a[63:0], b[63:0]) +ELSE + dst[63:0] := 0 +FI +dst[127:64] := a[127:64] +dst[MAX:128] := 0 +
immintrin.h
 - + Floating Point AVX512F - Elementary Math Functions - - - - Compute the exponential value of 2 raised to the power of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Special Math Functions + + + + + Compare the lower double-precision (64-bit) floating-point elements in "a" and "b", store the minimum value in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". -FOR j := 0 to 15 - i := j*32 - IF k[j] - dst[i+31:i] := 2^(a[i+31:i]) - ELSE - dst[i+31:i] := src[i+31:i] - FI -ENDFOR -dst[MAX:512] := 0 +IF k[0] + dst[63:0] := MIN(a[63:0], b[63:0]) +ELSE + dst[63:0] := 0 +FI +dst[127:64] := a[127:64] +dst[MAX:128] := 0 +
immintrin.h
 - + Floating Point AVX512F - Elementary Math Functions - - Compute the exponential value of "e" raised to the power of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst". + Special Math Functions + + + + + Compare the lower double-precision (64-bit) floating-point elements in "a" and "b", store the minimum value in the lower element of "dst" , and copy the upper element from "a" to the upper element of "dst". [sae_note] -FOR j := 0 to 7 - i := j*64 - dst[i+63:i] := e^(a[i+63:i]) -ENDFOR -dst[MAX:512] := 0 +dst[63:0] := MIN(a[63:0], b[63:0]) +dst[127:64] := a[127:64] +dst[MAX:128] := 0 +
immintrin.h
 - + Floating Point AVX512F - Elementary Math Functions - - - - Compute the exponential value of "e" raised to the power of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Special Math Functions + + + + + + + Compare the lower single-precision (32-bit) floating-point elements in "a" and "b", store the minimum value in the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". [sae_note] -FOR j := 0 to 7 - i := j*64 - IF k[j] - dst[i+63:i] := e^(a[i+63:i]) - ELSE - dst[i+63:i] := src[i+63:i] - FI -ENDFOR -dst[MAX:512] := 0 +IF k[0] + dst[31:0] := MIN(a[31:0], b[31:0]) +ELSE + dst[31:0] := src[31:0] +FI +dst[127:32] := a[127:32] +dst[MAX:128] := 0 +
immintrin.h
 - + Floating Point AVX512F - Elementary Math Functions - - Compute the exponential value of "e" raised to the power of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst". + Special Math Functions + + + + + + Compare the lower single-precision (32-bit) floating-point elements in "a" and "b", store the minimum value in the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". + +IF k[0] + dst[31:0] := MIN(a[31:0], b[31:0]) +ELSE + dst[31:0] := src[31:0] +FI +dst[127:32] := a[127:32] +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512F + Special Math Functions + + + + + + Compare the lower single-precision (32-bit) floating-point elements in "a" and "b", store the minimum value in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". [sae_note] + +IF k[0] + dst[31:0] := MIN(a[31:0], b[31:0]) +ELSE + dst[31:0] := 0 +FI +dst[127:32] := a[127:32] +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512F + Special Math Functions + + + + + Compare the lower single-precision (32-bit) floating-point elements in "a" and "b", store the minimum value in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". -FOR j := 0 to 15 - i := j*32 - dst[i+31:i] := e^(a[i+31:i]) -ENDFOR -dst[MAX:512] := 0 +IF k[0] + dst[31:0] := MIN(a[31:0], b[31:0]) +ELSE + dst[31:0] := 0 +FI +dst[127:32] := a[127:32] +dst[MAX:128] := 0 +
immintrin.h
 - + Floating Point AVX512F - Elementary Math Functions - - - - Compute the exponential value of "e" raised to the power of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Special Math Functions + + + + + Compare the lower single-precision (32-bit) floating-point elements in "a" and "b", store the minimum value in the lower element of "dst", and copy the upper 3 packed elements from "a" to the upper elements of "dst". [sae_note] -FOR j := 0 to 15 - i := j*32 - IF k[j] - dst[i+31:i] := e^(a[i+31:i]) - ELSE - dst[i+31:i] := src[i+31:i] - FI -ENDFOR -dst[MAX:512] := 0 +dst[31:0] := MIN(a[31:0], b[31:0]) +dst[127:32] := a[127:32] +dst[MAX:128] := 0 +
immintrin.h
 - + Floating Point AVX512F - Elementary Math Functions - - Compute the exponential value of "e" raised to the power of packed double-precision (64-bit) floating-point elements in "a", subtract one from each element, and store the results in "dst". + Load + + + + Load packed double-precision (64-bit) floating-point elements from memory into "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). "mem_addr" must be aligned on a 64-byte boundary or a general-protection exception may be generated. FOR j := 0 to 7 i := j*64 - dst[i+63:i] := e^(a[i+63:i]) - 1.0 + IF k[j] + dst[i+63:i] := MEM[mem_addr+i+63:mem_addr+i] + ELSE + dst[i+63:i] := 0 + FI ENDFOR dst[MAX:512] := 0 +
immintrin.h
 - + Floating Point AVX512F - Elementary Math Functions - - - - Compute the exponential value of "e" raised to the power of packed double-precision (64-bit) floating-point elements in "a", subtract one from each element, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Move + + + + Move packed double-precision (64-bit) floating-point elements from "a" into "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). FOR j := 0 to 7 i := j*64 IF k[j] - dst[i+63:i] := e^(a[i+63:i]) - 1.0 + dst[i+63:i] := a[i+63:i] ELSE - dst[i+63:i] := src[i+63:i] + dst[i+63:i] := 0 FI ENDFOR dst[MAX:512] := 0 +
immintrin.h
 - + Floating Point AVX512F - Elementary Math Functions - - Compute the exponential value of "e" raised to the power of packed single-precision (32-bit) floating-point elements in "a", subtract one from each element, and store the results in "dst". + Load + + + + Load packed single-precision (32-bit) floating-point elements from memory into "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). "mem_addr" must be aligned on a 64-byte boundary or a general-protection exception may be generated. FOR j := 0 to 15 i := j*32 - dst[i+31:i] := e^(a[i+31:i]) - 1.0 + IF k[j] + dst[i+31:i] := MEM[mem_addr+i+31:mem_addr+i] + ELSE + dst[i+31:i] := 0 + FI ENDFOR dst[MAX:512] := 0 +
immintrin.h
 - + Floating Point AVX512F - Elementary Math Functions - - - - Compute the exponential value of "e" raised to the power of packed single-precision (32-bit) floating-point elements in "a", subtract one from each element, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Move + + + + Move packed single-precision (32-bit) floating-point elements from "a" into "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). FOR j := 0 to 15 i := j*32 IF k[j] - dst[i+31:i] := e^(a[i+31:i]) - 1.0 + dst[i+31:i] := a[i+31:i] ELSE - dst[i+31:i] := src[i+31:i] + dst[i+31:i] := 0 FI ENDFOR dst[MAX:512] := 0 +
immintrin.h
 - + Floating Point AVX512F - Special Math Functions - - Round the packed double-precision (64-bit) floating-point elements in "a" down to an integer value, and store the results as packed double-precision floating-point elements in "dst". + Move + + + + + Duplicate even-indexed double-precision (64-bit) floating-point elements from "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). +tmp[63:0] := a[63:0] +tmp[127:64] := a[63:0] +tmp[191:128] := a[191:128] +tmp[255:192] := a[191:128] +tmp[319:256] := a[319:256] +tmp[383:320] := a[319:256] +tmp[447:384] := a[447:384] +tmp[511:448] := a[447:384] FOR j := 0 to 7 i := j*64 - dst[i+63:i] := FLOOR(a[i+63:i]) + IF k[j] + dst[i+63:i] := tmp[i+63:i] + ELSE + dst[i+63:i] := src[i+63:i] + FI ENDFOR dst[MAX:512] := 0 +
immintrin.h
 - + Floating Point AVX512F - Special Math Functions - - - - Round the packed double-precision (64-bit) floating-point elements in "a" down to an integer value, and store the results as packed double-precision floating-point elements in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Move + + + + Duplicate even-indexed double-precision (64-bit) floating-point elements from "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). +tmp[63:0] := a[63:0] +tmp[127:64] := a[63:0] +tmp[191:128] := a[191:128] +tmp[255:192] := a[191:128] +tmp[319:256] := a[319:256] +tmp[383:320] := a[319:256] +tmp[447:384] := a[447:384] +tmp[511:448] := a[447:384] FOR j := 0 to 7 i := j*64 IF k[j] - dst[i+63:i] := FLOOR(a[i+63:i]) + dst[i+63:i] := tmp[i+63:i] ELSE - dst[i+63:i] := src[i+63:i] + dst[i+63:i] := 0 FI ENDFOR dst[MAX:512] := 0 +
immintrin.h
 - + Floating Point AVX512F - Special Math Functions - - Round the packed single-precision (32-bit) floating-point elements in "a" down to an integer value, and store the results as packed single-precision floating-point elements in "dst". + Move + + + Duplicate even-indexed double-precision (64-bit) floating-point elements from "a", and store the results in "dst". + +dst[63:0] := a[63:0] +dst[127:64] := a[63:0] +dst[191:128] := a[191:128] +dst[255:192] := a[191:128] +dst[319:256] := a[319:256] +dst[383:320] := a[319:256] +dst[447:384] := a[447:384] +dst[511:448] := a[447:384] +dst[MAX:512] := 0 + + +
immintrin.h
+ + + Integer + AVX512F + Load + + + + Load packed 32-bit integers from memory into "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + "mem_addr" must be aligned on a 64-byte boundary or a general-protection exception may be generated. FOR j := 0 to 15 i := j*32 - dst[i+31:i] := FLOOR(a[i+31:i]) + IF k[j] + dst[i+31:i] := MEM[mem_addr+i+31:mem_addr+i] + ELSE + dst[i+31:i] := 0 + FI ENDFOR dst[MAX:512] := 0 +
immintrin.h
 - - Floating Point + + Integer AVX512F - Special Math Functions - - - - Round the packed single-precision (32-bit) floating-point elements in "a" down to an integer value, and store the results as packed single-precision floating-point elements in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Move + + + + Move packed 32-bit integers from "a" into "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). FOR j := 0 to 15 i := j*32 IF k[j] - dst[i+31:i] := FLOOR(a[i+31:i]) + dst[i+31:i] := a[i+31:i] ELSE - dst[i+31:i] := src[i+63:i] + dst[i+31:i] := 0 FI ENDFOR dst[MAX:512] := 0 +
immintrin.h
 - - Floating Point + + Integer AVX512F - Elementary Math Functions - - - Compute the length of the hypotenous of a right triangle, with the lengths of the other two sides of the triangle stored as packed double-precision (64-bit) floating-point elements in "a" and "b", and store the results in "dst". + Load + + + + Load packed 64-bit integers from memory into "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + "mem_addr" must be aligned on a 64-byte boundary or a general-protection exception may be generated. FOR j := 0 to 7 i := j*64 - dst[i+63:i] := SQRT(a[i+63:i]^2 + b[i+63:i]^2) + IF k[j] + dst[i+63:i] := MEM[mem_addr+i+63:mem_addr+i] + ELSE + dst[i+63:i] := 0 + FI ENDFOR dst[MAX:512] := 0 +
immintrin.h
 - - Floating Point + + Integer AVX512F - Elementary Math Functions - - - - - Compute the length of the hypotenous of a right triangle, with the lengths of the other two sides of the triangle stored as packed double-precision (64-bit) floating-point elements in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Move + + + + Move packed 64-bit integers from "a" into "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). FOR j := 0 to 7 i := j*64 IF k[j] - dst[i+63:i] := SQRT(a[i+63:i]^2 + b[i+63:i]^2) + dst[i+63:i] := a[i+63:i] ELSE - dst[i+63:i] := src[i+63:i] + dst[i+63:i] := 0 FI ENDFOR dst[MAX:512] := 0 +
immintrin.h
 - - Floating Point + + Integer AVX512F - Elementary Math Functions - - - Compute the length of the hypotenous of a right triangle, with the lengths of the other two sides of the triangle stored as packed single-precision (32-bit) floating-point elements in "a" and "b", and store the results in "dst". + Load + + + Load 512-bits of integer data from memory into "dst". + "mem_addr" does not need to be aligned on any particular boundary. -FOR j := 0 to 15 - i := j*32 - dst[i+31:i] := SQRT(a[i+31:i]^2 + b[i+31:i]^2) -ENDFOR +dst[511:0] := MEM[mem_addr+511:mem_addr] dst[MAX:512] := 0 +
immintrin.h
 - - Floating Point + + Integer AVX512F - Elementary Math Functions - - - - - Compute the length of the hypotenous of a right triangle, with the lengths of the other two sides of the triangle stored as packed single-precision (32-bit) floating-point elements in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Load + + + + + Load packed 32-bit integers from memory into "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + "mem_addr" does not need to be aligned on any particular boundary. FOR j := 0 to 15 i := j*32 IF k[j] - dst[i+31:i] := SQRT(a[i+31:i]^2 + b[i+31:i]^2) + dst[i+31:i] := MEM[mem_addr+i+31:mem_addr+i] ELSE dst[i+31:i] := src[i+31:i] FI ENDFOR dst[MAX:512] := 0 +
immintrin.h
 - + Integer AVX512F - Arithmetic - - - Divide packed 32-bit integers in "a" by packed elements in "b", and store the truncated results in "dst". + Store + + + + + Store packed 32-bit integers from "a" into memory using writemask "k". + "mem_addr" does not need to be aligned on any particular boundary. FOR j := 0 to 15 - i := 32*j - dst[i+31:i] := TRUNCATE(a[i+31:i] / b[i+31:i]) + i := j*32 + IF k[j] + MEM[mem_addr+i+31:mem_addr+i] := a[i+31:i] + FI ENDFOR -dst[MAX:512] := 0 +
immintrin.h
 - + Integer AVX512F - Arithmetic - - - - - Divide packed 32-bit integers in "a" by packed elements in "b", and store the truncated results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Load + + + + Load packed 32-bit integers from memory into "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + "mem_addr" does not need to be aligned on any particular boundary. FOR j := 0 to 15 - i := 32*j + i := j*32 IF k[j] - dst[i+31:i] := TRUNCATE(a[i+31:i] / b[i+31:i]) + dst[i+31:i] := MEM[mem_addr+i+31:mem_addr+i] ELSE - dst[i+31:i] := src[i+31:i] + dst[i+31:i] := 0 FI ENDFOR dst[MAX:512] := 0 +
immintrin.h
 - - Integer - AVX512F - Arithmetic - - - Divide packed 8-bit integers in "a" by packed elements in "b", and store the truncated results in "dst". - -FOR j := 0 to 63 - i := 8*j - dst[i+7:i] := TRUNCATE(a[i+7:i] / b[i+7:i]) -ENDFOR -dst[MAX:512] := 0 - -
immintrin.h
- - + Integer AVX512F - Arithmetic - - - Divide packed 16-bit integers in "a" by packed elements in "b", and store the truncated results in "dst". + Store + + + + Store 512-bits of integer data from "a" into memory. + "mem_addr" does not need to be aligned on any particular boundary. -FOR j := 0 to 31 - i := 16*j - dst[i+15:i] := TRUNCATE(a[i+15:i] / b[i+15:i]) -ENDFOR -dst[MAX:512] := 0 +MEM[mem_addr+511:mem_addr] := a[511:0] +
immintrin.h
 - + Integer AVX512F - Arithmetic - - - Divide packed 64-bit integers in "a" by packed elements in "b", and store the truncated results in "dst". + Load + + + + + Load packed 64-bit integers from memory into "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + "mem_addr" does not need to be aligned on any particular boundary. FOR j := 0 to 7 - i := 64*j - dst[i+63:i] := TRUNCATE(a[i+63:i] / b[i+63:i]) + i := j*64 + IF k[j] + dst[i+63:i] := MEM[mem_addr+i+63:mem_addr+i] + ELSE + dst[i+63:i] := src[i+63:i] + FI ENDFOR dst[MAX:512] := 0 +
immintrin.h
 - - Floating Point + + Integer AVX512F - Elementary Math Functions - - Compute the inverse square root of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst". + Store + + + + + Store packed 64-bit integers from "a" into memory using writemask "k". + "mem_addr" does not need to be aligned on any particular boundary. FOR j := 0 to 7 i := j*64 - dst[i+63:i] := InvSQRT(a[i+63:i]) + IF k[j] + MEM[mem_addr+i+63:mem_addr+i] := a[i+63:i] + FI ENDFOR -dst[MAX:512] := 0 +
immintrin.h
 - - Floating Point + + Integer AVX512F - Elementary Math Functions - - - - Compute the inverse square root of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Load + + + + Load packed 64-bit integers from memory into "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + "mem_addr" does not need to be aligned on any particular boundary. FOR j := 0 to 7 i := j*64 IF k[j] - dst[i+63:i] := InvSQRT(a[i+63:i]) + dst[i+63:i] := MEM[mem_addr+i+63:mem_addr+i] ELSE - dst[i+63:i] := src[i+63:i] + dst[i+63:i] := 0 FI ENDFOR dst[MAX:512] := 0 +
immintrin.h
 - - Floating Point - AVX512F - Elementary Math Functions - - Compute the inverse square root of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst". - -FOR j := 0 to 15 - i := j*32 - dst[i+31:i] := InvSQRT(a[i+31:i]) -ENDFOR -dst[MAX:512] := 0 - -
immintrin.h
- - - Floating Point + + Integer AVX512F - Elementary Math Functions - - - - Compute the inverse square root of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Load + + + Load 512-bits of integer data from memory into "dst" using a non-temporal memory hint. + "mem_addr" must be aligned on a 64-byte boundary or a general-protection exception may be generated. -FOR j := 0 to 15 - i := j*32 - IF k[j] - dst[i+31:i] := InvSQRT(a[i+31:i]) - ELSE - dst[i+31:i] := src[i+31:i] - FI -ENDFOR +dst[511:0] := MEM[mem_addr+511:mem_addr] dst[MAX:512] := 0 +
immintrin.h
 - + Integer AVX512F - Arithmetic - - - Divide packed 32-bit integers in "a" by packed elements in "b", and store the remainders as packed 32-bit integers in "dst". + Store + + + + Store 512-bits of integer data from "a" into memory using a non-temporal memory hint. + "mem_addr" must be aligned on a 64-byte boundary or a general-protection exception may be generated. -FOR j := 0 to 15 - i := 32*j - dst[i+31:i] := REMAINDER(a[i+31:i] / b[i+31:i]) -ENDFOR -dst[MAX:512] := 0 +MEM[mem_addr+511:mem_addr] := a[511:0] +
immintrin.h
 - - Integer + + Floating Point AVX512F - Arithmetic - - - - - Divide packed 32-bit integers in "a" by packed elements in "b", and store the remainders as packed 32-bit integers in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Store + + + + Store 512-bits (composed of 8 packed double-precision (64-bit) floating-point elements) from "a" into memory using a non-temporal memory hint. + "mem_addr" must be aligned on a 64-byte boundary or a general-protection exception may be generated. -FOR j := 0 to 15 - i := 32*j - IF k[j] - dst[i+31:i] := REMAINDER(a[i+31:i] / b[i+31:i]) - ELSE - dst[i+31:i] := src[i+31:i] - FI -ENDFOR -dst[MAX:512] := 0 +MEM[mem_addr+511:mem_addr] := a[511:0] +
immintrin.h
 - - Integer + + Floating Point AVX512F - Arithmetic - - - Divide packed 8-bit integers in "a" by packed elements in "b", and store the remainders as packed 32-bit integers in "dst". + Store + + + + Store 512-bits (composed of 16 packed single-precision (32-bit) floating-point elements) from "a" into memory using a non-temporal memory hint. + "mem_addr" must be aligned on a 64-byte boundary or a general-protection exception may be generated. -FOR j := 0 to 63 - i := 8*j - dst[i+7:i] := REMAINDER(a[i+7:i] / b[i+7:i]) -ENDFOR -dst[MAX:512] := 0 +MEM[mem_addr+511:mem_addr] := a[511:0] +
immintrin.h
 - - Integer + + Floating Point AVX512F - Arithmetic - - - Divide packed 16-bit integers in "a" by packed elements in "b", and store the remainders as packed 32-bit integers in "dst". + Load + + + + + Load a double-precision (64-bit) floating-point element from memory into the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and set the upper element of "dst" to zero. "mem_addr" must be aligned on a 16-byte boundary or a general-protection exception may be generated. -FOR j := 0 to 31 - i := 16*j - dst[i+15:i] := REMAINDER(a[i+15:i] / b[i+15:i]) -ENDFOR -dst[MAX:512] := 0 +IF k[0] + dst[63:0] := MEM[mem_addr+63:mem_addr] +ELSE + dst[63:0] := src[63:0] +FI +dst[MAX:64] := 0 +
immintrin.h
 - - Integer + + Floating Point AVX512F - Arithmetic - - - Divide packed 64-bit integers in "a" by packed elements in "b", and store the remainders as packed 32-bit integers in "dst". + Move + + + + + + Move the lower double-precision (64-bit) floating-point element from "b" to the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". -FOR j := 0 to 7 - i := 64*j - dst[i+63:i] := REMAINDER(a[i+63:i] / b[i+63:i]) -ENDFOR -dst[MAX:512] := 0 +IF k[0] + dst[63:0] := b[63:0] +ELSE + dst[63:0] := src[63:0] +FI +dst[127:64] := a[127:64] +dst[MAX:128] := 0 +
immintrin.h
 - + Floating Point AVX512F - Elementary Math Functions - - Compute the base-10 logarithm of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst". + Store + + + + + Store the lower double-precision (64-bit) floating-point element from "a" into memory using writemask "k". + "mem_addr" must be aligned on a 16-byte boundary or a general-protection exception may be generated. -FOR j := 0 to 7 - i := j*64 - dst[i+63:i] := log10(a[i+63:i]) -ENDFOR -dst[MAX:512] := 0 +IF k[0] + MEM[mem_addr+63:mem_addr] := a[63:0] +FI +
immintrin.h
 - + Floating Point AVX512F - Elementary Math Functions - - - - Compute the base-10 logarithm of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Load + + + + Load a double-precision (64-bit) floating-point element from memory into the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and set the upper element of "dst" to zero. "mem_addr" must be aligned on a 16-byte boundary or a general-protection exception may be generated. -FOR j := 0 to 7 - i := j*64 - IF k[j] - dst[i+63:i] := log10(a[i+63:i]) - ELSE - dst[i+63:i] := src[i+63:i] - FI -ENDFOR -dst[MAX:512] := 0 +IF k[0] + dst[63:0] := MEM[mem_addr+63:mem_addr] +ELSE + dst[63:0] := 0 +FI +dst[MAX:64] := 0 +
immintrin.h
 - + Floating Point AVX512F - Elementary Math Functions - - Compute the base-10 logarithm of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst". + Move + + + + + Move the lower double-precision (64-bit) floating-point element from "b" to the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". -FOR j := 0 to 15 - i := j*32 - dst[i+31:i] := log10(a[i+31:i]) -ENDFOR -dst[MAX:512] := 0 +IF k[0] + dst[63:0] := b[63:0] +ELSE + dst[63:0] := 0 +FI +dst[127:64] := a[127:64] +dst[MAX:128] := 0 +
immintrin.h
 - + Floating Point AVX512F - Elementary Math Functions - - - - Compute the base-10 logarithm of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Move + + + + + Duplicate odd-indexed single-precision (32-bit) floating-point elements from "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). +tmp[31:0] := a[63:32] +tmp[63:32] := a[63:32] +tmp[95:64] := a[127:96] +tmp[127:96] := a[127:96] +tmp[159:128] := a[191:160] +tmp[191:160] := a[191:160] +tmp[223:192] := a[255:224] +tmp[255:224] := a[255:224] +tmp[287:256] := a[319:288] +tmp[319:288] := a[319:288] +tmp[351:320] := a[383:352] +tmp[383:352] := a[383:352] +tmp[415:384] := a[447:416] +tmp[447:416] := a[447:416] +tmp[479:448] := a[511:480] +tmp[511:480] := a[511:480] FOR j := 0 to 15 i := j*32 IF k[j] - dst[i+31:i] := log10(a[i+31:i]) + dst[i+31:i] := tmp[i+31:i] ELSE dst[i+31:i] := src[i+31:i] FI ENDFOR dst[MAX:512] := 0 +
immintrin.h
 - + Floating Point AVX512F - Elementary Math Functions - - Compute the natural logarithm of one plus packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst". + Move + + + + Duplicate odd-indexed single-precision (32-bit) floating-point elements from "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := j*64 - dst[i+63:i] := ln(1.0 + a[i+63:i]) +tmp[31:0] := a[63:32] +tmp[63:32] := a[63:32] +tmp[95:64] := a[127:96] +tmp[127:96] := a[127:96] +tmp[159:128] := a[191:160] +tmp[191:160] := a[191:160] +tmp[223:192] := a[255:224] +tmp[255:224] := a[255:224] +tmp[287:256] := a[319:288] +tmp[319:288] := a[319:288] +tmp[351:320] := a[383:352] +tmp[383:352] := a[383:352] +tmp[415:384] := a[447:416] +tmp[447:416] := a[447:416] +tmp[479:448] := a[511:480] +tmp[511:480] := a[511:480] +FOR j := 0 to 15 + i := j*32 + IF k[j] + dst[i+31:i] := tmp[i+31:i] + ELSE + dst[i+31:i] := 0 + FI ENDFOR dst[MAX:512] := 0 +
immintrin.h
 - + Floating Point AVX512F - Elementary Math Functions - - - - Compute the natural logarithm of one plus packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Move + + + Duplicate odd-indexed single-precision (32-bit) floating-point elements from "a", and store the results in "dst". -FOR j := 0 to 7 - i := j*64 - IF k[j] - dst[i+63:i] := ln(1.0 + a[i+63:i]) - ELSE - dst[i+63:i] := src[i+63:i] - FI -ENDFOR +dst[31:0] := a[63:32] +dst[63:32] := a[63:32] +dst[95:64] := a[127:96] +dst[127:96] := a[127:96] +dst[159:128] := a[191:160] +dst[191:160] := a[191:160] +dst[223:192] := a[255:224] +dst[255:224] := a[255:224] +dst[287:256] := a[319:288] +dst[319:288] := a[319:288] +dst[351:320] := a[383:352] +dst[383:352] := a[383:352] +dst[415:384] := a[447:416] +dst[447:416] := a[447:416] +dst[479:448] := a[511:480] +dst[511:480] := a[511:480] dst[MAX:512] := 0 +
immintrin.h
 - + Floating Point AVX512F - Elementary Math Functions - - Compute the natural logarithm of one plus packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst". + Move + + + + + Duplicate even-indexed single-precision (32-bit) floating-point elements from "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). +tmp[31:0] := a[31:0] +tmp[63:32] := a[31:0] +tmp[95:64] := a[95:64] +tmp[127:96] := a[95:64] +tmp[159:128] := a[159:128] +tmp[191:160] := a[159:128] +tmp[223:192] := a[223:192] +tmp[255:224] := a[223:192] +tmp[287:256] := a[287:256] +tmp[319:288] := a[287:256] +tmp[351:320] := a[351:320] +tmp[383:352] := a[351:320] +tmp[415:384] := a[415:384] +tmp[447:416] := a[415:384] +tmp[479:448] := a[479:448] +tmp[511:480] := a[479:448] FOR j := 0 to 15 i := j*32 - dst[i+31:i] := ln(1.0 + a[i+31:i]) -ENDFOR + IF k[j] + dst[i+31:i] := tmp[i+31:i] + ELSE + dst[i+31:i] := src[i+31:i] + FI +ENDFOR dst[MAX:512] := 0 +
immintrin.h
 - + Floating Point AVX512F - Elementary Math Functions - - - - Compute the natural logarithm of one plus packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Move + + + + Duplicate even-indexed single-precision (32-bit) floating-point elements from "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). +tmp[31:0] := a[31:0] +tmp[63:32] := a[31:0] +tmp[95:64] := a[95:64] +tmp[127:96] := a[95:64] +tmp[159:128] := a[159:128] +tmp[191:160] := a[159:128] +tmp[223:192] := a[223:192] +tmp[255:224] := a[223:192] +tmp[287:256] := a[287:256] +tmp[319:288] := a[287:256] +tmp[351:320] := a[351:320] +tmp[383:352] := a[351:320] +tmp[415:384] := a[415:384] +tmp[447:416] := a[415:384] +tmp[479:448] := a[479:448] +tmp[511:480] := a[479:448] FOR j := 0 to 15 i := j*32 IF k[j] - dst[i+31:i] := ln(1.0 + a[i+31:i]) + dst[i+31:i] := tmp[i+31:i] ELSE - dst[i+31:i] := src[i+31:i] + dst[i+31:i] := 0 FI ENDFOR dst[MAX:512] := 0 +
immintrin.h
 - + Floating Point AVX512F - Elementary Math Functions - - Compute the base-2 logarithm of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst". + Move + + + Duplicate even-indexed single-precision (32-bit) floating-point elements from "a", and store the results in "dst". -FOR j := 0 to 7 - i := j*64 - dst[i+63:i] := log2(a[i+63:i]) -ENDFOR +dst[31:0] := a[31:0] +dst[63:32] := a[31:0] +dst[95:64] := a[95:64] +dst[127:96] := a[95:64] +dst[159:128] := a[159:128] +dst[191:160] := a[159:128] +dst[223:192] := a[223:192] +dst[255:224] := a[223:192] +dst[287:256] := a[287:256] +dst[319:288] := a[287:256] +dst[351:320] := a[351:320] +dst[383:352] := a[351:320] +dst[415:384] := a[415:384] +dst[447:416] := a[415:384] +dst[479:448] := a[479:448] +dst[511:480] := a[479:448] dst[MAX:512] := 0 +
immintrin.h
 - + Floating Point AVX512F - Elementary Math Functions - - - - Compute the base-2 logarithm of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Load + + + + + Load a single-precision (32-bit) floating-point element from memory into the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and set the upper elements of "dst" to zero. "mem_addr" must be aligned on a 16-byte boundary or a general-protection exception may be generated. -FOR j := 0 to 7 - i := j*64 - IF k[j] - dst[i+63:i] := log2(a[i+63:i]) - ELSE - dst[i+63:i] := src[i+63:i] - FI -ENDFOR -dst[MAX:512] := 0 +IF k[0] + dst[31:0] := MEM[mem_addr+31:mem_addr] +ELSE + dst[31:0] := src[31:0] +FI +dst[MAX:32] := 0 +
immintrin.h
 - + Floating Point - KNCNI - Elementary Math Functions - - Compute the base-2 logarithm of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst". + AVX512F + Move + + + + + + Move the lower single-precision (32-bit) floating-point element from "b" to the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". -FOR j := 0 to 15 - i := j*32 - dst[i+31:i] := log2(a[i+31:i]) -ENDFOR -dst[MAX:512] := 0 +IF k[0] + dst[31:0] := b[31:0] +ELSE + dst[31:0] := src[31:0] +FI +dst[127:32] := a[127:32] +dst[MAX:128] := 0 +
immintrin.h
- - + Floating Point - KNCNI - Elementary Math Functions - - - - Compute the base-2 logarithm of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + AVX512F + Store + + + + + Store the lower single-precision (32-bit) floating-point element from "a" into memory using writemask "k". + "mem_addr" must be aligned on a 16-byte boundary or a general-protection exception may be generated. -FOR j := 0 to 15 - i := j*32 - IF k[j] - dst[i+31:i] := log2(a[i+31:i]) - ELSE - dst[i+31:i] := src[i+31:i] - FI -ENDFOR -dst[MAX:512] := 0 +IF k[0] + MEM[mem_addr+31:mem_addr] := a[31:0] +FI - +
immintrin.h
 - + Floating Point AVX512F - Elementary Math Functions - - Compute the natural logarithm of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst". + Load + + + + Load a single-precision (32-bit) floating-point element from memory into the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and set the upper elements of "dst" to zero. "mem_addr" must be aligned on a 16-byte boundary or a general-protection exception may be generated. -FOR j := 0 to 7 - i := j*64 - dst[i+63:i] := ln(a[i+63:i]) -ENDFOR -dst[MAX:512] := 0 +IF k[0] + dst[31:0] := MEM[mem_addr+31:mem_addr] +ELSE + dst[31:0] := 0 +FI +dst[MAX:32] := 0 +
immintrin.h
 - + Floating Point AVX512F - Elementary Math Functions - - - - Compute the natural logarithm of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Move + + + + + Move the lower single-precision (32-bit) floating-point element from "b" to the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". -FOR j := 0 to 7 - i := j*64 - IF k[j] - dst[i+63:i] := ln(a[i+63:i]) - ELSE - dst[i+63:i] := src[i+63:i] - FI -ENDFOR -dst[MAX:512] := 0 +IF k[0] + dst[31:0] := b[31:0] +ELSE + dst[31:0] := 0 +FI +dst[127:32] := a[127:32] +dst[MAX:128] := 0 +
immintrin.h
 - + Floating Point AVX512F - Elementary Math Functions - - Compute the natural logarithm of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst". + Load + + + Load 512-bits (composed of 8 packed double-precision (64-bit) floating-point elements) from memory into "dst". + "mem_addr" does not need to be aligned on any particular boundary. -FOR j := 0 to 15 - i := j*32 - dst[i+31:i] := ln(a[i+31:i]) -ENDFOR +dst[511:0] := MEM[mem_addr+511:mem_addr] dst[MAX:512] := 0 +
immintrin.h
 - + Floating Point AVX512F - Elementary Math Functions - - - - Compute the natural logarithm of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Load + + + + + Load packed double-precision (64-bit) floating-point elements from memoy into "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + "mem_addr" does not need to be aligned on any particular boundary. -FOR j := 0 to 15 - i := j*32 +FOR j := 0 to 7 + i := j*64 IF k[j] - dst[i+31:i] := ln(a[i+31:i]) + dst[i+63:i] := MEM[mem_addr+i+63:mem_addr+i] ELSE - dst[i+31:i] := src[i+31:i] + dst[i+63:i] := src[i+63:i] FI ENDFOR dst[MAX:512] := 0 +
immintrin.h
 - + Floating Point AVX512F - Elementary Math Functions - - Convert the exponent of each packed double-precision (64-bit) floating-point element in "a" to a double-precision floating-point number representing the integer exponent, and store the results in "dst". This intrinsic essentially calculates "floor(log2(x))" for each element. + Store + + + + + Store packed double-precision (64-bit) floating-point elements from "a" into memory using writemask "k". + "mem_addr" does not need to be aligned on any particular boundary. FOR j := 0 to 7 i := j*64 - dst[i+63:i] := ConvertExpFP64(a[i+63:i]) + IF k[j] + MEM[mem_addr+i+63:mem_addr+i] := a[i+63:i] + FI ENDFOR -dst[MAX:512] := 0 +
immintrin.h
 - + Floating Point AVX512F - Elementary Math Functions - - - - Convert the exponent of each packed double-precision (64-bit) floating-point element in "a" to a double-precision floating-point number representing the integer exponent, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). This intrinsic essentially calculates "floor(log2(x))" for each element. + Load + + + + Load packed double-precision (64-bit) floating-point elements from memoy into "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + "mem_addr" does not need to be aligned on any particular boundary. FOR j := 0 to 7 i := j*64 IF k[j] - dst[i+63:i] := ConvertExpFP64(a[i+63:i]) + dst[i+63:i] := MEM[mem_addr+i+63:mem_addr+i] ELSE - dst[i+63:i] := src[i+63:i] + dst[i+63:i] := 0 FI ENDFOR dst[MAX:512] := 0 +
immintrin.h
 - - Floating Point - AVX512F - Elementary Math Functions - - Convert the exponent of each packed single-precision (32-bit) floating-point element in "a" to a single-precision floating-point number representing the integer exponent, and store the results in "dst". This intrinsic essentially calculates "floor(log2(x))" for each element. - -FOR j := 0 to 15 - i := j*32 - dst[i+31:i] := ConvertExpFP32(a[i+31:i]) -ENDFOR -dst[MAX:512] := 0 - -
immintrin.h
- - + Floating Point AVX512F - Elementary Math Functions - - - - Convert the exponent of each packed single-precision (32-bit) floating-point element in "a" to a single-precision floating-point number representing the integer exponent, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). This intrinsic essentially calculates "floor(log2(x))" for each element. + Store + + + + Store 512-bits (composed of 8 packed double-precision (64-bit) floating-point elements) from "a" into memory. + "mem_addr" does not need to be aligned on any particular boundary. -FOR j := 0 to 15 - i := j*32 - IF k[j] - dst[i+31:i] := ConvertExpFP32(a[i+31:i]) - ELSE - dst[i+31:i] := src[i+31:i] - FI -ENDFOR -dst[MAX:512] := 0 +MEM[mem_addr+511:mem_addr] := a[511:0] +
immintrin.h
 - + Floating Point AVX512F - Special Math Functions - - Rounds each packed double-precision (64-bit) floating-point element in "a" to the nearest integer value and stores the results as packed double-precision floating-point elements in "dst". + Load + + + Load 512-bits (composed of 16 packed single-precision (32-bit) floating-point elements) from memory into "dst". + "mem_addr" does not need to be aligned on any particular boundary. -FOR j := 0 to 7 - i := j*64 - dst[i+63:i] := NearbyInt(a[i+63:i]) -ENDFOR +dst[511:0] := MEM[mem_addr+511:mem_addr] dst[MAX:512] := 0 +
immintrin.h
 - + Floating Point AVX512F - Special Math Functions - - - - Rounds each packed double-precision (64-bit) floating-point element in "a" to the nearest integer value and stores the results as packed double-precision floating-point elements in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Load + + + + + Load packed single-precision (32-bit) floating-point elements from memory into "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + "mem_addr" does not need to be aligned on any particular boundary. -FOR j := 0 to 7 - i := j*64 +FOR j := 0 to 15 + i := j*32 IF k[j] - dst[i+63:i] := NearbyInt(a[i+63:i]) + dst[i+31:i] := MEM[mem_addr+i+31:mem_addr+i] ELSE - dst[i+63:i] := src[i+63:i] + dst[i+31:i] := src[i+31:i] FI ENDFOR dst[MAX:512] := 0 +
immintrin.h
 - + Floating Point AVX512F - Special Math Functions - - Rounds each packed single-precision (32-bit) floating-point element in "a" to the nearest integer value and stores the results as packed double-precision floating-point elements in "dst". + Store + + + + + Store packed single-precision (32-bit) floating-point elements from "a" into memory using writemask "k". + "mem_addr" does not need to be aligned on any particular boundary. FOR j := 0 to 15 i := j*32 - dst[i+31:i] := NearbyInt(a[i+31:i]) + IF k[j] + MEM[mem_addr+i+31:mem_addr+i] := a[i+31:i] + FI ENDFOR -dst[MAX:512] := 0 +
immintrin.h
 - + Floating Point AVX512F - Special Math Functions - - - - Rounds each packed single-precision (32-bit) floating-point element in "a" to the nearest integer value and stores the results as packed double-precision floating-point elements in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Load + + + + Load packed single-precision (32-bit) floating-point elements from memory into "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + "mem_addr" does not need to be aligned on any particular boundary. FOR j := 0 to 15 i := j*32 IF k[j] - dst[i+31:i] := NearbyInt(a[i+31:i]) + dst[i+31:i] := MEM[mem_addr+i+31:mem_addr+i] ELSE - dst[i+31:i] := src[i+31:i] + dst[i+31:i] := 0 FI ENDFOR dst[MAX:512] := 0 +
immintrin.h
 - + Floating Point AVX512F - Elementary Math Functions - - - Compute the exponential value of packed double-precision (64-bit) floating-point elements in "a" raised by packed elements in "b", and store the results in "dst". + Store + + + + Store 512-bits (composed of 16 packed single-precision (32-bit) floating-point elements) from "a" into memory. + "mem_addr" does not need to be aligned on any particular boundary. -FOR j := 0 to 7 - i := j*64 - dst[i+63:i] := (a[i+63:i])^(b[i+63:i]) -ENDFOR -dst[MAX:512] := 0 +MEM[mem_addr+511:mem_addr] := a[511:0] +
immintrin.h
 - + Floating Point AVX512F - Elementary Math Functions - - - - - Compute the exponential value of packed double-precision (64-bit) floating-point elements in "a" raised by packed elements in "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Arithmetic + + + + + Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). FOR j := 0 to 7 i := j*64 IF k[j] - dst[i+63:i] := (a[i+63:i])^(b[i+63:i]) + dst[i+63:i] := a[i+63:i] * b[i+63:i] ELSE - dst[i+63:i] := src[i+63:i] + dst[i+63:i] := 0 FI ENDFOR dst[MAX:512] := 0 +
immintrin.h
 - + Floating Point AVX512F - Elementary Math Functions - - - Compute the exponential value of packed single-precision (32-bit) floating-point elements in "a" raised by packed elements in "b", and store the results in "dst". + Arithmetic + + + + + + Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + [round_note] -FOR j := 0 to 15 - i := j*32 - dst[i+31:i] := (a[i+31:i])^(b[i+31:i]) +FOR j := 0 to 7 + i := j*64 + IF k[j] + dst[i+63:i] := a[i+63:i] * b[i+63:i] + ELSE + dst[i+63:i] := 0 + FI ENDFOR dst[MAX:512] := 0 +
immintrin.h
 - + Floating Point AVX512F - Elementary Math Functions - - - - - Compute the exponential value of packed single-precision (32-bit) floating-point elements in "a" raised by packed elements in "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Arithmetic + + + + + Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). FOR j := 0 to 15 i := j*32 IF k[j] - dst[i+31:i] := (a[i+31:i])^(b[i+31:i]) + dst[i+31:i] := a[i+31:i] * b[i+31:i] ELSE - dst[i+31:i] := src[i+31:i] + dst[i+31:i] := 0 FI ENDFOR dst[MAX:512] := 0 +
immintrin.h
 - + Floating Point AVX512F - Elementary Math Functions - - Computes the reciprocal of packed double-precision (64-bit) floating-point elements in "a", storing the results in "dst". + Arithmetic + + + + + + Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + [round_note] -FOR j := 0 to 7 - i := j*64 - dst[i+63:i] := (1 / a[i+63:i]) +FOR j := 0 to 15 + i := j*32 + IF k[j] + dst[i+31:i] := a[i+31:i] * b[i+31:i] + ELSE + dst[i+31:i] := 0 + FI ENDFOR dst[MAX:512] := 0 +
immintrin.h
 - + Floating Point AVX512F - Elementary Math Functions - - - - Computes the reciprocal of packed double-precision (64-bit) floating-point elements in "a", storing the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Arithmetic + + + + + + + Multiply the lower double-precision (64-bit) floating-point element in "a" and "b", store the result in the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". + [round_note] -FOR j := 0 to 7 - i := j*64 - IF k[j] - dst[i+63:i] := (1 / a[i+63:i]) - ELSE - dst[i+63:i] := src[i+63:i] - FI -ENDFOR -dst[MAX:512] := 0 +IF k[0] + dst[63:0] := a[63:0] * b[63:0] +ELSE + dst[63:0] := src[63:0] +FI +dst[127:64] := a[127:64] +dst[MAX:128] := 0 +
immintrin.h
 - + Floating Point AVX512F - Elementary Math Functions - - Computes the reciprocal of packed single-precision (32-bit) floating-point elements in "a", storing the results in "dst". + Arithmetic + + + + + + Multiply the lower double-precision (64-bit) floating-point element in "a" and "b", store the result in the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". -FOR j := 0 to 15 - i := j*32 - dst[i+31:i] := (1 / a[i+31:i]) -ENDFOR -dst[MAX:512] := 0 +IF k[0] + dst[63:0] := a[63:0] * b[63:0] +ELSE + dst[63:0] := src[63:0] +FI +dst[127:64] := a[127:64] +dst[MAX:128] := 0 +
immintrin.h
 - + Floating Point AVX512F - Elementary Math Functions - - - - Computes the reciprocal of packed single-precision (32-bit) floating-point elements in "a", storing the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Arithmetic + + + + + + Multiply the lower double-precision (64-bit) floating-point element in "a" and "b", store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". + [round_note] -FOR j := 0 to 15 - i := j*32 - IF k[j] - dst[i+31:i] := (1 / a[i+31:i]) - ELSE - dst[i+31:i] := src[i+31:i] - FI -ENDFOR -dst[MAX:512] := 0 +IF k[0] + dst[63:0] := a[63:0] * b[63:0] +ELSE + dst[63:0] := 0 +FI +dst[127:64] := a[127:64] +dst[MAX:128] := 0 +
immintrin.h
 - + Floating Point AVX512F - Special Math Functions - - Rounds the packed double-precision (64-bit) floating-point elements in "a" to the nearest even integer value and stores the results in "dst". + Arithmetic + + + + + Multiply the lower double-precision (64-bit) floating-point element in "a" and "b", store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". -FOR j := 0 to 7 - i := j*64 - dst[i+63:i] := RoundToNearestEven(a[i+63:i]) -ENDFOR -dst[MAX:512] := 0 +IF k[0] + dst[63:0] := a[63:0] * b[63:0] +ELSE + dst[63:0] := 0 +FI +dst[127:64] := a[127:64] +dst[MAX:128] := 0 +
immintrin.h
 - + Floating Point AVX512F - Special Math Functions - - - - Rounds the packed double-precision (64-bit) floating-point elements in "a" to the nearest even integer value and stores the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Arithmetic + + + + + Multiply the lower double-precision (64-bit) floating-point element in "a" and "b", store the result in the lower element of "dst", and copy the upper element from "a" to the upper element of "dst". + [round_note] -FOR j := 0 to 7 - i := j*64 - IF k[j] - dst[i+63:i] := RoundToNearestEven(a[i+63:i]) - ELSE - dst[i+63:i] := src[i+63:i] - FI -ENDFOR -dst[MAX:512] := 0 +dst[63:0] := a[63:0] * b[63:0] +dst[127:64] := a[127:64] +dst[MAX:128] := 0 +
immintrin.h
 - + Floating Point AVX512F - Special Math Functions - - Rounds the packed single-precision (32-bit) floating-point elements in "a" to the nearest even integer value and stores the results in "dst". + Arithmetic + + + + + + + Multiply the lower single-precision (32-bit) floating-point element in "a" and "b", store the result in the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". + [round_note] -FOR j := 0 to 15 - i := j*32 - dst[i+31:i] := RoundToNearestEven(a[i+31:i]) -ENDFOR -dst[MAX:512] := 0 +IF k[0] + dst[31:0] := a[31:0] * b[31:0] +ELSE + dst[31:0] := src[31:0] +FI +dst[127:32] := a[127:32] +dst[MAX:128] := 0 +
immintrin.h
 - + Floating Point AVX512F - Special Math Functions - - - - Rounds the packed single-precision (32-bit) floating-point elements in "a" to the nearest even integer value and stores the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Arithmetic + + + + + + Multiply the lower single-precision (32-bit) floating-point element in "a" and "b", store the result in the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". -FOR j := 0 to 15 - i := j*32 - IF k[j] - dst[i+31:i] := RoundToNearestEven(a[i+31:i]) - ELSE - dst[i+31:i] := src[i+31:i] - FI -ENDFOR -dst[MAX:512] := 0 +IF k[0] + dst[31:0] := a[31:0] * b[31:0] +ELSE + dst[31:0] := src[31:0] +FI +dst[127:32] := a[127:32] +dst[MAX:128] := 0 +
immintrin.h
 - + Floating Point AVX512F - Special Math Functions - - Round the packed double-precision (64-bit) floating-point elements in "a" to the nearest integer value, and store the results as packed double-precision floating-point elements in "dst". + Arithmetic + + + + + + Multiply the lower single-precision (32-bit) floating-point element in "a" and "b", store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". + [round_note] -FOR j := 0 to 7 - i := j*64 - dst[i+63:i] := ROUND(a[i+63:i]) -ENDFOR -dst[MAX:512] := 0 +IF k[0] + dst[31:0] := a[31:0] * b[31:0] +ELSE + dst[31:0] := 0 +FI +dst[127:32] := a[127:32] +dst[MAX:128] := 0 +
immintrin.h
 - + Floating Point AVX512F - Special Math Functions - - - - Round the packed double-precision (64-bit) floating-point elements in "a" to the nearest integer value, and store the results as packed double-precision floating-point elements in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - [round_note] - + Arithmetic + + + + + Multiply the lower single-precision (32-bit) floating-point element in "a" and "b", store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". -FOR j := 0 to 7 - i := j*64 - IF k[j] - dst[i+63:i] := ROUND(a[i+63:i]) - ELSE - dst[i+63:i] := src[i+63:i] - FI -ENDFOR -dst[MAX:512] := 0 +IF k[0] + dst[31:0] := a[31:0] * b[31:0] +ELSE + dst[31:0] := 0 +FI +dst[127:32] := a[127:32] +dst[MAX:128] := 0 +
immintrin.h
 - + Floating Point AVX512F - Trigonometry - - Compute the sine of packed double-precision (64-bit) floating-point elements in "a" expressed in radians, and store the results in "dst". + Arithmetic + + + + + Multiply the lower single-precision (32-bit) floating-point element in "a" and "b", store the result in the lower element of "dst", and copy the upper 3 packed elements from "a" to the upper elements of "dst". + [round_note] -FOR j := 0 to 7 - i := j*64 - dst[i+63:i] := SIN(a[i+63:i]) -ENDFOR -dst[MAX:512] := 0 +dst[31:0] := a[31:0] * b[31:0] +dst[127:32] := a[127:32] +dst[MAX:128] := 0 +
immintrin.h
 - - Floating Point + + Integer AVX512F - Trigonometry - - - - Compute the sine of packed double-precision (64-bit) floating-point elements in "a" expressed in radians, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Special Math Functions + + + Compute the absolute value of packed signed 32-bit integers in "a", and store the unsigned results in "dst". -FOR j := 0 to 7 - i := j*64 - IF k[j] - dst[i+63:i] := SIN(a[i+63:i]) - ELSE - dst[i+63:i] := src[i+63:i] - FI +FOR j := 0 to 15 + i := j*32 + dst[i+31:i] := ABS(a[i+31:i]) ENDFOR dst[MAX:512] := 0 +
immintrin.h
 - - Floating Point + + Integer AVX512F - Trigonometry - - Compute the sine of packed single-precision (32-bit) floating-point elements in "a" expressed in radians, and store the results in "dst". + Special Math Functions + + + + + Compute the absolute value of packed signed 32-bit integers in "a", and store the unsigned results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). FOR j := 0 to 15 i := j*32 - dst[i+31:i] := SIN(a[i+31:i]) + IF k[j] + dst[i+31:i] := ABS(a[i+31:i]) + ELSE + dst[i+31:i] := src[i+31:i] + FI ENDFOR dst[MAX:512] := 0 +
immintrin.h
 - - Floating Point + + Integer AVX512F - Trigonometry - - - - Compute the sine of packed single-precision (32-bit) floating-point elements in "a" expressed in radians, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Special Math Functions + + + + Compute the absolute value of packed signed 32-bit integers in "a", and store the unsigned results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). FOR j := 0 to 15 i := j*32 IF k[j] - dst[i+31:i] := SIN(a[i+31:i]) + dst[i+31:i] := ABS(a[i+31:i]) ELSE - dst[i+31:i] := src[i+31:i] + dst[i+31:i] := 0 FI ENDFOR dst[MAX:512] := 0 +
immintrin.h
 - - Floating Point + + Integer AVX512F - Trigonometry - - Compute the hyperbolic sine of packed double-precision (64-bit) floating-point elements in "a" expressed in radians, and store the results in "dst". + Special Math Functions + + + Compute the absolute value of packed signed 64-bit integers in "a", and store the unsigned results in "dst". FOR j := 0 to 7 i := j*64 - dst[i+63:i] := SINH(a[i+63:i]) + dst[i+63:i] := ABS(a[i+63:i]) ENDFOR dst[MAX:512] := 0 +
immintrin.h
 - - Floating Point + + Integer AVX512F - Trigonometry - - - - Compute the hyperbolic sine of packed double-precision (64-bit) floating-point elements in "a" expressed in radians, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Special Math Functions + + + + + Compute the absolute value of packed signed 64-bit integers in "a", and store the unsigned results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). FOR j := 0 to 7 i := j*64 IF k[j] - dst[i+63:i] := SINH(a[i+63:i]) + dst[i+63:i] := ABS(a[i+63:i]) ELSE dst[i+63:i] := src[i+63:i] FI ENDFOR dst[MAX:512] := 0 +
immintrin.h
 - - Floating Point + + Integer AVX512F - Trigonometry - - Compute the hyperbolic sine of packed single-precision (32-bit) floating-point elements in "a" expressed in radians, and store the results in "dst". + Special Math Functions + + + + Compute the absolute value of packed signed 64-bit integers in "a", and store the unsigned results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*32 - dst[i+31:i] := SINH(a[i+31:i]) +FOR j := 0 to 7 + i := j*64 + IF k[j] + dst[i+63:i] := ABS(a[i+63:i]) + ELSE + dst[i+63:i] := 0 + FI ENDFOR dst[MAX:512] := 0 +
immintrin.h
 - - Floating Point + + Integer AVX512F - Trigonometry - - - - Compute the hyperbolic sine of packed single-precision (32-bit) floating-point elements in "a" expressed in radians, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Arithmetic + + + + + Add packed 32-bit integers in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). FOR j := 0 to 15 i := j*32 IF k[j] - dst[i+31:i] := SINH(a[i+31:i]) + dst[i+31:i] := a[i+31:i] + b[i+31:i] ELSE - dst[i+31:i] := src[i+31:i] + dst[i+31:i] := 0 FI ENDFOR dst[MAX:512] := 0 +
immintrin.h
 - - Floating Point + + Integer AVX512F - Trigonometry - - Compute the sine of packed double-precision (64-bit) floating-point elements in "a" expressed in degrees, and store the results in "dst". + Arithmetic + + + + Add packed 64-bit integers in "a" and "b", and store the results in "dst". FOR j := 0 to 7 i := j*64 - dst[i+63:i] := SIND(a[i+63:i]) + dst[i+63:i] := a[i+63:i] + b[i+63:i] ENDFOR dst[MAX:512] := 0 +
immintrin.h
 - - Floating Point + + Integer AVX512F - Trigonometry - - - - Compute the sine of packed double-precision (64-bit) floating-point elements in "a" expressed in degrees, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Arithmetic + + + + + + Add packed 64-bit integers in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). FOR j := 0 to 7 i := j*64 IF k[j] - dst[i+63:i] := SIND(a[i+63:i]) + dst[i+63:i] := a[i+63:i] + b[i+63:i] ELSE dst[i+63:i] := src[i+63:i] FI ENDFOR dst[MAX:512] := 0 +
immintrin.h
 - - Floating Point + + Integer AVX512F - Trigonometry - - Compute the sine of packed single-precision (32-bit) floating-point elements in "a" expressed in degrees, and store the results in "dst". + Arithmetic + + + + + Add packed 64-bit integers in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + +FOR j := 0 to 7 + i := j*64 + IF k[j] + dst[i+63:i] := a[i+63:i] + b[i+63:i] + ELSE + dst[i+63:i] := 0 + FI +ENDFOR +dst[MAX:512] := 0 + + +
immintrin.h
+ + + Integer + AVX512F + Logical + + + + + Compute the bitwise AND of packed 32-bit integers in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). FOR j := 0 to 15 i := j*32 - dst[i+31:i] := SIND(a[i+31:i]) + IF k[j] + dst[i+31:i] := a[i+31:i] AND b[i+31:i] + ELSE + dst[i+31:i] := 0 + FI ENDFOR dst[MAX:512] := 0 +
immintrin.h
 - - Floating Point + + Integer AVX512F - Trigonometry - - - - Compute the sine of packed single-precision (32-bit) floating-point elements in "a" expressed in degrees, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Logical + + + + + Compute the bitwise NOT of packed 32-bit integers in "a" and then AND with "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). FOR j := 0 to 15 i := j*32 IF k[j] - dst[i+31:i] := SIND(a[i+31:i]) + dst[i+31:i] := (NOT a[i+31:i]) AND b[i+31:i] ELSE - dst[i+31:i] := src[i+31:i] + dst[i+31:i] := 0 FI ENDFOR dst[MAX:512] := 0 +
immintrin.h
 - - Floating Point + + Integer AVX512F - Trigonometry - - Compute the tangent of packed double-precision (64-bit) floating-point elements in "a" expressed in radians, and store the results in "dst". + Logical + + + + + Compute the bitwise NOT of packed 64-bit integers in "a" and then AND with "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). FOR j := 0 to 7 i := j*64 - dst[i+63:i] := TAN(a[i+63:i]) + IF k[j] + dst[i+63:i] := (NOT a[i+63:i]) AND b[i+63:i] + ELSE + dst[i+63:i] := 0 + FI ENDFOR dst[MAX:512] := 0 +
immintrin.h
 - - Floating Point + + Integer AVX512F - Trigonometry - - - - Compute the tangent of packed double-precision (64-bit) floating-point elements in "a" expressed in radians, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Logical + + + + + Compute the bitwise AND of packed 64-bit integers in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). FOR j := 0 to 7 i := j*64 IF k[j] - dst[i+63:i] := TAN(a[i+63:i]) + dst[i+63:i] := a[i+63:i] AND b[i+63:i] ELSE - dst[i+63:i] := src[i+63:i] + dst[i+63:i] := 0 FI ENDFOR dst[MAX:512] := 0 +
immintrin.h
 - - Floating Point + + Integer AVX512F - Trigonometry - - Compute the tangent of packed single-precision (32-bit) floating-point elements in "a" expressed in radians, and store the results in "dst". + Set + + + Broadcast 8-bit integer "a" to all elements of "dst". + +FOR j := 0 to 63 + i := j*8 + dst[i+7:i] := a[7:0] +ENDFOR +dst[MAX:512] := 0 + + +
immintrin.h
+ + + Integer + AVX512F + Swizzle + + + Broadcast the low packed 32-bit integer from "a" to all elements of "dst". FOR j := 0 to 15 i := j*32 - dst[i+31:i] := TAN(a[i+31:i]) + dst[i+31:i] := a[31:0] ENDFOR dst[MAX:512] := 0 +
immintrin.h
 - - Floating Point + + Integer AVX512F - Trigonometry - - - - Compute the tangent of packed single-precision (32-bit) floating-point elements in "a" expressed in radians, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Swizzle + + + + + Broadcast the low packed 32-bit integer from "a" to all elements of "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). FOR j := 0 to 15 i := j*32 IF k[j] - dst[i+31:i] := TAN(a[i+31:i]) + dst[i+31:i] := a[31:0] ELSE dst[i+31:i] := src[i+31:i] FI ENDFOR dst[MAX:512] := 0 +
immintrin.h
 - - Floating Point + + Integer AVX512F - Trigonometry - - Compute the tangent of packed double-precision (64-bit) floating-point elements in "a" expressed in degrees, and store the results in "dst". + Set + + + + + Broadcast 32-bit integer "a" to all elements of "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := j*64 - dst[i+63:i] := TAND(a[i+63:i]) +FOR j := 0 to 15 + i := j*32 + IF k[j] + dst[i+31:i] := a[31:0] + ELSE + dst[i+31:i] := src[i+31:i] + FI ENDFOR dst[MAX:512] := 0 +
immintrin.h
 - - Floating Point + + Integer AVX512F - Trigonometry - - - - Compute the tangent of packed double-precision (64-bit) floating-point elements in "a" expressed in degrees, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Swizzle + + + + Broadcast the low packed 32-bit integer from "a" to all elements of "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := j*64 +FOR j := 0 to 15 + i := j*32 IF k[j] - dst[i+63:i] := TAND(a[i+63:i]) + dst[i+31:i] := a[31:0] ELSE - dst[i+63:i] := src[i+63:i] + dst[i+31:i] := 0 FI ENDFOR dst[MAX:512] := 0 +
immintrin.h
 - - Floating Point + + Integer AVX512F - Trigonometry - - Compute the tangent of packed single-precision (32-bit) floating-point elements in "a" expressed in degrees, and store the results in "dst". + Set + + + + Broadcast 32-bit integer "a" to all elements of "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). FOR j := 0 to 15 i := j*32 - dst[i+31:i] := TAND(a[i+31:i]) + IF k[j] + dst[i+31:i] := a[31:0] + ELSE + dst[i+31:i] := 0 + FI ENDFOR dst[MAX:512] := 0 +
immintrin.h
 - - Floating Point + + Integer AVX512F - Trigonometry - - - - Compute the tangent of packed single-precision (32-bit) floating-point elements in "a" expressed in degrees, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Set + + + Broadcast 32-bit integer "a" to all elements of "dst". FOR j := 0 to 15 i := j*32 - IF k[j] - dst[i+31:i] := TAND(a[i+31:i]) - ELSE - dst[i+31:i] := src[i+31:i] - FI + dst[i+31:i] := a[31:0] ENDFOR dst[MAX:512] := 0 +
immintrin.h
 - - Floating Point + + Integer AVX512F - Trigonometry - - Compute the hyperbolic tangent of packed double-precision (64-bit) floating-point elements in "a" expressed in radians, and store the results in "dst". + Swizzle + + + Broadcast the low packed 64-bit integer from "a" to all elements of "dst". FOR j := 0 to 7 i := j*64 - dst[i+63:i] := TANH(a[i+63:i]) + dst[i+63:i] := a[63:0] ENDFOR dst[MAX:512] := 0 +
immintrin.h
 - - Floating Point + + Integer AVX512F - Trigonometry - - - - Compute the hyperbolic tangent of packed double-precision (64-bit) floating-point elements in "a" expressed in radians, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Swizzle + + + + + Broadcast the low packed 64-bit integer from "a" to all elements of "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). FOR j := 0 to 7 i := j*64 IF k[j] - dst[i+63:i] := TANH(a[i+63:i]) + dst[i+63:i] := a[63:0] ELSE dst[i+63:i] := src[i+63:i] FI ENDFOR dst[MAX:512] := 0 +
immintrin.h
 - - Floating Point + + Integer AVX512F - Trigonometry - - Compute the hyperbolic tangent of packed single-precision (32-bit) floating-point elements in "a" expressed in radians, and store the results in "dst". + Set + + + + + Broadcast 64-bit integer "a" to all elements of "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*32 - dst[i+31:i] := TANH(a[i+31:i]) +FOR j := 0 to 7 + i := j*64 + IF k[j] + dst[i+63:i] := a[63:0] + ELSE + dst[i+63:i] := src[i+63:i] + FI ENDFOR dst[MAX:512] := 0 +
immintrin.h
 - - Floating Point + + Integer AVX512F - Trigonometry - - - - Compute the hyperbolic tangent of packed single-precision (32-bit) floating-point elements in "a" expressed in radians, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Swizzle + + + + Broadcast the low packed 64-bit integer from "a" to all elements of "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*32 +FOR j := 0 to 7 + i := j*64 IF k[j] - dst[i+31:i] := TANH(a[i+31:i]) + dst[i+63:i] := a[63:0] ELSE - dst[i+31:i] := src[i+31:i] + dst[i+63:i] := 0 FI ENDFOR dst[MAX:512] := 0 +
immintrin.h
 - - Floating Point + + Integer AVX512F - Special Math Functions - - Truncate the packed double-precision (64-bit) floating-point elements in "a", and store the results as packed double-precision floating-point elements in "dst". + Set + + + + Broadcast 64-bit integer "a" to all elements of "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). FOR j := 0 to 7 i := j*64 - dst[i+63:i] := TRUNCATE(a[i+63:i]) + IF k[j] + dst[i+63:i] := a[63:0] + ELSE + dst[i+63:i] := 0 + FI ENDFOR dst[MAX:512] := 0 +
immintrin.h
 - - Floating Point + + Integer AVX512F - Special Math Functions - - - - Truncate the packed double-precision (64-bit) floating-point elements in "a", and store the results as packed double-precision floating-point elements in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Set + + + Broadcast 64-bit integer "a" to all elements of "dst". FOR j := 0 to 7 i := j*64 - IF k[j] - dst[i+63:i] := TRUNCATE(a[i+63:i]) - ELSE - dst[i+63:i] := src[i+63:i] - FI + dst[i+63:i] := a[63:0] ENDFOR dst[MAX:512] := 0 +
immintrin.h
 - - Floating Point + + Integer AVX512F - Special Math Functions - - Truncate the packed single-precision (32-bit) floating-point elements in "a", and store the results as packed single-precision floating-point elements in "dst". + Set + + + Broadcast the low packed 16-bit integer from "a" to all all elements of "dst". -FOR j := 0 to 15 - i := j*32 - dst[i+31:i] := TRUNCATE(a[i+31:i]) +FOR j := 0 to 31 + i := j*16 + dst[i+15:i] := a[15:0] ENDFOR dst[MAX:512] := 0 +
immintrin.h
 - - Floating Point + + Integer + Mask AVX512F - Special Math Functions - - - - Truncate the packed single-precision (32-bit) floating-point elements in "a", and store the results as packed single-precision floating-point elements in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Compare + + + + Compare packed signed 32-bit integers in "a" and "b" for less-than, and store the results in mask vector "k". FOR j := 0 to 15 i := j*32 - IF k[j] - dst[i+31:i] := TRUNCATE(a[i+31:i]) - ELSE - dst[i+31:i] := src[i+31:i] - FI + k[j] := ( a[i+31:i] < b[i+31:i] ) ? 1 : 0 ENDFOR -dst[MAX:512] := 0 +k[MAX:16] := 0 +
immintrin.h
 - + Integer + Mask AVX512F - Arithmetic - - - Divide packed unsigned 32-bit integers in "a" by packed elements in "b", and store the truncated results in "dst". + Compare + + + + + Compare packed signed 32-bit integers in "a" and "b" for less-than-or-equal, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). FOR j := 0 to 15 - i := 32*j - dst[i+31:i] := TRUNCATE(a[i+31:i] / b[i+31:i]) + i := j*32 + IF k1[j] + k[j] := ( a[i+31:i] < b[i+31:i] ) ? 1 : 0 + ELSE + k[j] := 0 + FI ENDFOR -dst[MAX:512] := 0 +k[MAX:16] := 0 +
immintrin.h
 - + Integer + Mask AVX512F - Arithmetic - - - - - Divide packed unsigned 32-bit integers in "a" by packed elements in "b", and store the truncated results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -FOR j := 0 to 15 - i := 32*j - IF k[j] - dst[i+31:i] := TRUNCATE(a[i+31:i] / b[i+31:i]) - ELSE - dst[i+31:i] := src[i+31:i] - FI + Compare + + + + + Compare packed signed 64-bit integers in "a" and "b" based on the comparison operand specified by "imm8", and store the results in mask vector "k". + CASE (imm8[2:0]) OF +0: OP := _MM_CMPINT_EQ +1: OP := _MM_CMPINT_LT +2: OP := _MM_CMPINT_LE +3: OP := _MM_CMPINT_FALSE +4: OP := _MM_CMPINT_NE +5: OP := _MM_CMPINT_NLT +6: OP := _MM_CMPINT_NLE +7: OP := _MM_CMPINT_TRUE +ESAC +FOR j := 0 to 7 + i := j*64 + k[j] := ( a[i+63:i] OP b[i+63:i] ) ? 1 : 0 ENDFOR -dst[MAX:512] := 0 +k[MAX:8] := 0 +
immintrin.h
 - + Integer + Mask AVX512F - Arithmetic - - - Divide packed unsigned 8-bit integers in "a" by packed elements in "b", and store the truncated results in "dst". + Compare + + + + Compare packed 64-bit integers in "a" and "b" for equality, and store the results in mask vector "k". -FOR j := 0 to 63 - i := 8*j - dst[i+7:i] := TRUNCATE(a[i+7:i] / b[i+7:i]) +FOR j := 0 to 7 + i := j*64 + k[j] := ( a[i+63:i] == b[i+63:i] ) ? 1 : 0 ENDFOR -dst[MAX:512] := 0 +k[MAX:8] := 0 +
immintrin.h
 - + Integer + Mask AVX512F - Arithmetic - - - Divide packed unsigned 16-bit integers in "a" by packed elements in "b", and store the truncated results in "dst". + Compare + + + + Compare packed signed 64-bit integers in "a" and "b" for greater-than-or-equal, and store the results in mask vector "k". -FOR j := 0 to 31 - i := 16*j - dst[i+15:i] := TRUNCATE(a[i+15:i] / b[i+15:i]) +FOR j := 0 to 7 + i := j*64 + k[j] := ( a[i+63:i] >= b[i+63:i] ) ? 1 : 0 ENDFOR -dst[MAX:512] := 0 +k[MAX:8] := 0 +
immintrin.h
 - + Integer + Mask AVX512F - Arithmetic - - - Divide packed unsigned 64-bit integers in "a" by packed elements in "b", and store the truncated results in "dst". + Compare + + + + Compare packed signed 64-bit integers in "a" and "b" for greater-than, and store the results in mask vector "k". FOR j := 0 to 7 - i := 64*j - dst[i+63:i] := TRUNCATE(a[i+63:i] / b[i+63:i]) + i := j*64 + k[j] := ( a[i+63:i] > b[i+63:i] ) ? 1 : 0 ENDFOR -dst[MAX:512] := 0 +k[MAX:8] := 0 +
immintrin.h
 - + Integer + Mask AVX512F - Arithmetic - - - Divide packed unsigned 32-bit integers in "a" by packed elements in "b", and store the remainders as packed unsigned 32-bit integers in "dst". + Compare + + + + Compare packed signed 64-bit integers in "a" and "b" for less-than-or-equal, and store the results in mask vector "k". -FOR j := 0 to 15 - i := 32*j - dst[i+31:i] := REMAINDER(a[i+31:i] / b[i+31:i]) +FOR j := 0 to 7 + i := j*64 + k[j] := ( a[i+63:i] <= b[i+63:i] ) ? 1 : 0 ENDFOR -dst[MAX:512] := 0 +k[MAX:8] := 0 +
immintrin.h
 - + Integer + Mask AVX512F - Arithmetic - - - - - Divide packed unsigned 32-bit integers in "a" by packed elements in "b", and store the remainders as packed unsigned 32-bit integers in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Compare + + + + Compare packed signed 64-bit integers in "a" and "b" for less-than, and store the results in mask vector "k". -FOR j := 0 to 15 - i := 32*j - IF k[j] - dst[i+31:i] := REMAINDER(a[i+31:i] / b[i+31:i]) - ELSE - dst[i+31:i] := src[i+31:i] - FI +FOR j := 0 to 7 + i := j*64 + k[j] := ( a[i+63:i] < b[i+63:i] ) ? 1 : 0 ENDFOR -dst[MAX:512] := 0 +k[MAX:8] := 0 +
immintrin.h
 - + Integer + Mask AVX512F - Arithmetic - - - Divide packed unsigned 8-bit integers in "a" by packed elements in "b", and store the remainders as packed unsigned 32-bit integers in "dst". + Compare + + + + Compare packed signed 64-bit integers in "a" and "b" for not-equal, and store the results in mask vector "k". -FOR j := 0 to 63 - i := 8*j - dst[i+7:i] := REMAINDER(a[i+7:i] / b[i+7:i]) +FOR j := 0 to 7 + i := j*64 + k[j] := ( a[i+63:i] != b[i+63:i] ) ? 1 : 0 ENDFOR -dst[MAX:512] := 0 +k[MAX:8] := 0 +
immintrin.h
 - + Integer + Mask AVX512F - Arithmetic - - - Divide packed unsigned 16-bit integers in "a" by packed elements in "b", and store the remainders as packed unsigned 32-bit integers in "dst". - -FOR j := 0 to 31 - i := 16*j - dst[i+15:i] := REMAINDER(a[i+15:i] / b[i+15:i]) + Compare + + + + + + Compare packed signed 64-bit integers in "a" and "b" based on the comparison operand specified by "imm8", and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). + CASE (imm8[2:0]) OF +0: OP := _MM_CMPINT_EQ +1: OP := _MM_CMPINT_LT +2: OP := _MM_CMPINT_LE +3: OP := _MM_CMPINT_FALSE +4: OP := _MM_CMPINT_NE +5: OP := _MM_CMPINT_NLT +6: OP := _MM_CMPINT_NLE +7: OP := _MM_CMPINT_TRUE +ESAC +FOR j := 0 to 7 + i := j*64 + IF k1[j] + k[j] := ( a[i+63:i] OP b[i+63:i] ) ? 1 : 0 + ELSE + k[j] := 0 + FI ENDFOR -dst[MAX:512] := 0 +k[MAX:8] := 0 +
immintrin.h
 - + Integer + Mask AVX512F - Arithmetic - - - Divide packed unsigned 64-bit integers in "a" by packed elements in "b", and store the remainders as packed unsigned 32-bit integers in "dst". + Compare + + + + + Compare packed 64-bit integers in "a" and "b" for equality, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). FOR j := 0 to 7 - i := 64*j - dst[i+63:i] := REMAINDER(a[i+63:i] / b[i+63:i]) + i := j*64 + IF k1[j] + k[j] := ( a[i+63:i] == b[i+63:i] ) ? 1 : 0 + ELSE + k[j] := 0 + FI ENDFOR -dst[MAX:512] := 0 +k[MAX:8] := 0 +
immintrin.h
 - - Floating Point - KNCNI - Special Math Functions - - Determines the minimum element of the packed single-precision (32-bit) floating-point elements stored in "a" and stores the result in "dst". + + Integer + Mask + AVX512F + Compare + + + + + Compare packed signed 64-bit integers in "a" and "b" for greater-than-or-equal, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). -min = a[31:0] -FOR j := 1 to 15 - i := j*32 - dst = FpMin(min, a[i+31:i]) +FOR j := 0 to 7 + i := j*64 + IF k1[j] + k[j] := ( a[i+63:i] >= b[i+63:i] ) ? 1 : 0 + ELSE + k[j] := 0 + FI ENDFOR -dst := min +k[MAX:8] := 0 +
immintrin.h
 - - Floating Point - KNCNI - Special Math Functions - - - Determines the minimum element of the packed single-precision (32-bit) floating-point elements stored in "a" and stores the result in "dst" using writemask "k" (elements are ignored when the corresponding mask bit is not set). + + Integer + Mask + AVX512F + Compare + + + + + Compare packed signed 64-bit integers in "a" and "b" for greater-than, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). -min = a[31:0] -FOR j := 1 to 15 - i := j*32 - IF k[j] - CONTINUE - ELSE - dst = FpMin(min, a[i+31:i]) +FOR j := 0 to 7 + i := j*64 + IF k1[j] + k[j] := ( a[i+63:i] > b[i+63:i] ) ? 1 : 0 + ELSE + k[j] := 0 FI ENDFOR -dst := min +k[MAX:8] := 0 +
immintrin.h
 - - Floating Point - KNCNI - Special Math Functions - - Determines the minimum element of the packed double-precision (64-bit) floating-point elements stored in "a" and stores the result in "dst". + + Integer + Mask + AVX512F + Compare + + + + + Compare packed signed 64-bit integers in "a" and "b" for less-than-or-equal, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). -min = a[63:0] -FOR j := 1 to 7 +FOR j := 0 to 7 i := j*64 - dst = FpMin(min, a[i+63:i]) + IF k1[j] + k[j] := ( a[i+63:i] <= b[i+63:i] ) ? 1 : 0 + ELSE + k[j] := 0 + FI ENDFOR -dst := min +k[MAX:8] := 0 +
immintrin.h
 - - Floating Point - KNCNI - Special Math Functions - - - Determines the minimum element of the packed double-precision (64-bit) floating-point elements stored in "a" and stores the result in "dst". Bitmask "k" is used to exclude certain elements (elements are ignored when the corresponding mask bit is not set). + + Integer + Mask + AVX512F + Compare + + + + + Compare packed signed 64-bit integers in "a" and "b" for less-than, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). -min = a[63:0] -FOR j := 1 to 7 +FOR j := 0 to 7 i := j*64 - IF k[j] - CONTINUE - ELSE - dst = FpMin(min, a[i+63:i]) + IF k1[j] + k[j] := ( a[i+63:i] < b[i+63:i] ) ? 1 : 0 + ELSE + k[j] := 0 FI ENDFOR -dst := min +k[MAX:8] := 0 +
immintrin.h
 - - Floating Point - KNCNI - Special Math Functions - - Determines the maximum element of the packed single-precision (32-bit) floating-point elements stored in "a" and stores the result in "dst". + + Integer + Mask + AVX512F + Compare + + + + + Compare packed signed 64-bit integers in "a" and "b" for not-equal, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). -max = a[31:0] -FOR j := 1 to 15 - i := j*32 - dst = FpMax(max, a[i+31:i]) +FOR j := 0 to 7 + i := j*64 + IF k1[j] + k[j] := ( a[i+63:i] != b[i+63:i] ) ? 1 : 0 + ELSE + k[j] := 0 + FI +ENDFOR +k[MAX:8] := 0 + + +
immintrin.h
+ + + Integer + Mask + AVX512F + Compare + + + + + Compare packed unsigned 64-bit integers in "a" and "b" based on the comparison operand specified by "imm8", and store the results in mask vector "k". + CASE (imm8[2:0]) OF +0: OP := _MM_CMPINT_EQ +1: OP := _MM_CMPINT_LT +2: OP := _MM_CMPINT_LE +3: OP := _MM_CMPINT_FALSE +4: OP := _MM_CMPINT_NE +5: OP := _MM_CMPINT_NLT +6: OP := _MM_CMPINT_NLE +7: OP := _MM_CMPINT_TRUE +ESAC +FOR j := 0 to 7 + i := j*64 + k[j] := ( a[i+63:i] OP b[i+63:i] ) ? 1 : 0 ENDFOR -dst := max +k[MAX:8] := 0 +
immintrin.h
 - - Floating Point - KNCNI - Special Math Functions - - - Determines the maximum element of the packed single-precision (32-bit) floating-point elements stored in "a" and stores the result in "dst". Bitmask "k" is used to exclude certain elements (elements are ignored when the corresponding mask bit is not set). + + Integer + Mask + AVX512F + Compare + + + + Compare packed unsigned 64-bit integers in "a" and "b" for equality, and store the results in mask vector "k". -max = a[31:0] -FOR j := 1 to 15 - i := j*32 - IF k[j] - CONTINUE - ELSE - dst = FpMax(max, a[i+31:i]) - FI +FOR j := 0 to 7 + i := j*64 + k[j] := ( a[i+63:i] == b[i+63:i] ) ? 1 : 0 ENDFOR -dst := max +k[MAX:8] := 0 +
immintrin.h
 - - Floating Point - KNCNI - Special Math Functions - - Determines the maximum element of the packed double-precision (64-bit) floating-point elements stored in "a" and stores the result in "dst". + + Integer + Mask + AVX512F + Compare + + + + Compare packed unsigned 64-bit integers in "a" and "b" for greater-than-or-equal, and store the results in mask vector "k". -max = a[63:0] -FOR j := 1 to 7 +FOR j := 0 to 7 i := j*64 - dst = FpMax(max, a[i+63:i]) + k[j] := ( a[i+63:i] >= b[i+63:i] ) ? 1 : 0 ENDFOR -dst := max +k[MAX:8] := 0 +
immintrin.h
 - - Floating Point - KNCNI - Special Math Functions - - - Determines the maximum element of the packed double-precision (64-bit) floating-point elements stored in "a" and stores the result in "dst". Bitmask "k" is used to exclude certain elements (elements are ignored when the corresponding mask bit is not set). + + Integer + Mask + AVX512F + Compare + + + + Compare packed unsigned 64-bit integers in "a" and "b" for greater-than, and store the results in mask vector "k". -max = a[63:0] -FOR j := 1 to 7 +FOR j := 0 to 7 i := j*64 - IF k[j] - CONTINUE - ELSE - dst = FpMax(max, a[i+63:i]) - FI + k[j] := ( a[i+63:i] > b[i+63:i] ) ? 1 : 0 ENDFOR -dst := max +k[MAX:8] := 0 +
immintrin.h
 - - KNCNI - Bit Manipulation - - - Counts the number of trailing bits in unsigned 32-bit integer "x" starting at bit "a" storing the result in "dst". + + Integer + Mask + AVX512F + Compare + + + + Compare packed unsigned 64-bit integers in "a" and "b" for less-than-or-equal, and store the results in mask vector "k". -count := 0 -FOR j := a to 31 - IF NOT(x[j] 1) - count := count + 1 - FI +FOR j := 0 to 7 + i := j*64 + k[j] := ( a[i+63:i] <= b[i+63:i] ) ? 1 : 0 ENDFOR -dst := count +k[MAX:8] := 0 - +
immintrin.h
 - - KNCNI - Bit Manipulation - - - Counts the number of trailing bits in unsigned 64-bit integer "x" starting at bit "a" storing the result in "dst". + + Integer + Mask + AVX512F + Compare + + + + Compare packed unsigned 64-bit integers in "a" and "b" for less-than, and store the results in mask vector "k". -count := 0 -FOR j := a to 63 - IF NOT(x[j] 1) - count := count + 1 - FI +FOR j := 0 to 7 + i := j*64 + k[j] := ( a[i+63:i] < b[i+63:i] ) ? 1 : 0 ENDFOR -dst := count +k[MAX:8] := 0 - +
immintrin.h
 - - KNCNI - General Support - - Stalls a thread without blocking other threads for 32-bit unsigned integer "r1" clock cycles. + + Integer + Mask + AVX512F + Compare + + + + Compare packed unsigned 64-bit integers in "a" and "b" for not-equal, and store the results in mask vector "k". -BlockThread(r1) +FOR j := 0 to 7 + i := j*64 + k[j] := ( a[i+63:i] != b[i+63:i] ) ? 1 : 0 +ENDFOR +k[MAX:8] := 0 - +
immintrin.h
 - - KNCNI - General Support - - Stalls a thread without blocking other threads for 64-bit unsigned integer "r1" clock cycles. - -BlockThread(r1) + + Integer + Mask + AVX512F + Compare + + + + + + Compare packed unsigned 64-bit integers in "a" and "b" based on the comparison operand specified by "imm8", and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). + CASE (imm8[2:0]) OF +0: OP := _MM_CMPINT_EQ +1: OP := _MM_CMPINT_LT +2: OP := _MM_CMPINT_LE +3: OP := _MM_CMPINT_FALSE +4: OP := _MM_CMPINT_NE +5: OP := _MM_CMPINT_NLT +6: OP := _MM_CMPINT_NLE +7: OP := _MM_CMPINT_TRUE +ESAC +FOR j := 0 to 7 + i := j*64 + IF k1[j] + k[j] := ( a[i+63:i] OP b[i+63:i] ) ? 1 : 0 + ELSE + k[j] := 0 + FI +ENDFOR +k[MAX:8] := 0 - +
immintrin.h
 - - KNCNI - General Support - - Set performance monitoring filtering mask to 32-bit unsigned integer "r1". + + Integer + Mask + AVX512F + Compare + + + + + Compare packed unsigned 64-bit integers in "a" and "b" for equality, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). -SetPerfMonMask(r1[31:0]) +FOR j := 0 to 7 + i := j*64 + IF k1[j] + k[j] := ( a[i+63:i] == b[i+63:i] ) ? 1 : 0 + ELSE + k[j] := 0 + FI +ENDFOR +k[MAX:8] := 0 - +
immintrin.h
 - - KNCNI - General Support - - Set performance monitoring filtering mask to 64-bit unsigned integer "r1". + + Integer + Mask + AVX512F + Compare + + + + + Compare packed unsigned 64-bit integers in "a" and "b" for greater-than-or-equal, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). -SetPerfMonMask(r1[63:0]) +FOR j := 0 to 7 + i := j*64 + IF k1[j] + k[j] := ( a[i+63:i] >= b[i+63:i] ) ? 1 : 0 + ELSE + k[j] := 0 + FI +ENDFOR +k[MAX:8] := 0 - +
immintrin.h
 - - KNCNI - General Support - - - Evicts the cache line containing the address "ptr" from cache level "level" (can be either 0 or 1). + + Integer + Mask + AVX512F + Compare + + + + + Compare packed unsigned 64-bit integers in "a" and "b" for greater-than, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). -CacheLineEvict(ptr, level) +FOR j := 0 to 7 + i := j*64 + IF k1[j] + k[j] := ( a[i+63:i] > b[i+63:i] ) ? 1 : 0 + ELSE + k[j] := 0 + FI +ENDFOR +k[MAX:8] := 0 - - +
immintrin.h
 - + + Integer Mask - KNCNI - Mask - - - Performs a bitwise AND operation between NOT of "k2" and "k1", storing the result in "dst". + AVX512F + Compare + + + + + Compare packed unsigned 64-bit integers in "a" and "b" for less-than-or-equal, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). -dst[15:0] := NOT(k2[15:0]) & k1[15:0] +FOR j := 0 to 7 + i := j*64 + IF k1[j] + k[j] := ( a[i+63:i] <= b[i+63:i] ) ? 1 : 0 + ELSE + k[j] := 0 + FI +ENDFOR +k[MAX:8] := 0 - +
immintrin.h
 - + + Integer Mask - KNCNI - Mask - - - Moves high byte from "k2" to low byte of "k1", and moves low byte of "k2" to high byte of "k1". + AVX512F + Compare + + + + + Compare packed unsigned 64-bit integers in "a" and "b" for less-than, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). -tmp[7:0] := k2[15:8] -k2[15:8] := k1[7:0] -k1[7:0] := tmp[7:0] - -tmp[7:0] := k2[7:0] -k2[7:0] := k1[15:8] -k1[15:8] := tmp[7:0] +FOR j := 0 to 7 + i := j*64 + IF k1[j] + k[j] := ( a[i+63:i] < b[i+63:i] ) ? 1 : 0 + ELSE + k[j] := 0 + FI +ENDFOR +k[MAX:8] := 0 - +
immintrin.h
 - + + Integer Mask - KNCNI - Mask - - - Performs bitwise OR between "k1" and "k2", storing the result in "dst". ZF flag is set if "dst" is 0. + AVX512F + Compare + + + + + Compare packed unsigned 64-bit integers in "a" and "b" for not-equal, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). -dst[15:0] := k1[15:0] | k2[15:0] -IF dst = 0 - SetZF() -FI +FOR j := 0 to 7 + i := j*64 + IF k1[j] + k[j] := ( a[i+63:i] != b[i+63:i] ) ? 1 : 0 + ELSE + k[j] := 0 + FI +ENDFOR +k[MAX:8] := 0 - +
immintrin.h
 - - Mask - KNCNI - Mask - - - Performs bitwise OR between "k1" and "k2", storing the result in "dst". CF flag is set if "dst" consists of all 1's. + + Integer + AVX512F + Swizzle + + + + + Contiguously store the active 32-bit integers in "a" (those with their respective bit set in writemask "k") to "dst", and pass through the remaining elements from "src". -dst[15:0] := k1[15:0] | k2[15:0] -IF PopCount(dst[15:0]) = 16 - SetCF() -FI +size := 32 +m := 0 +FOR j := 0 to 15 + i := j*32 + IF k[j] + dst[m+size-1:m] := a[i+31:i] + m := m + size + FI +ENDFOR +dst[511:m] := src[511:m] +dst[MAX:512] := 0 - +
immintrin.h
 - - Mask + + Integer AVX512F - Mask - - - Performs bitwise OR between "k1" and "k2", storing the result in "dst". ZF flag is set if "dst" is 0. + Store + Swizzle + + + + + Contiguously store the active 32-bit integers in "a" (those with their respective bit set in writemask "k") to unaligned memory at "base_addr". -dst[15:0] := k1[15:0] | k2[15:0] -IF dst = 0 - SetZF() -FI +size := 32 +m := base_addr +FOR j := 0 to 15 + i := j*32 + IF k[j] + MEM[m+size-1:m] := a[i+31:i] + m := m + size + FI +ENDFOR - +
immintrin.h
 - - Mask + + Integer AVX512F - Mask - - - Performs bitwise OR between "k1" and "k2", storing the result in "dst". CF flag is set if "dst" consists of all 1's. + Swizzle + + + + Contiguously store the active 32-bit integers in "a" (those with their respective bit set in zeromask "k") to "dst", and set the remaining elements to zero. -dst[15:0] := k1[15:0] | k2[15:0] -IF PopCount(dst[15:0]) = 16 - SetCF() -FI +size := 32 +m := 0 +FOR j := 0 to 15 + i := j*32 + IF k[j] + dst[m+size-1:m] := a[i+31:i] + m := m + size + FI +ENDFOR +dst[511:m] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - KNCNI - Mask - - Converts bit mask "k1" into an integer value, storing the results in "dst". + + Integer + AVX512F + Swizzle + + + + + Contiguously store the active 64-bit integers in "a" (those with their respective bit set in writemask "k") to "dst", and pass through the remaining elements from "src". -dst := ZeroExtend(k1) +size := 64 +m := 0 +FOR j := 0 to 7 + i := j*64 + IF k[j] + dst[m+size-1:m] := a[i+63:i] + m := m + size + FI +ENDFOR +dst[511:m] := src[511:m] +dst[MAX:512] := 0 +
immintrin.h
- - + + Integer AVX512F - Mask - - Converts bit mask "k1" into an integer value, storing the results in "dst". + Store + Swizzle + + + + + Contiguously store the active 64-bit integers in "a" (those with their respective bit set in writemask "k") to unaligned memory at "base_addr". -dst := ZeroExtend(k1) +size := 64 +m := base_addr +FOR j := 0 to 7 + i := j*64 + IF k[j] + MEM[m+size-1:m] := a[i+63:i] + m := m + size + FI +ENDFOR +
immintrin.h
- - - KNCNI - Mask - - Converts integer "mask" into bitmask, storing the result in "dst". + + Integer + AVX512F + Swizzle + + + + Contiguously store the active 64-bit integers in "a" (those with their respective bit set in zeromask "k") to "dst", and set the remaining elements to zero. -dst := mask[15:0] +size := 64 +m := 0 +FOR j := 0 to 7 + i := j*64 + IF k[j] + dst[m+size-1:m] := a[i+63:i] + m := m + size + FI +ENDFOR +dst[511:m] := 0 +dst[MAX:512] := 0 +
immintrin.h
- - + + Integer AVX512F - Mask - - Converts integer "mask" into bitmask, storing the result in "dst". + Swizzle + + + + + + Shuffle 32-bit integers in "a" across lanes using the corresponding index in "idx", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -dst := mask[15:0] +FOR j := 0 to 15 + i := j*32 + id := idx[i+3:i]*32 + IF k[j] + dst[i+31:i] := a[id+31:id] + ELSE + dst[i+31:i] := src[i+31:i] + FI +ENDFOR +dst[MAX:512] := 0 +
immintrin.h
- - - Mask - KNCNI - Mask - - - Packs masks "k1" and "k2" into the high 32 bits of "dst". The rest of "dst" is set to 0. + + Integer + AVX512F + Swizzle + + + + + Shuffle 32-bit integers in "a" across lanes using the corresponding index in "idx", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -dst[63:48] := k1[15:0] -dst[47:32] := k2[15:0] -dst[31:0] := 0 +FOR j := 0 to 15 + i := j*32 + id := idx[i+3:i]*32 + IF k[j] + dst[i+31:i] := a[id+31:id] + ELSE + dst[i+31:i] := 0 + FI +ENDFOR +dst[MAX:512] := 0 - +
immintrin.h
 - - Mask - KNCNI - Mask - - - Packs masks "k1" and "k2" into the low 32 bits of "dst". The rest of "dst" is set to 0. + + Integer + AVX512F + Swizzle + + + + Shuffle 32-bit integers in "a" across lanes using the corresponding index in "idx", and store the results in "dst". -dst[31:16] := k1[15:0] -dst[15:0] := k2[15:0] -dst[63:32] := 0 +FOR j := 0 to 15 + i := j*32 + id := idx[i+3:i]*32 + dst[i+31:i] := a[id+31:id] +ENDFOR +dst[MAX:512] := 0 - +
immintrin.h
 - - Mask - KNCNI - Mask - - - Extracts 16-bit value "b" from 64-bit integer "a", storing the result in "dst". + + Integer + AVX512F + Swizzle + + + + + + Shuffle 32-bit integers in "a" and "b" across lanes using the corresponding selector and index in "idx", and store the results in "dst" using writemask "k" (elements are copied from "idx" when the corresponding mask bit is not set). -CASE b of -0: dst[15:0] := a[63:48] -1: dst[15:0] := a[47:32] -2: dst[15:0] := a[31:16] -3: dst[15:0] := a[15:0] -ESAC -dst[MAX:15] := 0 +FOR j := 0 to 15 + i := j*32 + off := idx[i+3:i]*32 + IF k[j] + dst[i+31:i] := idx[i+4] ? b[off+31:off] : a[off+31:off] + ELSE + dst[i+31:i] := idx[i+31:i] + FI +ENDFOR +dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point - KNCNI - Arithmetic - - - Multiply packed single-precision (32-bit) floating-point elements in each 4-element set of "a" and by element 1 of the corresponding 4-element set from "b", add the intermediate result to element 0 of the corresponding 4-element set from "b", and store the results in "dst". + + Integer + AVX512F + Swizzle + + + + + + Shuffle 32-bit integers in "a" and "b" across lanes using the corresponding selector and index in "idx", and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). FOR j := 0 to 15 i := j*32 - base := (j & ~0x3) * 32 - scale[31:0] := b[base+63:base+32] - bias[31:0] := b[base+31:base] - dst[i+31:i] := (a[i+31:i] * scale[31:0]) + bias[31:0] + off := idx[i+3:i]*32 + IF k[j] + dst[i+31:i] := idx[i+4] ? b[off+31:off] : a[off+31:off] + ELSE + dst[i+31:i] := a[i+31:i] + FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point - KNCNI - Arithmetic - - - - - Multiply packed single-precision (32-bit) floating-point elements in each 4-element set of "a" and by element 1 of the corresponding 4-element set from "b", add the intermediate result to element 0 of the corresponding 4-element set from "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Integer + AVX512F + Swizzle + + + + + + Shuffle 32-bit integers in "a" and "b" across lanes using the corresponding selector and index in "idx", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). FOR j := 0 to 15 i := j*32 + off := idx[i+3:i]*32 IF k[j] - base := (j & ~0x3) * 32 - scale[31:0] := b[base+63:base+32] - bias[31:0] := b[base+31:base] - dst[i+31:i] := (a[i+31:i] * scale[31:0]) + bias[31:0] + dst[i+31:i] := (idx[i+4]) ? b[off+31:off] : a[off+31:off] ELSE - dst[i+31:i] := src[i+31:i] + dst[i+31:i] := 0 FI ENDFOR dst[MAX:512] := 0 - + +
immintrin.h
 - + Integer - KNCNI - Load - - - - - - Up-converts 8 single-precision (32-bit) memory locations starting at location "mv" at packed 64-bit integer indices stored in "index" scaled by "scale" using "conv" to 32-bit integer elements and stores them in "dst". "hint" indicates to the processor whether the data is non-temporal. + AVX512F + Swizzle + + + + + Shuffle 32-bit integers in "a" and "b" across lanes using the corresponding selector and index in "idx", and store the results in "dst". -FOR j := 0 to 7 - addr := MEM[mv + index[j] * scale] +FOR j := 0 to 15 i := j*32 - CASE conv OF - _MM_UPCONV_EPI32_NONE: - dst[i+31:i] := addr[i+31:i] - _MM_UPCONV_EPI32_UINT8: - n := j*8 - dst[i+31:i] := UInt8ToInt32(addr[n+7:n]) - _MM_UPCONV_EPI32_SINT8: - n := j*8 - dst[i+31:i] := SInt8ToInt32(addr[n+7:n]) - _MM_UPCONV_EPI32_UINT16: - n := j*16 - dst[i+31:i] := UInt16ToInt32(addr[n+15:n]) - _MM_UPCONV_EPI32_SINT16: - n := j*16 - dst[i+31:i] := SInt16ToInt32(addr[n+15:n]) - ESAC + off := idx[i+3:i]*32 + dst[i+31:i] := idx[i+4] ? b[off+31:off] : a[off+31:off] ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 + +
immintrin.h
 - - Integer - KNCNI - Load - - - - - - - - Up-converts 8 single-precision (32-bit) memory locations starting at location "mv" at packed 64-bit integer indices stored in "index" scaled by "scale" using "conv" to 32-bit integer elements and stores them in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). "hint" indicates to the processor whether the data is non-temporal. + + Floating Point + AVX512F + Swizzle + + + + + + Shuffle double-precision (64-bit) floating-point elements in "a" and "b" across lanes using the corresponding selector and index in "idx", and store the results in "dst" using writemask "k" (elements are copied from "idx" when the corresponding mask bit is not set) FOR j := 0 to 7 - addr := MEM[mv + index[j] * scale] - i := j*32 + i := j*64 + off := idx[i+2:i]*64 IF k[j] - CASE conv OF - _MM_UPCONV_EPI32_NONE: - dst[i+31:i] := addr[i+31:i] - _MM_UPCONV_EPI32_UINT8: - n := j*8 - dst[i+31:i] := UInt8ToInt32(addr[n+7:n]) - _MM_UPCONV_EPI32_SINT8: - n := j*8 - dst[i+31:i] := SInt8ToInt32(addr[n+7:n]) - _MM_UPCONV_EPI32_UINT16: - n := j*16 - dst[i+31:i] := UInt16ToInt32(addr[n+15:n]) - _MM_UPCONV_EPI32_SINT16: - n := j*16 - dst[i+31:i] := SInt16ToInt32(addr[n+15:n]) - ESAC + dst[i+63:i] := idx[i+3] ? b[off+63:off] : a[off+63:off] ELSE - dst[i+31:i] := src[i+31:i] + dst[i+63:i] := idx[i+63:i] FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 +
immintrin.h
 - - Integer - KNCNI - Load - - - - - - Up-converts 8 double-precision (64-bit) memory locations starting at location "mv" at packed 64-bit integer indices stored in "index" scaled by "scale" using "conv" to 64-bit integer elements and stores them in "dst". "hint" indicates to the processor whether the load is non-temporal. + + Floating Point + AVX512F + Swizzle + + + + + + Shuffle double-precision (64-bit) floating-point elements in "a" and "b" across lanes using the corresponding selector and index in "idx", and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). FOR j := 0 to 7 i := j*64 - addr := MEM[mv + index[j] * scale] - CASE conv OF - _MM_UPCONV_EPI64_NONE: dst[i+63:i] := addr[i+63:i] - ESAC + off := idx[i+2:i]*64 + IF k[j] + dst[i+63:i] := idx[i+3] ? b[off+63:off] : a[off+63:off] + ELSE + dst[i+63:i] := a[i+63:i] + FI ENDFOR dst[MAX:512] := 0 +
immintrin.h
 - - Integer - KNCNI - Load - - - - - - - - Up-converts 8 double-precision (64-bit) memory locations starting at location "mv" at packed 64-bit integer indices stored in "index" scaled by "scale" using "conv" to 64-bit integer elements and stores them in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). "hint" indicates to the processor whether the load is non-temporal. + + Floating Point + AVX512F + Swizzle + + + + + + Shuffle double-precision (64-bit) floating-point elements in "a" and "b" across lanes using the corresponding selector and index in "idx", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). FOR j := 0 to 7 i := j*64 + off := idx[i+2:i]*64 IF k[j] - addr := MEM[mv + index[j] * scale] - CASE conv OF - _MM_UPCONV_EPI64_NONE: dst[i+63:i] := addr[i+63:i] - ESAC + dst[i+63:i] := (idx[i+3]) ? b[off+63:off] : a[off+63:off] ELSE - dst[i+63:i] := src[i+63:i] + dst[i+63:i] := 0 FI ENDFOR dst[MAX:512] := 0 + +
immintrin.h
 - + Floating Point - KNCNI - Load - - - - - - Up-converts 8 memory locations starting at location "mv" at packed 64-bit integer indices stored in "index" scaled by "scale" using "conv" to single-precision (32-bit) floating-point elements and stores them in the lower half of "dst". "hint" indicates to the processor whether the load is non-temporal. + AVX512F + Swizzle + + + + + Shuffle double-precision (64-bit) floating-point elements in "a" and "b" across lanes using the corresponding selector and index in "idx", and store the results in "dst". FOR j := 0 to 7 - addr := MEM[mv + index[j] * scale] - i := j*32 - CASE conv OF - _MM_UPCONV_PS_NONE: - dst[i+31:i] := addr[i+31:i] - _MM_UPCONV_PS_FLOAT16: - n := j*16 - dst[i+31:i] := Float16ToFloat32(addr[n+15:n]) - _MM_UPCONV_PS_UINT8: - n := j*8 - dst[i+31:i] := UInt8ToFloat32(addr[n+7:n]) - _MM_UPCONV_PS_SINT8: - n := j*8 - dst[i+31:i] := SInt8ToFloat32(addr[n+7:n]) - _MM_UPCONV_PS_UINT16: - n := j*16 - dst[i+31:i] := UInt16ToFloat32(addr[n+15:n]) - _MM_UPCONV_PS_SINT16: - n := j*16 - dst[i+31:i] := SInt16ToFloat32(addr[n+15:n]) - ESAC + i := j*64 + off := idx[i+2:i]*64 + dst[i+63:i] := idx[i+3] ? b[off+63:off] : a[off+63:off] ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 + +
immintrin.h
 - + Floating Point - KNCNI - Load - - - - - - - - Up-converts 8 memory locations starting at location "mv" at packed 64-bit integer indices stored in "index" scaled by "scale" using "conv" to single-precision (32-bit) floating-point elements and stores them in the lower half of "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). "hint" indicates to the processor whether the load is non-temporal. + AVX512F + Swizzle + + + + + + Shuffle single-precision (32-bit) floating-point elements in "a" and "b" across lanes using the corresponding selector and index in "idx", and store the results in "dst" using writemask "k" (elements are copied from "idx" when the corresponding mask bit is not set). -FOR j := 0 to 7 - addr := MEM[mv + index[j] * scale] +FOR j := 0 to 15 i := j*32 + off := idx[i+3:i]*32 IF k[j] - CASE conv OF - _MM_UPCONV_PS_NONE: - dst[i+31:i] := addr[i+31:i] - _MM_UPCONV_PS_FLOAT16: - n := j*16 - dst[i+31:i] := Float16ToFloat32(addr[n+15:n]) - _MM_UPCONV_PS_UINT8: - n := j*8 - dst[i+31:i] := UInt8ToFloat32(addr[n+7:n]) - _MM_UPCONV_PS_SINT8: - n := j*8 - dst[i+31:i] := SInt8ToFloat32(addr[n+7:n]) - _MM_UPCONV_PS_UINT16: - n := j*16 - dst[i+31:i] := UInt16ToFloat32(addr[n+15:n]) - _MM_UPCONV_PS_SINT16: - n := j*16 - dst[i+31:i] := SInt16ToFloat32(addr[n+15:n]) - ESAC + dst[i+31:i] := idx[i+4] ? b[off+31:off] : a[off+31:off] ELSE - dst[i+31:i] := src[i+31:i] + dst[i+31:i] := idx[i+31:i] FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 +
immintrin.h
 - + Floating Point - KNCNI - Load - - - - - - Up-converts 8 double-precision (64-bit) floating-point elements stored in memory starting at location "mv" at packed 64-bit integer indices stored in "index" scaled by "scale" using "conv" to 64-bit floating-point elements and stores them in "dst". "hint" indicates to the processor whether the data is non-temporal. + AVX512F + Swizzle + + + + + + Shuffle single-precision (32-bit) floating-point elements in "a" and "b" across lanes using the corresponding selector and index in "idx", and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). -FOR j := 0 to 7 - addr := MEM[mv + index[j] * scale] - i := j*64 - CASE conv OF - _MM_UPCONV_PD_NONE: dst[i+63:i] := addr[i+63:i] - ESAC +FOR j := 0 to 15 + i := j*32 + off := idx[i+3:i]*32 + IF k[j] + dst[i+31:i] := idx[i+4] ? b[off+31:off] : a[off+31:off] + ELSE + dst[i+31:i] := a[i+31:i] + FI ENDFOR dst[MAX:512] := 0 +
immintrin.h
 - + Floating Point - KNCNI - Load - - - - - - - - Up-converts 8 double-precision (64-bit) floating-point elements stored in memory starting at location "mv" at packed 64-bit integer indices stored in "index" scaled by "scale" using "conv" to 64-bit floating-point elements and stores them in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). "hint" indicates to the processor whether the data is non-temporal. + AVX512F + Swizzle + + + + + + Shuffle single-precision (32-bit) floating-point elements in "a" and "b" across lanes using the corresponding selector and index in "idx", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 - addr := MEM[mv + index[j] * scale] - i := j*64 +FOR j := 0 to 15 + i := j*32 + off := idx[i+3:i]*32 IF k[j] - CASE conv OF - _MM_UPCONV_PD_NONE: dst[i+63:i] := addr[i+63:i] - ESAC + dst[i+31:i] := (idx[i+4]) ? b[off+31:off] : a[off+31:off] ELSE - dst[i+63:i] := src[i+63:i] + dst[i+31:i] := 0 FI ENDFOR dst[MAX:512] := 0 + +
immintrin.h
 - - Integer - AVX512F/KNCNI - Store - - - - - - - Down-converts 16 packed 32-bit integer elements in "v1" using "conv" and stores them in memory locations starting at location "mv" at packed 32-bit integer indices stored in "index" scaled by "scale". "hint" indicates to the processor whether the data is non-temporal. AVX512 supports _MM_DOWNCONV_EPI32_NONE. + + Floating Point + AVX512F + Swizzle + + + + + Shuffle single-precision (32-bit) floating-point elements in "a" and "b" across lanes using the corresponding selector and index in "idx", and store the results in "dst". FOR j := 0 to 15 - addr := MEM[mv + index[j] * scale] i := j*32 - CASE conv OF - _MM_DOWNCONV_EPI32_NONE: - addr[i+31:i] := v1[i+31:i] - _MM_DOWNCONV_EPI32_UINT8: - n := j*8 - addr[n+7:n] := UInt32ToUInt8(v1[i+31:i]) - _MM_DOWNCONV_EPI32_SINT8: - n := j*8 - addr[n+7:n] := SInt32ToSInt8(v1[i+31:i]) - _MM_DOWNCONV_EPI32_UINT16: - n := j*16 - addr[n+15:n] := UInt32ToUInt16(v1[i+31:i]) - _MM_DOWNCONV_EPI32_SINT16: - n := j*16 - addr[n+15:n] := SInt32ToSInt16(v1[n+15:n]) - ESAC + off := idx[i+3:i]*32 + dst[i+31:i] := idx[i+4] ? b[off+31:off] : a[off+31:off] ENDFOR +dst[MAX:512] := 0 - + +
immintrin.h
 - + Integer - AVX512F/KNCNI - Store - - - - - - - - Down-converts 16 packed 32-bit integer elements in "v1" using "conv" and stores them in memory locations starting at location "mv" at packed 32-bit integer indices stored in "index" scaled by "scale". Elements are written using writemask "k" (elements are only written when the corresponding mask bit is set; otherwise, elements are left unchanged in memory). "hint" indicates to the processor whether the data is non-temporal. AVX512 supports _MM_DOWNCONV_EPI32_NONE. + AVX512F + Swizzle + + + + + + Shuffle 64-bit integers in "a" and "b" across lanes using the corresponding selector and index in "idx", and store the results in "dst" using writemask "k" (elements are copied from "idx" when the corresponding mask bit is not set). -FOR j := 0 to 15 - addr := MEM[mv + index[j] * scale] - i := j*32 +FOR j := 0 to 7 + i := j*64 + off := idx[i+2:i]*64 IF k[j] - CASE conv OF - _MM_DOWNCONV_EPI32_NONE: - addr[i+31:i] := v1[i+31:i] - _MM_DOWNCONV_EPI32_UINT8: - n := j*8 - addr[n+7:n] := UInt32ToUInt8(v1[i+31:i]) - _MM_DOWNCONV_EPI32_SINT8: - n := j*8 - addr[n+7:n] := SInt32ToSInt8(v1[i+31:i]) - _MM_DOWNCONV_EPI32_UINT16: - n := j*16 - addr[n+15:n] := UInt32ToUInt16(v1[i+31:i]) - _MM_DOWNCONV_EPI32_SINT16: - n := j*16 - addr[n+15:n] := SInt32ToSInt16(v1[n+15:n]) - ESAC - FI + dst[i+63:i] := idx[i+3] ? b[off+63:off] : a[off+63:off] + ELSE + dst[i+63:i] := idx[i+63:i] + FI ENDFOR +dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point - KNCNI - Store - - - - - - - Down-converts 8 packed single-precision (32-bit) floating-point elements in "v1" using "conv" and stores them in memory locations starting at location "mv" at packed 64-bit integer indices stored in "index" scaled by "scale". "hint" indicates to the processor whether the data is non-temporal. + + Integer + AVX512F + Swizzle + + + + + + Shuffle 64-bit integers in "a" and "b" across lanes using the corresponding selector and index in "idx", and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). FOR j := 0 to 7 - addr := MEM[mv + index[j] * scale] - i := j*32 - CASE conv OF - _MM_DOWNCONV_PS_NONE: - addr[i+31:i] := v1[i+31:i] - _MM_DOWNCONV_PS_FLOAT16: - n := j*16 - addr[n+15:n] := Float32ToFloat16(v1[i+31:i]) - _MM_DOWNCONV_PS_UINT8: - n := j*8 - addr[n+7:n] := Float32ToUInt8(v1[i+31:i]) - _MM_DOWNCONV_PS_SINT8: - n := j*8 - addr[n+7:n] := Float32ToSInt8(v1[i+31:i]) - _MM_DOWNCONV_PS_UINT16: - n := j*16 - addr[n+15:n] := Float32ToUInt16(v1[i+31:i]) - _MM_DOWNCONV_PS_SINT16: - n := j*16 - addr[n+15:n] := Float32ToSInt16(v1[i+31:i]) - ESAC + i := j*64 + off := idx[i+2:i]*64 + IF k[j] + dst[i+63:i] := idx[i+3] ? b[off+63:off] : a[off+63:off] + ELSE + dst[i+63:i] := a[i+63:i] + FI ENDFOR +dst[MAX:512] := 0 +
immintrin.h
 - - Floating Point - KNCNI - Store - - - - - - - - Down-converts 8 packed single-precision (32-bit) floating-point elements in "v1" using "conv" and stores them in memory locations starting at location "mv" at packed 64-bit integer indices stored in "index" scaled by "scale". Elements are only written when the corresponding mask bit is set in "k"; otherwise, elements are unchanged in memory. "hint" indicates to the processor whether the data is non-temporal. + + Integer + AVX512F + Swizzle + + + + + + Shuffle 64-bit integers in "a" and "b" across lanes using the corresponding selector and index in "idx", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). FOR j := 0 to 7 - addr := MEM[mv + index[j] * scale] - i := j*32 + i := j*64 + off := idx[i+2:i]*64 IF k[j] - CASE conv OF - _MM_DOWNCONV_PS_NONE: - addr[i+31:i] := v[i+31:i] - _MM_DOWNCONV_PS_FLOAT16: - n := j*16 - addr[n+15:n] := Float32ToFloat16(v1[i+31:i]) - _MM_DOWNCONV_PS_UINT8: - n := j*8 - addr[n+7:n] := Float32ToUInt8(v1[i+31:i]) - _MM_DOWNCONV_PS_SINT8: - n := j*8 - addr[n+7:n] := Float32ToSInt8(v1[i+31:i]) - _MM_DOWNCONV_PS_UINT16: - n := j*16 - addr[n+15:n] := Float32ToUInt16(v1[i+31:i]) - _MM_DOWNCONV_PS_SINT16: - n := j*16 - addr[n+15:n] := Float32ToSInt16(v1[i+31:i]) - ESAC + dst[i+63:i] := (idx[i+3]) ? b[off+63:off] : a[off+63:off] + ELSE + dst[i+63:i] := 0 FI ENDFOR +dst[MAX:512] := 0 + +
immintrin.h
 - - Floating Point - KNCNI - Store - - - - - - - Down-converts 8 packed double-precision (64-bit) floating-point elements in "v1" using "conv" and stores them in memory locations starting at location "mv" at packed 64-bit integer indices stored in "index" scaled by "scale". "hint" indicates to the processor whether the data is non-temporal. + + Integer + AVX512F + Swizzle + + + + + Shuffle 64-bit integers in "a" and "b" across lanes using the corresponding selector and index in "idx", and store the results in "dst". FOR j := 0 to 7 - addr := MEM[mv + index[j] * scale] i := j*64 - CASE conv OF - _MM_DOWNCONV_EPI64_NONE: - addr[i+63:i] := v1[i+63:i] - ESAC + off := idx[i+2:i]*64 + dst[i+63:i] := idx[i+3] ? b[off+63:off] : a[off+63:off] ENDFOR +dst[MAX:512] := 0 + +
immintrin.h
 - + Floating Point - KNCNI - Store - - - - - - - - Down-converts 8 packed double-precision (64-bit) floating-point elements in "v1" using "conv" and stores them in memory locations starting at location "mv" at packed 64-bit integer indices stored in "index" scaled by "scale". Elements are written to memory using writemask "k" (elements are not stored to memory when the corresponding mask bit is not set; the memory location is left unchagned). "hint" indicates to the processor whether the data is non-temporal. + AVX512F + Swizzle + + + + + + Shuffle double-precision (64-bit) floating-point elements in "a" within 128-bit lanes using the control in "imm8", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). +IF (imm8[0] == 0) tmp_dst[63:0] := a[63:0]; FI +IF (imm8[0] == 1) tmp_dst[63:0] := a[127:64]; FI +IF (imm8[1] == 0) tmp_dst[127:64] := a[63:0]; FI +IF (imm8[1] == 1) tmp_dst[127:64] := a[127:64]; FI +IF (imm8[2] == 0) tmp_dst[191:128] := a[191:128]; FI +IF (imm8[2] == 1) tmp_dst[191:128] := a[255:192]; FI +IF (imm8[3] == 0) tmp_dst[255:192] := a[191:128]; FI +IF (imm8[3] == 1) tmp_dst[255:192] := a[255:192]; FI +IF (imm8[4] == 0) tmp_dst[319:256] := a[319:256]; FI +IF (imm8[4] == 1) tmp_dst[319:256] := a[383:320]; FI +IF (imm8[5] == 0) tmp_dst[383:320] := a[319:256]; FI +IF (imm8[5] == 1) tmp_dst[383:320] := a[383:320]; FI +IF (imm8[6] == 0) tmp_dst[447:384] := a[447:384]; FI +IF (imm8[6] == 1) tmp_dst[447:384] := a[511:448]; FI +IF (imm8[7] == 0) tmp_dst[511:448] := a[447:384]; FI +IF (imm8[7] == 1) tmp_dst[511:448] := a[511:448]; FI FOR j := 0 to 7 - addr := MEM[mv + index[j] * scale] i := j*64 IF k[j] - CASE conv OF - _MM_DOWNCONV_EPI64_NONE: - addr[i+63:i] := v1[i+63:i] - ESAC + dst[i+63:i] := tmp_dst[i+63:i] + ELSE + dst[i+63:i] := src[i+63:i] FI ENDFOR +dst[MAX:512] := 0 +
immintrin.h
 - - Integer - KNCNI - Store - - - - - - - Down-converts the low 8 packed 32-bit integer elements in "v1" using "conv" and stores them in memory locations starting at location "mv" at packed 64-bit integer indices stored in "index" scaled by "scale". "hint" indicates to the processor whether the data is non-temporal. + + Floating Point + AVX512F + Swizzle + + + + + + Shuffle double-precision (64-bit) floating-point elements in "a" within 128-bit lanes using the control in "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). +IF (b[1] == 0) tmp_dst[63:0] := a[63:0]; FI +IF (b[1] == 1) tmp_dst[63:0] := a[127:64]; FI +IF (b[65] == 0) tmp_dst[127:64] := a[63:0]; FI +IF (b[65] == 1) tmp_dst[127:64] := a[127:64]; FI +IF (b[129] == 0) tmp_dst[191:128] := a[191:128]; FI +IF (b[129] == 1) tmp_dst[191:128] := a[255:192]; FI +IF (b[193] == 0) tmp_dst[255:192] := a[191:128]; FI +IF (b[193] == 1) tmp_dst[255:192] := a[255:192]; FI +IF (b[257] == 0) tmp_dst[319:256] := a[319:256]; FI +IF (b[257] == 1) tmp_dst[319:256] := a[383:320]; FI +IF (b[321] == 0) tmp_dst[383:320] := a[319:256]; FI +IF (b[321] == 1) tmp_dst[383:320] := a[383:320]; FI +IF (b[385] == 0) tmp_dst[447:384] := a[447:384]; FI +IF (b[385] == 1) tmp_dst[447:384] := a[511:448]; FI +IF (b[449] == 0) tmp_dst[511:448] := a[447:384]; FI +IF (b[449] == 1) tmp_dst[511:448] := a[511:448]; FI FOR j := 0 to 7 - addr := MEM[mv + index[j] * scale] i := j*64 - CASE conv OF - _MM_DOWNCONV_EPI32_NONE: - addr[i+31:i] := v1[i+31:i] - _MM_DOWNCONV_EPI32_UINT8: - n := j*8 - addr[n+7:n] := UInt32ToUInt8(v1[i+31:i]) - _MM_DOWNCONV_EPI32_SINT8: - n := j*8 - addr[n+7:n] := SInt32ToSInt8(v1[i+31:i]) - _MM_DOWNCONV_EPI32_UINT16: - n := j*16 - addr[n+15:n] := UInt32ToUInt16(v1[i+31:i]) - _MM_DOWNCONV_EPI32_SINT16: - n := j*16 - addr[n+15:n] := SInt32ToSInt16(v1[n+15:n]) - ESAC + IF k[j] + dst[i+63:i] := tmp_dst[i+63:i] + ELSE + dst[i+63:i] := src[i+63:i] + FI ENDFOR +dst[MAX:512] := 0 +
immintrin.h
 - - Integer - KNCNI - Store - - - - - - - - Down-converts the low 8 packed 32-bit integer elements in "v1" using "conv" and stores them in memory locations starting at location "mv" at packed 64-bit integer indices stored in "index" scaled by "scale". Elements are written to memory using writemask "k" (elements are only written when the corresponding mask bit is set; otherwise, the memory location is left unchanged). "hint" indicates to the processor whether the data is non-temporal. + + Floating Point + AVX512F + Swizzle + + + + + Shuffle double-precision (64-bit) floating-point elements in "a" within 128-bit lanes using the control in "imm8", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). +IF (imm8[0] == 0) tmp_dst[63:0] := a[63:0]; FI +IF (imm8[0] == 1) tmp_dst[63:0] := a[127:64]; FI +IF (imm8[1] == 0) tmp_dst[127:64] := a[63:0]; FI +IF (imm8[1] == 1) tmp_dst[127:64] := a[127:64]; FI +IF (imm8[2] == 0) tmp_dst[191:128] := a[191:128]; FI +IF (imm8[2] == 1) tmp_dst[191:128] := a[255:192]; FI +IF (imm8[3] == 0) tmp_dst[255:192] := a[191:128]; FI +IF (imm8[3] == 1) tmp_dst[255:192] := a[255:192]; FI +IF (imm8[4] == 0) tmp_dst[319:256] := a[319:256]; FI +IF (imm8[4] == 1) tmp_dst[319:256] := a[383:320]; FI +IF (imm8[5] == 0) tmp_dst[383:320] := a[319:256]; FI +IF (imm8[5] == 1) tmp_dst[383:320] := a[383:320]; FI +IF (imm8[6] == 0) tmp_dst[447:384] := a[447:384]; FI +IF (imm8[6] == 1) tmp_dst[447:384] := a[511:448]; FI +IF (imm8[7] == 0) tmp_dst[511:448] := a[447:384]; FI +IF (imm8[7] == 1) tmp_dst[511:448] := a[511:448]; FI FOR j := 0 to 7 - addr := MEM[mv + index[j] * scale] i := j*64 IF k[j] - CASE conv OF - _MM_DOWNCONV_EPI32_NONE: - addr[i+31:i] := v1[i+31:i] - _MM_DOWNCONV_EPI32_UINT8: - n := j*8 - addr[n+7:n] := UInt32ToUInt8(v1[i+31:i]) - _MM_DOWNCONV_EPI32_SINT8: - n := j*8 - addr[n+7:n] := SInt32ToSInt8(v1[i+31:i]) - _MM_DOWNCONV_EPI32_UINT16: - n := j*16 - addr[n+15:n] := UInt32ToUInt16(v1[i+31:i]) - _MM_DOWNCONV_EPI32_SINT16: - n := j*16 - addr[n+15:n] := SInt32ToSInt16(v1[n+15:n]) - ESAC + dst[i+63:i] := tmp_dst[i+63:i] + ELSE + dst[i+63:i] := 0 FI ENDFOR +dst[MAX:512] := 0 +
immintrin.h
 - - Integer - KNCNI - Store - - - - - - - Down-converts 8 packed 64-bit integer elements in "v1" using "conv" and stores them in memory locations starting at location "mv" at packed 64-bit integer indices stored in "index" scaled by "scale". "hint" indicates to the processor whether the load is non-temporal. + + Floating Point + AVX512F + Swizzle + + + + + Shuffle double-precision (64-bit) floating-point elements in "a" within 128-bit lanes using the control in "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). +IF (b[1] == 0) tmp_dst[63:0] := a[63:0]; FI +IF (b[1] == 1) tmp_dst[63:0] := a[127:64]; FI +IF (b[65] == 0) tmp_dst[127:64] := a[63:0]; FI +IF (b[65] == 1) tmp_dst[127:64] := a[127:64]; FI +IF (b[129] == 0) tmp_dst[191:128] := a[191:128]; FI +IF (b[129] == 1) tmp_dst[191:128] := a[255:192]; FI +IF (b[193] == 0) tmp_dst[255:192] := a[191:128]; FI +IF (b[193] == 1) tmp_dst[255:192] := a[255:192]; FI +IF (b[257] == 0) tmp_dst[319:256] := a[319:256]; FI +IF (b[257] == 1) tmp_dst[319:256] := a[383:320]; FI +IF (b[321] == 0) tmp_dst[383:320] := a[319:256]; FI +IF (b[321] == 1) tmp_dst[383:320] := a[383:320]; FI +IF (b[385] == 0) tmp_dst[447:384] := a[447:384]; FI +IF (b[385] == 1) tmp_dst[447:384] := a[511:448]; FI +IF (b[449] == 0) tmp_dst[511:448] := a[447:384]; FI +IF (b[449] == 1) tmp_dst[511:448] := a[511:448]; FI FOR j := 0 to 7 - addr := MEM[mv + index[j] * scale] i := j*64 - CASE conv OF - _MM_DOWNCONV_EPI64_NONE: addr[i+63:i] := v1[i+63:i] - ESAC + IF k[j] + dst[i+63:i] := tmp_dst[i+63:i] + ELSE + dst[i+63:i] := 0 + FI ENDFOR +dst[MAX:512] := 0 +
immintrin.h
 - - Integer - KNCNI - Store - - - - - - - - Down-converts 8 packed 64-bit integer elements in "v1" using "conv" and stores them in memory locations starting at location "mv" at packed 64-bit integer indices stored in "index" scaled by "scale". Only those elements whose corresponding mask bit is set in writemask "k" are written to memory. + + Floating Point + AVX512F + Swizzle + + + + Shuffle double-precision (64-bit) floating-point elements in "a" within 128-bit lanes using the control in "imm8", and store the results in "dst". -FOR j := 0 to 7 - IF k[j] - addr := MEM[mv + index[j] * scale] - i := j*64 - CASE conv OF - _MM_DOWNCONV_EPI64_NONE: addr[i+63:i] := v1[i+63:i] - ESAC - FI -ENDFOR +IF (imm8[0] == 0) dst[63:0] := a[63:0]; FI +IF (imm8[0] == 1) dst[63:0] := a[127:64]; FI +IF (imm8[1] == 0) dst[127:64] := a[63:0]; FI +IF (imm8[1] == 1) dst[127:64] := a[127:64]; FI +IF (imm8[2] == 0) dst[191:128] := a[191:128]; FI +IF (imm8[2] == 1) dst[191:128] := a[255:192]; FI +IF (imm8[3] == 0) dst[255:192] := a[191:128]; FI +IF (imm8[3] == 1) dst[255:192] := a[255:192]; FI +IF (imm8[4] == 0) dst[319:256] := a[319:256]; FI +IF (imm8[4] == 1) dst[319:256] := a[383:320]; FI +IF (imm8[5] == 0) dst[383:320] := a[319:256]; FI +IF (imm8[5] == 1) dst[383:320] := a[383:320]; FI +IF (imm8[6] == 0) dst[447:384] := a[447:384]; FI +IF (imm8[6] == 1) dst[447:384] := a[511:448]; FI +IF (imm8[7] == 0) dst[511:448] := a[447:384]; FI +IF (imm8[7] == 1) dst[511:448] := a[511:448]; FI +dst[MAX:512] := 0 +
immintrin.h
 - - Integer + + Floating Point AVX512F - Store - - - Multiplies elements in packed 64-bit integer vectors "a" and "b" together, storing the lower 64 bits of the result in "dst". + Swizzle + + + + Shuffle double-precision (64-bit) floating-point elements in "a" within 128-bit lanes using the control in "b", and store the results in "dst". -FOR j := 0 to 7 - i := j*64 - dst[i+63:i] := a[i+63:i] * b[i+63:i] -ENDFOR +IF (b[1] == 0) dst[63:0] := a[63:0]; FI +IF (b[1] == 1) dst[63:0] := a[127:64]; FI +IF (b[65] == 0) dst[127:64] := a[63:0]; FI +IF (b[65] == 1) dst[127:64] := a[127:64]; FI +IF (b[129] == 0) dst[191:128] := a[191:128]; FI +IF (b[129] == 1) dst[191:128] := a[255:192]; FI +IF (b[193] == 0) dst[255:192] := a[191:128]; FI +IF (b[193] == 1) dst[255:192] := a[255:192]; FI +IF (b[257] == 0) dst[319:256] := a[319:256]; FI +IF (b[257] == 1) dst[319:256] := a[383:320]; FI +IF (b[321] == 0) dst[383:320] := a[319:256]; FI +IF (b[321] == 1) dst[383:320] := a[383:320]; FI +IF (b[385] == 0) dst[447:384] := a[447:384]; FI +IF (b[385] == 1) dst[447:384] := a[511:448]; FI +IF (b[449] == 0) dst[511:448] := a[447:384]; FI +IF (b[449] == 1) dst[511:448] := a[511:448]; FI dst[MAX:512] := 0 +
immintrin.h
 - - Integer + + Floating Point AVX512F - Store - - - - - Multiplies elements in packed 64-bit integer vectors "a" and "b" together, storing the lower 64 bits of the result in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Swizzle + + + + + + Shuffle single-precision (32-bit) floating-point elements in "a" within 128-bit lanes using the control in "imm8", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := j*64 +DEFINE SELECT4(src, control) { + CASE(control[1:0]) OF + 0: tmp[31:0] := src[31:0] + 1: tmp[31:0] := src[63:32] + 2: tmp[31:0] := src[95:64] + 3: tmp[31:0] := src[127:96] + ESAC + RETURN tmp[31:0] +} +tmp_dst[31:0] := SELECT4(a[127:0], imm8[1:0]) +tmp_dst[63:32] := SELECT4(a[127:0], imm8[3:2]) +tmp_dst[95:64] := SELECT4(a[127:0], imm8[5:4]) +tmp_dst[127:96] := SELECT4(a[127:0], imm8[7:6]) +tmp_dst[159:128] := SELECT4(a[255:128], imm8[1:0]) +tmp_dst[191:160] := SELECT4(a[255:128], imm8[3:2]) +tmp_dst[223:192] := SELECT4(a[255:128], imm8[5:4]) +tmp_dst[255:224] := SELECT4(a[255:128], imm8[7:6]) +tmp_dst[287:256] := SELECT4(a[383:256], imm8[1:0]) +tmp_dst[319:288] := SELECT4(a[383:256], imm8[3:2]) +tmp_dst[351:320] := SELECT4(a[383:256], imm8[5:4]) +tmp_dst[383:352] := SELECT4(a[383:256], imm8[7:6]) +tmp_dst[415:384] := SELECT4(a[511:384], imm8[1:0]) +tmp_dst[447:416] := SELECT4(a[511:384], imm8[3:2]) +tmp_dst[479:448] := SELECT4(a[511:384], imm8[5:4]) +tmp_dst[511:480] := SELECT4(a[511:384], imm8[7:6]) +FOR j := 0 to 15 + i := j*32 IF k[j] - dst[i+63:i] := a[i+63:i] * b[i+63:i] + dst[i+31:i] := tmp_dst[i+31:i] ELSE - dst[i+63:i] := src[i+63:i] + dst[i+31:i] := src[i+31:i] FI ENDFOR dst[MAX:512] := 0 +
immintrin.h
 - + Floating Point - KNCNI - Store - - - Permutes 128-bit blocks of the packed single-precision (32-bit) floating-point elements in "a" using constant "imm8". The results are stored in "dst". + AVX512F + Swizzle + + + + + + Shuffle single-precision (32-bit) floating-point elements in "a" within 128-bit lanes using the control in "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -SELECT4(src, control) { - CASE control[1:0] OF - 0: tmp[127:0] := src[127:0] - 1: tmp[127:0] := src[255:128] - 2: tmp[127:0] := src[383:256] - 3: tmp[127:0] := src[511:384] +DEFINE SELECT4(src, control) { + CASE(control[1:0]) OF + 0: tmp[31:0] := src[31:0] + 1: tmp[31:0] := src[63:32] + 2: tmp[31:0] := src[95:64] + 3: tmp[31:0] := src[127:96] ESAC - RETURN tmp[127:0] + RETURN tmp[31:0] } - -FOR j := 0 to 3 - i := j*128 - n := j*2 - dst[i+127:i] := SELECT4(a[511:0], imm8[n+1:n]) +tmp_dst[31:0] := SELECT4(a[127:0], b[1:0]) +tmp_dst[63:32] := SELECT4(a[127:0], b[33:32]) +tmp_dst[95:64] := SELECT4(a[127:0], b[65:64]) +tmp_dst[127:96] := SELECT4(a[127:0], b[97:96]) +tmp_dst[159:128] := SELECT4(a[255:128], b[129:128]) +tmp_dst[191:160] := SELECT4(a[255:128], b[161:160]) +tmp_dst[223:192] := SELECT4(a[255:128], b[193:192]) +tmp_dst[255:224] := SELECT4(a[255:128], b[225:224]) +tmp_dst[287:256] := SELECT4(a[383:256], b[257:256]) +tmp_dst[319:288] := SELECT4(a[383:256], b[289:288]) +tmp_dst[351:320] := SELECT4(a[383:256], b[321:320]) +tmp_dst[383:352] := SELECT4(a[383:256], b[353:352]) +tmp_dst[415:384] := SELECT4(a[511:384], b[385:384]) +tmp_dst[447:416] := SELECT4(a[511:384], b[417:416]) +tmp_dst[479:448] := SELECT4(a[511:384], b[449:448]) +tmp_dst[511:480] := SELECT4(a[511:384], b[481:480]) +FOR j := 0 to 15 + i := j*32 + IF k[j] + dst[i+31:i] := tmp_dst[i+31:i] + ELSE + dst[i+31:i] := src[i+31:i] + FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - KNCNI + AVX512F Swizzle - - - - - Permutes 128-bit blocks of the packed single-precision (32-bit) floating-point elements in "a" using constant "imm8". The results are stored in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + + + + Shuffle single-precision (32-bit) floating-point elements in "a" within 128-bit lanes using the control in "imm8", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -SELECT4(src, control) { - CASE control[1:0] OF - 0: tmp[127:0] := src[127:0] - 1: tmp[127:0] := src[255:128] - 2: tmp[127:0] := src[383:256] - 3: tmp[127:0] := src[511:384] +DEFINE SELECT4(src, control) { + CASE(control[1:0]) OF + 0: tmp[31:0] := src[31:0] + 1: tmp[31:0] := src[63:32] + 2: tmp[31:0] := src[95:64] + 3: tmp[31:0] := src[127:96] ESAC - RETURN tmp[127:0] + RETURN tmp[31:0] } - -tmp[511:0] := 0 -FOR j := 0 to 4 - i := j*128 - n := j*2 - tmp[i+127:i] := SELECT4(a[511:0], imm8[n+1:n]) +tmp_dst[31:0] := SELECT4(a[127:0], imm8[1:0]) +tmp_dst[63:32] := SELECT4(a[127:0], imm8[3:2]) +tmp_dst[95:64] := SELECT4(a[127:0], imm8[5:4]) +tmp_dst[127:96] := SELECT4(a[127:0], imm8[7:6]) +tmp_dst[159:128] := SELECT4(a[255:128], imm8[1:0]) +tmp_dst[191:160] := SELECT4(a[255:128], imm8[3:2]) +tmp_dst[223:192] := SELECT4(a[255:128], imm8[5:4]) +tmp_dst[255:224] := SELECT4(a[255:128], imm8[7:6]) +tmp_dst[287:256] := SELECT4(a[383:256], imm8[1:0]) +tmp_dst[319:288] := SELECT4(a[383:256], imm8[3:2]) +tmp_dst[351:320] := SELECT4(a[383:256], imm8[5:4]) +tmp_dst[383:352] := SELECT4(a[383:256], imm8[7:6]) +tmp_dst[415:384] := SELECT4(a[511:384], imm8[1:0]) +tmp_dst[447:416] := SELECT4(a[511:384], imm8[3:2]) +tmp_dst[479:448] := SELECT4(a[511:384], imm8[5:4]) +tmp_dst[511:480] := SELECT4(a[511:384], imm8[7:6]) +FOR j := 0 to 15 + i := j*32 + IF k[j] + dst[i+31:i] := tmp_dst[i+31:i] + ELSE + dst[i+31:i] := 0 + FI ENDFOR +dst[MAX:512] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512F + Swizzle + + + + + Shuffle single-precision (32-bit) floating-point elements in "a" within 128-bit lanes using the control in "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + +DEFINE SELECT4(src, control) { + CASE(control[1:0]) OF + 0: tmp[31:0] := src[31:0] + 1: tmp[31:0] := src[63:32] + 2: tmp[31:0] := src[95:64] + 3: tmp[31:0] := src[127:96] + ESAC + RETURN tmp[31:0] +} +tmp_dst[31:0] := SELECT4(a[127:0], b[1:0]) +tmp_dst[63:32] := SELECT4(a[127:0], b[33:32]) +tmp_dst[95:64] := SELECT4(a[127:0], b[65:64]) +tmp_dst[127:96] := SELECT4(a[127:0], b[97:96]) +tmp_dst[159:128] := SELECT4(a[255:128], b[129:128]) +tmp_dst[191:160] := SELECT4(a[255:128], b[161:160]) +tmp_dst[223:192] := SELECT4(a[255:128], b[193:192]) +tmp_dst[255:224] := SELECT4(a[255:128], b[225:224]) +tmp_dst[287:256] := SELECT4(a[383:256], b[257:256]) +tmp_dst[319:288] := SELECT4(a[383:256], b[289:288]) +tmp_dst[351:320] := SELECT4(a[383:256], b[321:320]) +tmp_dst[383:352] := SELECT4(a[383:256], b[353:352]) +tmp_dst[415:384] := SELECT4(a[511:384], b[385:384]) +tmp_dst[447:416] := SELECT4(a[511:384], b[417:416]) +tmp_dst[479:448] := SELECT4(a[511:384], b[449:448]) +tmp_dst[511:480] := SELECT4(a[511:384], b[481:480]) FOR j := 0 to 15 + i := j*32 IF k[j] - dst[i+31:i] := tmp[i+31:i] + dst[i+31:i] := tmp_dst[i+31:i] ELSE - dst[i+31:i] := src[i+31:i] + dst[i+31:i] := 0 FI ENDFOR dst[MAX:512] := 0 - + +
immintrin.h
+ + + Floating Point + AVX512F + Swizzle + + + + Shuffle single-precision (32-bit) floating-point elements in "a" within 128-bit lanes using the control in "imm8", and store the results in "dst". + +DEFINE SELECT4(src, control) { + CASE(control[1:0]) OF + 0: tmp[31:0] := src[31:0] + 1: tmp[31:0] := src[63:32] + 2: tmp[31:0] := src[95:64] + 3: tmp[31:0] := src[127:96] + ESAC + RETURN tmp[31:0] +} +dst[31:0] := SELECT4(a[127:0], imm8[1:0]) +dst[63:32] := SELECT4(a[127:0], imm8[3:2]) +dst[95:64] := SELECT4(a[127:0], imm8[5:4]) +dst[127:96] := SELECT4(a[127:0], imm8[7:6]) +dst[159:128] := SELECT4(a[255:128], imm8[1:0]) +dst[191:160] := SELECT4(a[255:128], imm8[3:2]) +dst[223:192] := SELECT4(a[255:128], imm8[5:4]) +dst[255:224] := SELECT4(a[255:128], imm8[7:6]) +dst[287:256] := SELECT4(a[383:256], imm8[1:0]) +dst[319:288] := SELECT4(a[383:256], imm8[3:2]) +dst[351:320] := SELECT4(a[383:256], imm8[5:4]) +dst[383:352] := SELECT4(a[383:256], imm8[7:6]) +dst[415:384] := SELECT4(a[511:384], imm8[1:0]) +dst[447:416] := SELECT4(a[511:384], imm8[3:2]) +dst[479:448] := SELECT4(a[511:384], imm8[5:4]) +dst[511:480] := SELECT4(a[511:384], imm8[7:6]) +dst[MAX:512] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512F + Swizzle + + + + Shuffle single-precision (32-bit) floating-point elements in "a" within 128-bit lanes using the control in "b", and store the results in "dst". + +DEFINE SELECT4(src, control) { + CASE(control[1:0]) OF + 0: tmp[31:0] := src[31:0] + 1: tmp[31:0] := src[63:32] + 2: tmp[31:0] := src[95:64] + 3: tmp[31:0] := src[127:96] + ESAC + RETURN tmp[31:0] +} +dst[31:0] := SELECT4(a[127:0], b[1:0]) +dst[63:32] := SELECT4(a[127:0], b[33:32]) +dst[95:64] := SELECT4(a[127:0], b[65:64]) +dst[127:96] := SELECT4(a[127:0], b[97:96]) +dst[159:128] := SELECT4(a[255:128], b[129:128]) +dst[191:160] := SELECT4(a[255:128], b[161:160]) +dst[223:192] := SELECT4(a[255:128], b[193:192]) +dst[255:224] := SELECT4(a[255:128], b[225:224]) +dst[287:256] := SELECT4(a[383:256], b[257:256]) +dst[319:288] := SELECT4(a[383:256], b[289:288]) +dst[351:320] := SELECT4(a[383:256], b[321:320]) +dst[383:352] := SELECT4(a[383:256], b[353:352]) +dst[415:384] := SELECT4(a[511:384], b[385:384]) +dst[447:416] := SELECT4(a[511:384], b[417:416]) +dst[479:448] := SELECT4(a[511:384], b[449:448]) +dst[511:480] := SELECT4(a[511:384], b[481:480]) +dst[MAX:512] := 0 + +
immintrin.h
 - + Floating Point AVX512F - Trigonometry - - - Computes the sine and cosine of the packed double-precision (64-bit) floating-point elements in "a" and stores the results of the sine computation in "dst" and the results of the cosine computation in "cos_res". + Swizzle + + + + + + Shuffle double-precision (64-bit) floating-point elements in "a" within 256-bit lanes using the control in "imm8", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). +DEFINE SELECT4(src, control) { + CASE(control[1:0]) OF + 0: tmp[63:0] := src[63:0] + 1: tmp[63:0] := src[127:64] + 2: tmp[63:0] := src[191:128] + 3: tmp[63:0] := src[255:192] + ESAC + RETURN tmp[63:0] +} +tmp_dst[63:0] := SELECT4(a[255:0], imm8[1:0]) +tmp_dst[127:64] := SELECT4(a[255:0], imm8[3:2]) +tmp_dst[191:128] := SELECT4(a[255:0], imm8[5:4]) +tmp_dst[255:192] := SELECT4(a[255:0], imm8[7:6]) +tmp_dst[319:256] := SELECT4(a[511:256], imm8[1:0]) +tmp_dst[383:320] := SELECT4(a[511:256], imm8[3:2]) +tmp_dst[447:384] := SELECT4(a[511:256], imm8[5:4]) +tmp_dst[511:448] := SELECT4(a[511:256], imm8[7:6]) FOR j := 0 to 7 i := j*64 - dst[i+63:i] := SIN(a[i+63:i]) - cos_res[i+63:i] := COS(a[i+63:i]) + IF k[j] + dst[i+63:i] := tmp_dst[i+63:i] + ELSE + dst[i+63:i] := src[i+63:i] + FI ENDFOR dst[MAX:512] := 0 -cos_res[MAX:512] := 0 +
immintrin.h
 - + Floating Point AVX512F - Trigonometry - - - - - - Computes the sine and cosine of the packed double-precision (64-bit) floating-point elements in "a" and stores the results of the sine computation in "dst" and the results of the cosine computation in "cos_res". Elements are written to their respective locations using writemask "k" (elements are copied from "sin_src" or "cos_src" when the corresponding mask bit is not set). + Swizzle + + + + + + Shuffle double-precision (64-bit) floating-point elements in "a" across lanes using the corresponding index in "idx", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). FOR j := 0 to 7 i := j*64 + id := idx[i+2:i]*64 IF k[j] - dst[i+63:i] := SIN(a[i+63:i]) - cos_res[i+63:i] := COS(a[i+63:i]) + dst[i+63:i] := a[id+63:id] ELSE - dst[i+63:i] := sin_src[i+63:i] - cos_res[i+63:i] := cos_src[i+63:i] + dst[i+63:i] := src[i+63:i] FI ENDFOR dst[MAX:512] := 0 -cos_res[MAX:512] := 0 +
immintrin.h
 - + Floating Point AVX512F - Trigonometry - - - Computes the sine and cosine of the packed single-precision (32-bit) floating-point elements in "a" and stores the results of the sine computation in "dst" and the results of the cosine computation in "cos_res". + Swizzle + + + + + Shuffle double-precision (64-bit) floating-point elements in "a" within 256-bit lanes using the control in "imm8", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*32 - dst[i+31:i] := SIN(a[i+31:i]) - cos_res[i+31:i] := COS(a[i+31:i]) +DEFINE SELECT4(src, control) { + CASE(control[1:0]) OF + 0: tmp[63:0] := src[63:0] + 1: tmp[63:0] := src[127:64] + 2: tmp[63:0] := src[191:128] + 3: tmp[63:0] := src[255:192] + ESAC + RETURN tmp[63:0] +} +tmp_dst[63:0] := SELECT4(a[255:0], imm8[1:0]) +tmp_dst[127:64] := SELECT4(a[255:0], imm8[3:2]) +tmp_dst[191:128] := SELECT4(a[255:0], imm8[5:4]) +tmp_dst[255:192] := SELECT4(a[255:0], imm8[7:6]) +tmp_dst[319:256] := SELECT4(a[511:256], imm8[1:0]) +tmp_dst[383:320] := SELECT4(a[511:256], imm8[3:2]) +tmp_dst[447:384] := SELECT4(a[511:256], imm8[5:4]) +tmp_dst[511:448] := SELECT4(a[511:256], imm8[7:6]) +FOR j := 0 to 7 + i := j*64 + IF k[j] + dst[i+63:i] := tmp_dst[i+63:i] + ELSE + dst[i+63:i] := 0 + FI ENDFOR dst[MAX:512] := 0 -cos_res[MAX:512] := 0 +
immintrin.h
 - + Floating Point AVX512F - Trigonometry - - - - - - Computes the sine and cosine of the packed single-precision (32-bit) floating-point elements in "a" and stores the results of the sine computation in "dst" and the results of the cosine computation in "cos_res". Elements are written to their respective locations using writemask "k" (elements are copied from "sin_src" or "cos_src" when the corresponding mask bit is not set). + Swizzle + + + + + Shuffle double-precision (64-bit) floating-point elements in "a" across lanes using the corresponding index in "idx", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*32 +FOR j := 0 to 7 + i := j*64 + id := idx[i+2:i]*64 IF k[j] - dst[i+31:i] := SIN(a[i+31:i]) - cos_res[i+31:i] := COS(a[i+31:i]) + dst[i+63:i] := a[id+63:id] ELSE - dst[i+31:i] := sin_src[i+31:i] - cos_res[i+31:i] := cos_src[i+31:i] + dst[i+63:i] := 0 FI ENDFOR dst[MAX:512] := 0 -cos_res[MAX:512] := 0 - -
immintrin.h
- - - Integer - KNCNI - Load - - - - Loads 8 32-bit integer memory locations starting at location "mv" at packed 64-bit integer indices stored in "index" scaled by "scale" to "dst". - -FOR j := 0 to 7 - addr := MEM[mv + index[j] * scale] - i := j*32 - dst[i+31:i] := addr[i+31:i] -ENDFOR -dst[MAX:256] := 0 +
immintrin.h
 - - Integer - KNCNI - Load - - - - - - Loads 8 32-bit integer memory locations starting at location "mv" at packed 64-bit integer indices stored in "index" scaled by "scale" to "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Floating Point + AVX512F + Swizzle + + + + Shuffle double-precision (64-bit) floating-point elements in "a" within 256-bit lanes using the control in "imm8", and store the results in "dst". -FOR j := 0 to 7 - i := j*32 - IF k[j] - addr := MEM[mv + index[j] * scale] - dst[i+31:i] := addr[i+31:i] - ELSE - dst[i+31:i] := src[i+31:i] - FI -ENDFOR -dst[MAX:256] := 0 +DEFINE SELECT4(src, control) { + CASE(control[1:0]) OF + 0: tmp[63:0] := src[63:0] + 1: tmp[63:0] := src[127:64] + 2: tmp[63:0] := src[191:128] + 3: tmp[63:0] := src[255:192] + ESAC + RETURN tmp[63:0] +} +dst[63:0] := SELECT4(a[255:0], imm8[1:0]) +dst[127:64] := SELECT4(a[255:0], imm8[3:2]) +dst[191:128] := SELECT4(a[255:0], imm8[5:4]) +dst[255:192] := SELECT4(a[255:0], imm8[7:6]) +dst[319:256] := SELECT4(a[511:256], imm8[1:0]) +dst[383:320] := SELECT4(a[511:256], imm8[3:2]) +dst[447:384] := SELECT4(a[511:256], imm8[5:4]) +dst[511:448] := SELECT4(a[511:256], imm8[7:6]) +dst[MAX:512] := 0 +
immintrin.h
 - + Floating Point - KNCNI - Load - - - - Loads 8 single-precision (32-bit) floating-point memory locations starting at location "mv" at packed 64-bit integer indices stored in "index" scaled by "scale" to "dst". + AVX512F + Swizzle + + + + Shuffle double-precision (64-bit) floating-point elements in "a" across lanes using the corresponding index in "idx", and store the results in "dst". FOR j := 0 to 7 - addr := MEM[mv + index[j] * scale] - i := j*32 - dst[i+31:i] := addr[i+31:i] + i := j*64 + id := idx[i+2:i]*64 + dst[i+63:i] := a[id+63:id] ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 +
immintrin.h
 - + Floating Point - KNCNI - Load - - - - - - Loads 8 single-precision (32-bit) floating-point memory locations starting at location "mv" at packed 64-bit integer indices stored in "index" scaled by "scale" to "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + AVX512F + Swizzle + + + + + + Shuffle single-precision (32-bit) floating-point elements in "a" across lanes using the corresponding index in "idx", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 7 +FOR j := 0 to 15 i := j*32 + id := idx[i+3:i]*32 IF k[j] - addr := MEM[mv + index[j] * scale] - dst[i+31:i] := addr[i+31:i] + dst[i+31:i] := a[id+31:id] ELSE dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:256] := 0 - -
immintrin.h
- - - Floating Point - KNCNI - Store - - - - - Stores 8 packed single-precision (32-bit) floating-point elements in "v" in memory locations starting at location "mv" at packed 64-bit integer indices stored in "index" scaled by "scale". - -FOR j := 0 to 7 - i := j*32 - addr := MEM[mv + index[j] * scale] - addr[i+31:i] := v[i+31:i] -ENDFOR +dst[MAX:512] := 0 +
immintrin.h
 - + Floating Point - KNCNI - Store - - - - - - Stores 8 packed single-precision (32-bit) floating-point elements in "v1" in memory locations starting at location "mv" at packed 64-bit integer indices stored in "index" scaled by "scale" using writemask "k" (elements are only written to memory when the corresponding mask bit is set). - -FOR j := 0 to 7 - i := j*32 - IF k[j] - addr := MEM[mv + index[j] * scale] - addr[i+31:i] := v1[i+31:i] - FI -ENDFOR - -
immintrin.h
- - - Integer - KNCNI - Store - - - - - Stores 8 packed 32-bit integer elements in "v1" in memory locations starting at location "mv" at packed 64-bit integer indices stored in "index" scaled by "scale". - -FOR j := 0 to 7 - i := j*32 - addr := MEM[mv + index[j] * scale] - addr[i+31:i] := v1[i+31:i] -ENDFOR - -
immintrin.h
- - - Integer - KNCNI - Store - - - - - - Stores 8 packed 32-bit integer elements in "v1" in memory locations starting at location "mv" at packed 64-bit integer indices stored in "index" scaled by "scale" using writemask "k" (elements are only written to memory when the corresponding mask bit is set). + AVX512F + Swizzle + + + + + Shuffle single-precision (32-bit) floating-point elements in "a" across lanes using the corresponding index in "idx", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 +FOR j := 0 to 15 i := j*32 + id := idx[i+3:i]*32 IF k[j] - addr := MEM[mv + index[j] * scale] - addr[i+31:i] := v1[i+31:i] - FI + dst[i+31:i] := a[id+31:id] + ELSE + dst[i+31:i] := 0 + FI ENDFOR +dst[MAX:512] := 0 +
immintrin.h
 - - Mask - KNCNI - Mask - - - Move the high element from "k1" to the low element of "k1", and insert the low element of "k2" into the high element of "k1". - -tmp[7:0] := k1[15:8] -k1[15:8] := k2[7:0] -k1[7:0] := tmp[7:0] - - -
immintrin.h
- - - Mask - KNCNI - Mask - - - Insert the low element of "k2" into the high element of "k1". - -k1[15:8] := k2[7:0] - - -
immintrin.h
- - - Floating Point - AVX512 - Convert - - Copy the lower single-precision (32-bit) floating-point element of "a" to "dst". - dst[31:0] := a[31:0] - -
immintrin.h
- - + Floating Point - AVX512 - Convert - - Copy the lower double-precision (64-bit) floating-point element of "a" to "dst". - dst[63:0] := a[63:0] - -
immintrin.h
- - - Integer - AVX512 - Convert - - Copy the lower 32-bit integer in "a" to "dst". - -dst[31:0] := a[31:0] - - -
immintrin.h
- - - Integer - AVX512_4VNNIW - Arithmetic - - - - - - - Compute 4 sequential operand source-block dot-products of two signed 16-bit element operands with 32-bit element accumulation, and store the results in "dst". + AVX512F + Swizzle + + + + Shuffle single-precision (32-bit) floating-point elements in "a" across lanes using the corresponding index in "idx". FOR j := 0 to 15 - FOR m := 0 to 3 - lim_base := m*32 - i := j*32 - tl := b[lim_base+15:lim_base] - tu := b[lim_base+31:lim_base+16] - lword := a{m}[i+15:i] * tl - uword := a{m}[i+31:i+16] * tu - dst[i+31:i] := src[i+31:i] + lword + uword - ENDFOR + i := j*32 + id := idx[i+3:i]*32 + dst[i+31:i] := a[id+31:id] ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX512_4VNNIW - Arithmetic - - - - - - - - Compute 4 sequential operand source-block dot-products of two signed 16-bit element operands with 32-bit element accumulation with mask, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set).. + AVX512F + Swizzle + + + + + + Shuffle 64-bit integers in "a" within 256-bit lanes using the control in "imm8", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 15 - IF mask[j] - FOR m := 0 to 3 - lim_base := m*32 - i := j*32 - tl := b[lim_base+15:lim_base] - tu := b[lim_base+31:lim_base+16] - lword := a{m}[i+15:i] * tl - uword := a{m}[i+31:i+16] * tu - dst[i+31:i] := src[i+31:i] + lword + uword - ENDFOR +DEFINE SELECT4(src, control) { + CASE(control[1:0]) OF + 0: tmp[63:0] := src[63:0] + 1: tmp[63:0] := src[127:64] + 2: tmp[63:0] := src[191:128] + 3: tmp[63:0] := src[255:192] + ESAC + RETURN tmp[63:0] +} +tmp_dst[63:0] := SELECT4(a[255:0], imm8[1:0]) +tmp_dst[127:64] := SELECT4(a[255:0], imm8[3:2]) +tmp_dst[191:128] := SELECT4(a[255:0], imm8[5:4]) +tmp_dst[255:192] := SELECT4(a[255:0], imm8[7:6]) +tmp_dst[319:256] := SELECT4(a[511:256], imm8[1:0]) +tmp_dst[383:320] := SELECT4(a[511:256], imm8[3:2]) +tmp_dst[447:384] := SELECT4(a[511:256], imm8[5:4]) +tmp_dst[511:448] := SELECT4(a[511:256], imm8[7:6]) +FOR j := 0 to 7 + i := j*64 + IF k[j] + dst[i+63:i] := tmp_dst[i+63:i] ELSE - dst[i+31:i] := src[i+31:i] + dst[i+63:i] := src[i+63:i] FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX512_4VNNIW - Arithmetic - - - - - - - - Compute 4 sequential operand source-block dot-products of two signed 16-bit element operands with 32-bit element accumulation with mask, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + AVX512F + Swizzle + + + + + + Shuffle 64-bit integers in "a" across lanes using the corresponding index in "idx", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 15 - IF mask[j] - FOR m := 0 to 3 - lim_base := m*32 - i := j*32 - tl := b[lim_base+15:lim_base] - tu := b[lim_base+31:lim_base+16] - lword := a{m}[i+15:i] * tl - uword := a{m}[i+31:i+16] * tu - dst[i+31:i] := src[i+31:i] + lword + uword - ENDFOR +FOR j := 0 to 7 + i := j*64 + id := idx[i+2:i]*64 + IF k[j] + dst[i+63:i] := a[id+63:id] ELSE - dst[i+31:i] := 0 + dst[i+63:i] := src[i+63:i] FI ENDFOR dst[MAX:512] := 0 - -
immintrin.h
- - - Integer - AVX512_4VNNIW - Arithmetic - - - - - - - Compute 4 sequential operand source-block dot-products of two signed 16-bit element operands with 32-bit element accumulation and signed saturation, and store the results in "dst". - -FOR j := 0 to 15 - FOR m := 0 to 3 - lim_base := m*32 - i := j*32 - tl := b[lim_base+15:lim_base] - tu := b[lim_base+31:lim_base+16] - lword := a{m}[i+15:i] * tl - uword := a{m}[i+31:i+16] * tu - dst[i+31:i] := SIGNED_DWORD_SATURATE(src[i+31:i] + lword + uword) - ENDFOR -ENDFOR -dst[MAX:512] := 0 - - +
immintrin.h
 - + Integer - AVX512_4VNNIW - Arithmetic - - - - - - - - Compute 4 sequential operand source-block dot-products of two signed 16-bit element operands with 32-bit element accumulation with mask and signed saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set).. + AVX512F + Swizzle + + + + + Shuffle 64-bit integers in "a" within 256-bit lanes using the control in "imm8", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 15 - IF mask[i] - FOR m := 0 to 3 - lim_base := m*32 - i := j*32 - tl := b[lim_base+15:lim_base] - tu := b[lim_base+31:lim_base+16] - lword := a{m}[i+15:i] * tl - uword := a{m}[i+31:i+16] * tu - dst[i+31:i] := SIGNED_DWORD_SATURATE(src[i+31:i] + lword + uword) - ENDFOR +DEFINE SELECT4(src, control) { + CASE(control[1:0]) OF + 0: tmp[63:0] := src[63:0] + 1: tmp[63:0] := src[127:64] + 2: tmp[63:0] := src[191:128] + 3: tmp[63:0] := src[255:192] + ESAC + RETURN tmp[63:0] +} +tmp_dst[63:0] := SELECT4(a[255:0], imm8[1:0]) +tmp_dst[127:64] := SELECT4(a[255:0], imm8[3:2]) +tmp_dst[191:128] := SELECT4(a[255:0], imm8[5:4]) +tmp_dst[255:192] := SELECT4(a[255:0], imm8[7:6]) +tmp_dst[319:256] := SELECT4(a[511:256], imm8[1:0]) +tmp_dst[383:320] := SELECT4(a[511:256], imm8[3:2]) +tmp_dst[447:384] := SELECT4(a[511:256], imm8[5:4]) +tmp_dst[511:448] := SELECT4(a[511:256], imm8[7:6]) +FOR j := 0 to 7 + i := j*64 + IF k[j] + dst[i+63:i] := tmp_dst[i+63:i] ELSE - dst[i+31:i] := src[i+31:i] + dst[i+63:i] := 0 FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX512_4VNNIW - Arithmetic - - - - - - - - Compute 4 sequential operand source-block dot-products of two signed 16-bit element operands with 32-bit element accumulation with mask and signed saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set).. + AVX512F + Swizzle + + + + + Shuffle 64-bit integers in "a" across lanes using the corresponding index in "idx", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 15 - IF mask[i] - FOR m := 0 to 3 - lim_base := m*32 - i := j*32 - tl := b[lim_base+15:lim_base] - tu := b[lim_base+31:lim_base+16] - lword := a{m}[i+15:i] * tl - uword := a{m}[i+31:i+16] * tu - dst[i+31:i] := SIGNED_DWORD_SATURATE(src[i+31:i] + lword + uword) - ENDFOR +FOR j := 0 to 7 + i := j*64 + id := idx[i+2:i]*64 + IF k[j] + dst[i+63:i] := a[id+63:id] ELSE - dst[i+31:i] := 0 + dst[i+63:i] := 0 FI ENDFOR dst[MAX:512] := 0 - -
immintrin.h
- - - Floating Point - AVX512_4FMAPS - Arithmetic - - - - - - - Multiply packed single-precision (32-bit) floating-point elements specified in 4 consecutive operands "b0" through "b3" by the 4 corresponding packed elements in "c", accumulating with the corresponding elements in "a". Store the results in "dst". - -dst := a -FOR m := 0 to 3 - FOR j := 0 to 15 - i = j*32 - n = m*32 - dst[i+31:i] := RoundFPControl_MXCSR(dst[i+31:i] + b{m}[i+31:i] * c[n+31:n]) - ENDFOR -ENDFOR -dst[MAX:512] := 0 - - -
immintrin.h
- - - Floating Point - AVX512_4FMAPS - Arithmetic - - - - - - - - Multiply packed single-precision (32-bit) floating-point elements specified in 4 consecutive operands "b0" through "b3" by the 4 corresponding packed elements in "c", accumulating with the corresponding elements in "a". Store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). - -dst := a -FOR m := 0 to 3 - FOR j := 0 to 15 - i = j*32 - n = m*32 - IF mask[j] - dst[i+31:i] := RoundFPControl_MXCSR(dst[i+31:i] + b{m}[i+31:i] * c[n+31:n]) - FI - ENDFOR -ENDFOR -dst[MAX:512] := 0 - - -
immintrin.h
- - - Floating Point - AVX512_4FMAPS - Arithmetic - - - - - - - - Multiply packed single-precision (32-bit) floating-point elements specified in 4 consecutive operands "b0" through "b3" by the 4 corresponding packed elements in "c", accumulating with the corresponding elements in "a". Store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - -dst := a -FOR m := 0 to 3 - FOR j := 0 to 15 - i = j*32 - n = m*32 - IF mask[j] - dst[i+31:i] := RoundFPControl_MXCSR(dst[i+31:i] + b{m}[i+31:i] * c[n+31:n]) - ELSE - dst[i+31:i] := 0 - FI - ENDFOR -ENDFOR -dst[MAX:512] := 0 - - -
immintrin.h
- - - Floating Point - AVX512_4FMAPS - Arithmetic - - - - - - - Multiply packed single-precision (32-bit) floating-point elements specified in 4 consecutive operands "b0" through "b3" by the 4 corresponding packed elements in "c", accumulating the negated intermediate result with the corresponding elements in "a". Store the results in "dst". - -dst := a -FOR m := 0 to 3 - FOR j := 0 to 15 - i = j*32 - n = m*32 - dst[i+31:i] := RoundFPControl_MXCSR(dst[i+31:i] - b{m}[i+31:i] * c[n+31:n]) - ENDFOR -ENDFOR -dst[MAX:512] := 0 - - -
immintrin.h
- - - Floating Point - AVX512_4FMAPS - Arithmetic - - - - - - - - Multiply packed single-precision (32-bit) floating-point elements specified in 4 consecutive operands "b0" through "b3" by the 4 corresponding packed elements in "c", accumulating the negated intermediate result with the corresponding elements in "a". Store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). - -dst := a -FOR m := 0 to 3 - FOR j := 0 to 15 - i = j*32 - n = m*32 - IF mask[j] - dst[i+31:i] := RoundFPControl_MXCSR(dst[i+31:i] - b{m}[i+31:i] * c[n+31:n]) - FI - ENDFOR -ENDFOR -dst[MAX:512] := 0 - - -
immintrin.h
- - - Floating Point - AVX512_4FMAPS - Arithmetic - - - - - - - - Multiply packed single-precision (32-bit) floating-point elements specified in 4 consecutive operands "b0" through "b3" by the 4 corresponding packed elements in "c", accumulating the negated intermediate result with the corresponding elements in "a". Store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - -dst := a -FOR m := 0 to 3 - FOR j := 0 to 15 - i = j*32 - n = m*32 - IF mask[j] - dst[i+31:i] := RoundFPControl_MXCSR(dst[i+31:i] - b{m}[i+31:i] * c[n+31:n]) - ELSE - dst[i+31:i] := 0 - FI - ENDFOR -ENDFOR -dst[MAX:512] := 0 - - -
immintrin.h
- - - Floating Point - AVX512_4FMAPS - Arithmetic - - - - - - - Multiply the lower single-precision (32-bit) floating-point elements specified in 4 consecutive operands "b0" through "b3" by corresponding element in "c", accumulating with the lower element in "a". Store the result in the lower element of "dst". - -dst := a -FOR j := 0 to 3 - i := j*32 - dst[31:0] := RoundFPControl_MXCSR(dst[31:0] + b{j}[31:0] * c[i+31:i]) -ENDFOR -dst[MAX:32] := 0 - - -
immintrin.h
- - - Floating Point - AVX512_4FMAPS - Arithmetic - - - - - - - - Multiply the lower single-precision (32-bit) floating-point elements specified in 4 consecutive operands "b0" through "b3" by corresponding element in "c", accumulating with the lower element in "a". Store the result in the lower element of "dst" using writemask "k" (the element is copied from "a" when mask bit 0 is not set). - -dst := a -IF k[0] - FOR j := 0 to 3 - i := j*32 - dst[31:0] := RoundFPControl_MXCSR(dst[31:0] + b{j}[31:0] * c[i+31:i]) - ENDFOR -FI -dst[MAX:32] := 0 - - -
immintrin.h
- - - Floating Point - AVX512_4FMAPS - Arithmetic - - - - - - - - Multiply the lower single-precision (32-bit) floating-point elements specified in 4 consecutive operands "b0" through "b3" by corresponding element in "c", accumulating with the lower element in "a". Store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set). - -dst := a -IF k[0] - FOR j := 0 to 3 - i := j*32 - dst[31:0] := RoundFPControl_MXCSR(dst[31:0] + b{j}[31:0] * c[i+31:i]) - ENDFOR -ELSE - dst[31:0] := 0 -FI -dst[MAX:32] := 0 - - -
immintrin.h
- - - Floating Point - AVX512_4FMAPS - Arithmetic - - - - - - - Multiply the lower single-precision (32-bit) floating-point elements specified in 4 consecutive operands "b0" through "b3" by corresponding element in "c", accumulating the negated intermediate result with the lower element in "a". Store the result in the lower element of "dst". - -dst := a -FOR j := 0 to 3 - i := j*32 - dst[31:0] := RoundFPControl_MXCSR(dst[31:0] - b{j}[31:0] * c[i+31:i]) -ENDFOR -dst[MAX:32] := 0 - - +
immintrin.h
 - - Floating Point - AVX512_4FMAPS - Arithmetic - - - - - - - - Multiply the lower single-precision (32-bit) floating-point elements specified in 4 consecutive operands "b0" through "b3" by corresponding element in "c", accumulating the negated intermediate result with the lower element in "a". Store the result in the lower element of "dst" using writemask "k" (the element is copied from "a" when mask bit 0 is not set). + + Integer + AVX512F + Swizzle + + + + Shuffle 64-bit integers in "a" within 256-bit lanes using the control in "imm8", and store the results in "dst". -dst := a -IF k[0] - FOR j := 0 to 3 - i := j*32 - dst[31:0] := RoundFPControl_MXCSR(dst[31:0] - b{j}[31:0] * c[i+31:i]) - ENDFOR -FI -dst[MAX:32] := 0 +DEFINE SELECT4(src, control) { + CASE(control[1:0]) OF + 0: tmp[63:0] := src[63:0] + 1: tmp[63:0] := src[127:64] + 2: tmp[63:0] := src[191:128] + 3: tmp[63:0] := src[255:192] + ESAC + RETURN tmp[63:0] +} +dst[63:0] := SELECT4(a[255:0], imm8[1:0]) +dst[127:64] := SELECT4(a[255:0], imm8[3:2]) +dst[191:128] := SELECT4(a[255:0], imm8[5:4]) +dst[255:192] := SELECT4(a[255:0], imm8[7:6]) +dst[319:256] := SELECT4(a[511:256], imm8[1:0]) +dst[383:320] := SELECT4(a[511:256], imm8[3:2]) +dst[447:384] := SELECT4(a[511:256], imm8[5:4]) +dst[511:448] := SELECT4(a[511:256], imm8[7:6]) +dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point - AVX512_4FMAPS - Arithmetic - - - - - - - - Multiply the lower single-precision (32-bit) floating-point elements specified in 4 consecutive operands "b0" through "b3" by corresponding element in "c", accumulating the negated intermediate result with the lower element in "a". Store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set). + + Integer + AVX512F + Swizzle + + + + Shuffle 64-bit integers in "a" across lanes using the corresponding index in "idx", and store the results in "dst". -dst := a -IF k[0] - FOR j := 0 to 3 - i := j*32 - dst[31:0] := RoundFPControl_MXCSR(dst[31:0] - b{j}[31:0] * c[i+31:i]) - ENDFOR -ELSE - dst[31:0] := 0 -FI -dst[MAX:32] := 0 +FOR j := 0 to 7 + i := j*64 + id := idx[i+2:i]*64 + dst[i+63:i] := a[id+63:id] +ENDFOR +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX512VPOPCNTDQ - Bit Manipulation - - Count the number of logical 1 bits in packed 32-bit integers in "a", and store the results in "dst". + AVX512F + Swizzle + + + + + Load contiguous active 32-bit integers from "a" (those with their respective bit set in mask "k"), and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). +m := 0 FOR j := 0 to 15 i := j*32 - dst[i+31:i] := a[i+31:i] + IF k[j] + dst[i+31:i] := a[m+31:m] + m := m + 32 + ELSE + dst[i+31:i] := src[i+31:i] + FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX512VPOPCNTDQ - Bit Manipulation - - - - Count the number of logical 1 bits in packed 32-bit integers in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + AVX512F + Load + Swizzle + + + + + Load contiguous active 32-bit integers from unaligned memory at "mem_addr" (those with their respective bit set in mask "k"), and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). +m := 0 FOR j := 0 to 15 i := j*32 IF k[j] - dst[i+31:i] := POPCNT(a[i+31:i]) + dst[i+31:i] := MEM[mem_addr+m+31:mem_addr+m] + m := m + 32 ELSE dst[i+31:i] := src[i+31:i] FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX512VPOPCNTDQ - Bit Manipulation - - - Count the number of logical 1 bits in packed 32-bit integers in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + AVX512F + Swizzle + + + + Load contiguous active 32-bit integers from "a" (those with their respective bit set in mask "k"), and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). +m := 0 FOR j := 0 to 15 i := j*32 IF k[j] - dst[i+31:i] := POPCNT(a[i+31:i]) + dst[i+31:i] := a[m+31:m] + m := m + 32 ELSE dst[i+31:i] := 0 FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX512VPOPCNTDQ - Bit Manipulation - - Count the number of logical 1 bits in packed 64-bit integers in "a", and store the results in "dst". + AVX512F + Load + Swizzle + + + + Load contiguous active 32-bit integers from unaligned memory at "mem_addr" (those with their respective bit set in mask "k"), and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := j*64 - dst[i+63:i] := a[i+63:i] +m := 0 +FOR j := 0 to 15 + i := j*32 + IF k[j] + dst[i+31:i] := MEM[mem_addr+m+31:mem_addr+m] + m := m + 32 + ELSE + dst[i+31:i] := 0 + FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX512VPOPCNTDQ - Bit Manipulation - - - - Count the number of logical 1 bits in packed 64-bit integers in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + AVX512F + Swizzle + + + + + Load contiguous active 64-bit integers from "a" (those with their respective bit set in mask "k"), and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). +m := 0 FOR j := 0 to 7 i := j*64 IF k[j] - dst[i+63:i] := POPCNT(a[i+63:i]) + dst[i+63:i] := a[m+63:m] + m := m + 64 ELSE dst[i+63:i] := src[i+63:i] FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX512VPOPCNTDQ - Bit Manipulation - - - Count the number of logical 1 bits in packed 64-bit integers in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + AVX512F + Load + Swizzle + + + + + Load contiguous active 64-bit integers from unaligned memory at "mem_addr" (those with their respective bit set in mask "k"), and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). +m := 0 FOR j := 0 to 7 i := j*64 IF k[j] - dst[i+63:i] := POPCNT(a[i+63:i]) + dst[i+63:i] := MEM[mem_addr+m+63:mem_addr+m] + m := m + 64 ELSE - dst[i+63:i] := 0 + dst[i+63:i] := src[i+63:i] FI ENDFOR dst[MAX:512] := 0 - -
immintrin.h
- - - Mask - AVX512BW - Miscellaneous - - - Unpack and interleave 32 bits from masks "a" and "b", and store the 64-bit result in "k". - -k[31:0] := a[31:0] -k[63:32] := b[31:0] -k[MAX:64] := 0 - - -
immintrin.h
- - - Mask - AVX512BW - Miscellaneous - - - Unpack and interleave 16 bits from masks "a" and "b", and store the 32-bit result in "k". - -k[15:0] := a[15:0] -k[31:16] := b[15:0] -k[MAX:32] := 0 - - +
immintrin.h
 - - Floating Point + + Integer AVX512F - AVX512VL - Arithmetic - - - - - Add packed double-precision (64-bit) floating-point elements in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Swizzle + + + + Load contiguous active 64-bit integers from "a" (those with their respective bit set in mask "k"), and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 3 +m := 0 +FOR j := 0 to 7 i := j*64 IF k[j] - dst[i+63:i] := a[i+63:i] + b[i+63:i] + dst[i+63:i] := a[m+63:m] + m := m + 64 ELSE - dst[i+63:i] := src[i+63:i] + dst[i+63:i] := 0 FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point + + Integer AVX512F - AVX512VL - Arithmetic - - - - Add packed double-precision (64-bit) floating-point elements in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Load + Swizzle + + + + Load contiguous active 64-bit integers from unaligned memory at "mem_addr" (those with their respective bit set in mask "k"), and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 3 +m := 0 +FOR j := 0 to 7 i := j*64 IF k[j] - dst[i+63:i] := a[i+63:i] + b[i+63:i] + dst[i+63:i] := MEM[mem_addr+m+63:mem_addr+m] + m := m + 64 ELSE dst[i+63:i] := 0 FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point + + Integer AVX512F - AVX512VL - Arithmetic - - - - - Add packed double-precision (64-bit) floating-point elements in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Load + + + + + Gather 64-bit integers from memory using 32-bit indices. 64-bit elements are loaded from addresses starting at "base_addr" and offset by each 32-bit element in "vindex" (each index is scaled by the factor in "scale"). Gathered elements are merged into "dst". "scale" should be 1, 2, 4 or 8. -FOR j := 0 to 1 +FOR j := 0 to 7 i := j*64 - IF k[j] - dst[i+63:i] := a[i+63:i] + b[i+63:i] - ELSE - dst[i+63:i] := src[i+63:i] - FI + m := j*32 + addr := base_addr + SignExtend64(vindex[m+31:m]) * ZeroExtend64(scale) * 8 + dst[i+63:i] := MEM[addr+63:addr] ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point + + Integer AVX512F - AVX512VL - Arithmetic - - - - Add packed double-precision (64-bit) floating-point elements in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Load + + + + + + + Gather 64-bit integers from memory using 32-bit indices. 64-bit elements are loaded from addresses starting at "base_addr" and offset by each 32-bit element in "vindex" (each index is scaled by the factor in "scale"). Gathered elements are merged into "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). "scale" should be 1, 2, 4 or 8. -FOR j := 0 to 1 +FOR j := 0 to 7 i := j*64 + m := j*32 IF k[j] - dst[i+63:i] := a[i+63:i] + b[i+63:i] + addr := base_addr + SignExtend64(vindex[m+31:m]) * ZeroExtend64(scale) * 8 + dst[i+63:i] := MEM[addr+63:addr] ELSE - dst[i+63:i] := 0 + dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point + + Integer AVX512F - AVX512VL - Arithmetic - - - - - Add packed single-precision (32-bit) floating-point elements in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Load + + + + + Gather 32-bit integers from memory using 64-bit indices. 32-bit elements are loaded from addresses starting at "base_addr" and offset by each 64-bit element in "vindex" (each index is scaled by the factor in "scale"). Gathered elements are merged into "dst". "scale" should be 1, 2, 4 or 8. FOR j := 0 to 7 i := j*32 - IF k[j] - dst[i+31:i] := a[i+31:i] + b[i+31:i] - ELSE - dst[i+31:i] := src[i+31:i] - FI + m := j*64 + addr := base_addr + vindex[m+63:m] * ZeroExtend64(scale) * 8 + dst[i+31:i] := MEM[addr+31:addr] ENDFOR dst[MAX:256] := 0 - +
immintrin.h
 - - Floating Point + + Integer AVX512F - AVX512VL - Arithmetic - - - - Add packed single-precision (32-bit) floating-point elements in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Load + + + + + + + Gather 32-bit integers from memory using 64-bit indices. 32-bit elements are loaded from addresses starting at "base_addr" and offset by each 64-bit element in "vindex" (each index is scaled by the factor in "scale"). Gathered elements are merged into "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). "scale" should be 1, 2, 4 or 8. FOR j := 0 to 7 i := j*32 + m := j*64 IF k[j] - dst[i+31:i] := a[i+31:i] + b[i+31:i] + addr := base_addr + vindex[m+63:m] * ZeroExtend64(scale) * 8 + dst[i+31:i] := MEM[addr+31:addr] ELSE - dst[i+31:i] := 0 + dst[i+31:i] := src[i+31:i] FI ENDFOR dst[MAX:256] := 0 - +
immintrin.h
 - - Floating Point + + Integer AVX512F - AVX512VL - Arithmetic - - - - - Add packed single-precision (32-bit) floating-point elements in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Load + + + + + Gather 64-bit integers from memory using 64-bit indices. 64-bit elements are loaded from addresses starting at "base_addr" and offset by each 64-bit element in "vindex" (each index is scaled by the factor in "scale"). Gathered elements are merged into "dst". "scale" should be 1, 2, 4 or 8. -FOR j := 0 to 3 - i := j*32 - IF k[j] - dst[i+31:i] := a[i+31:i] + b[i+31:i] - ELSE - dst[i+31:i] := src[i+31:i] - FI +FOR j := 0 to 7 + i := j*64 + m := j*64 + addr := base_addr + vindex[m+63:m] * ZeroExtend64(scale) * 8 + dst[i+63:i] := MEM[addr+63:addr] ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point + + Integer AVX512F - AVX512VL - Arithmetic - - - - Add packed single-precision (32-bit) floating-point elements in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Load + + + + + + + Gather 64-bit integers from memory using 64-bit indices. 64-bit elements are loaded from addresses starting at "base_addr" and offset by each 64-bit element in "vindex" (each index is scaled by the factor in "scale"). Gathered elements are merged into "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). "scale" should be 1, 2, 4 or 8. -FOR j := 0 to 3 - i := j*32 +FOR j := 0 to 7 + i := j*64 + m := j*64 IF k[j] - dst[i+31:i] := a[i+31:i] + b[i+31:i] + addr := base_addr + vindex[m+63:m] * ZeroExtend64(scale) * 8 + dst[i+63:i] := MEM[addr+63:addr] ELSE - dst[i+31:i] := 0 + dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer AVX512F - AVX512VL - Miscellaneous - - - - Concatenate "a" and "b" into a 64-byte immediate result, shift the result right by "count" 32-bit elements, and store the low 32 bytes (8 elements) in "dst". + Special Math Functions + + + + + Compare packed signed 32-bit integers in "a" and "b", and store packed maximum values in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -temp[511:256] := a[255:0] -temp[255:0] := b[255:0] -temp[511:0] := temp[511:0] >> (32*count) -dst[255:0] := temp[255:0] -dst[MAX:256] := 0 +FOR j := 0 to 15 + i := j*32 + IF k[j] + dst[i+31:i] := MAX(a[i+31:i], b[i+31:i]) + ELSE + dst[i+31:i] := 0 + FI +ENDFOR +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer AVX512F - AVX512VL - Miscellaneous - - - - - - Concatenate "a" and "b" into a 64-byte immediate result, shift the result right by "count" 32-bit elements, and store the low 32 bytes (8 elements) in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Special Math Functions + + + + + + Compare packed signed 64-bit integers in "a" and "b", and store packed maximum values in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -temp[511:256] := a[255:0] -temp[255:0] := b[255:0] -temp[511:0] := temp[511:0] >> (32*count) FOR j := 0 to 7 - i := j*32 + i := j*64 IF k[j] - dst[i+31:i] := temp[i+31:i] + dst[i+63:i] := MAX(a[i+63:i], b[i+63:i]) ELSE - dst[i+31:i] := src[i+31:i] + dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer AVX512F - AVX512VL - Miscellaneous - - - - - Concatenate "a" and "b" into a 64-byte immediate result, shift the result right by "count" 32-bit elements, and store the low 32 bytes (8 elements) in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Special Math Functions + + + + + Compare packed signed 64-bit integers in "a" and "b", and store packed maximum values in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -temp[511:256] := a[255:0] -temp[255:0] := b[255:0] -temp[511:0] := temp[511:0] >> (32*count) FOR j := 0 to 7 - i := j*32 + i := j*64 IF k[j] - dst[i+31:i] := temp[i+31:i] + dst[i+63:i] := MAX(a[i+63:i], b[i+63:i]) ELSE - dst[i+31:i] := 0 + dst[i+63:i] := 0 FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer AVX512F - AVX512VL - Miscellaneous - - - - Concatenate "a" and "b" into a 32-byte immediate result, shift the result right by "count" 32-bit elements, and store the low 16 bytes (4 elements) in "dst". + Special Math Functions + + + + Compare packed signed 64-bit integers in "a" and "b", and store packed maximum values in "dst". -temp[255:128] := a[127:0] -temp[127:0] := b[127:0] -temp[255:0] := temp[255:0] >> (32*count) -dst[127:0] := temp[127:0] -dst[MAX:128] := 0 +FOR j := 0 to 7 + i := j*64 + dst[i+63:i] := MAX(a[i+63:i], b[i+63:i]) +ENDFOR +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer AVX512F - AVX512VL - Miscellaneous - - - - - - Concatenate "a" and "b" into a 32-byte immediate result, shift the result right by "count" 32-bit elements, and store the low 16 bytes (4 elements) in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Special Math Functions + + + + + Compare packed unsigned 32-bit integers in "a" and "b", and store packed maximum values in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -temp[255:128] := a[127:0] -temp[127:0] := b[127:0] -temp[255:0] := temp[255:0] >> (32*count) -FOR j := 0 to 3 +FOR j := 0 to 15 i := j*32 IF k[j] - dst[i+31:i] := temp[i+31:i] + dst[i+31:i] := MAX(a[i+31:i], b[i+31:i]) ELSE - dst[i+31:i] := src[i+31:i] + dst[i+31:i] := 0 FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer AVX512F - AVX512VL - Miscellaneous - - - - - Concatenate "a" and "b" into a 32-byte immediate result, shift the result right by "count" 32-bit elements, and store the low 16 bytes (4 elements) in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Special Math Functions + + + + + + Compare packed unsigned 64-bit integers in "a" and "b", and store packed maximum values in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -temp[255:128] := a[127:0] -temp[127:0] := b[127:0] -temp[255:0] := temp[255:0] >> (32*count) -FOR j := 0 to 3 - i := j*32 +FOR j := 0 to 7 + i := j*64 IF k[j] - dst[i+31:i] := temp[i+31:i] + dst[i+63:i] := MAX(a[i+63:i], b[i+63:i]) ELSE - dst[i+31:i] := 0 + dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:128] := 0 - - -
immintrin.h
- - - Integer - AVX512F - AVX512VL - Miscellaneous - - - - Concatenate "a" and "b" into a 64-byte immediate result, shift the result right by "count" 64-bit elements, and store the low 32 bytes (4 elements) in "dst". - -temp[511:256] := a[255:0] -temp[255:0] := b[255:0] -temp[511:0] := temp[511:0] >> (64*count) -dst[255:0] := temp[255:0] -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer AVX512F - AVX512VL - Miscellaneous - - - - - - Concatenate "a" and "b" into a 64-byte immediate result, shift the result right by "count" 64-bit elements, and store the low 32 bytes (4 elements) in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Special Math Functions + + + + + Compare packed unsigned 64-bit integers in "a" and "b", and store packed maximum values in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -temp[511:256] := a[255:0] -temp[255:0] := b[255:0] -temp[511:0] := temp[511:0] >> (64*count) -FOR j := 0 to 3 +FOR j := 0 to 7 i := j*64 IF k[j] - dst[i+63:i] := temp[i+63:i] + dst[i+63:i] := MAX(a[i+63:i], b[i+63:i]) ELSE - dst[i+63:i] := src[i+63:i] + dst[i+63:i] := 0 FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer AVX512F - AVX512VL - Miscellaneous - - - - - Concatenate "a" and "b" into a 64-byte immediate result, shift the result right by "count" 64-bit elements, and store the low 32 bytes (4 elements) in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Special Math Functions + + + + Compare packed unsigned 64-bit integers in "a" and "b", and store packed maximum values in "dst". -temp[511:256] := a[255:0] -temp[255:0] := b[255:0] -temp[511:0] := temp[511:0] >> (64*count) -FOR j := 0 to 3 +FOR j := 0 to 7 i := j*64 - IF k[j] - dst[i+63:i] := temp[i+63:i] - ELSE - dst[i+63:i] := 0 - FI + dst[i+63:i] := MAX(a[i+63:i], b[i+63:i]) ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer AVX512F - AVX512VL - Miscellaneous - - - - Concatenate "a" and "b" into a 32-byte immediate result, shift the result right by "count" 64-bit elements, and store the low 16 bytes (2 elements) in "dst". + Special Math Functions + + + + + Compare packed signed 32-bit integers in "a" and "b", and store packed minimum values in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -temp[255:128] := a[127:0] -temp[127:0] := b[127:0] -temp[255:0] := temp[255:0] >> (64*count) -dst[127:0] := temp[127:0] -dst[MAX:128] := 0 +FOR j := 0 to 15 + i := j*32 + IF k[j] + dst[i+31:i] := MIN(a[i+31:i], b[i+31:i]) + ELSE + dst[i+31:i] := 0 + FI +ENDFOR +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer AVX512F - AVX512VL - Miscellaneous - - - - - - Concatenate "a" and "b" into a 32-byte immediate result, shift the result right by "count" 64-bit elements, and store the low 16 bytes (2 elements) in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Special Math Functions + + + + + + Compare packed signed 64-bit integers in "a" and "b", and store packed minimum values in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -temp[255:128] := a[127:0] -temp[127:0] := b[127:0] -temp[255:0] := temp[255:0] >> (64*count) -FOR j := 0 to 1 +FOR j := 0 to 7 i := j*64 IF k[j] - dst[i+63:i] := temp[i+63:i] + dst[i+63:i] := MIN(a[i+63:i], b[i+63:i]) ELSE dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer AVX512F - AVX512VL - Miscellaneous - - - - - Concatenate "a" and "b" into a 32-byte immediate result, shift the result right by "count" 64-bit elements, and store the low 16 bytes (2 elements) in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Special Math Functions + + + + + Compare packed signed 64-bit integers in "a" and "b", and store packed minimum values in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -temp[255:128] := a[127:0] -temp[127:0] := b[127:0] -temp[255:0] := temp[255:0] >> (64*count) -FOR j := 0 to 1 +FOR j := 0 to 7 i := j*64 IF k[j] - dst[i+63:i] := temp[i+63:i] + dst[i+63:i] := MIN(a[i+63:i], b[i+63:i]) ELSE dst[i+63:i] := 0 FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point - AVX512VL - AVX512DQ - Logical - - - - - Compute the bitwise NOT of packed double-precision (64-bit) floating-point elements in "a" and then AND with "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Integer + AVX512F + Special Math Functions + + + + Compare packed signed 64-bit integers in "a" and "b", and store packed minimum values in "dst". -FOR j := 0 to 3 +FOR j := 0 to 7 i := j*64 - IF k[j] - dst[i+63:i] := ((NOT a[i+63:i]) AND b[i+63:i]) - ELSE - dst[i+63:i] := src[i+63:i] - FI + dst[i+63:i] := MIN(a[i+63:i], b[i+63:i]) ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point - AVX512VL - AVX512DQ - Logical - - - - Compute the bitwise NOT of packed double-precision (64-bit) floating-point elements in "a" and then AND with "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Integer + AVX512F + Special Math Functions + + + + + Compare packed unsigned 32-bit integers in "a" and "b", and store packed minimum values in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*64 +FOR j := 0 to 15 + i := j*32 IF k[j] - dst[i+63:i] := ((NOT a[i+63:i]) AND b[i+63:i]) + dst[i+31:i] := MIN(a[i+31:i], b[i+31:i]) ELSE - dst[i+63:i] := 0 + dst[i+31:i] := 0 FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point - AVX512DQ - Logical - - - Compute the bitwise NOT of packed double-precision (64-bit) floating-point elements in "a" and then AND with "b", and store the results in "dst". + + Integer + AVX512F + Special Math Functions + + + + + + Compare packed unsigned 64-bit integers in "a" and "b", and store packed minimum values in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). FOR j := 0 to 7 i := j*64 - dst[i+63:i] := ((NOT a[i+63:i]) AND b[i+63:i]) + IF k[j] + dst[i+63:i] := MIN(a[i+63:i], b[i+63:i]) + ELSE + dst[i+63:i] := src[i+63:i] + FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point - AVX512DQ - Logical - - - - - Compute the bitwise NOT of packed double-precision (64-bit) floating-point elements in "a" and then AND with "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Integer + AVX512F + Special Math Functions + + + + + Compare packed unsigned 64-bit integers in "a" and "b", and store packed minimum values in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). FOR j := 0 to 7 i := j*64 IF k[j] - dst[i+63:i] := ((NOT a[i+63:i]) AND b[i+63:i]) + dst[i+63:i] := MIN(a[i+63:i], b[i+63:i]) ELSE - dst[i+63:i] := src[i+63:i] + dst[i+63:i] := 0 FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point - AVX512DQ - Logical - - - - Compute the bitwise NOT of packed double-precision (64-bit) floating-point elements in "a" and then AND with "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Integer + AVX512F + Special Math Functions + + + + Compare packed unsigned 64-bit integers in "a" and "b", and store packed minimum values in "dst". FOR j := 0 to 7 i := j*64 - IF k[j] - dst[i+63:i] := ((NOT a[i+63:i]) AND b[i+63:i]) - ELSE - dst[i+63:i] := 0 - FI + dst[i+63:i] := MIN(a[i+63:i], b[i+63:i]) ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point - AVX512VL - AVX512DQ - Logical - - - - - Compute the bitwise NOT of packed double-precision (64-bit) floating-point elements in "a" and then AND with "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Integer + AVX512F + Convert + + + Convert packed 32-bit integers in "a" to packed 8-bit integers with truncation, and store the results in "dst". -FOR j := 0 to 1 - i := j*64 - IF k[j] - dst[i+63:i] := ((NOT a[i+63:i]) AND b[i+63:i]) - ELSE - dst[i+63:i] := src[i+63:i] - FI +FOR j := 0 to 15 + i := 32*j + k := 8*j + dst[k+7:k] := Truncate8(a[i+31:i]) ENDFOR dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point - AVX512VL - AVX512DQ - Logical - - - - Compute the bitwise NOT of packed double-precision (64-bit) floating-point elements in "a" and then AND with "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Integer + AVX512F + Convert + + + + + Convert packed 32-bit integers in "a" to packed 8-bit integers with truncation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 1 - i := j*64 +FOR j := 0 to 15 + i := 32*j + l := 8*j IF k[j] - dst[i+63:i] := ((NOT a[i+63:i]) AND b[i+63:i]) + dst[l+7:l] := Truncate8(a[i+31:i]) ELSE - dst[i+63:i] := 0 + dst[l+7:l] := src[l+7:l] FI ENDFOR dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point - AVX512VL - AVX512DQ - Logical - - - - - Compute the bitwise NOT of packed single-precision (32-bit) floating-point elements in "a" and then AND with "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Integer + AVX512F + Convert + Store + + + + + Convert packed 32-bit integers in "a" to packed 8-bit integers with truncation, and store the active results (those with their respective bit set in writemask "k") to unaligned memory at "base_addr". -FOR j := 0 to 7 - i := j*32 +FOR j := 0 to 15 + i := 32*j + l := 8*j IF k[j] - dst[i+31:i] := ((NOT a[i+31:i]) AND b[i+31:i]) - ELSE - dst[i+31:i] := src[i+31:i] + MEM[base_addr+l+7:base_addr+l] := Truncate8(a[i+31:i]) FI ENDFOR -dst[MAX:256] := 0 - +
immintrin.h
 - - Floating Point - AVX512VL - AVX512DQ - Logical - - - - Compute the bitwise NOT of packed single-precision (32-bit) floating-point elements in "a" and then AND with "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Integer + AVX512F + Convert + + + + Convert packed 32-bit integers in "a" to packed 8-bit integers with truncation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := j*32 +FOR j := 0 to 15 + i := 32*j + l := 8*j IF k[j] - dst[i+31:i] := ((NOT a[i+31:i]) AND b[i+31:i]) + dst[l+7:l] := Truncate8(a[i+31:i]) ELSE - dst[i+31:i] := 0 + dst[l+7:l] := 0 FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point - AVX512DQ - Logical - - - Compute the bitwise NOT of packed single-precision (32-bit) floating-point elements in "a" and then AND with "b", and store the results in "dst". + + Integer + AVX512F + Convert + + + Convert packed 32-bit integers in "a" to packed 16-bit integers with truncation, and store the results in "dst". FOR j := 0 to 15 - i := j*32 - dst[i+31:i] := ((NOT a[i+31:i]) AND b[i+31:i]) + i := 32*j + k := 16*j + dst[k+15:k] := Truncate16(a[i+31:i]) ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - - Floating Point - AVX512DQ - Logical - - - - - Compute the bitwise NOT of packed single-precision (32-bit) floating-point elements in "a" and then AND with "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Integer + AVX512F + Convert + + + + + Convert packed 32-bit integers in "a" to packed 16-bit integers with truncation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). FOR j := 0 to 15 - i := j*32 + i := 32*j + l := 16*j IF k[j] - dst[i+31:i] := ((NOT a[i+31:i]) AND b[i+31:i]) + dst[l+15:l] := Truncate16(a[i+31:i]) ELSE - dst[i+31:i] := src[i+31:i] + dst[l+15:l] := src[l+15:l] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - - Floating Point - AVX512DQ - Logical - - - - Compute the bitwise NOT of packed single-precision (32-bit) floating-point elements in "a" and then AND with "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Integer + AVX512F + Convert + Store + + + + + Convert packed 32-bit integers in "a" to packed 16-bit integers with truncation, and store the active results (those with their respective bit set in writemask "k") to unaligned memory at "base_addr". FOR j := 0 to 15 - i := j*32 + i := 32*j + l := 16*j IF k[j] - dst[i+31:i] := ((NOT a[i+31:i]) AND b[i+31:i]) - ELSE - dst[i+31:i] := 0 + MEM[base_addr+l+15:base_addr+l] := Truncate16(a[i+31:i]) FI ENDFOR -dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point - AVX512VL - AVX512DQ - Logical - - - - - Compute the bitwise NOT of packed single-precision (32-bit) floating-point elements in "a" and then AND with "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Integer + AVX512F + Convert + + + + Convert packed 32-bit integers in "a" to packed 16-bit integers with truncation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*32 +FOR j := 0 to 15 + i := 32*j + l := 16*j IF k[j] - dst[i+31:i] := ((NOT a[i+31:i]) AND b[i+31:i]) + dst[l+15:l] := Truncate16(a[i+31:i]) ELSE - dst[i+31:i] := src[i+31:i] + dst[l+15:l] := 0 FI ENDFOR +dst[MAX:256] := 0 + + +
immintrin.h
+ + + Integer + AVX512F + Convert + + + Convert packed 64-bit integers in "a" to packed 8-bit integers with truncation, and store the results in "dst". + +FOR j := 0 to 7 + i := 64*j + k := 8*j + dst[k+7:k] := Truncate8(a[i+63:i]) +ENDFOR dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point - AVX512VL - AVX512DQ - Logical - - - - Compute the bitwise NOT of packed single-precision (32-bit) floating-point elements in "a" and then AND with "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Integer + AVX512F + Convert + + + + + Convert packed 64-bit integers in "a" to packed 8-bit integers with truncation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*32 +FOR j := 0 to 7 + i := 64*j + l := 8*j IF k[j] - dst[i+31:i] := ((NOT a[i+31:i]) AND b[i+31:i]) + dst[l+7:l] := Truncate8(a[i+63:i]) ELSE - dst[i+31:i] := 0 + dst[l+7:l] := src[l+7:l] FI ENDFOR dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point - AVX512VL - AVX512DQ - Logical - - - - - Compute the bitwise AND of packed double-precision (64-bit) floating-point elements in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Integer + AVX512F + Convert + Store + + + + + Convert packed 64-bit integers in "a" to packed 8-bit integers with truncation, and store the active results (those with their respective bit set in writemask "k") to unaligned memory at "base_addr". -FOR j := 0 to 3 - i := j*64 +FOR j := 0 to 7 + i := 64*j + l := 8*j IF k[j] - dst[i+63:i] := (a[i+63:i] AND b[i+63:i]) - ELSE - dst[i+63:i] := src[i+63:i] + MEM[base_addr+l+7:base_addr+l] := Truncate8(a[i+63:i]) FI ENDFOR -dst[MAX:256] := 0 - +
immintrin.h
 - - Floating Point - AVX512VL - AVX512DQ - Logical - - - - Compute the bitwise AND of packed double-precision (64-bit) floating-point elements in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Integer + AVX512F + Convert + + + + Convert packed 64-bit integers in "a" to packed 8-bit integers with truncation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*64 +FOR j := 0 to 7 + i := 64*j + l := 8*j IF k[j] - dst[i+63:i] := (a[i+63:i] AND b[i+63:i]) + dst[l+7:l] := Truncate8(a[i+63:i]) ELSE - dst[i+63:i] := 0 + dst[l+7:l] := 0 FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point - AVX512DQ - Logical - - - Compute the bitwise AND of packed double-precision (64-bit) floating-point elements in "a" and "b", and store the results in "dst". + + Integer + AVX512F + Convert + + + Convert packed 64-bit integers in "a" to packed 32-bit integers with truncation, and store the results in "dst". FOR j := 0 to 7 - i := j*64 - dst[i+63:i] := (a[i+63:i] AND b[i+63:i]) + i := 64*j + k := 32*j + dst[k+31:k] := Truncate32(a[i+63:i]) ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - - Floating Point - AVX512DQ - Logical - - - - - Compute the bitwise AND of packed double-precision (64-bit) floating-point elements in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Integer + AVX512F + Convert + + + + + Convert packed 64-bit integers in "a" to packed 32-bit integers with truncation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). FOR j := 0 to 7 - i := j*64 + i := 64*j + l := 32*j IF k[j] - dst[i+63:i] := (a[i+63:i] AND b[i+63:i]) + dst[l+31:l] := Truncate32(a[i+63:i]) ELSE - dst[i+63:i] := src[i+63:i] + dst[l+31:l] := src[l+31:l] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - - Floating Point - AVX512DQ - Logical - - - - Compute the bitwise AND of packed double-precision (64-bit) floating-point elements in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Integer + AVX512F + Convert + Store + + + + + Convert packed 64-bit integers in "a" to packed 32-bit integers with truncation, and store the active results (those with their respective bit set in writemask "k") to unaligned memory at "base_addr". FOR j := 0 to 7 - i := j*64 + i := 64*j + l := 32*j IF k[j] - dst[i+63:i] := (a[i+63:i] AND b[i+63:i]) - ELSE - dst[i+63:i] := 0 + MEM[base_addr+l+31:base_addr+l] := Truncate32(a[i+63:i]) FI ENDFOR -dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point - AVX512VL - AVX512DQ - Logical - - - - - Compute the bitwise AND of packed double-precision (64-bit) floating-point elements in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Integer + AVX512F + Convert + + + + Convert packed 64-bit integers in "a" to packed 32-bit integers with truncation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 1 - i := j*64 +FOR j := 0 to 7 + i := 64*j + l := 32*j IF k[j] - dst[i+63:i] := (a[i+63:i] AND b[i+63:i]) + dst[l+31:l] := Truncate32(a[i+63:i]) ELSE - dst[i+63:i] := src[i+63:i] + dst[l+31:l] := 0 FI ENDFOR +dst[MAX:256] := 0 + + +
immintrin.h
+ + + Integer + AVX512F + Convert + + + Convert packed 64-bit integers in "a" to packed 16-bit integers with truncation, and store the results in "dst". + +FOR j := 0 to 7 + i := 64*j + k := 16*j + dst[k+15:k] := Truncate16(a[i+63:i]) +ENDFOR dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point - AVX512VL - AVX512DQ - Logical - - - - Compute the bitwise AND of packed double-precision (64-bit) floating-point elements in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Integer + AVX512F + Convert + + + + + Convert packed 64-bit integers in "a" to packed 16-bit integers with truncation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 1 - i := j*64 +FOR j := 0 to 7 + i := 64*j + l := 16*j IF k[j] - dst[i+63:i] := (a[i+63:i] AND b[i+63:i]) + dst[l+15:l] := Truncate16(a[i+63:i]) ELSE - dst[i+63:i] := 0 + dst[l+15:l] := src[l+15:l] FI ENDFOR dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point - AVX512VL - AVX512DQ - Logical - - - - - Compute the bitwise AND of packed single-precision (32-bit) floating-point elements in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Integer + AVX512F + Convert + Store + + + + + Convert packed 64-bit integers in "a" to packed 16-bit integers with truncation, and store the active results (those with their respective bit set in writemask "k") to unaligned memory at "base_addr". FOR j := 0 to 7 - i := j*32 + i := 64*j + l := 16*j IF k[j] - dst[i+31:i] := (a[i+31:i] AND b[i+31:i]) - ELSE - dst[i+31:i] := src[i+31:i] + MEM[base_addr+l+15:base_addr+l] := Truncate16(a[i+63:i]) FI ENDFOR -dst[MAX:256] := 0 - +
immintrin.h
 - - Floating Point - AVX512VL - AVX512DQ - Logical - - - - Compute the bitwise AND of packed single-precision (32-bit) floating-point elements in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Integer + AVX512F + Convert + + + + Convert packed 64-bit integers in "a" to packed 16-bit integers with truncation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). FOR j := 0 to 7 - i := j*32 + i := 64*j + l := 16*j IF k[j] - dst[i+31:i] := (a[i+31:i] AND b[i+31:i]) + dst[l+15:l] := Truncate16(a[i+63:i]) ELSE - dst[i+31:i] := 0 + dst[l+15:l] := 0 FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point - AVX512DQ - Logical - - - Compute the bitwise AND of packed single-precision (32-bit) floating-point elements in "a" and "b", and store the results in "dst". + + Integer + AVX512F + Convert + + + Convert packed signed 32-bit integers in "a" to packed 8-bit integers with signed saturation, and store the results in "dst". FOR j := 0 to 15 - i := j*32 - dst[i+31:i] := (a[i+31:i] AND b[i+31:i]) + i := 32*j + k := 8*j + dst[k+7:k] := Saturate8(a[i+31:i]) ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point - AVX512DQ - Logical - - - - - Compute the bitwise AND of packed single-precision (32-bit) floating-point elements in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Integer + AVX512F + Convert + + + + + Convert packed signed 32-bit integers in "a" to packed 8-bit integers with signed saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). FOR j := 0 to 15 - i := j*32 + i := 32*j + l := 8*j IF k[j] - dst[i+31:i] := (a[i+31:i] AND b[i+31:i]) + dst[l+7:l] := Saturate8(a[i+31:i]) ELSE - dst[i+31:i] := src[i+31:i] + dst[l+7:l] := src[l+7:l] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point - AVX512DQ - Logical - - - - Compute the bitwise AND of packed single-precision (32-bit) floating-point elements in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Integer + AVX512F + Convert + Store + + + + + Convert packed signed 32-bit integers in "a" to packed 8-bit integers with signed saturation, and store the active results (those with their respective bit set in writemask "k") to unaligned memory at "base_addr". FOR j := 0 to 15 - i := j*32 + i := 32*j + l := 8*j IF k[j] - dst[i+31:i] := (a[i+31:i] AND b[i+31:i]) - ELSE - dst[i+31:i] := 0 + MEM[base_addr+l+7:base_addr+l] := Saturate8(a[i+31:i]) FI ENDFOR -dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point - AVX512VL - AVX512DQ - Logical - - - - - Compute the bitwise AND of packed single-precision (32-bit) floating-point elements in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Integer + AVX512F + Convert + + + + Convert packed signed 32-bit integers in "a" to packed 8-bit integers with signed saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*32 +FOR j := 0 to 15 + i := 32*j + l := 8*j IF k[j] - dst[i+31:i] := (a[i+31:i] AND b[i+31:i]) + dst[l+7:l] := Saturate8(a[i+31:i]) ELSE - dst[i+31:i] := src[i+31:i] + dst[l+7:l] := 0 FI ENDFOR dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point - AVX512VL - AVX512DQ - Logical - - - - Compute the bitwise AND of packed single-precision (32-bit) floating-point elements in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Integer + AVX512F + Convert + + + Convert packed signed 32-bit integers in "a" to packed 16-bit integers with signed saturation, and store the results in "dst". -FOR j := 0 to 3 - i := j*32 - IF k[j] - dst[i+31:i] := (a[i+31:i] AND b[i+31:i]) - ELSE - dst[i+31:i] := 0 - FI +FOR j := 0 to 15 + i := 32*j + k := 16*j + dst[k+15:k] := Saturate16(a[i+31:i]) ENDFOR -dst[MAX:128] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - - Floating Point - AVX512VL + + Integer AVX512F - Miscellaneous - - - - Blend packed double-precision (64-bit) floating-point elements from "a" and "b" using control mask "k", and store the results in "dst". + Convert + + + + + Convert packed signed 32-bit integers in "a" to packed 16-bit integers with signed saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*64 +FOR j := 0 to 15 + i := 32*j + l := 16*j IF k[j] - dst[i+63:i] := b[i+63:i] + dst[l+15:l] := Saturate16(a[i+31:i]) ELSE - dst[i+63:i] := a[i+63:i] + dst[l+15:l] := src[l+15:l] FI ENDFOR dst[MAX:256] := 0 - +
immintrin.h
 - - Floating Point - AVX512VL + + Integer AVX512F - Miscellaneous - - - - Blend packed double-precision (64-bit) floating-point elements from "a" and "b" using control mask "k", and store the results in "dst". + Convert + Store + + + + + Convert packed signed 32-bit integers in "a" to packed 16-bit integers with signed saturation, and store the active results (those with their respective bit set in writemask "k") to unaligned memory at "base_addr". -FOR j := 0 to 1 - i := j*64 +FOR j := 0 to 15 + i := 32*j + l := 16*j IF k[j] - dst[i+63:i] := b[i+63:i] - ELSE - dst[i+63:i] := a[i+63:i] + MEM[base_addr+l+15:base_addr+l] := Saturate16(a[i+31:i]) FI ENDFOR -dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point - AVX512VL + + Integer AVX512F - Miscellaneous - - - - Blend packed single-precision (32-bit) floating-point elements from "a" and "b" using control mask "k", and store the results in "dst". + Convert + + + + Convert packed signed 32-bit integers in "a" to packed 16-bit integers with signed saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := j*32 +FOR j := 0 to 15 + i := 32*j + l := 16*j IF k[j] - dst[i+31:i] := b[i+31:i] + dst[l+15:l] := Saturate16(a[i+31:i]) ELSE - dst[i+31:i] := a[i+31:i] + dst[l+15:l] := 0 FI ENDFOR dst[MAX:256] := 0 - +
immintrin.h
 - - Floating Point - AVX512VL + + Integer AVX512F - Miscellaneous - - - - Blend packed single-precision (32-bit) floating-point elements from "a" and "b" using control mask "k", and store the results in "dst". + Convert + + + Convert packed signed 64-bit integers in "a" to packed 8-bit integers with signed saturation, and store the results in "dst". -FOR j := 0 to 3 - i := j*32 +FOR j := 0 to 7 + i := 64*j + k := 8*j + dst[k+7:k] := Saturate8(a[i+63:i]) +ENDFOR +dst[MAX:64] := 0 + + +
immintrin.h
+ + + Integer + AVX512F + Convert + + + + + Convert packed signed 64-bit integers in "a" to packed 8-bit integers with signed saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +FOR j := 0 to 7 + i := 64*j + l := 8*j IF k[j] - dst[i+31:i] := b[i+31:i] + dst[l+7:l] := Saturate8(a[i+63:i]) ELSE - dst[i+31:i] := a[i+31:i] + dst[l+7:l] := src[l+7:l] FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:64] := 0 - +
immintrin.h
 - - Floating Point - AVX512VL - AVX512DQ - Miscellaneous - - Broadcast the lower 2 packed single-precision (32-bit) floating-point elements from "a" to all elements of "dst". + + Integer + AVX512F + Convert + Store + + + + + Convert packed signed 64-bit integers in "a" to packed 8-bit integers with signed saturation, and store the active results (those with their respective bit set in writemask "k") to unaligned memory at "base_addr". FOR j := 0 to 7 - i := j*32 - n := (j mod 2)*32 - dst[i+31:i] := a[n+31:n] + i := 64*j + l := 8*j + IF k[j] + MEM[base_addr+l+7:base_addr+l] := Saturate8(a[i+63:i]) + FI ENDFOR -dst[MAX:256] := 0 - +
immintrin.h
 - - Floating Point - AVX512VL - AVX512DQ - Miscellaneous - - - - Broadcast the lower 2 packed single-precision (32-bit) floating-point elements from "a" to all elements of "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Integer + AVX512F + Convert + + + + Convert packed signed 64-bit integers in "a" to packed 8-bit integers with signed saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). FOR j := 0 to 7 - i := j*32 - n := (j mod 2)*32 + i := 64*j + l := 8*j IF k[j] - dst[i+31:i] := a[n+31:n] + dst[l+7:l] := Saturate8(a[i+63:i]) ELSE - dst[i+31:i] := src[i+31:i] + dst[l+7:l] := 0 FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:64] := 0 - +
immintrin.h
 - - Floating Point - AVX512VL - AVX512DQ - Miscellaneous - - - Broadcast the lower 2 packed single-precision (32-bit) floating-point elements from "a" to all elements of "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Integer + AVX512F + Convert + + + Convert packed signed 64-bit integers in "a" to packed 32-bit integers with signed saturation, and store the results in "dst". FOR j := 0 to 7 - i := j*32 - n := (j mod 2)*32 - IF k[j] - dst[i+31:i] := a[n+31:n] - ELSE - dst[i+31:i] := 0 - FI + i := 64*j + k := 32*j + dst[k+31:k] := Saturate32(a[i+63:i]) ENDFOR dst[MAX:256] := 0 - +
immintrin.h
 - - Floating Point - AVX512DQ - Miscellaneous - - Broadcast the lower 2 packed single-precision (32-bit) floating-point elements from "a" to all elements of "dst". + + Integer + AVX512F + Convert + + + + + Convert packed signed 64-bit integers in "a" to packed 32-bit integers with signed saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*32 - n := (j mod 2)*32 - dst[i+31:i] := a[n+31:n] +FOR j := 0 to 7 + i := 64*j + l := 32*j + IF k[j] + dst[l+31:l] := Saturate32(a[i+63:i]) + ELSE + dst[l+31:l] := src[l+31:l] + FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - - Floating Point - AVX512DQ - Miscellaneous - - - - Broadcast the lower 2 packed single-precision (32-bit) floating-point elements from "a" to all elements of "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Integer + AVX512F + Convert + Store + + + + + Convert packed signed 64-bit integers in "a" to packed 32-bit integers with signed saturation, and store the active results (those with their respective bit set in writemask "k") to unaligned memory at "base_addr". -FOR j := 0 to 15 - i := j*32 - n := (j mod 2)*32 +FOR j := 0 to 7 + i := 64*j + l := 32*j IF k[j] - dst[i+31:i] := a[n+31:n] - ELSE - dst[i+31:i] := src[i+31:i] + MEM[base_addr+l+31:base_addr+l] := Saturate32(a[i+63:i]) FI ENDFOR -dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point - AVX512DQ - Miscellaneous - - - Broadcast the lower 2 packed single-precision (32-bit) floating-point elements from "a" to all elements of "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Integer + AVX512F + Convert + + + + Convert packed signed 64-bit integers in "a" to packed 32-bit integers with signed saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*32 - n := (j mod 2)*32 +FOR j := 0 to 7 + i := 64*j + l := 32*j IF k[j] - dst[i+31:i] := a[n+31:n] + dst[l+31:l] := Saturate32(a[i+63:i]) ELSE - dst[i+31:i] := 0 + dst[l+31:l] := 0 FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - - Floating Point - AVX512VL + + Integer AVX512F - Miscellaneous - - Broadcast the 4 packed single-precision (32-bit) floating-point elements from "a" to all elements of "dst". + Convert + + + Convert packed signed 64-bit integers in "a" to packed 16-bit integers with signed saturation, and store the results in "dst". FOR j := 0 to 7 - i := j*32 - n := (j mod 4)*32 - dst[i+31:i] := a[n+31:n] + i := 64*j + k := 16*j + dst[k+15:k] := Saturate16(a[i+63:i]) ENDFOR -dst[MAX:256] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point - AVX512VL + + Integer AVX512F - Miscellaneous - - - - Broadcast the 4 packed single-precision (32-bit) floating-point elements from "a" to all elements of "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Convert + + + + + Convert packed signed 64-bit integers in "a" to packed 16-bit integers with signed saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). FOR j := 0 to 7 - i := j*32 - n := (j mod 4)*32 + i := 64*j + l := 16*j IF k[j] - dst[i+31:i] := a[n+31:n] + dst[l+15:l] := Saturate16(a[i+63:i]) ELSE - dst[i+31:i] := src[i+31:i] + dst[l+15:l] := src[l+15:l] FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point - AVX512VL + + Integer AVX512F - Miscellaneous - - - Broadcast the 4 packed single-precision (32-bit) floating-point elements from "a" to all elements of "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Convert + Store + + + + + Convert packed signed 64-bit integers in "a" to packed 16-bit integers with signed saturation, and store the active results (those with their respective bit set in writemask "k") to unaligned memory at "base_addr". FOR j := 0 to 7 - i := j*32 - n := (j mod 4)*32 + i := 64*j + l := 16*j IF k[j] - dst[i+31:i] := a[n+31:n] + MEM[base_addr+l+15:base_addr+l] := Saturate16(a[i+63:i]) + FI +ENDFOR + + +
immintrin.h
+ + + Integer + AVX512F + Convert + + + + Convert packed signed 64-bit integers in "a" to packed 16-bit integers with signed saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + +FOR j := 0 to 7 + i := 64*j + l := 16*j + IF k[j] + dst[l+15:l] := Saturate16(a[i+63:i]) ELSE - dst[i+31:i] := 0 + dst[l+15:l] := 0 FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point - AVX512DQ - Miscellaneous - - Broadcast the 8 packed single-precision (32-bit) floating-point elements from "a" to all elements of "dst". + + Integer + AVX512F + Convert + + + Sign extend packed 8-bit integers in "a" to packed 32-bit integers, and store the results in "dst". FOR j := 0 to 15 - i := j*32 - n := (j mod 8)*32 - dst[i+31:i] := a[n+31:n] + i := 32*j + k := 8*j + dst[i+31:i] := SignExtend32(a[k+7:k]) ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point - AVX512DQ - Miscellaneous - - - - Broadcast the 8 packed single-precision (32-bit) floating-point elements from "a" to all elements of "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Integer + AVX512F + Convert + + + + + Sign extend packed 8-bit integers in "a" to packed 32-bit integers, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). FOR j := 0 to 15 - i := j*32 - n := (j mod 8)*32 + i := 32*j + l := 8*j IF k[j] - dst[i+31:i] := a[n+31:n] + dst[i+31:i] := SignExtend32(a[l+7:l]) ELSE dst[i+31:i] := src[i+31:i] FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point - AVX512DQ - Miscellaneous - - - Broadcast the 8 packed single-precision (32-bit) floating-point elements from "a" to all elements of "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Integer + AVX512F + Convert + + + + Sign extend packed 8-bit integers in "a" to packed 32-bit integers, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). FOR j := 0 to 15 - i := j*32 - n := (j mod 8)*32 + i := 32*j + l := 8*j IF k[j] - dst[i+31:i] := a[n+31:n] + dst[i+31:i] := SignExtend32(a[l+7:l]) ELSE dst[i+31:i] := 0 FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point - AVX512VL - AVX512DQ - Miscellaneous - - Broadcast the 2 packed double-precision (64-bit) floating-point elements from "a" to all elements of "dst". + + Integer + AVX512F + Convert + + + Sign extend packed 8-bit integers in the low 8 bytes of "a" to packed 64-bit integers, and store the results in "dst". -FOR j := 0 to 3 - i := j*64 - n := (j mod 2)*64 - dst[i+63:i] := a[n+63:n] +FOR j := 0 to 7 + i := 64*j + k := 8*j + dst[i+63:i] := SignExtend64(a[k+7:k]) ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point - AVX512VL - AVX512DQ - Miscellaneous - - - - Broadcast the 2 packed double-precision (64-bit) floating-point elements from "a" to all elements of "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Integer + AVX512F + Convert + + + + + Sign extend packed 8-bit integers in the low 8 bytes of "a" to packed 64-bit integers, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*64 - n := (j mod 2)*64 +FOR j := 0 to 7 + i := 64*j + l := 8*j IF k[j] - dst[i+63:i] := a[n+63:n] + dst[i+63:i] := SignExtend64(a[l+7:l]) ELSE - dst[i+63:i] := src[n+63:n] + dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point - AVX512VL - AVX512DQ - Miscellaneous - - - Broadcast the 2 packed double-precision (64-bit) floating-point elements from "a" to all elements of "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Integer + AVX512F + Convert + + + + Sign extend packed 8-bit integers in the low 8 bytes of "a" to packed 64-bit integers, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*64 - n := (j mod 2)*64 +FOR j := 0 to 7 + i := 64*j + l := 8*j IF k[j] - dst[i+63:i] := a[n+63:n] + dst[i+63:i] := SignExtend64(a[l+7:l]) ELSE dst[i+63:i] := 0 FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point - AVX512DQ - Miscellaneous - - Broadcast the 2 packed double-precision (64-bit) floating-point elements from "a" to all elements of "dst". + + Integer + AVX512F + Convert + + + Sign extend packed 32-bit integers in "a" to packed 64-bit integers, and store the results in "dst". FOR j := 0 to 7 - i := j*64 - n := (j mod 2)*64 - dst[i+63:i] := a[n+63:n] + i := 64*j + k := 32*j + dst[i+63:i] := SignExtend64(a[k+31:k]) ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point - AVX512DQ - Miscellaneous - - - - Broadcast the 2 packed double-precision (64-bit) floating-point elements from "a" to all elements of "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Integer + AVX512F + Convert + + + + + Sign extend packed 32-bit integers in "a" to packed 64-bit integers, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). FOR j := 0 to 7 - i := j*64 - n := (j mod 2)*64 + i := 64*j + l := 32*j IF k[j] - dst[i+63:i] := a[n+63:n] + dst[i+63:i] := SignExtend64(a[l+31:l]) ELSE - dst[i+63:i] := src[n+63:n] + dst[i+63:i] := src[i+63:i] FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point - AVX512DQ - Miscellaneous - - - Broadcast the 2 packed double-precision (64-bit) floating-point elements from "a" to all elements of "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Integer + AVX512F + Convert + + + + Sign extend packed 32-bit integers in "a" to packed 64-bit integers, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). FOR j := 0 to 7 - i := j*64 - n := (j mod 2)*64 + i := 64*j + l := 32*j IF k[j] - dst[i+63:i] := a[n+63:n] + dst[i+63:i] := SignExtend64(a[l+31:l]) ELSE dst[i+63:i] := 0 FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - - AVX512VL - AVX512DQ - Miscellaneous - - Broadcast the lower 2 packed 32-bit integers from "a" to all elements of "dst. + + Integer + AVX512F + Convert + + + Sign extend packed 16-bit integers in "a" to packed 32-bit integers, and store the results in "dst". -FOR j := 0 to 7 - i := j*32 - n := (j mod 2)*32 - dst[i+31:i] := a[n+31:n] +FOR j := 0 to 15 + i := 32*j + k := 16*j + dst[i+31:i] := SignExtend32(a[k+15:k]) ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - AVX512VL - AVX512DQ - Miscellaneous - - - - Broadcast the lower 2 packed 32-bit integers from "a" to all elements of "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Integer + AVX512F + Convert + + + + + Sign extend packed 16-bit integers in "a" to packed 32-bit integers, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 7 +FOR j := 0 to 15 i := j*32 - n := (j mod 2)*32 + l := j*16 IF k[j] - dst[i+31:i] := a[n+31:n] + dst[i+31:i] := SignExtend32(a[l+15:l]) ELSE - dst[i+31:i] := src[n+31:n] + dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - AVX512VL - AVX512DQ - Miscellaneous - - - Broadcast the lower 2 packed 32-bit integers from "a" to all elements of "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Integer + AVX512F + Convert + + + + Sign extend packed 16-bit integers in "a" to packed 32-bit integers, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := j*32 - n := (j mod 2)*32 +FOR j := 0 to 15 + i := 32*j + l := 16*j IF k[j] - dst[i+31:i] := a[n+31:n] + dst[i+31:i] := SignExtend32(a[l+15:l]) ELSE dst[i+31:i] := 0 FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - AVX512DQ - Miscellaneous - - Broadcast the lower 2 packed 32-bit integers from "a" to all elements of "dst. + + Integer + AVX512F + Convert + + + Sign extend packed 16-bit integers in "a" to packed 64-bit integers, and store the results in "dst". -FOR j := 0 to 15 - i := j*32 - n := (j mod 2)*32 - dst[i+31:i] := a[n+31:n] +FOR j := 0 to 7 + i := 64*j + k := 16*j + dst[i+63:i] := SignExtend64(a[k+15:k]) ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - - AVX512DQ - Miscellaneous - - - - Broadcast the lower 2 packed 32-bit integers from "a" to all elements of "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Integer + AVX512F + Convert + + + + + Sign extend packed 16-bit integers in "a" to packed 64-bit integers, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*32 - n := (j mod 2)*32 +FOR j := 0 to 7 + i := 64*j + l := 16*j IF k[j] - dst[i+31:i] := a[n+31:n] + dst[i+63:i] := SignExtend64(a[l+15:l]) ELSE - dst[i+31:i] := src[n+31:n] + dst[i+63:i] := src[i+63:i] FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - - AVX512DQ - Miscellaneous - - - Broadcast the lower 2 packed 32-bit integers from "a" to all elements of "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Integer + AVX512F + Convert + + + + Sign extend packed 16-bit integers in "a" to packed 64-bit integers, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*32 - n := (j mod 2)*32 +FOR j := 0 to 7 + i := 64*j + l := 16*j IF k[j] - dst[i+31:i] := a[n+31:n] + dst[i+63:i] := SignExtend64(a[l+15:l]) ELSE - dst[i+31:i] := 0 + dst[i+63:i] := 0 FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - - AVX512VL - AVX512DQ - Miscellaneous - - Broadcast the lower 2 packed 32-bit integers from "a" to all elements of "dst. + + Integer + AVX512F + Convert + + + Convert packed unsigned 32-bit integers in "a" to packed unsigned 8-bit integers with unsigned saturation, and store the results in "dst". -FOR j := 0 to 3 - i := j*32 - n := (j mod 2)*32 - dst[i+31:i] := a[n+31:n] +FOR j := 0 to 15 + i := 32*j + k := 8*j + dst[k+7:k] := SaturateU8(a[i+31:i]) ENDFOR dst[MAX:128] := 0 - +
immintrin.h
 - - AVX512VL - AVX512DQ - Miscellaneous - - - - Broadcast the lower 2 packed 32-bit integers from "a" to all elements of "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Integer + AVX512F + Convert + + + + + Convert packed unsigned 32-bit integers in "a" to packed unsigned 8-bit integers with unsigned saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*32 - n := (j mod 2)*32 +FOR j := 0 to 15 + i := 32*j + l := 8*j IF k[j] - dst[i+31:i] := a[n+31:n] + dst[l+7:l] := SaturateU8(a[i+31:i]) ELSE - dst[i+31:i] := src[n+31:n] + dst[l+7:l] := src[l+7:l] FI ENDFOR dst[MAX:128] := 0 - +
immintrin.h
 - - AVX512VL - AVX512DQ - Miscellaneous - - - Broadcast the lower 2 packed 32-bit integers from "a" to all elements of "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Integer + AVX512F + Convert + Store + + + + + Convert packed unsigned 32-bit integers in "a" to packed 8-bit integers with unsigned saturation, and store the active results (those with their respective bit set in writemask "k") to unaligned memory at "base_addr". -FOR j := 0 to 3 - i := j*32 - n := (j mod 2)*32 +FOR j := 0 to 15 + i := 32*j + l := 8*j IF k[j] - dst[i+31:i] := a[n+31:n] - ELSE - dst[i+31:i] := 0 + MEM[base_addr+l+7:base_addr+l] := SaturateU8(a[i+31:i]) FI ENDFOR -dst[MAX:128] := 0 - +
immintrin.h
 - - AVX512VL + + Integer AVX512F - Miscellaneous - - Broadcast the 4 packed 32-bit integers from "a" to all elements of "dst". + Convert + + + + Convert packed unsigned 32-bit integers in "a" to packed unsigned 8-bit integers with unsigned saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := j*32 - n := (j mod 4)*32 - dst[i+31:i] := a[n+31:n] +FOR j := 0 to 15 + i := 32*j + l := 8*j + IF k[j] + dst[l+7:l] := SaturateU8(a[i+31:i]) + ELSE + dst[l+7:l] := 0 + FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - - AVX512VL + + Integer AVX512F - Miscellaneous - - - - Broadcast the 4 packed 32-bit integers from "a" to all elements of "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Convert + + + Convert packed unsigned 32-bit integers in "a" to packed unsigned 16-bit integers with unsigned saturation, and store the results in "dst". -FOR j := 0 to 7 - i := j*32 - n := (j mod 4)*32 - IF k[j] - dst[i+31:i] := a[n+31:n] - ELSE - dst[i+31:i] := src[n+31:n] - FI +FOR j := 0 to 15 + i := 32*j + k := 16*j + dst[k+15:k] := SaturateU16(a[i+31:i]) ENDFOR dst[MAX:256] := 0 - +
immintrin.h
 - - AVX512VL + + Integer AVX512F - Miscellaneous - - - Broadcast the 4 packed 32-bit integers from "a" to all elements of "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Convert + + + + + Convert packed unsigned 32-bit integers in "a" to packed unsigned 16-bit integers with unsigned saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := j*32 - n := (j mod 4)*32 +FOR j := 0 to 15 + i := 32*j + l := 16*j IF k[j] - dst[i+31:i] := a[n+31:n] + dst[l+15:l] := SaturateU16(a[i+31:i]) ELSE - dst[i+31:i] := 0 + dst[l+15:l] := src[l+15:l] FI ENDFOR dst[MAX:256] := 0 - +
immintrin.h
 - - AVX512DQ - Miscellaneous - - Broadcast the 8 packed 32-bit integers from "a" to all elements of "dst". + + Integer + AVX512F + Convert + Store + + + + + Convert packed unsigned 32-bit integers in "a" to packed 16-bit integers with unsigned saturation, and store the active results (those with their respective bit set in writemask "k") to unaligned memory at "base_addr". FOR j := 0 to 15 - i := j*32 - n := (j mod 8)*32 - dst[i+31:i] := a[n+31:n] + i := 32*j + l := 16*j + IF k[j] + MEM[base_addr+l+15:base_addr+l] := SaturateU16(a[i+31:i]) + FI ENDFOR -dst[MAX:512] := 0 - +
immintrin.h
 - - AVX512DQ - Miscellaneous - - - - Broadcast the 8 packed 32-bit integers from "a" to all elements of "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Integer + AVX512F + Convert + + + + Convert packed unsigned 32-bit integers in "a" to packed unsigned 16-bit integers with unsigned saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). FOR j := 0 to 15 - i := j*32 - n := (j mod 8)*32 + i := 32*j + l := 16*j IF k[j] - dst[i+31:i] := a[n+31:n] + dst[l+15:l] := SaturateU16(a[i+31:i]) ELSE - dst[i+31:i] := src[n+31:n] + dst[l+15:l] := 0 FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - - AVX512DQ - Miscellaneous - - - Broadcast the 8 packed 32-bit integers from "a" to all elements of "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Integer + AVX512F + Convert + + + Convert packed unsigned 64-bit integers in "a" to packed unsigned 8-bit integers with unsigned saturation, and store the results in "dst". -FOR j := 0 to 15 - i := j*32 - n := (j mod 8)*32 +FOR j := 0 to 7 + i := 64*j + k := 8*j + dst[k+7:k] := SaturateU8(a[i+63:i]) +ENDFOR +dst[MAX:64] := 0 + + +
immintrin.h
+ + + Integer + AVX512F + Convert + + + + + Convert packed unsigned 64-bit integers in "a" to packed unsigned 8-bit integers with unsigned saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +FOR j := 0 to 7 + i := 64*j + l := 8*j IF k[j] - dst[i+31:i] := a[n+31:n] + dst[l+7:l] := SaturateU8(a[i+63:i]) ELSE - dst[i+31:i] := 0 + dst[l+7:l] := src[l+7:l] FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:64] := 0 - +
immintrin.h
 - - AVX512VL - AVX512DQ - Miscellaneous - - Broadcast the 2 packed 64-bit integers from "a" to all elements of "dst". + + Integer + AVX512F + Convert + Store + + + + + Convert packed unsigned 64-bit integers in "a" to packed 8-bit integers with unsigned saturation, and store the active results (those with their respective bit set in writemask "k") to unaligned memory at "base_addr". -FOR j := 0 to 3 - i := j*64 - n := (j mod 2)*64 - dst[i+63:i] := a[n+63:n] +FOR j := 0 to 7 + i := 64*j + l := 8*j + IF k[j] + MEM[base_addr+l+7:base_addr+l] := SaturateU8(a[i+63:i]) + FI ENDFOR -dst[MAX:256] := 0 - +
immintrin.h
 - - AVX512VL - AVX512DQ - Miscellaneous - - - - Broadcast the 2 packed 64-bit integers from "a" to all elements of "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Integer + AVX512F + Convert + + + + Convert packed unsigned 64-bit integers in "a" to packed unsigned 8-bit integers with unsigned saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*64 - n := (j mod 2)*64 +FOR j := 0 to 7 + i := 64*j + l := 8*j IF k[j] - dst[i+63:i] := a[n+63:n] + dst[l+7:l] := SaturateU8(a[i+63:i]) ELSE - dst[i+63:i] := src[n+63:n] + dst[l+7:l] := 0 FI ENDFOR +dst[MAX:64] := 0 + + +
immintrin.h
+ + + Integer + AVX512F + Convert + + + Convert packed unsigned 64-bit integers in "a" to packed unsigned 32-bit integers with unsigned saturation, and store the results in "dst". + +FOR j := 0 to 7 + i := 64*j + k := 32*j + dst[k+31:k] := SaturateU32(a[i+63:i]) +ENDFOR dst[MAX:256] := 0 - +
immintrin.h
 - - AVX512VL - AVX512DQ - Miscellaneous - - - Broadcast the 2 packed 64-bit integers from "a" to all elements of "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Integer + AVX512F + Convert + + + + + Convert packed unsigned 64-bit integers in "a" to packed unsigned 32-bit integers with unsigned saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*64 - n := (j mod 2)*64 +FOR j := 0 to 7 + i := 64*j + l := 32*j IF k[j] - dst[i+63:i] := a[n+63:n] + dst[l+31:l] := SaturateU32(a[i+63:i]) ELSE - dst[i+63:i] := 0 + dst[l+31:l] := src[l+31:l] FI ENDFOR dst[MAX:256] := 0 - +
immintrin.h
 - - AVX512DQ - Miscellaneous - - Broadcast the 2 packed 64-bit integers from "a" to all elements of "dst". + + Integer + AVX512F + Convert + Store + + + + + Convert packed unsigned 64-bit integers in "a" to packed 32-bit integers with unsigned saturation, and store the active results (those with their respective bit set in writemask "k") to unaligned memory at "base_addr". FOR j := 0 to 7 - i := j*64 - n := (j mod 2)*64 - dst[i+63:i] := a[n+63:n] + i := 64*j + l := 32*j + IF k[j] + MEM[base_addr+l+31:base_addr+l] := SaturateU32(a[i+63:i]) + FI ENDFOR -dst[MAX:512] := 0 - +
immintrin.h
 - - AVX512DQ - Miscellaneous - - - - Broadcast the 2 packed 64-bit integers from "a" to all elements of "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Integer + AVX512F + Convert + + + + Convert packed unsigned 64-bit integers in "a" to packed unsigned 32-bit integers with unsigned saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). FOR j := 0 to 7 - i := j*64 - n := (j mod 2)*64 + i := 64*j + l := 32*j IF k[j] - dst[i+63:i] := a[n+63:n] + dst[l+31:l] := SaturateU32(a[i+63:i]) ELSE - dst[i+63:i] := src[n+63:n] + dst[l+31:l] := 0 FI ENDFOR -dst[MAX:512] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - - AVX512DQ - Miscellaneous - - - Broadcast the 2 packed 64-bit integers from "a" to all elements of "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Integer + AVX512F + Convert + + + Convert packed unsigned 64-bit integers in "a" to packed unsigned 16-bit integers with unsigned saturation, and store the results in "dst". FOR j := 0 to 7 - i := j*64 - n := (j mod 2)*64 - IF k[j] - dst[i+63:i] := a[n+63:n] - ELSE - dst[i+63:i] := 0 - FI + i := 64*j + k := 16*j + dst[k+15:k] := SaturateU16(a[i+63:i]) ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - +
immintrin.h
- - - Floating Point - AVX512VL + + + Integer AVX512F - Miscellaneous - - - - Broadcast the low double-precision (64-bit) floating-point element from "a" to all elements of "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Convert + + + + + Convert packed unsigned 64-bit integers in "a" to packed unsigned 16-bit integers with unsigned saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*64 +FOR j := 0 to 7 + i := 64*j + l := 16*j IF k[j] - dst[i+63:i] := a[63:0] + dst[l+15:l] := SaturateU16(a[i+63:i]) ELSE - dst[i+63:i] := src[i+63:i] + dst[l+15:l] := src[l+15:l] FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point - AVX512VL + + Integer AVX512F - Miscellaneous - - - Broadcast the low double-precision (64-bit) floating-point element from "a" to all elements of "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Convert + Store + + + + + Convert packed unsigned 64-bit integers in "a" to packed 16-bit integers with unsigned saturation, and store the active results (those with their respective bit set in writemask "k") to unaligned memory at "base_addr". -FOR j := 0 to 3 - i := j*64 +FOR j := 0 to 7 + i := 64*j + l := 16*j IF k[j] - dst[i+63:i] := a[63:0] - ELSE - dst[i+63:i] := 0 + MEM[base_addr+l+15:base_addr+l] := SaturateU16(a[i+63:i]) FI ENDFOR -dst[MAX:256] := 0 - +
immintrin.h
 - - Floating Point - AVX512VL + + Integer AVX512F - Miscellaneous - - - - Broadcast the low single-precision (32-bit) floating-point element from "a" to all elements of "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Convert + + + + Convert packed unsigned 64-bit integers in "a" to packed unsigned 16-bit integers with unsigned saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). FOR j := 0 to 7 - i := j*32 + i := 64*j + l := 16*j IF k[j] - dst[i+31:i] := a[31:0] + dst[l+15:l] := SaturateU16(a[i+63:i]) ELSE - dst[i+31:i] := src[i+31:i] + dst[l+15:l] := 0 FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point - AVX512VL + + Integer AVX512F - Miscellaneous - - - Broadcast the low single-precision (32-bit) floating-point element from "a" to all elements of "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Convert + + + Zero extend packed unsigned 8-bit integers in "a" to packed 32-bit integers, and store the results in "dst". -FOR j := 0 to 7 - i := j*32 - IF k[j] - dst[i+31:i] := a[31:0] - ELSE - dst[i+31:i] := 0 - FI +FOR j := 0 to 15 + i := 32*j + k := 8*j + dst[i+31:i] := ZeroExtend32(a[k+7:k]) ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point - AVX512VL + + Integer AVX512F - Miscellaneous - - - - Broadcast the low single-precision (32-bit) floating-point element from "a" to all elements of "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Convert + + + + + Zero extend packed unsigned 8-bit integers in "a" to packed 32-bit integers, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*32 +FOR j := 0 to 15 + i := 32*j + l := 8*j IF k[j] - dst[i+31:i] := a[31:0] + dst[i+31:i] := ZeroExtend32(a[l+7:l]) ELSE dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point - AVX512VL + + Integer AVX512F - Miscellaneous - - - Broadcast the low single-precision (32-bit) floating-point element from "a" to all elements of "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Convert + + + + Zero extend packed unsigned 8-bit integers in "a" to packed 32-bit integers, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*32 +FOR j := 0 to 15 + i := 32*j + l := 8*j IF k[j] - dst[i+31:i] := a[31:0] + dst[i+31:i] := ZeroExtend32(a[l+7:l]) ELSE dst[i+31:i] := 0 FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point - Mask - AVX512VL + + Integer AVX512F - Compare - - - - Compare packed double-precision (64-bit) floating-point elements in "a" and "b" based on the comparison operand specified by "imm8", and store the results in mask vector "k". + Convert + + + Zero extend packed unsigned 8-bit integers in the low 8 byte sof "a" to packed 64-bit integers, and store the results in "dst". -CASE (imm8[7:0]) OF -0: OP := _CMP_EQ_OQ -1: OP := _CMP_LT_OS -2: OP := _CMP_LE_OS -3: OP := _CMP_UNORD_Q -4: OP := _CMP_NEQ_UQ -5: OP := _CMP_NLT_US -6: OP := _CMP_NLE_US -7: OP := _CMP_ORD_Q -8: OP := _CMP_EQ_UQ -9: OP := _CMP_NGE_US -10: OP := _CMP_NGT_US -11: OP := _CMP_FALSE_OQ -12: OP := _CMP_NEQ_OQ -13: OP := _CMP_GE_OS -14: OP := _CMP_GT_OS -15: OP := _CMP_TRUE_UQ -16: OP := _CMP_EQ_OS -17: OP := _CMP_LT_OQ -18: OP := _CMP_LE_OQ -19: OP := _CMP_UNORD_S -20: OP := _CMP_NEQ_US -21: OP := _CMP_NLT_UQ -22: OP := _CMP_NLE_UQ -23: OP := _CMP_ORD_S -24: OP := _CMP_EQ_US -25: OP := _CMP_NGE_UQ -26: OP := _CMP_NGT_UQ -27: OP := _CMP_FALSE_OS -28: OP := _CMP_NEQ_OS -29: OP := _CMP_GE_OQ -30: OP := _CMP_GT_OQ -31: OP := _CMP_TRUE_US -ESAC -FOR j := 0 to 3 - i := j*64 - k[j] := (a[i+63:i] OP b[i+63:i]) ? 1 : 0 +FOR j := 0 to 7 + i := 64*j + k := 8*j + dst[i+63:i] := ZeroExtend64(a[k+7:k]) ENDFOR -k[MAX:4] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point - Mask - AVX512VL + + Integer AVX512F - Compare - - - - - Compare packed double-precision (64-bit) floating-point elements in "a" and "b" based on the comparison operand specified by "imm8", and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). + Convert + + + + + Zero extend packed unsigned 8-bit integers in the low 8 bytes of "a" to packed 64-bit integers, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -CASE (imm8[7:0]) OF -0: OP := _CMP_EQ_OQ -1: OP := _CMP_LT_OS -2: OP := _CMP_LE_OS -3: OP := _CMP_UNORD_Q -4: OP := _CMP_NEQ_UQ -5: OP := _CMP_NLT_US -6: OP := _CMP_NLE_US -7: OP := _CMP_ORD_Q -8: OP := _CMP_EQ_UQ -9: OP := _CMP_NGE_US -10: OP := _CMP_NGT_US -11: OP := _CMP_FALSE_OQ -12: OP := _CMP_NEQ_OQ -13: OP := _CMP_GE_OS -14: OP := _CMP_GT_OS -15: OP := _CMP_TRUE_UQ -16: OP := _CMP_EQ_OS -17: OP := _CMP_LT_OQ -18: OP := _CMP_LE_OQ -19: OP := _CMP_UNORD_S -20: OP := _CMP_NEQ_US -21: OP := _CMP_NLT_UQ -22: OP := _CMP_NLE_UQ -23: OP := _CMP_ORD_S -24: OP := _CMP_EQ_US -25: OP := _CMP_NGE_UQ -26: OP := _CMP_NGT_UQ -27: OP := _CMP_FALSE_OS -28: OP := _CMP_NEQ_OS -29: OP := _CMP_GE_OQ -30: OP := _CMP_GT_OQ -31: OP := _CMP_TRUE_US -ESAC -FOR j := 0 to 3 - i := j*64 - IF k1[j] - k[j] := ( a[i+63:i] OP b[i+63:i] ) ? 1 : 0 - ELSE - k[j] := 0 +FOR j := 0 to 7 + i := 64*j + l := 8*j + IF k[j] + dst[i+63:i] := ZeroExtend64(a[l+7:l]) + ELSE + dst[i+63:i] := src[i+63:i] FI ENDFOR -k[MAX:4] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point - Mask - AVX512VL + + Integer AVX512F - Compare - - - - Compare packed double-precision (64-bit) floating-point elements in "a" and "b" based on the comparison operand specified by "imm8", and store the results in mask vector "k". + Convert + + + + Zero extend packed unsigned 8-bit integers in the low 8 bytes of "a" to packed 64-bit integers, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -CASE (imm8[7:0]) OF -0: OP := _CMP_EQ_OQ -1: OP := _CMP_LT_OS -2: OP := _CMP_LE_OS -3: OP := _CMP_UNORD_Q -4: OP := _CMP_NEQ_UQ -5: OP := _CMP_NLT_US -6: OP := _CMP_NLE_US -7: OP := _CMP_ORD_Q -8: OP := _CMP_EQ_UQ -9: OP := _CMP_NGE_US -10: OP := _CMP_NGT_US -11: OP := _CMP_FALSE_OQ -12: OP := _CMP_NEQ_OQ -13: OP := _CMP_GE_OS -14: OP := _CMP_GT_OS -15: OP := _CMP_TRUE_UQ -16: OP := _CMP_EQ_OS -17: OP := _CMP_LT_OQ -18: OP := _CMP_LE_OQ -19: OP := _CMP_UNORD_S -20: OP := _CMP_NEQ_US -21: OP := _CMP_NLT_UQ -22: OP := _CMP_NLE_UQ -23: OP := _CMP_ORD_S -24: OP := _CMP_EQ_US -25: OP := _CMP_NGE_UQ -26: OP := _CMP_NGT_UQ -27: OP := _CMP_FALSE_OS -28: OP := _CMP_NEQ_OS -29: OP := _CMP_GE_OQ -30: OP := _CMP_GT_OQ -31: OP := _CMP_TRUE_US -ESAC -FOR j := 0 to 1 - i := j*64 - k[j] := ( a[i+63:i] OP b[i+63:i] ) ? 1 : 0 +FOR j := 0 to 7 + i := 64*j + l := 8*j + IF k[j] + dst[i+63:i] := ZeroExtend64(a[l+7:l]) + ELSE + dst[i+63:i] := 0 + FI ENDFOR -k[MAX:2] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point - Mask - AVX512VL + + Integer AVX512F - Compare - - - - - Compare packed double-precision (64-bit) floating-point elements in "a" and "b" based on the comparison operand specified by "imm8", and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). + Convert + + + Zero extend packed unsigned 32-bit integers in "a" to packed 64-bit integers, and store the results in "dst". -CASE (imm8[7:0]) OF -0: OP := _CMP_EQ_OQ -1: OP := _CMP_LT_OS -2: OP := _CMP_LE_OS -3: OP := _CMP_UNORD_Q -4: OP := _CMP_NEQ_UQ -5: OP := _CMP_NLT_US -6: OP := _CMP_NLE_US -7: OP := _CMP_ORD_Q -8: OP := _CMP_EQ_UQ -9: OP := _CMP_NGE_US -10: OP := _CMP_NGT_US -11: OP := _CMP_FALSE_OQ -12: OP := _CMP_NEQ_OQ -13: OP := _CMP_GE_OS -14: OP := _CMP_GT_OS -15: OP := _CMP_TRUE_UQ -16: OP := _CMP_EQ_OS -17: OP := _CMP_LT_OQ -18: OP := _CMP_LE_OQ -19: OP := _CMP_UNORD_S -20: OP := _CMP_NEQ_US -21: OP := _CMP_NLT_UQ -22: OP := _CMP_NLE_UQ -23: OP := _CMP_ORD_S -24: OP := _CMP_EQ_US -25: OP := _CMP_NGE_UQ -26: OP := _CMP_NGT_UQ -27: OP := _CMP_FALSE_OS -28: OP := _CMP_NEQ_OS -29: OP := _CMP_GE_OQ -30: OP := _CMP_GT_OQ -31: OP := _CMP_TRUE_US -ESAC -FOR j := 0 to 1 - i := j*64 - IF k1[j] - k[j] := ( a[i+63:i] OP b[i+63:i] ) ? 1 : 0 - ELSE - k[j] := 0 - FI +FOR j := 0 to 7 + i := 64*j + k := 32*j + dst[i+63:i] := ZeroExtend64(a[k+31:k]) ENDFOR -k[MAX:2] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point - Mask - AVX512VL + + Integer AVX512F - Compare - - - - Compare packed single-precision (32-bit) floating-point elements in "a" and "b" based on the comparison operand specified by "imm8", and store the results in mask vector "k". + Convert + + + + + Zero extend packed unsigned 32-bit integers in "a" to packed 64-bit integers, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -CASE (imm8[7:0]) OF -0: OP := _CMP_EQ_OQ -1: OP := _CMP_LT_OS -2: OP := _CMP_LE_OS -3: OP := _CMP_UNORD_Q -4: OP := _CMP_NEQ_UQ -5: OP := _CMP_NLT_US -6: OP := _CMP_NLE_US -7: OP := _CMP_ORD_Q -8: OP := _CMP_EQ_UQ -9: OP := _CMP_NGE_US -10: OP := _CMP_NGT_US -11: OP := _CMP_FALSE_OQ -12: OP := _CMP_NEQ_OQ -13: OP := _CMP_GE_OS -14: OP := _CMP_GT_OS -15: OP := _CMP_TRUE_UQ -16: OP := _CMP_EQ_OS -17: OP := _CMP_LT_OQ -18: OP := _CMP_LE_OQ -19: OP := _CMP_UNORD_S -20: OP := _CMP_NEQ_US -21: OP := _CMP_NLT_UQ -22: OP := _CMP_NLE_UQ -23: OP := _CMP_ORD_S -24: OP := _CMP_EQ_US -25: OP := _CMP_NGE_UQ -26: OP := _CMP_NGT_UQ -27: OP := _CMP_FALSE_OS -28: OP := _CMP_NEQ_OS -29: OP := _CMP_GE_OQ -30: OP := _CMP_GT_OQ -31: OP := _CMP_TRUE_US -ESAC FOR j := 0 to 7 - i := j*32 - k[j] := (a[i+31:i] OP b[i+31:i]) ? 1 : 0 + i := 64*j + l := 32*j + IF k[j] + dst[i+63:i] := ZeroExtend64(a[l+31:l]) + ELSE + dst[i+63:i] := src[i+63:i] + FI ENDFOR -k[MAX:8] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point - Mask - AVX512VL + + Integer AVX512F - Compare - - - - - Compare packed single-precision (32-bit) floating-point elements in "a" and "b" based on the comparison operand specified by "imm8", and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). + Convert + + + + Zero extend packed unsigned 32-bit integers in "a" to packed 64-bit integers, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -CASE (imm8[7:0]) OF -0: OP := _CMP_EQ_OQ -1: OP := _CMP_LT_OS -2: OP := _CMP_LE_OS -3: OP := _CMP_UNORD_Q -4: OP := _CMP_NEQ_UQ -5: OP := _CMP_NLT_US -6: OP := _CMP_NLE_US -7: OP := _CMP_ORD_Q -8: OP := _CMP_EQ_UQ -9: OP := _CMP_NGE_US -10: OP := _CMP_NGT_US -11: OP := _CMP_FALSE_OQ -12: OP := _CMP_NEQ_OQ -13: OP := _CMP_GE_OS -14: OP := _CMP_GT_OS -15: OP := _CMP_TRUE_UQ -16: OP := _CMP_EQ_OS -17: OP := _CMP_LT_OQ -18: OP := _CMP_LE_OQ -19: OP := _CMP_UNORD_S -20: OP := _CMP_NEQ_US -21: OP := _CMP_NLT_UQ -22: OP := _CMP_NLE_UQ -23: OP := _CMP_ORD_S -24: OP := _CMP_EQ_US -25: OP := _CMP_NGE_UQ -26: OP := _CMP_NGT_UQ -27: OP := _CMP_FALSE_OS -28: OP := _CMP_NEQ_OS -29: OP := _CMP_GE_OQ -30: OP := _CMP_GT_OQ -31: OP := _CMP_TRUE_US -ESAC FOR j := 0 to 7 - i := j*32 - IF k1[j] - k[j] := ( a[i+31:i] OP b[i+31:i] ) ? 1 : 0 + i := 64*j + l := 32*j + IF k[j] + dst[i+63:i] := ZeroExtend64(a[l+31:l]) ELSE - k[j] := 0 + dst[i+63:i] := 0 FI ENDFOR -k[MAX:8] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point - Mask - AVX512VL + + Integer AVX512F - Compare - - - - Compare packed single-precision (32-bit) floating-point elements in "a" and "b" based on the comparison operand specified by "imm8", and store the results in mask vector "k". + Convert + + + Zero extend packed unsigned 16-bit integers in "a" to packed 32-bit integers, and store the results in "dst". -CASE (imm8[7:0]) OF -0: OP := _CMP_EQ_OQ -1: OP := _CMP_LT_OS -2: OP := _CMP_LE_OS -3: OP := _CMP_UNORD_Q -4: OP := _CMP_NEQ_UQ -5: OP := _CMP_NLT_US -6: OP := _CMP_NLE_US -7: OP := _CMP_ORD_Q -8: OP := _CMP_EQ_UQ -9: OP := _CMP_NGE_US -10: OP := _CMP_NGT_US -11: OP := _CMP_FALSE_OQ -12: OP := _CMP_NEQ_OQ -13: OP := _CMP_GE_OS -14: OP := _CMP_GT_OS -15: OP := _CMP_TRUE_UQ -16: OP := _CMP_EQ_OS -17: OP := _CMP_LT_OQ -18: OP := _CMP_LE_OQ -19: OP := _CMP_UNORD_S -20: OP := _CMP_NEQ_US -21: OP := _CMP_NLT_UQ -22: OP := _CMP_NLE_UQ -23: OP := _CMP_ORD_S -24: OP := _CMP_EQ_US -25: OP := _CMP_NGE_UQ -26: OP := _CMP_NGT_UQ -27: OP := _CMP_FALSE_OS -28: OP := _CMP_NEQ_OS -29: OP := _CMP_GE_OQ -30: OP := _CMP_GT_OQ -31: OP := _CMP_TRUE_US -ESAC -FOR j := 0 to 3 - i := j*32 - k[j] := ( a[i+31:i] OP b[i+31:i] ) ? 1 : 0 +FOR j := 0 to 15 + i := 32*j + k := 16*j + dst[i+31:i] := ZeroExtend32(a[k+15:k]) ENDFOR -k[MAX:4] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point - Mask - AVX512VL + + Integer AVX512F - Compare - - - - - Compare packed single-precision (32-bit) floating-point elements in "a" and "b" based on the comparison operand specified by "imm8", and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). + Convert + + + + + Zero extend packed unsigned 16-bit integers in "a" to packed 32-bit integers, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -CASE (imm8[7:0]) OF -0: OP := _CMP_EQ_OQ -1: OP := _CMP_LT_OS -2: OP := _CMP_LE_OS -3: OP := _CMP_UNORD_Q -4: OP := _CMP_NEQ_UQ -5: OP := _CMP_NLT_US -6: OP := _CMP_NLE_US -7: OP := _CMP_ORD_Q -8: OP := _CMP_EQ_UQ -9: OP := _CMP_NGE_US -10: OP := _CMP_NGT_US -11: OP := _CMP_FALSE_OQ -12: OP := _CMP_NEQ_OQ -13: OP := _CMP_GE_OS -14: OP := _CMP_GT_OS -15: OP := _CMP_TRUE_UQ -16: OP := _CMP_EQ_OS -17: OP := _CMP_LT_OQ -18: OP := _CMP_LE_OQ -19: OP := _CMP_UNORD_S -20: OP := _CMP_NEQ_US -21: OP := _CMP_NLT_UQ -22: OP := _CMP_NLE_UQ -23: OP := _CMP_ORD_S -24: OP := _CMP_EQ_US -25: OP := _CMP_NGE_UQ -26: OP := _CMP_NGT_UQ -27: OP := _CMP_FALSE_OS -28: OP := _CMP_NEQ_OS -29: OP := _CMP_GE_OQ -30: OP := _CMP_GT_OQ -31: OP := _CMP_TRUE_US -ESAC -FOR j := 0 to 3 - i := j*32 - IF k1[j] - k[j] := ( a[i+31:i] OP b[i+31:i] ) ? 1 : 0 - ELSE - k[j] := 0 +FOR j := 0 to 15 + i := 32*j + l := 16*j + IF k[j] + dst[i+31:i] := ZeroExtend32(a[l+15:l]) + ELSE + dst[i+31:i] := src[i+31:i] + FI +ENDFOR +dst[MAX:512] := 0 + + +
immintrin.h
+ + + Integer + AVX512F + Convert + + + + Zero extend packed unsigned 16-bit integers in "a" to packed 32-bit integers, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + +FOR j := 0 to 15 + i := 32*j + l := 16*j + IF k[j] + dst[i+31:i] := ZeroExtend32(a[l+15:l]) + ELSE + dst[i+31:i] := 0 FI ENDFOR -k[MAX:4] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point - AVX512VL + + Integer AVX512F - Miscellaneous - - - - Contiguously store the active double-precision (64-bit) floating-point elements in "a" (those with their respective bit set in writemask "k") to "dst", and pass through the remaining elements from "src". + Convert + + + Zero extend packed unsigned 16-bit integers in "a" to packed 64-bit integers, and store the results in "dst". -size := 64 -m := 0 -FOR j := 0 to 3 - i := j*64 - IF k[j] - dst[m+size-1:m] := a[i+63:i] - m := m + size - FI +FOR j := 0 to 7 + i := 64*j + k := 16*j + dst[i+63:i] := ZeroExtend64(a[k+15:k]) ENDFOR -dst[255:m] := src[255:m] -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point - AVX512VL + + Integer AVX512F - Store - - - - Contiguously store the active double-precision (64-bit) floating-point elements in "a" (those with their respective bit set in writemask "k") to unaligned memory at "base_addr". + Convert + + + + + Zero extend packed unsigned 16-bit integers in "a" to packed 64-bit integers, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -size := 64 -m := base_addr -FOR j := 0 to 3 - i := j*64 +FOR j := 0 to 7 + i := 64*j + l := 16*j IF k[j] - MEM[m+size-1:m] := a[i+63:i] - m := m + size + dst[i+63:i] := ZeroExtend64(a[l+15:l]) + ELSE + dst[i+63:i] := src[i+63:i] FI ENDFOR +dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point - AVX512VL + + Integer AVX512F - Miscellaneous - - - Contiguously store the active double-precision (64-bit) floating-point elements in "a" (those with their respective bit set in zeromask "k") to "dst", and set the remaining elements to zero. + Convert + + + + Zero extend packed unsigned 16-bit integers in "a" to packed 64-bit integers, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -size := 64 -m := 0 -FOR j := 0 to 3 - i := j*64 +FOR j := 0 to 7 + i := 64*j + l := 16*j IF k[j] - dst[m+size-1:m] := a[i+63:i] - m := m + size + dst[i+63:i] := ZeroExtend64(a[l+15:l]) + ELSE + dst[i+63:i] := 0 FI ENDFOR -dst[255:m] := 0 -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point - AVX512VL + + Integer AVX512F - Miscellaneous - - - - Contiguously store the active double-precision (64-bit) floating-point elements in "a" (those with their respective bit set in writemask "k") to "dst", and pass through the remaining elements from "src". + Arithmetic + + + + + + Multiply the low signed 32-bit integers from each packed 64-bit element in "a" and "b", and store the signed 64-bit results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -size := 64 -m := 0 -FOR j := 0 to 1 +FOR j := 0 to 7 i := j*64 IF k[j] - dst[m+size-1:m] := a[i+63:i] - m := m + size + dst[i+63:i] := SignExtend64(a[i+31:i]) * SignExtend64(b[i+31:i]) + ELSE + dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[127:m] := src[127:m] -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point - AVX512VL + + Integer AVX512F - Store - - - - Contiguously store the active double-precision (64-bit) floating-point elements in "a" (those with their respective bit set in writemask "k") to unaligned memory at "base_addr". + Arithmetic + + + + + Multiply the low signed 32-bit integers from each packed 64-bit element in "a" and "b", and store the signed 64-bit results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -size := 64 -m := base_addr -FOR j := 0 to 1 +FOR j := 0 to 7 i := j*64 IF k[j] - MEM[m+size-1:m] := a[i+63:i] - m := m + size + dst[i+63:i] := SignExtend64(a[i+31:i]) * SignExtend64(b[i+31:i]) + ELSE + dst[i+63:i] := 0 FI ENDFOR +dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point - AVX512VL + + Integer AVX512F - Miscellaneous - - - Contiguously store the active double-precision (64-bit) floating-point elements in "a" (those with their respective bit set in zeromask "k") to "dst", and set the remaining elements to zero. + Arithmetic + + + + Multiply the low signed 32-bit integers from each packed 64-bit element in "a" and "b", and store the signed 64-bit results in "dst". -size := 64 -m := 0 -FOR j := 0 to 1 +FOR j := 0 to 7 i := j*64 - IF k[j] - dst[m+size-1:m] := a[i+63:i] - m := m + size - FI + dst[i+63:i] := SignExtend64(a[i+31:i]) * SignExtend64(b[i+31:i]) ENDFOR -dst[127:m] := 0 -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point - AVX512VL + + Integer AVX512F - Miscellaneous - - - - Contiguously store the active single-precision (32-bit) floating-point elements in "a" (those with their respective bit set in writemask "k") to "dst", and pass through the remaining elements from "src". + Arithmetic + + + + + + Multiply the low unsigned 32-bit integers from each packed 64-bit element in "a" and "b", and store the unsigned 64-bit results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -size := 32 -m := 0 FOR j := 0 to 7 - i := j*32 + i := j*64 IF k[j] - dst[m+size-1:m] := a[i+31:i] - m := m + size + dst[i+63:i] := a[i+31:i] * b[i+31:i] + ELSE + dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[255:m] := src[255:m] -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point - AVX512VL + + Integer AVX512F - Store - - - - Contiguously store the active single-precision (32-bit) floating-point elements in "a" (those with their respective bit set in writemask "k") to unaligned memory at "base_addr". + Arithmetic + + + + + Multiply the low unsigned 32-bit integers from each packed 64-bit element in "a" and "b", and store the unsigned 64-bit results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -size := 32 -m := base_addr FOR j := 0 to 7 - i := j*32 + i := j*64 IF k[j] - MEM[m+size-1:m] := a[i+31:i] - m := m + size + dst[i+63:i] := a[i+31:i] * b[i+31:i] + ELSE + dst[i+63:i] := 0 FI ENDFOR +dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point - AVX512VL + + Integer AVX512F - Miscellaneous - - - Contiguously store the active single-precision (32-bit) floating-point elements in "a" (those with their respective bit set in zeromask "k") to "dst", and set the remaining elements to zero. + Arithmetic + + + + Multiply the low unsigned 32-bit integers from each packed 64-bit element in "a" and "b", and store the unsigned 64-bit results in "dst". -size := 32 -m := 0 FOR j := 0 to 7 - i := j*32 - IF k[j] - dst[m+size-1:m] := a[i+31:i] - m := m + size - FI + i := j*64 + dst[i+63:i] := a[i+31:i] * b[i+31:i] ENDFOR -dst[255:m] := 0 -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point - AVX512VL + + Integer AVX512F - Miscellaneous - - - - Contiguously store the active single-precision (32-bit) floating-point elements in "a" (those with their respective bit set in writemask "k") to "dst", and pass through the remaining elements from "src". + Logical + + + + + Compute the bitwise OR of packed 32-bit integers in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -size := 32 -m := 0 -FOR j := 0 to 3 +FOR j := 0 to 15 i := j*32 IF k[j] - dst[m+size-1:m] := a[i+31:i] - m := m + size + dst[i+31:i] := a[i+31:i] OR b[i+31:i] + ELSE + dst[i+31:i] := 0 FI ENDFOR -dst[127:m] := src[127:m] -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point - AVX512VL + + Integer AVX512F - Store - - - - Contiguously store the active single-precision (32-bit) floating-point elements in "a" (those with their respective bit set in writemask "k") to unaligned memory at "base_addr". + Logical + + + + + Compute the bitwise OR of packed 64-bit integers in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -size := 32 -m := base_addr -FOR j := 0 to 3 - i := j*32 +FOR j := 0 to 7 + i := j*64 IF k[j] - MEM[m+size-1:m] := a[i+31:i] - m := m + size + dst[i+63:i] := a[i+63:i] OR b[i+63:i] + ELSE + dst[i+63:i] := 0 FI ENDFOR +dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point - AVX512VL + + Integer AVX512F - Miscellaneous - - - Contiguously store the active single-precision (32-bit) floating-point elements in "a" (those with their respective bit set in zeromask "k") to "dst", and set the remaining elements to zero. - -size := 32 -m := 0 -FOR j := 0 to 3 + Shift + + + + + + Rotate the bits in each packed 32-bit integer in "a" to the left by the number of bits specified in "imm8", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +DEFINE LEFT_ROTATE_DWORDS(src, count_src) { + count := count_src % 32 + RETURN (src << count) OR (src >> (32 - count)) +} +FOR j := 0 to 15 i := j*32 IF k[j] - dst[m+size-1:m] := a[i+31:i] - m := m + size + dst[i+31:i] := LEFT_ROTATE_DWORDS(a[i+31:i], imm8[7:0]) + ELSE + dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[127:m] := 0 -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point + Integer - AVX512VL AVX512F - Convert - - - - Convert packed 32-bit integers in "a" to packed double-precision (64-bit) floating-point elements, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -FOR j := 0 to 3 + Shift + + + + + Rotate the bits in each packed 32-bit integer in "a" to the left by the number of bits specified in "imm8", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + +DEFINE LEFT_ROTATE_DWORDS(src, count_src) { + count := count_src % 32 + RETURN (src << count) OR (src >> (32 - count)) +} +FOR j := 0 to 15 i := j*32 - m := j*64 IF k[j] - dst[m+63:m] := Convert_Int32_To_FP64(a[i+31:i]) + dst[i+31:i] := LEFT_ROTATE_DWORDS(a[i+31:i], imm8[7:0]) ELSE - dst[m+63:m] := src[m+63:m] + dst[i+31:i] := 0 FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point + Integer - AVX512VL AVX512F - Convert - - - Convert packed 32-bit integers in "a" to packed double-precision (64-bit) floating-point elements, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Shift + + + + Rotate the bits in each packed 32-bit integer in "a" to the left by the number of bits specified in "imm8", and store the results in "dst". -FOR j := 0 to 3 +DEFINE LEFT_ROTATE_DWORDS(src, count_src) { + count := count_src % 32 + RETURN (src << count) OR (src >> (32 - count)) +} +FOR j := 0 to 15 i := j*32 - m := j*64 - IF k[j] - dst[m+63:m] := Convert_Int32_To_FP64(a[i+31:i]) - ELSE - dst[m+63:m] := 0 - FI + dst[i+31:i] := LEFT_ROTATE_DWORDS(a[i+31:i], imm8[7:0]) ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point + Integer - AVX512VL AVX512F - Convert - - - - Convert packed 32-bit integers in "a" to packed double-precision (64-bit) floating-point elements, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -FOR j := 0 to 1 - i := j*32 - m := j*64 + Shift + + + + + + Rotate the bits in each packed 64-bit integer in "a" to the left by the number of bits specified in "imm8", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +DEFINE LEFT_ROTATE_QWORDS(src, count_src) { + count := count_src % 64 + RETURN (src << count) OR (src >> (64 - count)) +} +FOR j := 0 to 7 + i := j*64 IF k[j] - dst[m+63:m] := Convert_Int32_To_FP64(a[i+31:i]) + dst[i+63:i] := LEFT_ROTATE_QWORDS(a[i+63:i], imm8[7:0]) ELSE - dst[m+63:m] := src[m+63:m] + dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point + Integer - AVX512VL AVX512F - Convert - - - Convert packed 32-bit integers in "a" to packed double-precision (64-bit) floating-point elements, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - -FOR j := 0 to 1 - i := j*32 - m := j*64 + Shift + + + + + Rotate the bits in each packed 64-bit integer in "a" to the left by the number of bits specified in "imm8", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + +DEFINE LEFT_ROTATE_QWORDS(src, count_src) { + count := count_src % 64 + RETURN (src << count) OR (src >> (64 - count)) +} +FOR j := 0 to 7 + i := j*64 IF k[j] - dst[m+63:m] := Convert_Int32_To_FP64(a[i+31:i]) + dst[i+63:i] := LEFT_ROTATE_QWORDS(a[i+63:i], imm8[7:0]) ELSE - dst[m+63:m] := 0 + dst[i+63:i] := 0 FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point + Integer - AVX512VL AVX512F - Convert - - - - Convert packed 32-bit integers in "a" to packed single-precision (32-bit) floating-point elements, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Shift + + + + Rotate the bits in each packed 64-bit integer in "a" to the left by the number of bits specified in "imm8", and store the results in "dst". +DEFINE LEFT_ROTATE_QWORDS(src, count_src) { + count := count_src % 64 + RETURN (src << count) OR (src >> (64 - count)) +} FOR j := 0 to 7 + i := j*64 + dst[i+63:i] := LEFT_ROTATE_QWORDS(a[i+63:i], imm8[7:0]) +ENDFOR +dst[MAX:512] := 0 + + +
immintrin.h
+ + + Integer + AVX512F + Shift + + + + + + Rotate the bits in each packed 32-bit integer in "a" to the left by the number of bits specified in the corresponding element of "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +DEFINE LEFT_ROTATE_DWORDS(src, count_src) { + count := count_src % 32 + RETURN (src << count) OR (src >> (32 - count)) +} +FOR j := 0 to 15 i := j*32 IF k[j] - dst[i+31:i] := Convert_Int32_To_FP32(a[i+31:i]) + dst[i+31:i] := LEFT_ROTATE_DWORDS(a[i+31:i], b[i+31:i]) ELSE dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point + Integer - AVX512VL AVX512F - Convert - - - Convert packed 32-bit integers in "a" to packed single-precision (32-bit) floating-point elements, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - -FOR j := 0 to 7 - i := 32*j + Shift + + + + + Rotate the bits in each packed 32-bit integer in "a" to the left by the number of bits specified in the corresponding element of "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + +DEFINE LEFT_ROTATE_DWORDS(src, count_src) { + count := count_src % 32 + RETURN (src << count) OR (src >> (32 - count)) +} +FOR j := 0 to 15 + i := j*32 IF k[j] - dst[i+31:i] := Convert_Int32_To_FP32(a[i+31:i]) + dst[i+31:i] := LEFT_ROTATE_DWORDS(a[i+31:i], b[i+31:i]) ELSE dst[i+31:i] := 0 FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point + Integer - AVX512VL AVX512F - Convert - - - - Convert packed 32-bit integers in "a" to packed single-precision (32-bit) floating-point elements, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Shift + + + + Rotate the bits in each packed 32-bit integer in "a" to the left by the number of bits specified in the corresponding element of "b", and store the results in "dst". -FOR j := 0 to 3 +DEFINE LEFT_ROTATE_DWORDS(src, count_src) { + count := count_src % 32 + RETURN (src << count) OR (src >> (32 - count)) +} +FOR j := 0 to 15 i := j*32 + dst[i+31:i] := LEFT_ROTATE_DWORDS(a[i+31:i], b[i+31:i]) +ENDFOR +dst[MAX:512] := 0 + + +
immintrin.h
+ + + Integer + AVX512F + Shift + + + + + + Rotate the bits in each packed 64-bit integer in "a" to the left by the number of bits specified in the corresponding element of "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +DEFINE LEFT_ROTATE_QWORDS(src, count_src) { + count := count_src % 64 + RETURN (src << count) OR (src >> (64 - count)) +} +FOR j := 0 to 7 + i := j*64 IF k[j] - dst[i+31:i] := Convert_Int32_To_FP32(a[i+31:i]) + dst[i+63:i] := LEFT_ROTATE_QWORDS(a[i+63:i], b[i+63:i]) ELSE - dst[i+31:i] := src[i+31:i] + dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point + Integer - AVX512VL AVX512F - Convert - - - Convert packed 32-bit integers in "a" to packed single-precision (32-bit) floating-point elements, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - -FOR j := 0 to 3 - i := 32*j + Shift + + + + + Rotate the bits in each packed 64-bit integer in "a" to the left by the number of bits specified in the corresponding element of "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + +DEFINE LEFT_ROTATE_QWORDS(src, count_src) { + count := count_src % 64 + RETURN (src << count) OR (src >> (64 - count)) +} +FOR j := 0 to 7 + i := j*64 IF k[j] - dst[i+31:i] := Convert_Int32_To_FP32(a[i+31:i]) + dst[i+63:i] := LEFT_ROTATE_QWORDS(a[i+63:i], b[i+63:i]) ELSE - dst[i+31:i] := 0 + dst[i+63:i] := 0 FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point + Integer - AVX512VL AVX512F - Convert - - - - Convert packed double-precision (64-bit) floating-point elements in "a" to packed 32-bit integers, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Shift + + + + Rotate the bits in each packed 64-bit integer in "a" to the left by the number of bits specified in the corresponding element of "b", and store the results in "dst". -FOR j := 0 to 3 +DEFINE LEFT_ROTATE_QWORDS(src, count_src) { + count := count_src % 64 + RETURN (src << count) OR (src >> (64 - count)) +} +FOR j := 0 to 7 + i := j*64 + dst[i+63:i] := LEFT_ROTATE_QWORDS(a[i+63:i], b[i+63:i]) +ENDFOR +dst[MAX:512] := 0 + + +
immintrin.h
+ + + Integer + AVX512F + Shift + + + + + + Rotate the bits in each packed 32-bit integer in "a" to the right by the number of bits specified in "imm8", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +DEFINE RIGHT_ROTATE_DWORDS(src, count_src) { + count := count_src % 32 + RETURN (src >>count) OR (src << (32 - count)) +} +FOR j := 0 to 15 i := j*32 - l := j*64 IF k[j] - dst[i+31:i] := Convert_FP64_To_Int32(a[l+63:l]) + dst[i+31:i] := RIGHT_ROTATE_DWORDS(a[i+31:i], imm8[7:0]) ELSE dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point + Integer - AVX512VL AVX512F - Convert - - - Convert packed double-precision (64-bit) floating-point elements in "a" to packed 32-bit integers, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - -FOR j := 0 to 3 - i := 32*j - l := 64*j + Shift + + + + + Rotate the bits in each packed 32-bit integer in "a" to the right by the number of bits specified in "imm8", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + +DEFINE RIGHT_ROTATE_DWORDS(src, count_src) { + count := count_src % 32 + RETURN (src >>count) OR (src << (32 - count)) +} +FOR j := 0 to 15 + i := j*32 IF k[j] - dst[i+31:i] := Convert_FP64_To_Int32(a[l+63:l]) + dst[i+31:i] := RIGHT_ROTATE_DWORDS(a[i+31:i], imm8[7:0]) ELSE dst[i+31:i] := 0 FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point + Integer - AVX512VL AVX512F - Convert - - - - Convert packed double-precision (64-bit) floating-point elements in "a" to packed 32-bit integers, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Shift + + + + Rotate the bits in each packed 32-bit integer in "a" to the right by the number of bits specified in "imm8", and store the results in "dst". -FOR j := 0 to 1 +DEFINE RIGHT_ROTATE_DWORDS(src, count_src) { + count := count_src % 32 + RETURN (src >>count) OR (src << (32 - count)) +} +FOR j := 0 to 15 i := j*32 - l := j*64 - IF k[j] - dst[i+31:i] := Convert_FP64_To_Int32(a[l+63:l]) - ELSE - dst[i+31:i] := src[i+31:i] - FI + dst[i+31:i] := RIGHT_ROTATE_DWORDS(a[i+31:i], imm8[7:0]) ENDFOR -dst[MAX:64] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point + Integer - AVX512VL AVX512F - Convert - - - Convert packed double-precision (64-bit) floating-point elements in "a" to packed 32-bit integers, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - -FOR j := 0 to 1 - i := 32*j - l := 64*j + Shift + + + + + + Rotate the bits in each packed 64-bit integer in "a" to the right by the number of bits specified in "imm8", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +DEFINE RIGHT_ROTATE_QWORDS(src, count_src) { + count := count_src % 64 + RETURN (src >> count) OR (src << (64 - count)) +} +FOR j := 0 to 7 + i := j*64 IF k[j] - dst[i+31:i] := Convert_FP64_To_Int32(a[l+63:l]) + dst[i+63:i] := RIGHT_ROTATE_QWORDS(a[i+63:i], imm8[7:0]) ELSE - dst[i+31:i] := 0 + dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:64] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point - AVX512VL + + Integer AVX512F - Convert - - - - Convert packed double-precision (64-bit) floating-point elements in "a" to packed single-precision (32-bit) floating-point elements, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -FOR j := 0 to 3 - i := 32*j - l := 64*j + Shift + + + + + Rotate the bits in each packed 64-bit integer in "a" to the right by the number of bits specified in "imm8", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + +DEFINE RIGHT_ROTATE_QWORDS(src, count_src) { + count := count_src % 64 + RETURN (src >> count) OR (src << (64 - count)) +} +FOR j := 0 to 7 + i := j*64 IF k[j] - dst[i+31:i] := Convert_FP64_To_FP32(a[l+63:l]) + dst[i+63:i] := RIGHT_ROTATE_QWORDS(a[i+63:i], imm8[7:0]) ELSE - dst[i+31:i] := src[i+31:i] + dst[i+63:i] := 0 FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point - AVX512VL + + Integer AVX512F - Convert - - - Convert packed double-precision (64-bit) floating-point elements in "a" to packed single-precision (32-bit) floating-point elements, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Shift + + + + Rotate the bits in each packed 64-bit integer in "a" to the right by the number of bits specified in "imm8", and store the results in "dst". -FOR j := 0 to 3 - i := j*32 - l := j*64 - IF k[j] - dst[i+31:i] := Convert_FP64_To_FP32(a[l+63:l]) - ELSE - dst[i+31:i] := 0 - FI +DEFINE RIGHT_ROTATE_QWORDS(src, count_src) { + count := count_src % 64 + RETURN (src >> count) OR (src << (64 - count)) +} +FOR j := 0 to 7 + i := j*64 + dst[i+63:i] := RIGHT_ROTATE_QWORDS(a[i+63:i], imm8[7:0]) ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point - AVX512VL + + Integer AVX512F - Convert - - - - Convert packed double-precision (64-bit) floating-point elements in "a" to packed single-precision (32-bit) floating-point elements, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -FOR j := 0 to 1 - i := 32*j - l := 64*j + Shift + + + + + + Rotate the bits in each packed 32-bit integer in "a" to the right by the number of bits specified in the corresponding element of "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +DEFINE RIGHT_ROTATE_DWORDS(src, count_src) { + count := count_src % 32 + RETURN (src >>count) OR (src << (32 - count)) +} +FOR j := 0 to 15 + i := j*32 IF k[j] - dst[i+31:i] := Convert_FP64_To_FP32(a[l+63:l]) + dst[i+31:i] := RIGHT_ROTATE_DWORDS(a[i+31:i], b[i+31:i]) ELSE dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:64] := 0 +dst[MAX:512] := 0 - +
immintrin.h
- - - Floating Point - AVX512VL + + + Integer AVX512F - Convert - - - Convert packed double-precision (64-bit) floating-point elements in "a" to packed single-precision (32-bit) floating-point elements, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Miscellaneous + + + + + Rotate the bits in each packed 32-bit integer in "a" to the right by the number of bits specified in the corresponding element of "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 1 +DEFINE RIGHT_ROTATE_DWORDS(src, count_src) { + count := count_src % 32 + RETURN (src >>count) OR (src << (32 - count)) +} +FOR j := 0 to 15 i := j*32 - l := j*64 IF k[j] - dst[i+31:i] := Convert_FP64_To_FP32(a[l+63:l]) + dst[i+31:i] := RIGHT_ROTATE_DWORDS(a[i+31:i], b[i+31:i]) ELSE dst[i+31:i] := 0 FI ENDFOR -dst[MAX:64] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point + Integer - AVX512VL - AVX512DQ - Convert - - Convert packed double-precision (64-bit) floating-point elements in "a" to packed 64-bit integers, and store the results in "dst". + AVX512F + Shift + + + + Rotate the bits in each packed 32-bit integer in "a" to the right by the number of bits specified in the corresponding element of "b", and store the results in "dst". -FOR j := 0 to 3 - i := j*64 - dst[i+63:i] := Convert_FP64_To_Int64(a[i+63:i]) +DEFINE RIGHT_ROTATE_DWORDS(src, count_src) { + count := count_src % 32 + RETURN (src >>count) OR (src << (32 - count)) +} +FOR j := 0 to 15 + i := j*32 + dst[i+31:i] := RIGHT_ROTATE_DWORDS(a[i+31:i], b[i+31:i]) ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point + Integer - AVX512VL - AVX512DQ - Convert - - - - Convert packed double-precision (64-bit) floating-point elements in "a" to packed 64-bit integers, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -FOR j := 0 to 3 + AVX512F + Shift + + + + + + Rotate the bits in each packed 64-bit integer in "a" to the right by the number of bits specified in the corresponding element of "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +DEFINE RIGHT_ROTATE_QWORDS(src, count_src) { + count := count_src % 64 + RETURN (src >> count) OR (src << (64 - count)) +} +FOR j := 0 to 7 i := j*64 IF k[j] - dst[i+63:i] := Convert_FP64_To_Int64(a[i+63:i]) + dst[i+63:i] := RIGHT_ROTATE_QWORDS(a[i+63:i], b[i+63:i]) ELSE dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point + Integer - AVX512VL - AVX512DQ - Convert - - - Convert packed double-precision (64-bit) floating-point elements in "a" to packed 64-bit integers, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - -FOR j := 0 to 3 + AVX512F + Shift + + + + + Rotate the bits in each packed 64-bit integer in "a" to the right by the number of bits specified in the corresponding element of "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + +DEFINE RIGHT_ROTATE_QWORDS(src, count_src) { + count := count_src % 64 + RETURN (src >> count) OR (src << (64 - count)) +} +FOR j := 0 to 7 i := j*64 IF k[j] - dst[i+63:i] := Convert_FP64_To_Int64(a[i+63:i]) + dst[i+63:i] := RIGHT_ROTATE_QWORDS(a[i+63:i], b[i+63:i]) ELSE dst[i+63:i] := 0 FI ENDFOR -dst[MAX:256] := 0 - - -
immintrin.h
- - - Floating Point - Integer - AVX512DQ - Convert - - - Convert packed double-precision (64-bit) floating-point elements in "a" to packed 64-bit integers, and store the results in "dst". - [round_note] - - -FOR j := 0 to 7 - i := j*64 - dst[i+63:i] := Convert_FP64_To_Int64(a[i+63:i]) -ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point + Integer - AVX512DQ - Convert - - Convert packed double-precision (64-bit) floating-point elements in "a" to packed 64-bit integers, and store the results in "dst". + AVX512F + Shift + + + + Rotate the bits in each packed 64-bit integer in "a" to the right by the number of bits specified in the corresponding element of "b", and store the results in "dst". +DEFINE RIGHT_ROTATE_QWORDS(src, count_src) { + count := count_src % 64 + RETURN (src >> count) OR (src << (64 - count)) +} FOR j := 0 to 7 i := j*64 - dst[i+63:i] := Convert_FP64_To_Int64(a[i+63:i]) + dst[i+63:i] := RIGHT_ROTATE_QWORDS(a[i+63:i], b[i+63:i]) ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point + Integer - AVX512DQ - Convert - - - - - Convert packed double-precision (64-bit) floating-point elements in "a" to packed 64-bit integers, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - [round_note] - + AVX512F + Store + + + + + + Scatter 64-bit integers from "a" into memory using 32-bit indices. 64-bit elements are stored at addresses starting at "base_addr" and offset by each 32-bit element in "vindex" (each index is scaled by the factor in "scale"). "scale" should be 1, 2, 4 or 8. FOR j := 0 to 7 i := j*64 - IF k[j] - dst[i+63:i] := Convert_FP64_To_Int64(a[i+63:i]) - ELSE - dst[i+63:i] := src[i+63:i] - FI + m := j*32 + addr := base_addr + SignExtend64(vindex[m+31:m]) * ZeroExtend64(scale) * 8 + MEM[addr+63:addr] := a[i+63:i] ENDFOR -dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point + Integer - AVX512DQ - Convert - - - - Convert packed double-precision (64-bit) floating-point elements in "a" to packed 64-bit integers, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + AVX512F + Store + + + + + + + Scatter 64-bit integers from "a" into memory using 32-bit indices. 64-bit elements are stored at addresses starting at "base_addr" and offset by each 32-bit element in "vindex" (each index is scaled by the factor in "scale") subject to mask "k" (elements are not stored when the corresponding mask bit is not set). "scale" should be 1, 2, 4 or 8. FOR j := 0 to 7 i := j*64 + m := j*32 IF k[j] - dst[i+63:i] := Convert_FP64_To_Int64(a[i+63:i]) - ELSE - dst[i+63:i] := src[i+63:i] + addr := base_addr + SignExtend64(vindex[m+31:m]) * ZeroExtend64(scale) * 8 + MEM[addr+63:addr] := a[i+63:i] FI ENDFOR -dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point + Integer - AVX512DQ - Convert - - - - Convert packed double-precision (64-bit) floating-point elements in "a" to packed 64-bit integers, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - [round_note] - + AVX512F + Store + + + + + + Scatter 32-bit integers from "a" into memory using 64-bit indices. 32-bit elements are stored at addresses starting at "base_addr" and offset by each 64-bit element in "vindex" (each index is scaled by the factor in "scale"). "scale" should be 1, 2, 4 or 8. FOR j := 0 to 7 - i := j*64 - IF k[j] - dst[i+63:i] := Convert_FP64_To_Int64(a[i+63:i]) - ELSE - dst[i+63:i] := 0 - FI + i := j*32 + m := j*64 + addr := base_addr + vindex[m+63:m] * ZeroExtend64(scale) * 8 + MEM[addr+31:addr] := a[i+31:i] ENDFOR -dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point + Integer - AVX512DQ - Convert - - - Convert packed double-precision (64-bit) floating-point elements in "a" to packed 64-bit integers, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + AVX512F + Store + + + + + + + Scatter 32-bit integers from "a" into memory using 64-bit indices. 32-bit elements are stored at addresses starting at "base_addr" and offset by each 64-bit element in "vindex" (each index is scaled by the factor in "scale") subject to mask "k" (elements are not stored when the corresponding mask bit is not set). "scale" should be 1, 2, 4 or 8. FOR j := 0 to 7 - i := j*64 + i := j*32 + m := j*64 IF k[j] - dst[i+63:i] := Convert_FP64_To_Int64(a[i+63:i]) - ELSE - dst[i+63:i] := 0 + addr := base_addr + vindex[m+63:m] * ZeroExtend64(scale) * 8 + MEM[addr+31:addr] := a[i+31:i] FI ENDFOR -dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point + Integer - AVX512VL - AVX512DQ - Convert - - Convert packed double-precision (64-bit) floating-point elements in "a" to packed 64-bit integers, and store the results in "dst". + AVX512F + Store + + + + + + Scatter 64-bit integers from "a" into memory using 64-bit indices. 64-bit elements are stored at addresses starting at "base_addr" and offset by each 64-bit element in "vindex" (each index is scaled by the factor in "scale"). "scale" should be 1, 2, 4 or 8. -FOR j := 0 to 1 +FOR j := 0 to 7 i := j*64 - dst[i+63:i] := Convert_FP64_To_Int64(a[i+63:i]) + m := j*64 + addr := base_addr + vindex[m+63:m] * ZeroExtend64(scale) * 8 + MEM[addr+63:addr] := a[i+63:i] ENDFOR -dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point + Integer - AVX512VL - AVX512DQ - Convert - - - - Convert packed double-precision (64-bit) floating-point elements in "a" to packed 64-bit integers, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + AVX512F + Store + + + + + + + Scatter 64-bit integers from "a" into memory using 64-bit indices. 64-bit elements are stored at addresses starting at "base_addr" and offset by each 64-bit element in "vindex" (each index is scaled by the factor in "scale") subject to mask "k" (elements are not stored when the corresponding mask bit is not set). "scale" should be 1, 2, 4 or 8. -FOR j := 0 to 1 +FOR j := 0 to 7 i := j*64 + m := j*64 IF k[j] - dst[i+63:i] := Convert_FP64_To_Int64(a[i+63:i]) - ELSE - dst[i+63:i] := src[i+63:i] + addr := base_addr + vindex[m+63:m] * ZeroExtend64(scale) * 8 + MEM[addr+63:addr] := a[i+63:i] FI ENDFOR -dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point + Integer - AVX512VL - AVX512DQ - Convert - - - Convert packed double-precision (64-bit) floating-point elements in "a" to packed 64-bit integers, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + AVX512F + Swizzle + + + + + Shuffle 32-bit integers in "a" within 128-bit lanes using the control in "imm8", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 1 - i := j*64 +DEFINE SELECT4(src, control) { + CASE(control[1:0]) OF + 0: tmp[31:0] := src[31:0] + 1: tmp[31:0] := src[63:32] + 2: tmp[31:0] := src[95:64] + 3: tmp[31:0] := src[127:96] + ESAC + RETURN tmp[31:0] +} +tmp_dst[31:0] := SELECT4(a[127:0], imm8[1:0]) +tmp_dst[63:32] := SELECT4(a[127:0], imm8[3:2]) +tmp_dst[95:64] := SELECT4(a[127:0], imm8[5:4]) +tmp_dst[127:96] := SELECT4(a[127:0], imm8[7:6]) +tmp_dst[159:128] := SELECT4(a[255:128], imm8[1:0]) +tmp_dst[191:160] := SELECT4(a[255:128], imm8[3:2]) +tmp_dst[223:192] := SELECT4(a[255:128], imm8[5:4]) +tmp_dst[255:224] := SELECT4(a[255:128], imm8[7:6]) +tmp_dst[287:256] := SELECT4(a[383:256], imm8[1:0]) +tmp_dst[319:288] := SELECT4(a[383:256], imm8[3:2]) +tmp_dst[351:320] := SELECT4(a[383:256], imm8[5:4]) +tmp_dst[383:352] := SELECT4(a[383:256], imm8[7:6]) +tmp_dst[415:384] := SELECT4(a[511:384], imm8[1:0]) +tmp_dst[447:416] := SELECT4(a[511:384], imm8[3:2]) +tmp_dst[479:448] := SELECT4(a[511:384], imm8[5:4]) +tmp_dst[511:480] := SELECT4(a[511:384], imm8[7:6]) +FOR j := 0 to 15 + i := j*32 IF k[j] - dst[i+63:i] := Convert_FP64_To_Int64(a[i+63:i]) + dst[i+31:i] := tmp_dst[i+31:i] ELSE - dst[i+63:i] := 0 + dst[i+31:i] := 0 FI ENDFOR -dst[MAX:128] := 0 - - -
immintrin.h
- - - Floating Point - Integer - AVX512VL - AVX512F - Convert - - Convert packed double-precision (64-bit) floating-point elements in "a" to packed unsigned 32-bit integers, and store the results in "dst". - -FOR j := 0 to 3 - i := 32*j - k := 64*j - dst[i+31:i] := Convert_FP64_To_UnsignedInt32(a[k+63:k]) -ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point + Integer - AVX512VL AVX512F - Convert - - - - Convert packed double-precision (64-bit) floating-point elements in "a" to packed unsigned 32-bit integers, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Shift + + + + + + Shift packed 32-bit integers in "a" left by "count" while shifting in zeros, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 3 +FOR j := 0 to 15 i := j*32 - l := j*64 IF k[j] - dst[i+31:i] := Convert_FP64_To_UnsignedInt32(a[l+63:l]) + IF count[63:0] > 31 + dst[i+31:i] := 0 + ELSE + dst[i+31:i] := ZeroExtend32(a[i+31:i] << count[63:0]) + FI ELSE dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point + Integer - AVX512VL AVX512F - Convert - - - Convert packed double-precision (64-bit) floating-point elements in "a" to packed unsigned 32-bit integers, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Shift + + + + + Shift packed 32-bit integers in "a" left by "count" while shifting in zeros, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := 32*j - l := 64*j +FOR j := 0 to 15 + i := j*32 IF k[j] - dst[i+31:i] := Convert_FP64_To_UnsignedInt32(a[l+63:l]) + IF count[63:0] > 31 + dst[i+31:i] := 0 + ELSE + dst[i+31:i] := ZeroExtend32(a[i+31:i] << count[63:0]) + FI ELSE dst[i+31:i] := 0 FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point + Integer - AVX512VL AVX512F - Convert - - Convert packed double-precision (64-bit) floating-point elements in "a" to packed unsigned 32-bit integers, and store the results in "dst". + Shift + + + + + Shift packed 32-bit integers in "a" left by "imm8" while shifting in zeros, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 1 - i := 32*j - k := 64*j - dst[i+31:i] := Convert_FP64_To_UnsignedInt32(a[k+63:k]) +FOR j := 0 to 15 + i := j*32 + IF k[j] + IF imm8[7:0] > 31 + dst[i+31:i] := 0 + ELSE + dst[i+31:i] := ZeroExtend32(a[i+31:i] << imm8[7:0]) + FI + ELSE + dst[i+31:i] := 0 + FI ENDFOR -dst[MAX:64] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point + Integer - AVX512VL AVX512F - Convert - - - - Convert packed double-precision (64-bit) floating-point elements in "a" to packed unsigned 32-bit integers, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Shift + + + + Shift packed 32-bit integers in "a" left by "count" while shifting in zeros, and store the results in "dst". -FOR j := 0 to 1 +FOR j := 0 to 15 i := j*32 - l := j*64 - IF k[j] - dst[i+31:i] := Convert_FP64_To_UnsignedInt32(a[l+63:l]) + IF count[63:0] > 31 + dst[i+31:i] := 0 ELSE - dst[i+31:i] := src[i+31:i] + dst[i+31:i] := ZeroExtend32(a[i+31:i] << count[63:0]) FI ENDFOR -dst[MAX:64] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point + Integer - AVX512VL AVX512F - Convert - - - Convert packed double-precision (64-bit) floating-point elements in "a" to packed unsigned 32-bit integers, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Shift + + + + + + Shift packed 64-bit integers in "a" left by "count" while shifting in zeros, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 1 - i := 32*j - l := 64*j +FOR j := 0 to 7 + i := j*64 IF k[j] - dst[i+31:i] := Convert_FP64_To_UnsignedInt32(a[l+63:l]) + IF count[63:0] > 63 + dst[i+63:i] := 0 + ELSE + dst[i+63:i] := ZeroExtend64(a[i+63:i] << count[63:0]) + FI ELSE - dst[i+31:i] := 0 + dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:64] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point + Integer - AVX512VL - AVX512DQ - Convert - - Convert packed double-precision (64-bit) floating-point elements in "a" to packed unsigned 64-bit integers, and store the results in "dst". + AVX512F + Shift + + + + + + Shift packed 64-bit integers in "a" left by "imm8" while shifting in zeros, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 3 +FOR j := 0 to 7 i := j*64 - dst[i+63:i] := Convert_FP64_To_UnsignedInt64(a[i+63:i]) + IF k[j] + IF imm8[7:0] > 63 + dst[i+63:i] := 0 + ELSE + dst[i+63:i] := ZeroExtend64(a[i+63:i] << imm8[7:0]) + FI + ELSE + dst[i+63:i] := src[i+63:i] + FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point + Integer - AVX512VL - AVX512DQ - Convert - - - - Convert packed double-precision (64-bit) floating-point elements in "a" to packed unsigned 64-bit integers, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + AVX512F + Shift + + + + + Shift packed 64-bit integers in "a" left by "count" while shifting in zeros, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 3 +FOR j := 0 to 7 i := j*64 IF k[j] - dst[i+63:i] := Convert_FP64_To_UnsignedInt64(a[i+63:i]) + IF count[63:0] > 63 + dst[i+63:i] := 0 + ELSE + dst[i+63:i] := ZeroExtend64(a[i+63:i] << count[63:0]) + FI ELSE - dst[i+63:i] := src[i+63:i] + dst[i+63:i] := 0 FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point + Integer - AVX512VL - AVX512DQ - Convert - - - Convert packed double-precision (64-bit) floating-point elements in "a" to packed unsigned 64-bit integers, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + AVX512F + Shift + + + + + Shift packed 64-bit integers in "a" left by "imm8" while shifting in zeros, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 3 +FOR j := 0 to 7 i := j*64 IF k[j] - dst[i+63:i] := Convert_FP64_To_UnsignedInt64(a[i+63:i]) + IF imm8[7:0] > 63 + dst[i+63:i] := 0 + ELSE + dst[i+63:i] := ZeroExtend64(a[i+63:i] << imm8[7:0]) + FI ELSE dst[i+63:i] := 0 FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point + Integer - AVX512DQ - Convert - - - Convert packed double-precision (64-bit) floating-point elements in "a" to packed unsigned 64-bit integers, and store the results in "dst". - [round_note] - + AVX512F + Shift + + + + Shift packed 64-bit integers in "a" left by "count" while shifting in zeros, and store the results in "dst". FOR j := 0 to 7 i := j*64 - dst[i+63:i] := Convert_FP64_To_UnsignedInt64(a[i+63:i]) + IF count[63:0] > 63 + dst[i+63:i] := 0 + ELSE + dst[i+63:i] := ZeroExtend64(a[i+63:i] << count[63:0]) + FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point + Integer - AVX512DQ - Convert - - Convert packed double-precision (64-bit) floating-point elements in "a" to packed unsigned 64-bit integers, and store the results in "dst". + AVX512F + Shift + + + + Shift packed 64-bit integers in "a" left by "imm8" while shifting in zeros, and store the results in "dst". FOR j := 0 to 7 i := j*64 - dst[i+63:i] := Convert_FP64_To_UnsignedInt64(a[i+63:i]) + IF imm8[7:0] > 63 + dst[i+63:i] := 0 + ELSE + dst[i+63:i] := ZeroExtend64(a[i+63:i] << imm8[7:0]) + FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point + Integer - AVX512DQ - Convert - - - - - Convert packed double-precision (64-bit) floating-point elements in "a" to packed unsigned 64-bit integers, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - [round_note] - + AVX512F + Shift + + + + + Shift packed 32-bit integers in "a" left by the amount specified by the corresponding element in "count" while shifting in zeros, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := j*64 +FOR j := 0 to 15 + i := j*32 IF k[j] - dst[i+63:i] := Convert_FP64_To_UnsignedInt64(a[i+63:i]) + IF count[i+31:i] < 32 + dst[i+31:i] := ZeroExtend32(a[i+31:i] << count[i+31:i]) + ELSE + dst[i+31:i] := 0 + FI ELSE - dst[i+63:i] := src[i+63:i] - FI + dst[i+31:i] := 0 + FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point + Integer - AVX512DQ - Convert - - - - Convert packed double-precision (64-bit) floating-point elements in "a" to packed unsigned 64-bit integers, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + AVX512F + Shift + + + + + + Shift packed 64-bit integers in "a" left by the amount specified by the corresponding element in "count" while shifting in zeros, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). FOR j := 0 to 7 i := j*64 IF k[j] - dst[i+63:i] := Convert_FP64_To_UnsignedInt64(a[i+63:i]) + IF count[i+63:i] < 64 + dst[i+63:i] := ZeroExtend64(a[i+63:i] << count[i+63:i]) + ELSE + dst[i+63:i] := 0 + FI ELSE dst[i+63:i] := src[i+63:i] - FI + FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point + Integer - AVX512DQ - Convert - - - - Convert packed double-precision (64-bit) floating-point elements in "a" to packed unsigned 64-bit integers, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - [round_note] - + AVX512F + Shift + + + + + Shift packed 64-bit integers in "a" left by the amount specified by the corresponding element in "count" while shifting in zeros, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). FOR j := 0 to 7 i := j*64 IF k[j] - dst[i+63:i] := Convert_FP64_To_UnsignedInt64(a[i+63:i]) + IF count[i+63:i] < 64 + dst[i+63:i] := ZeroExtend64(a[i+63:i] << count[i+63:i]) + ELSE + dst[i+63:i] := 0 + FI ELSE dst[i+63:i] := 0 - FI + FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point + Integer - AVX512DQ - Convert - - - Convert packed double-precision (64-bit) floating-point elements in "a" to packed unsigned 64-bit integers, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + AVX512F + Shift + + + + Shift packed 64-bit integers in "a" left by the amount specified by the corresponding element in "count" while shifting in zeros, and store the results in "dst". FOR j := 0 to 7 i := j*64 - IF k[j] - dst[i+63:i] := Convert_FP64_To_UnsignedInt64(a[i+63:i]) + IF count[i+63:i] < 64 + dst[i+63:i] := ZeroExtend64(a[i+63:i] << count[i+63:i]) ELSE dst[i+63:i] := 0 FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point + Integer - AVX512VL - AVX512DQ - Convert - - Convert packed double-precision (64-bit) floating-point elements in "a" to packed unsigned 64-bit integers, and store the results in "dst". + AVX512F + Shift + + + + + + Shift packed 32-bit integers in "a" right by "count" while shifting in sign bits, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 1 - i := j*64 - dst[i+63:i] := Convert_FP64_To_UnsignedInt64(a[i+63:i]) +FOR j := 0 to 15 + i := j*32 + IF k[j] + IF count[63:0] > 31 + dst[i+31:i] := (a[i+31] ? 0xFFFFFFFF : 0x0) + ELSE + dst[i+31:i] := SignExtend32(a[i+31:i] >> count[63:0]) + FI + ELSE + dst[i+31:i] := src[i+31:i] + FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point + Integer - AVX512VL - AVX512DQ - Convert - - - - Convert packed double-precision (64-bit) floating-point elements in "a" to packed unsigned 64-bit integers, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + AVX512F + Shift + + + + + Shift packed 32-bit integers in "a" right by "count" while shifting in sign bits, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 1 - i := j*64 +FOR j := 0 to 15 + i := j*32 IF k[j] - dst[i+63:i] := Convert_FP64_To_UnsignedInt64(a[i+63:i]) + IF count[63:0] > 31 + dst[i+31:i] := (a[i+31] ? 0xFFFFFFFF : 0x0) + ELSE + dst[i+31:i] := SignExtend32(a[i+31:i] >> count[63:0]) + FI ELSE - dst[i+63:i] := src[i+63:i] + dst[i+31:i] := 0 FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point + Integer - AVX512VL - AVX512DQ - Convert - - - Convert packed double-precision (64-bit) floating-point elements in "a" to packed unsigned 64-bit integers, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + AVX512F + Shift + + + + + Shift packed 32-bit integers in "a" right by "imm8" while shifting in sign bits, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 1 - i := j*64 +FOR j := 0 to 15 + i := j*32 IF k[j] - dst[i+63:i] := Convert_FP64_To_UnsignedInt64(a[i+63:i]) + IF imm8[7:0] > 31 + dst[i+31:i] := (a[i+31] ? 0xFFFFFFFF : 0x0) + ELSE + dst[i+31:i] := SignExtend32(a[i+31:i] >> imm8[7:0]) + FI ELSE - dst[i+63:i] := 0 + dst[i+31:i] := 0 FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point - AVX512VL + + Integer AVX512F - Convert - - - - Convert packed half-precision (16-bit) floating-point elements in "a" to packed single-precision (32-bit) floating-point elements, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Shift + + + + Shift packed 32-bit integers in "a" right by "count" while shifting in sign bits, and store the results in "dst". -FOR j := 0 to 7 +FOR j := 0 to 15 i := j*32 - m := j*16 - IF k[j] - dst[i+31:i] := Convert_FP16_To_FP32(a[m+15:m]) + IF count[63:0] > 31 + dst[i+31:i] := (a[i+31] ? 0xFFFFFFFF : 0x0) ELSE - dst[i+31:i] := src[i+31:i] + dst[i+31:i] := SignExtend32(a[i+31:i] >> count[63:0]) FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point - AVX512VL + + Integer AVX512F - Convert - - - Convert packed half-precision (16-bit) floating-point elements in "a" to packed single-precision (32-bit) floating-point elements, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Shift + + + + + + Shift packed 64-bit integers in "a" right by "count" while shifting in sign bits, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). FOR j := 0 to 7 - i := j*32 - m := j*16 + i := j*64 IF k[j] - dst[i+31:i] := Convert_FP16_To_FP32(a[m+15:m]) + IF count[63:0] > 63 + dst[i+63:i] := (a[i+63] ? 0xFFFFFFFFFFFFFFFF : 0x0) + ELSE + dst[i+63:i] := SignExtend64(a[i+63:i] >> count[63:0]) + FI ELSE - dst[i+31:i] := 0 + dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point - AVX512VL + + Integer AVX512F - Convert - - - - Convert packed half-precision (16-bit) floating-point elements in "a" to packed single-precision (32-bit) floating-point elements, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Shift + + + + + + Shift packed 64-bit integers in "a" right by "imm8" while shifting in sign bits, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*32 - m := j*16 +FOR j := 0 to 7 + i := j*64 IF k[j] - dst[i+31:i] := Convert_FP16_To_FP32(a[m+15:m]) + IF imm8[7:0] > 63 + dst[i+63:i] := (a[i+63] ? 0xFFFFFFFFFFFFFFFF : 0x0) + ELSE + dst[i+63:i] := SignExtend64(a[i+63:i] >> imm8[7:0]) + FI ELSE - dst[i+31:i] := src[i+31:i] + dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point - AVX512VL + + Integer AVX512F - Convert - - - Convert packed half-precision (16-bit) floating-point elements in "a" to packed single-precision (32-bit) floating-point elements, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Shift + + + + + Shift packed 64-bit integers in "a" right by "count" while shifting in sign bits, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*32 - m := j*16 +FOR j := 0 to 7 + i := j*64 IF k[j] - dst[i+31:i] := Convert_FP16_To_FP32(a[m+15:m]) + IF count[63:0] > 63 + dst[i+63:i] := (a[i+63] ? 0xFFFFFFFFFFFFFFFF : 0x0) + ELSE + dst[i+63:i] := SignExtend64(a[i+63:i] >> count[63:0]) + FI ELSE - dst[i+31:i] := 0 + dst[i+63:i] := 0 FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point + Integer - AVX512VL AVX512F - Convert - - - - Convert packed single-precision (32-bit) floating-point elements in "a" to packed 32-bit integers, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Shift + + + + + Shift packed 64-bit integers in "a" right by "imm8" while shifting in sign bits, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). FOR j := 0 to 7 - i := j*32 + i := j*64 IF k[j] - dst[i+31:i] := Convert_FP32_To_Int32(a[i+31:i]) + IF imm8[7:0] > 63 + dst[i+63:i] := (a[i+63] ? 0xFFFFFFFFFFFFFFFF : 0x0) + ELSE + dst[i+63:i] := SignExtend64(a[i+63:i] >> imm8[7:0]) + FI ELSE - dst[i+31:i] := src[i+31:i] + dst[i+63:i] := 0 FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point + Integer - AVX512VL AVX512F - Convert - - - Convert packed single-precision (32-bit) floating-point elements in "a" to packed 32-bit integers, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Shift + + + + Shift packed 64-bit integers in "a" right by "count" while shifting in sign bits, and store the results in "dst". FOR j := 0 to 7 - i := 32*j - IF k[j] - dst[i+31:i] := Convert_FP32_To_Int32(a[i+31:i]) + i := j*64 + IF count[63:0] > 63 + dst[i+63:i] := (a[i+63] ? 0xFFFFFFFFFFFFFFFF : 0x0) ELSE - dst[i+31:i] := 0 + dst[i+63:i] := SignExtend64(a[i+63:i] >> count[63:0]) FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point + Integer - AVX512VL AVX512F - Convert - - - - Convert packed single-precision (32-bit) floating-point elements in "a" to packed 32-bit integers, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Shift + + + + Shift packed 64-bit integers in "a" right by "imm8" while shifting in sign bits, and store the results in "dst". -FOR j := 0 to 3 - i := j*32 - IF k[j] - dst[i+31:i] := Convert_FP32_To_Int32(a[i+31:i]) +FOR j := 0 to 7 + i := j*64 + IF imm8[7:0] > 63 + dst[i+63:i] := (a[i+63] ? 0xFFFFFFFFFFFFFFFF : 0x0) ELSE - dst[i+31:i] := src[i+31:i] + dst[i+63:i] := SignExtend64(a[i+63:i] >> imm8[7:0]) FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point + Integer - AVX512VL AVX512F - Convert - - - Convert packed single-precision (32-bit) floating-point elements in "a" to packed 32-bit integers, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Shift + + + + + Shift packed 32-bit integers in "a" right by the amount specified by the corresponding element in "count" while shifting in sign bits, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := 32*j +FOR j := 0 to 15 + i := j*32 IF k[j] - dst[i+31:i] := Convert_FP32_To_Int32(a[i+31:i]) + IF count[i+31:i] < 32 + dst[i+31:i] := SignExtend32(a[i+31:i] >> count[i+31:i]) + ELSE + dst[i+31:i] := (a[i+31] ? 0xFFFFFFFF : 0) + FI ELSE dst[i+31:i] := 0 - FI + FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point - AVX512VL + + Integer AVX512F - Convert - - - - - Convert packed single-precision (32-bit) floating-point elements in "a" to packed half-precision (16-bit) floating-point elements, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - [round_note] - + Shift + + + + + + Shift packed 64-bit integers in "a" right by the amount specified by the corresponding element in "count" while shifting in sign bits, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). FOR j := 0 to 7 - i := 16*j - l := 32*j + i := j*64 IF k[j] - dst[i+15:i] := Convert_FP32_To_FP16FP(a[l+31:l]) + IF count[i+63:i] < 64 + dst[i+63:i] := SignExtend64(a[i+63:i] >> count[i+63:i]) + ELSE + dst[i+63:i] := (a[i+63] ? 0xFFFFFFFFFFFFFFFF : 0) + FI ELSE - dst[i+15:i] := src[i+15:i] - FI + dst[i+63:i] := src[i+63:i] + FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point - AVX512VL + + Integer AVX512F - Convert - - - - - Convert packed single-precision (32-bit) floating-point elements in "a" to packed half-precision (16-bit) floating-point elements, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - [round_note] - + Shift + + + + + Shift packed 64-bit integers in "a" right by the amount specified by the corresponding element in "count" while shifting in sign bits, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). FOR j := 0 to 7 - i := 16*j - l := 32*j + i := j*64 IF k[j] - dst[i+15:i] := Convert_FP32_To_FP16FP(a[l+31:l]) + IF count[i+63:i] < 64 + dst[i+63:i] := SignExtend64(a[i+63:i] >> count[i+63:i]) + ELSE + dst[i+63:i] := (a[i+63] ? 0xFFFFFFFFFFFFFFFF : 0) + FI ELSE - dst[i+15:i] := src[i+15:i] - FI + dst[i+63:i] := 0 + FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point - AVX512VL + + Integer AVX512F - Convert - - - - Convert packed single-precision (32-bit) floating-point elements in "a" to packed half-precision (16-bit) floating-point elements, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - [round_note] - + Shift + + + + Shift packed 64-bit integers in "a" right by the amount specified by the corresponding element in "count" while shifting in sign bits, and store the results in "dst". FOR j := 0 to 7 - i := 16*j - l := 32*j - IF k[j] - dst[i+15:i] := Convert_FP32_To_FP16FP(a[l+31:l]) + i := j*64 + IF count[i+63:i] < 64 + dst[i+63:i] := SignExtend64(a[i+63:i] >> count[i+63:i]) ELSE - dst[i+15:i] := 0 + dst[i+63:i] := (a[i+63] ? 0xFFFFFFFFFFFFFFFF : 0) FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point - AVX512VL + + Integer AVX512F - Convert - - - - Convert packed single-precision (32-bit) floating-point elements in "a" to packed half-precision (16-bit) floating-point elements, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - [round_note] - + Shift + + + + + + Shift packed 32-bit integers in "a" right by "count" while shifting in zeros, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := 16*j - l := 32*j +FOR j := 0 to 15 + i := j*32 IF k[j] - dst[i+15:i] := Convert_FP32_To_FP16FP(a[l+31:l]) + IF count[63:0] > 31 + dst[i+31:i] := 0 + ELSE + dst[i+31:i] := ZeroExtend32(a[i+31:i] >> count[63:0]) + FI ELSE - dst[i+15:i] := 0 + dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point - AVX512VL + + Integer AVX512F - Convert - - - - - Convert packed single-precision (32-bit) floating-point elements in "a" to packed half-precision (16-bit) floating-point elements, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - [round_note] - + Shift + + + + + Shift packed 32-bit integers in "a" right by "count" while shifting in zeros, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := 16*j - l := 32*j +FOR j := 0 to 15 + i := j*32 IF k[j] - dst[i+15:i] := Convert_FP32_To_FP16FP(a[l+31:l]) + IF count[63:0] > 31 + dst[i+31:i] := 0 + ELSE + dst[i+31:i] := ZeroExtend32(a[i+31:i] >> count[63:0]) + FI ELSE - dst[i+15:i] := src[i+15:i] + dst[i+31:i] := 0 FI ENDFOR -dst[MAX:64] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point - AVX512VL + + Integer AVX512F - Convert - - - - - Convert packed single-precision (32-bit) floating-point elements in "a" to packed half-precision (16-bit) floating-point elements, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - [round_note] - + Shift + + + + + Shift packed 32-bit integers in "a" right by "imm8" while shifting in zeros, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := 16*j - l := 32*j +FOR j := 0 to 15 + i := j*32 IF k[j] - dst[i+15:i] := Convert_FP32_To_FP16FP(a[l+31:l]) + IF imm8[7:0] > 31 + dst[i+31:i] := 0 + ELSE + dst[i+31:i] := ZeroExtend32(a[i+31:i] >> imm8[7:0]) + FI ELSE - dst[i+15:i] := src[i+15:i] + dst[i+31:i] := 0 FI ENDFOR -dst[MAX:64] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point - AVX512VL + + Integer AVX512F - Convert - - - - Convert packed single-precision (32-bit) floating-point elements in "a" to packed half-precision (16-bit) floating-point elements, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - [round_note] - + Shift + + + + Shift packed 32-bit integers in "a" right by "count" while shifting in zeros, and store the results in "dst". -FOR j := 0 to 3 - i := 16*j - l := 32*j - IF k[j] - dst[i+15:i] := Convert_FP32_To_FP16FP(a[l+31:l]) +FOR j := 0 to 15 + i := j*32 + IF count[63:0] > 31 + dst[i+31:i] := 0 ELSE - dst[i+15:i] := 0 + dst[i+31:i] := ZeroExtend32(a[i+31:i] >> count[63:0]) FI ENDFOR -dst[MAX:64] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point - AVX512VL + + Integer AVX512F - Convert - - - - Convert packed single-precision (32-bit) floating-point elements in "a" to packed half-precision (16-bit) floating-point elements, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - [round_note] - + Shift + + + + + + Shift packed 64-bit integers in "a" right by "count" while shifting in zeros, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := 16*j - l := 32*j +FOR j := 0 to 7 + i := j*64 IF k[j] - dst[i+15:i] := Convert_FP32_To_FP16FP(a[l+31:l]) + IF count[63:0] > 63 + dst[i+63:i] := 0 + ELSE + dst[i+63:i] := ZeroExtend64(a[i+63:i] >> count[63:0]) + FI ELSE - dst[i+15:i] := 0 + dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:64] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point + Integer - AVX512VL - AVX512DQ - Convert - - Convert packed single-precision (32-bit) floating-point elements in "a" to packed 64-bit integers, and store the results in "dst". + AVX512F + Shift + + + + + + Shift packed 64-bit integers in "a" right by "imm8" while shifting in zeros, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 3 +FOR j := 0 to 7 i := j*64 - l := j*32 - dst[i+63:i] := Convert_FP32_To_Int64(a[l+31:l]) + IF k[j] + IF imm8[7:0] > 63 + dst[i+63:i] := 0 + ELSE + dst[i+63:i] := ZeroExtend64(a[i+63:i] >> imm8[7:0]) + FI + ELSE + dst[i+63:i] := src[i+63:i] + FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point + Integer - AVX512VL - AVX512DQ - Convert - - - - Convert packed single-precision (32-bit) floating-point elements in "a" to packed 64-bit integers, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + AVX512F + Shift + + + + + Shift packed 64-bit integers in "a" right by "count" while shifting in zeros, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 3 +FOR j := 0 to 7 i := j*64 - l := j*32 IF k[j] - dst[i+63:i] := Convert_FP32_To_Int64(a[l+31:l]) + IF count[63:0] > 63 + dst[i+63:i] := 0 + ELSE + dst[i+63:i] := ZeroExtend64(a[i+63:i] >> count[63:0]) + FI ELSE - dst[i+63:i] := src[i+63:i] + dst[i+63:i] := 0 FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point + Integer - AVX512VL - AVX512DQ - Convert - - - Convert packed single-precision (32-bit) floating-point elements in "a" to packed 64-bit integers, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + AVX512F + Shift + + + + + Shift packed 64-bit integers in "a" right by "imm8" while shifting in zeros, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 3 +FOR j := 0 to 7 i := j*64 - l := j*32 IF k[j] - dst[i+63:i] := Convert_FP32_To_Int64(a[l+31:l]) + IF imm8[7:0] > 63 + dst[i+63:i] := 0 + ELSE + dst[i+63:i] := ZeroExtend64(a[i+63:i] >> imm8[7:0]) + FI ELSE dst[i+63:i] := 0 FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point + Integer - AVX512DQ - Convert - - - Convert packed single-precision (32-bit) floating-point elements in "a" to packed 64-bit integers, and store the results in "dst". - [round_note] - + AVX512F + Shift + + + + Shift packed 64-bit integers in "a" right by "count" while shifting in zeros, and store the results in "dst". FOR j := 0 to 7 i := j*64 - l := j*32 - dst[i+63:i] := Convert_FP32_To_Int64(a[l+31:l]) + IF count[63:0] > 63 + dst[i+63:i] := 0 + ELSE + dst[i+63:i] := ZeroExtend64(a[i+63:i] >> count[63:0]) + FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point + Integer - AVX512DQ - Convert - - Convert packed single-precision (32-bit) floating-point elements in "a" to packed 64-bit integers, and store the results in "dst". + AVX512F + Shift + + + + Shift packed 64-bit integers in "a" right by "imm8" while shifting in zeros, and store the results in "dst". FOR j := 0 to 7 i := j*64 - l := j*32 - dst[i+63:i] := Convert_FP32_To_Int64(a[l+31:l]) + IF imm8[7:0] > 63 + dst[i+63:i] := 0 + ELSE + dst[i+63:i] := ZeroExtend64(a[i+63:i] >> imm8[7:0]) + FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point + Integer - AVX512DQ - Convert - - - - - Convert packed single-precision (32-bit) floating-point elements in "a" to packed 64-bit integers, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - [round_note] - + AVX512F + Shift + + + + + Shift packed 32-bit integers in "a" right by the amount specified by the corresponding element in "count" while shifting in zeros, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := j*64 - l := j*32 +FOR j := 0 to 15 + i := j*32 IF k[j] - dst[i+63:i] := Convert_FP32_To_Int64(a[l+31:l]) + IF count[i+31:i] < 32 + dst[i+31:i] := ZeroExtend32(a[i+31:i] >> count[i+31:i]) + ELSE + dst[i+31:i] := 0 + FI ELSE - dst[i+63:i] := src[i+63:i] - FI + dst[i+31:i] := 0 + FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point + Integer - AVX512DQ - Convert - - - - Convert packed single-precision (32-bit) floating-point elements in "a" to packed 64-bit integers, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + AVX512F + Shift + + + + + + Shift packed 64-bit integers in "a" right by the amount specified by the corresponding element in "count" while shifting in zeros, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). FOR j := 0 to 7 i := j*64 - l := j*32 IF k[j] - dst[i+63:i] := Convert_FP32_To_Int64(a[l+31:l]) + IF count[i+63:i] < 64 + dst[i+63:i] := ZeroExtend64(a[i+63:i] >> count[i+63:i]) + ELSE + dst[i+63:i] := 0 + FI ELSE dst[i+63:i] := src[i+63:i] - FI + FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point + Integer - AVX512DQ - Convert - - - - Convert packed single-precision (32-bit) floating-point elements in "a" to packed 64-bit integers, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - [round_note] - + AVX512F + Shift + + + + + Shift packed 64-bit integers in "a" right by the amount specified by the corresponding element in "count" while shifting in zeros, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). FOR j := 0 to 7 i := j*64 - l := j*32 IF k[j] - dst[i+63:i] := Convert_FP32_To_Int64(a[l+31:l]) + IF count[i+63:i] < 64 + dst[i+63:i] := ZeroExtend64(a[i+63:i] >> count[i+63:i]) + ELSE + dst[i+63:i] := 0 + FI ELSE dst[i+63:i] := 0 - FI + FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point + Integer - AVX512DQ - Convert - - - Convert packed single-precision (32-bit) floating-point elements in "a" to packed 64-bit integers, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + AVX512F + Shift + + + + Shift packed 64-bit integers in "a" right by the amount specified by the corresponding element in "count" while shifting in zeros, and store the results in "dst". FOR j := 0 to 7 i := j*64 - l := j*32 - IF k[j] - dst[i+63:i] := Convert_FP32_To_Int64(a[l+31:l]) + IF count[i+63:i] < 64 + dst[i+63:i] := ZeroExtend64(a[i+63:i] >> count[i+63:i]) ELSE dst[i+63:i] := 0 FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point + Integer - AVX512VL - AVX512DQ - Convert - - Convert packed single-precision (32-bit) floating-point elements in "a" to packed 64-bit integers, and store the results in "dst". + AVX512F + Arithmetic + + + + + Subtract packed 32-bit integers in "b" from packed 32-bit integers in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 1 - i := j*64 - l := j*32 - dst[i+63:i] := Convert_FP32_To_Int64(a[l+31:l]) +FOR j := 0 to 15 + i := j*32 + IF k[j] + dst[i+31:i] := a[i+31:i] - b[i+31:i] + ELSE + dst[i+31:i] := 0 + FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point + Integer - AVX512VL - AVX512DQ - Convert - - - - Convert packed single-precision (32-bit) floating-point elements in "a" to packed 64-bit integers, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + AVX512F + Arithmetic + + + + + + Subtract packed 64-bit integers in "b" from packed 64-bit integers in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 1 +FOR j := 0 to 7 i := j*64 - l := j*32 IF k[j] - dst[i+63:i] := Convert_FP32_To_Int64(a[l+31:l]) + dst[i+63:i] := a[i+63:i] - b[i+63:i] ELSE dst[i+63:i] := src[i+63:i] - FI + FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point + Integer - AVX512VL - AVX512DQ - Convert - - - Convert packed single-precision (32-bit) floating-point elements in "a" to packed 64-bit integers, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + AVX512F + Arithmetic + + + + + Subtract packed 64-bit integers in "b" from packed 64-bit integers in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 1 +FOR j := 0 to 7 i := j*64 - l := j*32 IF k[j] - dst[i+63:i] := Convert_FP32_To_Int64(a[l+31:l]) + dst[i+63:i] := a[i+63:i] - b[i+63:i] ELSE dst[i+63:i] := 0 - FI + FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point + Integer - AVX512VL AVX512F - Convert - - Convert packed single-precision (32-bit) floating-point elements in "a" to packed unsigned 32-bit integers, and store the results in "dst". + Arithmetic + + + + Subtract packed 64-bit integers in "b" from packed 64-bit integers in "a", and store the results in "dst". FOR j := 0 to 7 - i := 32*j - dst[i+31:i] := Convert_FP32_To_UnsignedInt32(a[i+31:i]) + i := j*64 + dst[i+63:i] := a[i+63:i] - b[i+63:i] ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point + Integer - AVX512VL AVX512F - Convert - - - - Convert packed single-precision (32-bit) floating-point elements in "a" to packed unsigned 32-bit integers, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Logical + + + + + + + Bitwise ternary logic that provides the capability to implement any three-operand binary function; the specific binary function is specified by value in "imm8". For each bit in each packed 32-bit integer, the corresponding bit from "src", "a", and "b" are used to form a 3 bit index into "imm8", and the value at that bit in "imm8" is written to the corresponding bit in "dst" using writemask "k" at 32-bit granularity (32-bit elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := 32*j +FOR j := 0 to 15 + i := j*32 IF k[j] - dst[i+31:i] := Convert_FP32_To_UnsignedInt32(a[i+31:i]) + FOR h := 0 to 31 + index[2:0] := (src[i+h] << 2) OR (a[i+h] << 1) OR b[i+h] + dst[i+h] := imm8[index[2:0]] + ENDFOR ELSE dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point + Integer - AVX512VL AVX512F - Convert - - - Convert packed single-precision (32-bit) floating-point elements in "a" to packed unsigned 32-bit integers, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Logical + + + + + + + Bitwise ternary logic that provides the capability to implement any three-operand binary function; the specific binary function is specified by value in "imm8". For each bit in each packed 32-bit integer, the corresponding bit from "a", "b", and "c" are used to form a 3 bit index into "imm8", and the value at that bit in "imm8" is written to the corresponding bit in "dst" using zeromask "k" at 32-bit granularity (32-bit elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := 32*j +FOR j := 0 to 15 + i := j*32 IF k[j] - dst[i+31:i] := Convert_FP32_To_UnsignedInt32(a[i+31:i]) + FOR h := 0 to 31 + index[2:0] := (a[i+h] << 2) OR (b[i+h] << 1) OR c[i+h] + dst[i+h] := imm8[index[2:0]] + ENDFOR ELSE dst[i+31:i] := 0 FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point + Integer - AVX512VL AVX512F - Convert - - Convert packed single-precision (32-bit) floating-point elements in "a" to packed unsigned 32-bit integers, and store the results in "dst". + Logical + + + + + + Bitwise ternary logic that provides the capability to implement any three-operand binary function; the specific binary function is specified by value in "imm8". For each bit in each packed 32-bit integer, the corresponding bit from "a", "b", and "c" are used to form a 3 bit index into "imm8", and the value at that bit in "imm8" is written to the corresponding bit in "dst". -FOR j := 0 to 3 - i := 32*j - dst[i+31:i] := Convert_FP32_To_UnsignedInt32(a[i+31:i]) +FOR j := 0 to 15 + i := j*32 + FOR h := 0 to 31 + index[2:0] := (a[i+h] << 2) OR (b[i+h] << 1) OR c[i+h] + dst[i+h] := imm8[index[2:0]] + ENDFOR ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point + Integer - AVX512VL AVX512F - Convert - - - - Convert packed single-precision (32-bit) floating-point elements in "a" to packed unsigned 32-bit integers, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Logical + + + + + + + Bitwise ternary logic that provides the capability to implement any three-operand binary function; the specific binary function is specified by value in "imm8". For each bit in each packed 64-bit integer, the corresponding bit from "src", "a", and "b" are used to form a 3 bit index into "imm8", and the value at that bit in "imm8" is written to the corresponding bit in "dst" using writemask "k" at 64-bit granularity (64-bit elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := 32*j +FOR j := 0 to 7 + i := j*64 IF k[j] - dst[i+31:i] := Convert_FP32_To_UnsignedInt32(a[i+31:i]) + FOR h := 0 to 63 + index[2:0] := (src[i+h] << 2) OR (a[i+h] << 1) OR b[i+h] + dst[i+h] := imm8[index[2:0]] + ENDFOR ELSE - dst[i+31:i] := src[i+31:i] + dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point + Integer - AVX512VL AVX512F - Convert - - - Convert packed single-precision (32-bit) floating-point elements in "a" to packed unsigned 32-bit integers, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Logical + + + + + + + Bitwise ternary logic that provides the capability to implement any three-operand binary function; the specific binary function is specified by value in "imm8". For each bit in each packed 64-bit integer, the corresponding bit from "a", "b", and "c" are used to form a 3 bit index into "imm8", and the value at that bit in "imm8" is written to the corresponding bit in "dst" using zeromask "k" at 64-bit granularity (64-bit elements are zeroed out when the corresponding mask bit is not set). + +FOR j := 0 to 7 + i := j*64 + IF k[j] + FOR h := 0 to 63 + index[2:0] := (a[i+h] << 2) OR (b[i+h] << 1) OR c[i+h] + dst[i+h] := imm8[index[2:0]] + ENDFOR + ELSE + dst[i+63:i] := 0 + FI +ENDFOR +dst[MAX:512] := 0 + + +
immintrin.h
+ + + Integer + AVX512F + Logical + + + + + + Bitwise ternary logic that provides the capability to implement any three-operand binary function; the specific binary function is specified by value in "imm8". For each bit in each packed 64-bit integer, the corresponding bit from "a", "b", and "c" are used to form a 3 bit index into "imm8", and the value at that bit in "imm8" is written to the corresponding bit in "dst". + +FOR j := 0 to 7 + i := j*64 + FOR h := 0 to 63 + index[2:0] := (a[i+h] << 2) OR (b[i+h] << 1) OR c[i+h] + dst[i+h] := imm8[index[2:0]] + ENDFOR +ENDFOR +dst[MAX:512] := 0 + + +
immintrin.h
+ + + Integer + Mask + AVX512F + Logical + + + + + Compute the bitwise AND of packed 64-bit integers in "a" and "b", producing intermediate 64-bit values, and set the corresponding bit in result mask "k" (subject to writemask "k") if the intermediate value is non-zero. -FOR j := 0 to 3 - i := 32*j - IF k[j] - dst[i+31:i] := Convert_FP32_To_UnsignedInt32(a[i+31:i]) +FOR j := 0 to 7 + i := j*64 + IF k1[j] + k[j] := ((a[i+63:i] AND b[i+63:i]) != 0) ? 1 : 0 ELSE - dst[i+31:i] := 0 + k[j] := 0 FI ENDFOR -dst[MAX:128] := 0 +k[MAX:8] := 0 - +
immintrin.h
 - - Floating Point + Integer - AVX512VL - AVX512DQ - Convert - - Convert packed single-precision (32-bit) floating-point elements in "a" to packed unsigned 64-bit integers, and store the results in "dst". + Mask + AVX512F + Logical + + + + Compute the bitwise AND of packed 64-bit integers in "a" and "b", producing intermediate 64-bit values, and set the corresponding bit in result mask "k" if the intermediate value is non-zero. -FOR j := 0 to 3 +FOR j := 0 to 7 i := j*64 - l := j*32 - dst[i+63:i] := Convert_FP32_To_UnsignedInt64(a[l+31:l]) + k[j] := ((a[i+63:i] AND b[i+63:i]) != 0) ? 1 : 0 ENDFOR -dst[MAX:256] := 0 +k[MAX:8] := 0 - +
immintrin.h
 - - Floating Point + Integer - AVX512VL - AVX512DQ - Convert - - - - Convert packed single-precision (32-bit) floating-point elements in "a" to packed unsigned 64-bit integers, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - + Mask + AVX512F + Logical + + + + + Compute the bitwise NAND of packed 32-bit integers in "a" and "b", producing intermediate 32-bit values, and set the corresponding bit in result mask "k" (subject to writemask "k") if the intermediate value is zero. -FOR j := 0 to 3 - i := j*64 - l := j*32 - IF k[j] - dst[i+63:i] := Convert_FP32_To_UnsignedInt64(a[l+31:l]) +FOR j := 0 to 15 + i := j*32 + IF k1[j] + k[j] := ((a[i+31:i] AND b[i+31:i]) == 0) ? 1 : 0 ELSE - dst[i+63:i] := src[i+63:i] + k[j] := 0 FI ENDFOR -dst[MAX:256] := 0 +k[MAX:16] := 0 - +
immintrin.h
 - - Floating Point + Integer - AVX512VL - AVX512DQ - Convert - - - Convert packed single-precision (32-bit) floating-point elements in "a" to packed unsigned 64-bit integers, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Mask + AVX512F + Logical + + + + Compute the bitwise NAND of packed 32-bit integers in "a" and "b", producing intermediate 32-bit values, and set the corresponding bit in result mask "k" if the intermediate value is zero. -FOR j := 0 to 3 - i := j*64 - l := j*32 - IF k[j] - dst[i+63:i] := Convert_FP32_To_UnsignedInt64(a[l+31:l]) - ELSE - dst[i+63:i] := 0 - FI +FOR j := 0 to 15 + i := j*32 + k[j] := ((a[i+31:i] AND b[i+31:i]) == 0) ? 1 : 0 ENDFOR -dst[MAX:256] := 0 +k[MAX:16] := 0 - +
immintrin.h
 - - Floating Point + Integer - AVX512DQ - Convert - - - Convert packed single-precision (32-bit) floating-point elements in "a" to packed unsigned 64-bit integers, and store the results in "dst". - [round_note] - + Mask + AVX512F + Logical + + + + + Compute the bitwise NAND of packed 64-bit integers in "a" and "b", producing intermediate 64-bit values, and set the corresponding bit in result mask "k" (subject to writemask "k") if the intermediate value is zero. FOR j := 0 to 7 i := j*64 - l := j*32 - dst[i+63:i] := Convert_FP32_To_UnsignedInt64(a[l+31:l]) + IF k1[j] + k[j] := ((a[i+63:i] AND b[i+63:i]) == 0) ? 1 : 0 + ELSE + k[j] := 0 + FI ENDFOR -dst[MAX:512] := 0 +k[MAX:8] := 0 - +
immintrin.h
 - - Floating Point + Integer - AVX512DQ - Convert - - Convert packed single-precision (32-bit) floating-point elements in "a" to packed unsigned 64-bit integers, and store the results in "dst". + Mask + AVX512F + Logical + + + + Compute the bitwise NAND of packed 64-bit integers in "a" and "b", producing intermediate 64-bit values, and set the corresponding bit in result mask "k" if the intermediate value is zero. FOR j := 0 to 7 i := j*64 - l := j*32 - dst[i+63:i] := Convert_FP32_To_UnsignedInt64(a[l+31:l]) + k[j] := ((a[i+63:i] AND b[i+63:i]) == 0) ? 1 : 0 ENDFOR -dst[MAX:512] := 0 +k[MAX:8] := 0 - +
immintrin.h
 - - Floating Point + Integer - AVX512DQ - Convert - - - - - Convert packed single-precision (32-bit) floating-point elements in "a" to packed unsigned 64-bit integers, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - [round_note] - - -FOR j := 0 to 7 - i := j*64 - l := j*32 + AVX512F + Swizzle + + + + + + Unpack and interleave 32-bit integers from the high half of each 128-bit lane in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +DEFINE INTERLEAVE_HIGH_DWORDS(src1[127:0], src2[127:0]) { + dst[31:0] := src1[95:64] + dst[63:32] := src2[95:64] + dst[95:64] := src1[127:96] + dst[127:96] := src2[127:96] + RETURN dst[127:0] +} +tmp_dst[127:0] := INTERLEAVE_HIGH_DWORDS(a[127:0], b[127:0]) +tmp_dst[255:128] := INTERLEAVE_HIGH_DWORDS(a[255:128], b[255:128]) +tmp_dst[383:256] := INTERLEAVE_HIGH_DWORDS(a[383:256], b[383:256]) +tmp_dst[511:384] := INTERLEAVE_HIGH_DWORDS(a[511:384], b[511:384]) +FOR j := 0 to 15 + i := j*32 IF k[j] - dst[i+63:i] := Convert_FP32_To_UnsignedInt64(a[l+31:l]) + dst[i+31:i] := tmp_dst[i+31:i] ELSE - dst[i+63:i] := src[i+63:i] + dst[i+31:i] := src[i+31:i] FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point + Integer - AVX512DQ - Convert - - - - Convert packed single-precision (32-bit) floating-point elements in "a" to packed unsigned 64-bit integers, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - + AVX512F + Swizzle + + + + + Unpack and interleave 32-bit integers from the high half of each 128-bit lane in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := j*64 - l := j*32 +DEFINE INTERLEAVE_HIGH_DWORDS(src1[127:0], src2[127:0]) { + dst[31:0] := src1[95:64] + dst[63:32] := src2[95:64] + dst[95:64] := src1[127:96] + dst[127:96] := src2[127:96] + RETURN dst[127:0] +} +tmp_dst[127:0] := INTERLEAVE_HIGH_DWORDS(a[127:0], b[127:0]) +tmp_dst[255:128] := INTERLEAVE_HIGH_DWORDS(a[255:128], b[255:128]) +tmp_dst[383:256] := INTERLEAVE_HIGH_DWORDS(a[383:256], b[383:256]) +tmp_dst[511:384] := INTERLEAVE_HIGH_DWORDS(a[511:384], b[511:384]) +FOR j := 0 to 15 + i := j*32 IF k[j] - dst[i+63:i] := Convert_FP32_To_UnsignedInt64(a[l+31:l]) + dst[i+31:i] := tmp_dst[i+31:i] ELSE - dst[i+63:i] := src[i+63:i] + dst[i+31:i] := 0 FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point + Integer - AVX512DQ - Convert - - - - Convert packed single-precision (32-bit) floating-point elements in "a" to packed unsigned 64-bit integers, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - [round_note] - + AVX512F + Swizzle + + + + Unpack and interleave 32-bit integers from the high half of each 128-bit lane in "a" and "b", and store the results in "dst". +DEFINE INTERLEAVE_HIGH_DWORDS(src1[127:0], src2[127:0]) { + dst[31:0] := src1[95:64] + dst[63:32] := src2[95:64] + dst[95:64] := src1[127:96] + dst[127:96] := src2[127:96] + RETURN dst[127:0] +} +dst[127:0] := INTERLEAVE_HIGH_DWORDS(a[127:0], b[127:0]) +dst[255:128] := INTERLEAVE_HIGH_DWORDS(a[255:128], b[255:128]) +dst[383:256] := INTERLEAVE_HIGH_DWORDS(a[383:256], b[383:256]) +dst[511:384] := INTERLEAVE_HIGH_DWORDS(a[511:384], b[511:384]) +dst[MAX:512] := 0 + + +
immintrin.h
+ + + Integer + AVX512F + Swizzle + + + + + + Unpack and interleave 64-bit integers from the high half of each 128-bit lane in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +DEFINE INTERLEAVE_HIGH_QWORDS(src1[127:0], src2[127:0]) { + dst[63:0] := src1[127:64] + dst[127:64] := src2[127:64] + RETURN dst[127:0] +} +tmp_dst[127:0] := INTERLEAVE_HIGH_QWORDS(a[127:0], b[127:0]) +tmp_dst[255:128] := INTERLEAVE_HIGH_QWORDS(a[255:128], b[255:128]) +tmp_dst[383:256] := INTERLEAVE_HIGH_QWORDS(a[383:256], b[383:256]) +tmp_dst[511:384] := INTERLEAVE_HIGH_QWORDS(a[511:384], b[511:384]) FOR j := 0 to 7 i := j*64 - l := j*32 IF k[j] - dst[i+63:i] := Convert_FP32_To_UnsignedInt64(a[l+31:l]) + dst[i+63:i] := tmp_dst[i+63:i] ELSE - dst[i+63:i] := 0 + dst[i+63:i] := src[i+63:i] FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point + Integer - AVX512DQ - Convert - - - Convert packed single-precision (32-bit) floating-point elements in "a" to packed unsigned 64-bit integers, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + AVX512F + Swizzle + + + + + Unpack and interleave 64-bit integers from the high half of each 128-bit lane in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). +DEFINE INTERLEAVE_HIGH_QWORDS(src1[127:0], src2[127:0]) { + dst[63:0] := src1[127:64] + dst[127:64] := src2[127:64] + RETURN dst[127:0] +} +tmp_dst[127:0] := INTERLEAVE_HIGH_QWORDS(a[127:0], b[127:0]) +tmp_dst[255:128] := INTERLEAVE_HIGH_QWORDS(a[255:128], b[255:128]) +tmp_dst[383:256] := INTERLEAVE_HIGH_QWORDS(a[383:256], b[383:256]) +tmp_dst[511:384] := INTERLEAVE_HIGH_QWORDS(a[511:384], b[511:384]) FOR j := 0 to 7 i := j*64 - l := j*32 IF k[j] - dst[i+63:i] := Convert_FP32_To_UnsignedInt64(a[l+31:l]) + dst[i+63:i] := tmp_dst[i+63:i] ELSE dst[i+63:i] := 0 FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point + Integer - AVX512VL - AVX512DQ - Convert - - Convert packed single-precision (32-bit) floating-point elements in "a" to packed unsigned 64-bit integers, and store the results in "dst". + AVX512F + Swizzle + + + + Unpack and interleave 64-bit integers from the high half of each 128-bit lane in "a" and "b", and store the results in "dst". -FOR j := 0 to 1 - i := j*64 - l := j*32 - dst[i+63:i] := Convert_FP32_To_UnsignedInt64(a[l+31:l]) -ENDFOR -dst[MAX:128] := 0 +DEFINE INTERLEAVE_HIGH_QWORDS(src1[127:0], src2[127:0]) { + dst[63:0] := src1[127:64] + dst[127:64] := src2[127:64] + RETURN dst[127:0] +} +dst[127:0] := INTERLEAVE_HIGH_QWORDS(a[127:0], b[127:0]) +dst[255:128] := INTERLEAVE_HIGH_QWORDS(a[255:128], b[255:128]) +dst[383:256] := INTERLEAVE_HIGH_QWORDS(a[383:256], b[383:256]) +dst[511:384] := INTERLEAVE_HIGH_QWORDS(a[511:384], b[511:384]) +dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point + Integer - AVX512VL - AVX512DQ - Convert - - - - Convert packed single-precision (32-bit) floating-point elements in "a" to packed unsigned 64-bit integers, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - - -FOR j := 0 to 1 - i := j*64 - l := j*32 + AVX512F + Swizzle + + + + + + Unpack and interleave 32-bit integers from the low half of each 128-bit lane in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +DEFINE INTERLEAVE_DWORDS(src1[127:0], src2[127:0]) { + dst[31:0] := src1[31:0] + dst[63:32] := src2[31:0] + dst[95:64] := src1[63:32] + dst[127:96] := src2[63:32] + RETURN dst[127:0] +} +tmp_dst[127:0] := INTERLEAVE_DWORDS(a[127:0], b[127:0]) +tmp_dst[255:128] := INTERLEAVE_DWORDS(a[255:128], b[255:128]) +tmp_dst[383:256] := INTERLEAVE_DWORDS(a[383:256], b[383:256]) +tmp_dst[511:384] := INTERLEAVE_DWORDS(a[511:384], b[511:384]) +FOR j := 0 to 15 + i := j*32 IF k[j] - dst[i+63:i] := Convert_FP32_To_UnsignedInt64(a[l+31:l]) + dst[i+31:i] := tmp_dst[i+31:i] ELSE - dst[i+63:i] := src[i+63:i] + dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point + Integer - AVX512VL - AVX512DQ - Convert - - - Convert packed single-precision (32-bit) floating-point elements in "a" to packed unsigned 64-bit integers, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + AVX512F + Swizzle + + + + + Unpack and interleave 32-bit integers from the low half of each 128-bit lane in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 1 - i := j*64 - l := j*32 +DEFINE INTERLEAVE_DWORDS(src1[127:0], src2[127:0]) { + dst[31:0] := src1[31:0] + dst[63:32] := src2[31:0] + dst[95:64] := src1[63:32] + dst[127:96] := src2[63:32] + RETURN dst[127:0] +} +tmp_dst[127:0] := INTERLEAVE_DWORDS(a[127:0], b[127:0]) +tmp_dst[255:128] := INTERLEAVE_DWORDS(a[255:128], b[255:128]) +tmp_dst[383:256] := INTERLEAVE_DWORDS(a[383:256], b[383:256]) +tmp_dst[511:384] := INTERLEAVE_DWORDS(a[511:384], b[511:384]) +FOR j := 0 to 15 + i := j*32 IF k[j] - dst[i+63:i] := Convert_FP32_To_UnsignedInt64(a[l+31:l]) + dst[i+31:i] := tmp_dst[i+31:i] ELSE - dst[i+63:i] := 0 + dst[i+31:i] := 0 FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point + Integer - AVX512VL - AVX512DQ - Convert - - Convert packed 64-bit integers in "a" to packed double-precision (64-bit) floating-point elements, and store the results in "dst". + AVX512F + Swizzle + + + + Unpack and interleave 32-bit integers from the low half of each 128-bit lane in "a" and "b", and store the results in "dst". -FOR j := 0 to 3 - i := j*64 - dst[i+63:i] := Convert_Int64_To_FP64(a[i+63:i]) -ENDFOR -dst[MAX:256] := 0 +DEFINE INTERLEAVE_DWORDS(src1[127:0], src2[127:0]) { + dst[31:0] := src1[31:0] + dst[63:32] := src2[31:0] + dst[95:64] := src1[63:32] + dst[127:96] := src2[63:32] + RETURN dst[127:0] +} +dst[127:0] := INTERLEAVE_DWORDS(a[127:0], b[127:0]) +dst[255:128] := INTERLEAVE_DWORDS(a[255:128], b[255:128]) +dst[383:256] := INTERLEAVE_DWORDS(a[383:256], b[383:256]) +dst[511:384] := INTERLEAVE_DWORDS(a[511:384], b[511:384]) +dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point + Integer - AVX512VL - AVX512DQ - Convert - - - - Convert packed 64-bit integers in "a" to packed double-precision (64-bit) floating-point elements, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -FOR j := 0 to 3 + AVX512F + Swizzle + + + + + + Unpack and interleave 64-bit integers from the low half of each 128-bit lane in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +DEFINE INTERLEAVE_QWORDS(src1[127:0], src2[127:0]) { + dst[63:0] := src1[63:0] + dst[127:64] := src2[63:0] + RETURN dst[127:0] +} +tmp_dst[127:0] := INTERLEAVE_QWORDS(a[127:0], b[127:0]) +tmp_dst[255:128] := INTERLEAVE_QWORDS(a[255:128], b[255:128]) +tmp_dst[383:256] := INTERLEAVE_QWORDS(a[383:256], b[383:256]) +tmp_dst[511:384] := INTERLEAVE_QWORDS(a[511:384], b[511:384]) +FOR j := 0 to 7 i := j*64 IF k[j] - dst[i+63:i] := Convert_Int64_To_FP64(a[i+63:i]) + dst[i+63:i] := tmp_dst[i+63:i] ELSE dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point + Integer - AVX512VL - AVX512DQ - Convert - - - Convert packed 64-bit integers in "a" to packed double-precision (64-bit) floating-point elements, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + AVX512F + Swizzle + + + + + Unpack and interleave 64-bit integers from the low half of each 128-bit lane in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 3 +DEFINE INTERLEAVE_QWORDS(src1[127:0], src2[127:0]) { + dst[63:0] := src1[63:0] + dst[127:64] := src2[63:0] + RETURN dst[127:0] +} +tmp_dst[127:0] := INTERLEAVE_QWORDS(a[127:0], b[127:0]) +tmp_dst[255:128] := INTERLEAVE_QWORDS(a[255:128], b[255:128]) +tmp_dst[383:256] := INTERLEAVE_QWORDS(a[383:256], b[383:256]) +tmp_dst[511:384] := INTERLEAVE_QWORDS(a[511:384], b[511:384]) +FOR j := 0 to 7 i := j*64 IF k[j] - dst[i+63:i] := Convert_Int64_To_FP64(a[i+63:i]) + dst[i+63:i] := tmp_dst[i+63:i] ELSE dst[i+63:i] := 0 FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point + Integer - AVX512DQ - Convert - - - Convert packed 64-bit integers in "a" to packed double-precision (64-bit) floating-point elements, and store the results in "dst". - [round_note] - + AVX512F + Swizzle + + + + Unpack and interleave 64-bit integers from the low half of each 128-bit lane in "a" and "b", and store the results in "dst". -FOR j := 0 to 7 - i := j*64 - dst[i+63:i] := Convert_Int64_To_FP64(a[i+63:i]) -ENDFOR +DEFINE INTERLEAVE_QWORDS(src1[127:0], src2[127:0]) { + dst[63:0] := src1[63:0] + dst[127:64] := src2[63:0] + RETURN dst[127:0] +} +dst[127:0] := INTERLEAVE_QWORDS(a[127:0], b[127:0]) +dst[255:128] := INTERLEAVE_QWORDS(a[255:128], b[255:128]) +dst[383:256] := INTERLEAVE_QWORDS(a[383:256], b[383:256]) +dst[511:384] := INTERLEAVE_QWORDS(a[511:384], b[511:384]) dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point + Integer - AVX512DQ - Convert - - Convert packed 64-bit integers in "a" to packed double-precision (64-bit) floating-point elements, and store the results in "dst". + AVX512F + Logical + + + + + Compute the bitwise XOR of packed 32-bit integers in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := j*64 - dst[i+63:i] := Convert_Int64_To_FP64(a[i+63:i]) +FOR j := 0 to 15 + i := j*32 + IF k[j] + dst[i+31:i] := a[i+31:i] XOR b[i+31:i] + ELSE + dst[i+31:i] := 0 + FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point + Integer - AVX512DQ - Convert - - - - - Convert packed 64-bit integers in "a" to packed double-precision (64-bit) floating-point elements, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - [round_note] - + AVX512F + Logical + + + + + Compute the bitwise XOR of packed 64-bit integers in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). FOR j := 0 to 7 i := j*64 IF k[j] - dst[i+63:i] := Convert_Int64_To_FP64(a[i+63:i]) + dst[i+63:i] := a[i+63:i] XOR b[i+63:i] ELSE - dst[i+63:i] := src[i+63:i] + dst[i+63:i] := 0 FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - Integer - AVX512DQ - Convert - - - - Convert packed 64-bit integers in "a" to packed double-precision (64-bit) floating-point elements, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + AVX512F + Elementary Math Functions + + + + + Compute the approximate reciprocal of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). The maximum relative error for this approximation is less than 2^-14. FOR j := 0 to 7 i := j*64 IF k[j] - dst[i+63:i] := Convert_Int64_To_FP64(a[i+63:i]) + dst[i+63:i] := (1.0 / a[i+63:i]) ELSE dst[i+63:i] := src[i+63:i] FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - Integer - AVX512DQ - Convert - - - - Convert packed 64-bit integers in "a" to packed double-precision (64-bit) floating-point elements, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - [round_note] - + AVX512F + Elementary Math Functions + + + + Compute the approximate reciprocal of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). The maximum relative error for this approximation is less than 2^-14. FOR j := 0 to 7 i := j*64 IF k[j] - dst[i+63:i] := Convert_Int64_To_FP64(a[i+63:i]) + dst[i+63:i] := (1.0 / a[i+63:i]) ELSE dst[i+63:i] := 0 FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - Integer - AVX512DQ - Convert - - - Convert packed 64-bit integers in "a" to packed double-precision (64-bit) floating-point elements, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + AVX512F + Elementary Math Functions + + + Compute the approximate reciprocal of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst". The maximum relative error for this approximation is less than 2^-14. FOR j := 0 to 7 i := j*64 - IF k[j] - dst[i+63:i] := Convert_Int64_To_FP64(a[i+63:i]) - ELSE - dst[i+63:i] := 0 - FI + dst[i+63:i] := (1.0 / a[i+63:i]) ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - Integer - AVX512VL - AVX512DQ - Convert - - Convert packed 64-bit integers in "a" to packed double-precision (64-bit) floating-point elements, and store the results in "dst". + AVX512F + Elementary Math Functions + + + + + Compute the approximate reciprocal of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). The maximum relative error for this approximation is less than 2^-14. -FOR j := 0 to 1 - i := j*64 - dst[i+63:i] := Convert_Int64_To_FP64(a[i+63:i]) +FOR j := 0 to 15 + i := j*32 + IF k[j] + dst[i+31:i] := (1.0 / a[i+31:i]) + ELSE + dst[i+31:i] := src[i+31:i] + FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - Integer - AVX512VL - AVX512DQ - Convert - - - - Convert packed 64-bit integers in "a" to packed double-precision (64-bit) floating-point elements, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + AVX512F + Elementary Math Functions + + + + Compute the approximate reciprocal of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). The maximum relative error for this approximation is less than 2^-14. -FOR j := 0 to 1 - i := j*64 +FOR j := 0 to 15 + i := j*32 IF k[j] - dst[i+63:i] := Convert_Int64_To_FP64(a[i+63:i]) + dst[i+31:i] := (1.0 / a[i+31:i]) ELSE - dst[i+63:i] := src[i+63:i] + dst[i+31:i] := 0 FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - Integer - AVX512VL - AVX512DQ - Convert - - - Convert packed 64-bit integers in "a" to packed double-precision (64-bit) floating-point elements, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + AVX512F + Elementary Math Functions + + + Compute the approximate reciprocal of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst". The maximum relative error for this approximation is less than 2^-14. -FOR j := 0 to 1 - i := j*64 - IF k[j] - dst[i+63:i] := Convert_Int64_To_FP64(a[i+63:i]) - ELSE - dst[i+63:i] := 0 - FI +FOR j := 0 to 15 + i := j*32 + dst[i+31:i] := (1.0 / a[i+31:i]) ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - Integer - AVX512VL - AVX512DQ - Convert - - Convert packed 64-bit integers in "a" to packed single-precision (32-bit) floating-point elements, and store the results in "dst". + AVX512F + Elementary Math Functions + + + + + + Compute the approximate reciprocal of the lower double-precision (64-bit) floating-point element in "b", store the result in the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". The maximum relative error for this approximation is less than 2^-14. -FOR j := 0 to 3 - i := j*64 - l := j*32 - dst[l+31:l] := Convert_Int64_To_FP32(a[i+63:i]) -ENDFOR +IF k[0] + dst[63:0] := (1.0 / b[63:0]) +ELSE + dst[63:0] := src[63:0] +FI +dst[127:64] := a[127:64] dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point - Integer - AVX512VL - AVX512DQ - Convert - - - - Convert packed 64-bit integers in "a" to packed single-precision (32-bit) floating-point elements, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + AVX512F + Elementary Math Functions + + + + + Compute the approximate reciprocal of the lower double-precision (64-bit) floating-point element in "b", store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". The maximum relative error for this approximation is less than 2^-14. -FOR j := 0 to 3 - i := j*64 - l := j*32 - IF k[j] - dst[l+31:l] := Convert_Int64_To_FP32(a[i+63:i]) - ELSE - dst[l+31:l] := src[l+31:l] - FI -ENDFOR +IF k[0] + dst[63:0] := (1.0 / b[63:0]) +ELSE + dst[63:0] := 0 +FI +dst[127:64] := a[127:64] dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point - Integer - AVX512VL - AVX512DQ - Convert - - - Convert packed 64-bit integers in "a" to packed single-precision (32-bit) floating-point elements, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + AVX512F + Elementary Math Functions + + + + Compute the approximate reciprocal of the lower double-precision (64-bit) floating-point element in "b", store the result in the lower element of "dst", and copy the upper element from "a" to the upper element of "dst". The maximum relative error for this approximation is less than 2^-14. -FOR j := 0 to 3 - i := j*64 - l := j*32 - IF k[j] - dst[l+31:l] := Convert_Int64_To_FP32(a[i+63:i]) - ELSE - dst[l+31:l] := 0 - FI -ENDFOR +dst[63:0] := (1.0 / b[63:0]) +dst[127:64] := a[127:64] +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512F + Elementary Math Functions + + + + + + Compute the approximate reciprocal of the lower single-precision (32-bit) floating-point element in "b", store the result in the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". The maximum relative error for this approximation is less than 2^-14. + +IF k[0] + dst[31:0] := (1.0 / b[31:0]) +ELSE + dst[31:0] := src[31:0] +FI +dst[127:32] := a[127:32] dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point - Integer - AVX512DQ - Convert - - - Convert packed 64-bit integers in "a" to packed single-precision (32-bit) floating-point elements, and store the results in "dst". - [round_note] - + AVX512F + Elementary Math Functions + + + + + Compute the approximate reciprocal of the lower single-precision (32-bit) floating-point element in "b", store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". The maximum relative error for this approximation is less than 2^-14. -FOR j := 0 to 7 - i := j*64 - l := j*32 - dst[l+31:l] := Convert_Int64_To_FP32(a[i+63:i]) -ENDFOR -dst[MAX:256] := 0 +IF k[0] + dst[31:0] := (1.0 / b[31:0]) +ELSE + dst[31:0] := 0 +FI +dst[127:32] := a[127:32] +dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point - Integer - AVX512DQ - Convert - - Convert packed 64-bit integers in "a" to packed single-precision (32-bit) floating-point elements, and store the results in "dst". + AVX512F + Elementary Math Functions + + + + Compute the approximate reciprocal of the lower single-precision (32-bit) floating-point element in "b", store the result in the lower element of "dst", and copy the upper 3 packed elements from "a" to the upper elements of "dst". The maximum relative error for this approximation is less than 2^-14. -FOR j := 0 to 7 - i := j*64 - l := j*32 - dst[l+31:l] := Convert_Int64_To_FP32(a[i+63:i]) -ENDFOR -dst[MAX:256] := 0 +dst[31:0] := (1.0 / b[31:0]) +dst[127:32] := a[127:32] +dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point - Integer - AVX512DQ - Convert - - - - - Convert packed 64-bit integers in "a" to packed single-precision (32-bit) floating-point elements, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - [round_note] - + AVX512F + Miscellaneous + + + + + + Round packed double-precision (64-bit) floating-point elements in "a" to the number of fraction bits specified by "imm8", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). [round_imm_note] +DEFINE RoundScaleFP64(src1[63:0], imm8[7:0]) { + m[63:0] := FP64(imm8[7:4]) // number of fraction bits after the binary point to be preserved + tmp[63:0] := POW(2.0, -m) * ROUND(POW(2.0, m) * src1[63:0], imm8[3:0]) + IF IsInf(tmp[63:0]) + tmp[63:0] := src1[63:0] + FI + RETURN tmp[63:0] +} FOR j := 0 to 7 i := j*64 - l := j*32 IF k[j] - dst[l+31:l] := Convert_Int64_To_FP32(a[i+63:i]) + dst[i+63:i] := RoundScaleFP64(a[i+63:i], imm8[7:0]) ELSE - dst[l+31:l] := src[l+31:l] + dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - Integer - AVX512DQ - Convert - - - - Convert packed 64-bit integers in "a" to packed single-precision (32-bit) floating-point elements, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + AVX512F + Miscellaneous + + + + + + + Round packed double-precision (64-bit) floating-point elements in "a" to the number of fraction bits specified by "imm8", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). [round_imm_note][sae_note] +DEFINE RoundScaleFP64(src1[63:0], imm8[7:0]) { + m[63:0] := FP64(imm8[7:4]) // number of fraction bits after the binary point to be preserved + tmp[63:0] := POW(2.0, -m) * ROUND(POW(2.0, m) * src1[63:0], imm8[3:0]) + IF IsInf(tmp[63:0]) + tmp[63:0] := src1[63:0] + FI + RETURN tmp[63:0] +} FOR j := 0 to 7 i := j*64 - l := j*32 IF k[j] - dst[l+31:l] := Convert_Int64_To_FP32(a[i+63:i]) + dst[i+63:i] := RoundScaleFP64(a[i+63:i], imm8[7:0]) ELSE - dst[l+31:l] := src[l+31:l] + dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - Integer - AVX512DQ - Convert - - - - Convert packed 64-bit integers in "a" to packed single-precision (32-bit) floating-point elements, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - [round_note] - + AVX512F + Miscellaneous + + + + + Round packed double-precision (64-bit) floating-point elements in "a" to the number of fraction bits specified by "imm8", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). [round_imm_note] +DEFINE RoundScaleFP64(src1[63:0], imm8[7:0]) { + m[63:0] := FP64(imm8[7:4]) // number of fraction bits after the binary point to be preserved + tmp[63:0] := POW(2.0, -m) * ROUND(POW(2.0, m) * src1[63:0], imm8[3:0]) + IF IsInf(tmp[63:0]) + tmp[63:0] := src1[63:0] + FI + RETURN tmp[63:0] +} FOR j := 0 to 7 i := j*64 - l := j*32 IF k[j] - dst[l+31:l] := Convert_Int64_To_FP32(a[i+63:i]) + dst[i+63:i] := RoundScaleFP64(a[i+63:i], imm8[7:0]) ELSE - dst[l+31:l] := 0 + dst[i+63:i] := 0 FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - Integer - AVX512DQ - Convert - - - Convert packed 64-bit integers in "a" to packed single-precision (32-bit) floating-point elements, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + AVX512F + Miscellaneous + + + + + + Round packed double-precision (64-bit) floating-point elements in "a" to the number of fraction bits specified by "imm8", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). [round_imm_note][sae_note] +DEFINE RoundScaleFP64(src1[63:0], imm8[7:0]) { + m[63:0] := FP64(imm8[7:4]) // number of fraction bits after the binary point to be preserved + tmp[63:0] := POW(2.0, -m) * ROUND(POW(2.0, m) * src1[63:0], imm8[3:0]) + IF IsInf(tmp[63:0]) + tmp[63:0] := src1[63:0] + FI + RETURN tmp[63:0] +} FOR j := 0 to 7 i := j*64 - l := j*32 IF k[j] - dst[l+31:l] := Convert_Int64_To_FP32(a[i+63:i]) + dst[i+63:i] := RoundScaleFP64(a[i+63:i], imm8[7:0]) ELSE - dst[l+31:l] := 0 + dst[i+63:i] := 0 FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - Integer - AVX512VL - AVX512DQ - Convert - - Convert packed 64-bit integers in "a" to packed single-precision (32-bit) floating-point elements, and store the results in "dst". + AVX512F + Miscellaneous + + + + Round packed double-precision (64-bit) floating-point elements in "a" to the number of fraction bits specified by "imm8", and store the results in "dst". [round_imm_note] -FOR j := 0 to 1 +DEFINE RoundScaleFP64(src1[63:0], imm8[7:0]) { + m[63:0] := FP64(imm8[7:4]) // number of fraction bits after the binary point to be preserved + tmp[63:0] := POW(2.0, -m) * ROUND(POW(2.0, m) * src1[63:0], imm8[3:0]) + IF IsInf(tmp[63:0]) + tmp[63:0] := src1[63:0] + FI + RETURN tmp[63:0] +} +FOR j := 0 to 7 i := j*64 - l := j*32 - dst[l+31:l] := Convert_Int64_To_FP32(a[i+63:i]) + dst[i+63:i] := RoundScaleFP64(a[i+63:i], imm8[7:0]) ENDFOR -dst[MAX:64] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - Integer - AVX512VL - AVX512DQ - Convert - - - - Convert packed 64-bit integers in "a" to packed single-precision (32-bit) floating-point elements, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + AVX512F + Miscellaneous + + + + + Round packed double-precision (64-bit) floating-point elements in "a" to the number of fraction bits specified by "imm8", and store the results in "dst". [round_imm_note][sae_note] -FOR j := 0 to 1 - i := j*64 - l := j*32 - IF k[j] - dst[l+31:l] := Convert_Int64_To_FP32(a[i+63:i]) - ELSE - dst[l+31:l] := src[l+31:l] +DEFINE RoundScaleFP64(src1[63:0], imm8[7:0]) { + m[63:0] := FP64(imm8[7:4]) // number of fraction bits after the binary point to be preserved + tmp[63:0] := POW(2.0, -m) * ROUND(POW(2.0, m) * src1[63:0], imm8[3:0]) + IF IsInf(tmp[63:0]) + tmp[63:0] := src1[63:0] FI + RETURN tmp[63:0] +} +FOR j := 0 to 7 + i := j*64 + dst[i+63:i] := RoundScaleFP64(a[i+63:i], imm8[7:0]) ENDFOR -dst[MAX:64] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - Integer - AVX512VL - AVX512DQ - Convert - - - Convert packed 64-bit integers in "a" to packed single-precision (32-bit) floating-point elements, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + AVX512F + Miscellaneous + + + + + + Round packed single-precision (32-bit) floating-point elements in "a" to the number of fraction bits specified by "imm8", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). [round_imm_note] -FOR j := 0 to 1 - i := j*64 - l := j*32 +DEFINE RoundScaleFP32(src1[31:0], imm8[7:0]) { + m[31:0] := FP32(imm8[7:4]) // number of fraction bits after the binary point to be preserved + tmp[31:0] := POW(FP32(2.0), -m) * ROUND(POW(FP32(2.0), m) * src1[31:0], imm8[3:0]) + IF IsInf(tmp[31:0]) + tmp[31:0] := src1[31:0] + FI + RETURN tmp[31:0] +} +FOR j := 0 to 15 + i := j*32 IF k[j] - dst[l+31:l] := Convert_Int64_To_FP32(a[i+63:i]) + dst[i+31:i] := RoundScaleFP32(a[i+31:i], imm8[7:0]) ELSE - dst[l+31:l] := 0 + dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:64] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - Integer - AVX512VL AVX512F - Convert - - - - Convert packed double-precision (64-bit) floating-point elements in "a" to packed 32-bit integers with truncation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Miscellaneous + + + + + + + Round packed single-precision (32-bit) floating-point elements in "a" to the number of fraction bits specified by "imm8", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). [round_imm_note][sae_note] -FOR j := 0 to 3 - i := 32*j - l := 64*j +DEFINE RoundScaleFP32(src1[31:0], imm8[7:0]) { + m[31:0] := FP32(imm8[7:4]) // number of fraction bits after the binary point to be preserved + tmp[31:0] := POW(FP32(2.0), -m) * ROUND(POW(FP32(2.0), m) * src1[31:0], imm8[3:0]) + IF IsInf(tmp[31:0]) + tmp[31:0] := src1[31:0] + FI + RETURN tmp[31:0] +} +FOR j := 0 to 15 + i := j*32 IF k[j] - dst[i+31:i] := Convert_FP64_To_Int32_Truncate(a[l+63:l]) + dst[i+31:i] := RoundScaleFP32(a[i+31:i], imm8[7:0]) ELSE dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - Integer - AVX512VL AVX512F - Convert - - - Convert packed double-precision (64-bit) floating-point elements in "a" to packed 32-bit integers with truncation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Miscellaneous + + + + + Round packed single-precision (32-bit) floating-point elements in "a" to the number of fraction bits specified by "imm8", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). [round_imm_note] -FOR j := 0 to 3 - i := 32*j - l := 64*j +DEFINE RoundScaleFP32(src1[31:0], imm8[7:0]) { + m[31:0] := FP32(imm8[7:4]) // number of fraction bits after the binary point to be preserved + tmp[31:0] := POW(FP32(2.0), -m) * ROUND(POW(FP32(2.0), m) * src1[31:0], imm8[3:0]) + IF IsInf(tmp[31:0]) + tmp[31:0] := src1[31:0] + FI + RETURN tmp[31:0] +} +FOR j := 0 to 15 + i := j*32 IF k[j] - dst[i+31:i] := Convert_FP64_To_Int32_Truncate(a[l+63:l]) + dst[i+31:i] := RoundScaleFP32(a[i+31:i], imm8[7:0]) ELSE dst[i+31:i] := 0 FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - Integer - AVX512VL AVX512F - Convert - - - - Convert packed double-precision (64-bit) floating-point elements in "a" to packed 32-bit integers with truncation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Miscellaneous + + + + + + Round packed single-precision (32-bit) floating-point elements in "a" to the number of fraction bits specified by "imm8", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). [round_imm_note][sae_note] -FOR j := 0 to 1 - i := 32*j - l := 64*j +DEFINE RoundScaleFP32(src1[31:0], imm8[7:0]) { + m[31:0] := FP32(imm8[7:4]) // number of fraction bits after the binary point to be preserved + tmp[31:0] := POW(FP32(2.0), -m) * ROUND(POW(FP32(2.0), m) * src1[31:0], imm8[3:0]) + IF IsInf(tmp[31:0]) + tmp[31:0] := src1[31:0] + FI + RETURN tmp[31:0] +} +FOR j := 0 to 15 + i := j*32 IF k[j] - dst[i+31:i] := Convert_FP64_To_Int32_Truncate(a[l+63:l]) + dst[i+31:i] := RoundScaleFP32(a[i+31:i], imm8[7:0]) ELSE - dst[i+31:i] := src[i+31:i] + dst[i+31:i] := 0 FI ENDFOR -dst[MAX:64] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - Integer - AVX512VL AVX512F - Convert - - - Convert packed double-precision (64-bit) floating-point elements in "a" to packed 32-bit integers with truncation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Miscellaneous + + + + Round packed single-precision (32-bit) floating-point elements in "a" to the number of fraction bits specified by "imm8", and store the results in "dst". [round_imm_note] -FOR j := 0 to 1 - i := 32*j - l := 64*j - IF k[j] - dst[i+31:i] := Convert_FP64_To_Int32_Truncate(a[l+63:l]) - ELSE - dst[i+31:i] := 0 +DEFINE RoundScaleFP32(src1[31:0], imm8[7:0]) { + m[31:0] := FP32(imm8[7:4]) // number of fraction bits after the binary point to be preserved + tmp[31:0] := POW(FP32(2.0), -m) * ROUND(POW(FP32(2.0), m) * src1[31:0], imm8[3:0]) + IF IsInf(tmp[31:0]) + tmp[31:0] := src1[31:0] FI + RETURN tmp[31:0] +} +FOR j := 0 to 15 + i := j*32 + dst[i+31:i] := RoundScaleFP32(a[i+31:i], imm8[7:0]) ENDFOR -dst[MAX:64] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - Integer - AVX512VL - AVX512DQ - Convert - - Convert packed double-precision (64-bit) floating-point elements in "a" to packed 64-bit integers with truncation, and store the results in "dst". + AVX512F + Miscellaneous + + + + + Round packed single-precision (32-bit) floating-point elements in "a" to the number of fraction bits specified by "imm8", and store the results in "dst". [round_imm_note][sae_note] -FOR j := 0 to 3 - i := j*64 - dst[i+63:i] := Convert_FP64_To_Int64_Truncate(a[i+63:i]) +DEFINE RoundScaleFP32(src1[31:0], imm8[7:0]) { + m[31:0] := FP32(imm8[7:4]) // number of fraction bits after the binary point to be preserved + tmp[31:0] := POW(FP32(2.0), -m) * ROUND(POW(FP32(2.0), m) * src1[31:0], imm8[3:0]) + IF IsInf(tmp[31:0]) + tmp[31:0] := src1[31:0] + FI + RETURN tmp[31:0] +} +FOR j := 0 to 15 + i := j*32 + dst[i+31:i] := RoundScaleFP32(a[i+31:i], imm8[7:0]) ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - Integer - AVX512VL - AVX512DQ - Convert - - - - Convert packed double-precision (64-bit) floating-point elements in "a" to packed 64-bit integers with truncation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + AVX512F + Miscellaneous + + + + + + + + Round the lower double-precision (64-bit) floating-point element in "b" to the number of fraction bits specified by "imm8", store the result in the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". [round_imm_note][sae_note] -FOR j := 0 to 3 - i := j*64 - IF k[j] - dst[i+63:i] := Convert_FP64_To_Int64_Truncate(a[i+63:i]) - ELSE - dst[i+63:i] := src[i+63:i] +DEFINE RoundScaleFP64(src1[63:0], imm8[7:0]) { + m[63:0] := FP64(imm8[7:4]) // number of fraction bits after the binary point to be preserved + tmp[63:0] := POW(2.0, -m) * ROUND(POW(2.0, m) * src1[63:0], imm8[3:0]) + IF IsInf(tmp[63:0]) + tmp[63:0] := src1[63:0] FI -ENDFOR -dst[MAX:256] := 0 + RETURN tmp[63:0] +} +IF k[0] + dst[63:0] := RoundScaleFP64(b[63:0], imm8[7:0]) +ELSE + dst[63:0] := src[63:0] +FI +dst[127:64] := a[127:64] +dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point - Integer - AVX512VL - AVX512DQ - Convert - - - Convert packed double-precision (64-bit) floating-point elements in "a" to packed 64-bit integers with truncation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - + AVX512F + Miscellaneous + + + + + + + Round the lower double-precision (64-bit) floating-point element in "b" to the number of fraction bits specified by "imm8", store the result in the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". [round_imm_note] -FOR j := 0 to 3 - i := j*64 - IF k[j] - dst[i+63:i] := Convert_FP64_To_Int64_Truncate(a[i+63:i]) - ELSE - dst[i+63:i] := 0 +DEFINE RoundScaleFP64(src1[63:0], imm8[7:0]) { + m[63:0] := FP64(imm8[7:4]) // number of fraction bits after the binary point to be preserved + tmp[63:0] := POW(2.0, -m) * ROUND(POW(2.0, m) * src1[63:0], imm8[3:0]) + IF IsInf(tmp[63:0]) + tmp[63:0] := src1[63:0] FI -ENDFOR -dst[MAX:256] := 0 + RETURN tmp[63:0] +} +IF k[0] + dst[63:0] := RoundScaleFP64(b[63:0], imm8[7:0]) +ELSE + dst[63:0] := src[63:0] +FI +dst[127:64] := a[127:64] +dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point - Integer - AVX512DQ - Convert - - - Convert packed double-precision (64-bit) floating-point elements in "a" to packed 64-bit integers with truncation, and store the results in "dst". Pass __MM_FROUND_NO_EXC to "sae" to suppress all exceptions. + AVX512F + Miscellaneous + + + + + + + Round the lower double-precision (64-bit) floating-point element in "b" to the number of fraction bits specified by "imm8", store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". [round_imm_note][sae_note] -FOR j := 0 to 7 - i := j*64 - dst[i+63:i] := Convert_FP64_To_Int64_Truncate(a[i+63:i]) -ENDFOR -dst[MAX:512] := 0 +DEFINE RoundScaleFP64(src1[63:0], imm8[7:0]) { + m[63:0] := FP64(imm8[7:4]) // number of fraction bits after the binary point to be preserved + tmp[63:0] := POW(2.0, -m) * ROUND(POW(2.0, m) * src1[63:0], imm8[3:0]) + IF IsInf(tmp[63:0]) + tmp[63:0] := src1[63:0] + FI + RETURN tmp[63:0] +} +IF k[0] + dst[63:0] := RoundScaleFP64(b[63:0], imm8[7:0]) +ELSE + dst[63:0] := 0 +FI +dst[127:64] := a[127:64] +dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point - Integer - AVX512DQ - Convert - - Convert packed double-precision (64-bit) floating-point elements in "a" to packed 64-bit integers with truncation, and store the results in "dst". + AVX512F + Miscellaneous + + + + + + Round the lower double-precision (64-bit) floating-point element in "b" to the number of fraction bits specified by "imm8", store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". [round_imm_note] -FOR j := 0 to 7 - i := j*64 - dst[i+63:i] := Convert_FP64_To_Int64_Truncate(a[i+63:i]) -ENDFOR -dst[MAX:512] := 0 +DEFINE RoundScaleFP64(src1[63:0], imm8[7:0]) { + m[63:0] := FP64(imm8[7:4]) // number of fraction bits after the binary point to be preserved + tmp[63:0] := POW(2.0, -m) * ROUND(POW(2.0, m) * src1[63:0], imm8[3:0]) + IF IsInf(tmp[63:0]) + tmp[63:0] := src1[63:0] + FI + RETURN tmp[63:0] +} +IF k[0] + dst[63:0] := RoundScaleFP64(b[63:0], imm8[7:0]) +ELSE + dst[63:0] := 0 +FI +dst[127:64] := a[127:64] +dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point - Integer - AVX512DQ - Convert - - - - - Convert packed double-precision (64-bit) floating-point elements in "a" to packed 64-bit integers with truncation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - Pass __MM_FROUND_NO_EXC to "sae" to suppress all exceptions. + AVX512F + Miscellaneous + + + + + + Round the lower double-precision (64-bit) floating-point element in "b" to the number of fraction bits specified by "imm8", store the result in the lower element of "dst", and copy the upper element from "a" to the upper element of "dst". [round_imm_note][sae_note] -FOR j := 0 to 7 - i := j*64 - IF k[j] - dst[i+63:i] := Convert_FP64_To_Int64_Truncate(a[i+63:i]) - ELSE - dst[i+63:i] := src[i+63:i] +DEFINE RoundScaleFP64(src1[63:0], imm8[7:0]) { + m[63:0] := FP64(imm8[7:4]) // number of fraction bits after the binary point to be preserved + tmp[63:0] := POW(2.0, -m) * ROUND(POW(2.0, m) * src1[63:0], imm8[3:0]) + IF IsInf(tmp[63:0]) + tmp[63:0] := src1[63:0] FI -ENDFOR -dst[MAX:512] := 0 + RETURN tmp[63:0] +} +dst[63:0] := RoundScaleFP64(b[63:0], imm8[7:0]) +dst[127:64] := a[127:64] +dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point - Integer - AVX512DQ - Convert - - - - Convert packed double-precision (64-bit) floating-point elements in "a" to packed 64-bit integers with truncation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + AVX512F + Miscellaneous + + + + + Round the lower double-precision (64-bit) floating-point element in "b" to the number of fraction bits specified by "imm8", store the result in the lower element of "dst", and copy the upper element from "a" to the upper element of "dst". [round_imm_note] -FOR j := 0 to 7 - i := j*64 - IF k[j] - dst[i+63:i] := Convert_FP64_To_Int64_Truncate(a[i+63:i]) - ELSE - dst[i+63:i] := src[i+63:i] +DEFINE RoundScaleFP64(src1[63:0], imm8[7:0]) { + m[63:0] := FP64(imm8[7:4]) // number of fraction bits after the binary point to be preserved + tmp[63:0] := POW(2.0, -m) * ROUND(POW(2.0, m) * src1[63:0], imm8[3:0]) + IF IsInf(tmp[63:0]) + tmp[63:0] := src1[63:0] FI -ENDFOR -dst[MAX:512] := 0 + RETURN tmp[63:0] +} +dst[63:0] := RoundScaleFP64(b[63:0], imm8[7:0]) +dst[127:64] := a[127:64] +dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point - Integer - AVX512DQ - Convert - - - - Convert packed double-precision (64-bit) floating-point elements in "a" to packed 64-bit integers with truncation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). Pass __MM_FROUND_NO_EXC to "sae" to suppress all exceptions. - + AVX512F + Miscellaneous + + + + + + + + Round the lower single-precision (32-bit) floating-point element in "b" to the number of fraction bits specified by "imm8", store the result in the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". [round_imm_note][sae_note] -FOR j := 0 to 7 - i := j*64 - IF k[j] - dst[i+63:i] := Convert_FP64_To_Int64_Truncate(a[i+63:i]) - ELSE - dst[i+63:i] := 0 +DEFINE RoundScaleFP32(src1[31:0], imm8[7:0]) { + m[31:0] := FP32(imm8[7:4]) // number of fraction bits after the binary point to be preserved + tmp[31:0] := POW(FP32(2.0), -m) * ROUND(POW(FP32(2.0), m) * src1[31:0], imm8[3:0]) + IF IsInf(tmp[31:0]) + tmp[31:0] := src1[31:0] FI -ENDFOR -dst[MAX:512] := 0 + RETURN tmp[31:0] +} +IF k[0] + dst[31:0] := RoundScaleFP32(b[31:0], imm8[7:0]) +ELSE + dst[31:0] := src[31:0] +FI +dst[127:32] := a[127:32] +dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point - Integer - AVX512DQ - Convert - - - Convert packed double-precision (64-bit) floating-point elements in "a" to packed 64-bit integers with truncation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - + AVX512F + Miscellaneous + + + + + + + Round the lower single-precision (32-bit) floating-point element in "b" to the number of fraction bits specified by "imm8", store the result in the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". [round_imm_note] -FOR j := 0 to 7 - i := j*64 - IF k[j] - dst[i+63:i] := Convert_FP64_To_Int64_Truncate(a[i+63:i]) - ELSE - dst[i+63:i] := 0 +DEFINE RoundScaleFP32(src1[31:0], imm8[7:0]) { + m[31:0] := FP32(imm8[7:4]) // number of fraction bits after the binary point to be preserved + tmp[31:0] := POW(FP32(2.0), -m) * ROUND(POW(FP32(2.0), m) * src1[31:0], imm8[3:0]) + IF IsInf(tmp[31:0]) + tmp[31:0] := src1[31:0] FI -ENDFOR -dst[MAX:512] := 0 + RETURN tmp[31:0] +} +IF k[0] + dst[31:0] := RoundScaleFP32(b[31:0], imm8[7:0]) +ELSE + dst[31:0] := src[31:0] +FI +dst[127:32] := a[127:32] +dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point - Integer - AVX512VL - AVX512DQ - Convert - - Convert packed double-precision (64-bit) floating-point elements in "a" to packed 64-bit integers with truncation, and store the results in "dst". + AVX512F + Miscellaneous + + + + + + + Round the lower single-precision (32-bit) floating-point element in "b" to the number of fraction bits specified by "imm8", store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". [round_imm_note][sae_note] -FOR j := 0 to 1 - i := j*64 - dst[i+63:i] := Convert_FP64_To_Int64_Truncate(a[i+63:i]) -ENDFOR +DEFINE RoundScaleFP32(src1[31:0], imm8[7:0]) { + m[31:0] := FP32(imm8[7:4]) // number of fraction bits after the binary point to be preserved + tmp[31:0] := POW(FP32(2.0), -m) * ROUND(POW(FP32(2.0), m) * src1[31:0], imm8[3:0]) + IF IsInf(tmp[31:0]) + tmp[31:0] := src1[31:0] + FI + RETURN tmp[31:0] +} +IF k[0] + dst[31:0] := RoundScaleFP32(b[31:0], imm8[7:0]) +ELSE + dst[31:0] := 0 +FI +dst[127:32] := a[127:32] dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point - Integer - AVX512VL - AVX512DQ - Convert - - - - Convert packed double-precision (64-bit) floating-point elements in "a" to packed 64-bit integers with truncation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + AVX512F + Miscellaneous + + + + + + Round the lower single-precision (32-bit) floating-point element in "b" to the number of fraction bits specified by "imm8", store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". [round_imm_note] -FOR j := 0 to 1 - i := j*64 - IF k[j] - dst[i+63:i] := Convert_FP64_To_Int64_Truncate(a[i+63:i]) - ELSE - dst[i+63:i] := src[i+63:i] +DEFINE RoundScaleFP32(src1[31:0], imm8[7:0]) { + m[31:0] := FP32(imm8[7:4]) // number of fraction bits after the binary point to be preserved + tmp[31:0] := POW(FP32(2.0), -m) * ROUND(POW(FP32(2.0), m) * src1[31:0], imm8[3:0]) + IF IsInf(tmp[31:0]) + tmp[31:0] := src1[31:0] FI -ENDFOR + RETURN tmp[31:0] +} +IF k[0] + dst[31:0] := RoundScaleFP32(b[31:0], imm8[7:0]) +ELSE + dst[31:0] := 0 +FI +dst[127:32] := a[127:32] dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point - Integer - AVX512VL - AVX512DQ - Convert - - - Convert packed double-precision (64-bit) floating-point elements in "a" to packed 64-bit integers with truncation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - + AVX512F + Miscellaneous + + + + + + Round the lower single-precision (32-bit) floating-point element in "b" to the number of fraction bits specified by "imm8", store the result in the lower element of "dst", and copy the upper 3 packed elements from "a" to the upper elements of "dst". [round_imm_note][sae_note] -FOR j := 0 to 1 - i := j*64 - IF k[j] - dst[i+63:i] := Convert_FP64_To_Int64_Truncate(a[i+63:i]) - ELSE - dst[i+63:i] := 0 +DEFINE RoundScaleFP32(src1[31:0], imm8[7:0]) { + m[31:0] := FP32(imm8[7:4]) // number of fraction bits after the binary point to be preserved + tmp[31:0] := POW(FP32(2.0), -m) * ROUND(POW(FP32(2.0), m) * src1[31:0], imm8[3:0]) + IF IsInf(tmp[31:0]) + tmp[31:0] := src1[31:0] FI -ENDFOR + RETURN tmp[31:0] +} +dst[31:0] := RoundScaleFP32(b[31:0], imm8[7:0]) +dst[127:32] := a[127:32] dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point - Integer - AVX512VL AVX512F - Convert - - Convert packed double-precision (64-bit) floating-point elements in "a" to packed unsigned 32-bit integers with truncation, and store the results in "dst". + Miscellaneous + + + + + Round the lower single-precision (32-bit) floating-point element in "b" to the number of fraction bits specified by "imm8", store the result in the lower element of "dst", and copy the upper 3 packed elements from "a" to the upper elements of "dst". [round_imm_note] -FOR j := 0 to 3 - i := 32*j - k := 64*j - dst[i+31:i] := Convert_FP64_To_UnsignedInt32_Truncate(a[k+63:k]) -ENDFOR +DEFINE RoundScaleFP32(src1[31:0], imm8[7:0]) { + m[31:0] := FP32(imm8[7:4]) // number of fraction bits after the binary point to be preserved + tmp[31:0] := POW(FP32(2.0), -m) * ROUND(POW(FP32(2.0), m) * src1[31:0], imm8[3:0]) + IF IsInf(tmp[31:0]) + tmp[31:0] := src1[31:0] + FI + RETURN tmp[31:0] +} +dst[31:0] := RoundScaleFP32(b[31:0], imm8[7:0]) +dst[127:32] := a[127:32] dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point - Integer - AVX512VL AVX512F - Convert - - - - Convert packed double-precision (64-bit) floating-point elements in "a" to packed unsigned 32-bit integers with truncation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Elementary Math Functions + + + + + Compute the approximate reciprocal square root of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). The maximum relative error for this approximation is less than 2^-14. -FOR j := 0 to 3 - i := 32*j - l := 64*j +FOR j := 0 to 7 + i := j*64 IF k[j] - dst[i+31:i] := Convert_FP64_To_UnsignedInt32_Truncate(a[l+63:l]) + dst[i+63:i] := (1.0 / SQRT(a[i+63:i])) ELSE - dst[i+31:i] := src[i+31:i] + dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - Integer - AVX512VL AVX512F - Convert - - - Convert packed double-precision (64-bit) floating-point elements in "a" to packed unsigned 32-bit integers with truncation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Elementary Math Functions + + + + Compute the approximate reciprocal square root of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). The maximum relative error for this approximation is less than 2^-14. -FOR j := 0 to 3 - i := 32*j - l := 64*j +FOR j := 0 to 7 + i := j*64 IF k[j] - dst[i+31:i] := Convert_FP64_To_UnsignedInt32_Truncate(a[l+63:l]) + dst[i+63:i] := (1.0 / SQRT(a[i+63:i])) ELSE - dst[i+31:i] := 0 + dst[i+63:i] := 0 FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - Integer - AVX512VL AVX512F - Convert - - Convert packed double-precision (64-bit) floating-point elements in "a" to packed unsigned 32-bit integers with truncation, and store the results in "dst". + Elementary Math Functions + + + Compute the approximate reciprocal square root of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst". The maximum relative error for this approximation is less than 2^-14. -FOR j := 0 to 1 - i := 32*j - k := 64*j - dst[i+31:i] := Convert_FP64_To_UnsignedInt32_Truncate(a[k+63:k]) +FOR j := 0 to 7 + i := j*64 + dst[i+63:i] := (1.0 / SQRT(a[i+63:i])) ENDFOR -dst[MAX:64] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - Integer - AVX512VL AVX512F - Convert - - - - Convert packed double-precision (64-bit) floating-point elements in "a" to packed unsigned 32-bit integers with truncation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Elementary Math Functions + + + + + Compute the approximate reciprocal square root of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). The maximum relative error for this approximation is less than 2^-14. -FOR j := 0 to 1 - i := 32*j - l := 64*j +FOR j := 0 to 15 + i := j*32 IF k[j] - dst[i+31:i] := Convert_FP64_To_UnsignedInt32_Truncate(a[l+63:l]) + dst[i+31:i] := (1.0 / SQRT(a[i+31:i])) ELSE dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:64] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - Integer - AVX512VL AVX512F - Convert - - - Convert packed double-precision (64-bit) floating-point elements in "a" to packed unsigned 32-bit integers with truncation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Elementary Math Functions + + + + Compute the approximate reciprocal square root of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). The maximum relative error for this approximation is less than 2^-14. -FOR j := 0 to 1 - i := 32*j - l := 64*j +FOR j := 0 to 15 + i := j*32 IF k[j] - dst[i+31:i] := Convert_FP64_To_UnsignedInt32_Truncate(a[l+63:l]) + dst[i+31:i] := (1.0 / SQRT(a[i+31:i])) ELSE dst[i+31:i] := 0 FI ENDFOR -dst[MAX:64] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - Integer - AVX512VL - AVX512DQ - Convert - - Convert packed double-precision (64-bit) floating-point elements in "a" to packed unsigned 64-bit integers with truncation, and store the results in "dst". + AVX512F + Elementary Math Functions + + + Compute the approximate reciprocal square root of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst". The maximum relative error for this approximation is less than 2^-14. -FOR j := 0 to 3 - i := j*64 - dst[i+63:i] := Convert_FP64_To_UnsignedInt64_Truncate(a[i+63:i]) +FOR j := 0 to 15 + i := j*32 + dst[i+31:i] := (1.0 / SQRT(a[i+31:i])) ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - Integer - AVX512VL - AVX512DQ - Convert - - - - Convert packed double-precision (64-bit) floating-point elements in "a" to packed unsigned 64-bit integers with truncation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + AVX512F + Elementary Math Functions + + + + + + Compute the approximate reciprocal square root of the lower double-precision (64-bit) floating-point element in "b", store the result in the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". The maximum relative error for this approximation is less than 2^-14. -FOR j := 0 to 3 - i := j*64 - IF k[j] - dst[i+63:i] := Convert_FP64_To_UnsignedInt64_Truncate(a[i+63:i]) - ELSE - dst[i+63:i] := src[i+63:i] - FI -ENDFOR -dst[MAX:256] := 0 +IF k[0] + dst[63:0] := (1.0 / SQRT(b[63:0])) +ELSE + dst[63:0] := src[63:0] +FI +dst[127:64] := a[127:64] +dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point - Integer - AVX512VL - AVX512DQ - Convert - - - Convert packed double-precision (64-bit) floating-point elements in "a" to packed unsigned 64-bit integers with truncation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - + AVX512F + Elementary Math Functions + + + + + Compute the approximate reciprocal square root of the lower double-precision (64-bit) floating-point element in "b", store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". The maximum relative error for this approximation is less than 2^-14. -FOR j := 0 to 3 - i := j*64 - IF k[j] - dst[i+63:i] := Convert_FP64_To_UnsignedInt64_Truncate(a[i+63:i]) - ELSE - dst[i+63:i] := 0 - FI -ENDFOR -dst[MAX:256] := 0 +IF k[0] + dst[63:0] := (1.0 / SQRT(b[63:0])) +ELSE + dst[63:0] := 0 +FI +dst[127:64] := a[127:64] +dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point - Integer - AVX512DQ - Convert - - - Convert packed double-precision (64-bit) floating-point elements in "a" to packed unsigned 64-bit integers with truncation, and store the results in "dst". Pass __MM_FROUND_NO_EXC to "sae" to suppress all exceptions. + AVX512F + Elementary Math Functions + + + + Compute the approximate reciprocal square root of the lower double-precision (64-bit) floating-point element in "b", store the result in the lower element of "dst", and copy the upper element from "a" to the upper element of "dst". The maximum relative error for this approximation is less than 2^-14. -FOR j := 0 to 7 - i := j*64 - dst[i+63:i] := Convert_FP64_To_UnsignedInt64_Truncate(a[i+63:i]) -ENDFOR -dst[MAX:512] := 0 +dst[63:0] := (1.0 / SQRT(b[63:0])) +dst[127:64] := a[127:64] +dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point - Integer - AVX512DQ - Convert - - Convert packed double-precision (64-bit) floating-point elements in "a" to packed unsigned 64-bit integers with truncation, and store the results in "dst". + AVX512F + Elementary Math Functions + + + + + + Compute the approximate reciprocal square root of the lower single-precision (32-bit) floating-point element in "b", store the result in the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". The maximum relative error for this approximation is less than 2^-14. -FOR j := 0 to 7 - i := j*64 - dst[i+63:i] := Convert_FP64_To_UnsignedInt64_Truncate(a[i+63:i]) -ENDFOR -dst[MAX:512] := 0 +IF k[0] + dst[31:0] := (1.0 / SQRT(b[31:0])) +ELSE + dst[31:0] := src[31:0] +FI +dst[127:32] := a[127:32] +dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point - Integer - AVX512DQ - Convert - - - - - Convert packed double-precision (64-bit) floating-point elements in "a" to packed unsigned 64-bit integers with truncation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - Pass __MM_FROUND_NO_EXC to "sae" to suppress all exceptions. + AVX512F + Elementary Math Functions + + + + + Compute the approximate reciprocal square root of the lower single-precision (32-bit) floating-point element in "b", store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". The maximum relative error for this approximation is less than 2^-14. + +IF k[0] + dst[31:0] := (1.0 / SQRT(b[31:0])) +ELSE + dst[31:0] := 0 +FI +dst[127:32] := a[127:32] +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512F + Elementary Math Functions + + + + Compute the approximate reciprocal square root of the lower single-precision (32-bit) floating-point element in "b", store the result in the lower element of "dst", and copy the upper 3 packed elements from "a" to the upper elements of "dst". The maximum relative error for this approximation is less than 2^-14. +dst[31:0] := (1.0 / SQRT(b[31:0])) +dst[127:32] := a[127:32] +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512F + Miscellaneous + + + + + + Scale the packed double-precision (64-bit) floating-point elements in "a" using values from "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + DEFINE SCALE(src1, src2) { + IF (src2 == NaN) + IF (src2 == SNaN) + RETURN QNAN(src2) + FI + ELSE IF (src1 == NaN) + IF (src1 == SNaN) + RETURN QNAN(src1) + FI + IF (src2 != INF) + RETURN QNAN(src1) + FI + ELSE + tmp_src2 := src2 + tmp_src1 := src1 + IF (IS_DENORMAL(src2) AND MXCSR.DAZ) + tmp_src2 := 0 + FI + IF (IS_DENORMAL(src1) AND MXCSR.DAZ) + tmp_src1 := 0 + FI + FI + dst[63:0] := tmp_src1[63:0] * POW(2.0, FLOOR(tmp_src2[63:0])) + RETURN dst[63:0] +} FOR j := 0 to 7 i := j*64 IF k[j] - dst[i+63:i] := Convert_FP64_To_UnsignedInt64_Truncate(a[i+63:i]) + dst[i+63:i] := SCALE(a[i+63:0], b[i+63:i]) ELSE dst[i+63:i] := src[i+63:i] FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - Integer - AVX512DQ - Convert - - - - Convert packed double-precision (64-bit) floating-point elements in "a" to packed unsigned 64-bit integers with truncation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - + AVX512F + Miscellaneous + + + + + + + Scale the packed double-precision (64-bit) floating-point elements in "a" using values from "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + [round_note] + DEFINE SCALE(src1, src2) { + IF (src2 == NaN) + IF (src2 == SNaN) + RETURN QNAN(src2) + FI + ELSE IF (src1 == NaN) + IF (src1 == SNaN) + RETURN QNAN(src1) + FI + IF (src2 != INF) + RETURN QNAN(src1) + FI + ELSE + tmp_src2 := src2 + tmp_src1 := src1 + IF (IS_DENORMAL(src2) AND MXCSR.DAZ) + tmp_src2 := 0 + FI + IF (IS_DENORMAL(src1) AND MXCSR.DAZ) + tmp_src1 := 0 + FI + FI + dst[63:0] := tmp_src1[63:0] * POW(2.0, FLOOR(tmp_src2[63:0])) + RETURN dst[63:0] +} FOR j := 0 to 7 i := j*64 IF k[j] - dst[i+63:i] := Convert_FP64_To_UnsignedInt64_Truncate(a[i+63:i]) + dst[i+63:i] := SCALE(a[i+63:0], b[i+63:i]) ELSE dst[i+63:i] := src[i+63:i] FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - Integer - AVX512DQ - Convert - - - - Convert packed double-precision (64-bit) floating-point elements in "a" to packed unsigned 64-bit integers with truncation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). Pass __MM_FROUND_NO_EXC to "sae" to suppress all exceptions. - - + AVX512F + Miscellaneous + + + + + Scale the packed double-precision (64-bit) floating-point elements in "a" using values from "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + DEFINE SCALE(src1, src2) { + IF (src2 == NaN) + IF (src2 == SNaN) + RETURN QNAN(src2) + FI + ELSE IF (src1 == NaN) + IF (src1 == SNaN) + RETURN QNAN(src1) + FI + IF (src2 != INF) + RETURN QNAN(src1) + FI + ELSE + tmp_src2 := src2 + tmp_src1 := src1 + IF (IS_DENORMAL(src2) AND MXCSR.DAZ) + tmp_src2 := 0 + FI + IF (IS_DENORMAL(src1) AND MXCSR.DAZ) + tmp_src1 := 0 + FI + FI + dst[63:0] := tmp_src1[63:0] * POW(2.0, FLOOR(tmp_src2[63:0])) + RETURN dst[63:0] +} FOR j := 0 to 7 i := j*64 IF k[j] - dst[i+63:i] := Convert_FP64_To_UnsignedInt64_Truncate(a[i+63:i]) + dst[i+63:i] := SCALE(a[i+63:0], b[i+63:i]) ELSE dst[i+63:i] := 0 FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - Integer - AVX512DQ - Convert - - - Convert packed double-precision (64-bit) floating-point elements in "a" to packed unsigned 64-bit integers with truncation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - - + AVX512F + Miscellaneous + + + + + + Scale the packed double-precision (64-bit) floating-point elements in "a" using values from "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + [round_note] + DEFINE SCALE(src1, src2) { + IF (src2 == NaN) + IF (src2 == SNaN) + RETURN QNAN(src2) + FI + ELSE IF (src1 == NaN) + IF (src1 == SNaN) + RETURN QNAN(src1) + FI + IF (src2 != INF) + RETURN QNAN(src1) + FI + ELSE + tmp_src2 := src2 + tmp_src1 := src1 + IF (IS_DENORMAL(src2) AND MXCSR.DAZ) + tmp_src2 := 0 + FI + IF (IS_DENORMAL(src1) AND MXCSR.DAZ) + tmp_src1 := 0 + FI + FI + dst[63:0] := tmp_src1[63:0] * POW(2.0, FLOOR(tmp_src2[63:0])) + RETURN dst[63:0] +} FOR j := 0 to 7 i := j*64 IF k[j] - dst[i+63:i] := Convert_FP64_To_UnsignedInt64_Truncate(a[i+63:i]) + dst[i+63:i] := SCALE(a[i+63:0], b[i+63:i]) ELSE dst[i+63:i] := 0 FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - Integer - AVX512VL - AVX512DQ - Convert - - Convert packed double-precision (64-bit) floating-point elements in "a" to packed unsigned 64-bit integers with truncation, and store the results in "dst". - -FOR j := 0 to 1 + AVX512F + Miscellaneous + + + + Scale the packed double-precision (64-bit) floating-point elements in "a" using values from "b", and store the results in "dst". + DEFINE SCALE(src1, src2) { + IF (src2 == NaN) + IF (src2 == SNaN) + RETURN QNAN(src2) + FI + ELSE IF (src1 == NaN) + IF (src1 == SNaN) + RETURN QNAN(src1) + FI + IF (src2 != INF) + RETURN QNAN(src1) + FI + ELSE + tmp_src2 := src2 + tmp_src1 := src1 + IF (IS_DENORMAL(src2) AND MXCSR.DAZ) + tmp_src2 := 0 + FI + IF (IS_DENORMAL(src1) AND MXCSR.DAZ) + tmp_src1 := 0 + FI + FI + dst[63:0] := tmp_src1[63:0] * POW(2.0, FLOOR(tmp_src2[63:0])) + RETURN dst[63:0] +} +FOR j := 0 to 7 i := j*64 - dst[i+63:i] := Convert_FP64_To_UnsignedInt64_Truncate(a[i+63:i]) + dst[i+63:i] := SCALE(a[i+63:0], b[i+63:i]) ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - Integer - AVX512VL - AVX512DQ - Convert - - - - Convert packed double-precision (64-bit) floating-point elements in "a" to packed unsigned 64-bit integers with truncation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -FOR j := 0 to 1 - i := j*64 - IF k[j] - dst[i+63:i] := Convert_FP64_To_UnsignedInt64_Truncate(a[i+63:i]) + AVX512F + Miscellaneous + + + + + Scale the packed double-precision (64-bit) floating-point elements in "a" using values from "b", and store the results in "dst". + [round_note] + DEFINE SCALE(src1, src2) { + IF (src2 == NaN) + IF (src2 == SNaN) + RETURN QNAN(src2) + FI + ELSE IF (src1 == NaN) + IF (src1 == SNaN) + RETURN QNAN(src1) + FI + IF (src2 != INF) + RETURN QNAN(src1) + FI ELSE - dst[i+63:i] := src[i+63:i] + tmp_src2 := src2 + tmp_src1 := src1 + IF (IS_DENORMAL(src2) AND MXCSR.DAZ) + tmp_src2 := 0 + FI + IF (IS_DENORMAL(src1) AND MXCSR.DAZ) + tmp_src1 := 0 + FI FI + dst[63:0] := tmp_src1[63:0] * POW(2.0, FLOOR(tmp_src2[63:0])) + RETURN dst[63:0] +} +FOR j := 0 to 7 + i := j*64 + dst[i+63:i] := SCALE(a[i+63:0], b[i+63:i]) ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - Integer - AVX512VL - AVX512DQ - Convert - - - Convert packed double-precision (64-bit) floating-point elements in "a" to packed unsigned 64-bit integers with truncation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - - -FOR j := 0 to 1 - i := j*64 + AVX512F + Miscellaneous + + + + + + Scale the packed single-precision (32-bit) floating-point elements in "a" using values from "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + DEFINE SCALE(src1, src2) { + IF (src2 == NaN) + IF (src2 == SNaN) + RETURN QNAN(src2) + FI + ELSE IF (src1 == NaN) + IF (src1 == SNaN) + RETURN QNAN(src1) + FI + IF (src2 != INF) + RETURN QNAN(src1) + FI + ELSE + tmp_src2 := src2 + tmp_src1 := src1 + IF (IS_DENORMAL(src2) AND MXCSR.DAZ) + tmp_src2 := 0 + FI + IF (IS_DENORMAL(src1) AND MXCSR.DAZ) + tmp_src1 := 0 + FI + FI + dst[31:0] := tmp_src1[31:0] * POW(2.0, FLOOR(tmp_src2[31:0])) + RETURN dst[31:0] +} +FOR j := 0 to 15 + i := j*32 IF k[j] - dst[i+63:i] := Convert_FP64_To_UnsignedInt64_Truncate(a[i+63:i]) + dst[i+31:i] := SCALE(a[i+31:0], b[i+31:i]) ELSE - dst[i+63:i] := 0 + dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - Integer - AVX512VL AVX512F - Convert - - - - Convert packed single-precision (32-bit) floating-point elements in "a" to packed 32-bit integers with truncation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -FOR j := 0 to 7 - i := 32*j + Miscellaneous + + + + + + + Scale the packed single-precision (32-bit) floating-point elements in "a" using values from "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + [round_note] + DEFINE SCALE(src1, src2) { + IF (src2 == NaN) + IF (src2 == SNaN) + RETURN QNAN(src2) + FI + ELSE IF (src1 == NaN) + IF (src1 == SNaN) + RETURN QNAN(src1) + FI + IF (src2 != INF) + RETURN QNAN(src1) + FI + ELSE + tmp_src2 := src2 + tmp_src1 := src1 + IF (IS_DENORMAL(src2) AND MXCSR.DAZ) + tmp_src2 := 0 + FI + IF (IS_DENORMAL(src1) AND MXCSR.DAZ) + tmp_src1 := 0 + FI + FI + dst[31:0] := tmp_src1[31:0] * POW(2.0, FLOOR(tmp_src2[31:0])) + RETURN dst[31:0] +} +FOR j := 0 to 15 + i := j*32 IF k[j] - dst[i+31:i] := Convert_FP32_To_Int32_Truncate(a[i+31:i]) + dst[i+31:i] := SCALE(a[i+31:0], b[i+31:i]) ELSE dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - Integer - AVX512VL AVX512F - Convert - - - Convert packed single-precision (32-bit) floating-point elements in "a" to packed 32-bit integers with truncation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - -FOR j := 0 to 7 - i := 32*i + Miscellaneous + + + + + Scale the packed single-precision (32-bit) floating-point elements in "a" using values from "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + DEFINE SCALE(src1, src2) { + IF (src2 == NaN) + IF (src2 == SNaN) + RETURN QNAN(src2) + FI + ELSE IF (src1 == NaN) + IF (src1 == SNaN) + RETURN QNAN(src1) + FI + IF (src2 != INF) + RETURN QNAN(src1) + FI + ELSE + tmp_src2 := src2 + tmp_src1 := src1 + IF (IS_DENORMAL(src2) AND MXCSR.DAZ) + tmp_src2 := 0 + FI + IF (IS_DENORMAL(src1) AND MXCSR.DAZ) + tmp_src1 := 0 + FI + FI + dst[31:0] := tmp_src1[31:0] * POW(2.0, FLOOR(tmp_src2[31:0])) + RETURN dst[31:0] +} +FOR j := 0 to 15 + i := j*32 IF k[j] - dst[i+31:i] := Convert_FP32_To_IntegerTruncate(a[i+31:i]) + dst[i+31:i] := SCALE(a[i+31:0], b[i+31:i]) ELSE dst[i+31:i] := 0 FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - Integer - AVX512VL AVX512F - Convert - - - - Convert packed single-precision (32-bit) floating-point elements in "a" to packed 32-bit integers with truncation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -FOR j := 0 to 3 - i := 32*j - IF k[j] - dst[i+31:i] := Convert_FP32_To_Int32_Truncate(a[i+31:i]) + Miscellaneous + + + + + + Scale the packed single-precision (32-bit) floating-point elements in "a" using values from "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + [round_note] + DEFINE SCALE(src1, src2) { + IF (src2 == NaN) + IF (src2 == SNaN) + RETURN QNAN(src2) + FI + ELSE IF (src1 == NaN) + IF (src1 == SNaN) + RETURN QNAN(src1) + FI + IF (src2 != INF) + RETURN QNAN(src1) + FI ELSE - dst[i+31:i] := src[i+31:i] + tmp_src2 := src2 + tmp_src1 := src1 + IF (IS_DENORMAL(src2) AND MXCSR.DAZ) + tmp_src2 := 0 + FI + IF (IS_DENORMAL(src1) AND MXCSR.DAZ) + tmp_src1 := 0 + FI FI -ENDFOR -dst[MAX:128] := 0 - - -
immintrin.h
- - - Floating Point - Integer - AVX512VL - AVX512F - Convert - - - Convert packed single-precision (32-bit) floating-point elements in "a" to packed 32-bit integers with truncation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - -FOR j := 0 to 3 - i := 32*i + dst[31:0] := tmp_src1[31:0] * POW(2.0, FLOOR(tmp_src2[31:0])) + RETURN dst[31:0] +} +FOR j := 0 to 15 + i := j*32 IF k[j] - dst[i+31:i] := Convert_FP32_To_IntegerTruncate(a[i+31:i]) + dst[i+31:i] := SCALE(a[i+31:0], b[i+31:i]) ELSE dst[i+31:i] := 0 FI ENDFOR -dst[MAX:128] := 0 - - -
immintrin.h
- - - Floating Point - Integer - AVX512VL - AVX512DQ - Convert - - Convert packed single-precision (32-bit) floating-point elements in "a" to packed 64-bit integers with truncation, and store the results in "dst". - -FOR j := 0 to 3 - i := j*64 - l := j*32 - dst[i+63:i] := Convert_FP32_To_Int64_Truncate(a[l+31:l]) -ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - Integer - AVX512VL - AVX512DQ - Convert - - - - Convert packed single-precision (32-bit) floating-point elements in "a" to packed 64-bit integers with truncation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -FOR j := 0 to 3 - i := j*64 - l := j*32 - IF k[j] - dst[i+63:i] := Convert_FP32_To_Int64_Truncate(a[l+31:l]) + AVX512F + Miscellaneous + + + + Scale the packed single-precision (32-bit) floating-point elements in "a" using values from "b", and store the results in "dst". + DEFINE SCALE(src1, src2) { + IF (src2 == NaN) + IF (src2 == SNaN) + RETURN QNAN(src2) + FI + ELSE IF (src1 == NaN) + IF (src1 == SNaN) + RETURN QNAN(src1) + FI + IF (src2 != INF) + RETURN QNAN(src1) + FI ELSE - dst[i+63:i] := src[i+63:i] + tmp_src2 := src2 + tmp_src1 := src1 + IF (IS_DENORMAL(src2) AND MXCSR.DAZ) + tmp_src2 := 0 + FI + IF (IS_DENORMAL(src1) AND MXCSR.DAZ) + tmp_src1 := 0 + FI FI + dst[31:0] := tmp_src1[31:0] * POW(2.0, FLOOR(tmp_src2[31:0])) + RETURN dst[31:0] +} +FOR j := 0 to 15 + i := j*32 + dst[i+31:i] := SCALE(a[i+31:0], b[i+31:i]) ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - Integer - AVX512VL - AVX512DQ - Convert - - - Convert packed single-precision (32-bit) floating-point elements in "a" to packed 64-bit integers with truncation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - -FOR j := 0 to 3 - i := j*64 - l := j*32 - IF k[j] - dst[i+63:i] := Convert_FP32_To_Int64_Truncate(a[l+31:l]) + AVX512F + Miscellaneous + + + + + Scale the packed single-precision (32-bit) floating-point elements in "a" using values from "b", and store the results in "dst". + [round_note] + DEFINE SCALE(src1, src2) { + IF (src2 == NaN) + IF (src2 == SNaN) + RETURN QNAN(src2) + FI + ELSE IF (src1 == NaN) + IF (src1 == SNaN) + RETURN QNAN(src1) + FI + IF (src2 != INF) + RETURN QNAN(src1) + FI ELSE - dst[i+63:i] := 0 + tmp_src2 := src2 + tmp_src1 := src1 + IF (IS_DENORMAL(src2) AND MXCSR.DAZ) + tmp_src2 := 0 + FI + IF (IS_DENORMAL(src1) AND MXCSR.DAZ) + tmp_src1 := 0 + FI FI -ENDFOR -dst[MAX:256] := 0 - - -
immintrin.h
- - - Floating Point - Integer - AVX512DQ - Convert - - - Convert packed single-precision (32-bit) floating-point elements in "a" to packed 64-bit integers with truncation, and store the results in "dst". Pass __MM_FROUND_NO_EXC to "sae" to suppress all exceptions. - -FOR j := 0 to 7 - i := j*64 - l := j*32 - dst[i+63:i] := Convert_FP32_To_Int64_Truncate(a[l+31:l]) + dst[31:0] := tmp_src1[31:0] * POW(2.0, FLOOR(tmp_src2[31:0])) + RETURN dst[31:0] +} +FOR j := 0 to 15 + i := j*32 + dst[i+31:i] := SCALE(a[i+31:0], b[i+31:i]) ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - Integer - AVX512DQ - Convert - - Convert packed single-precision (32-bit) floating-point elements in "a" to packed 64-bit integers with truncation, and store the results in "dst". - -FOR j := 0 to 7 - i := j*64 - l := j*32 - dst[i+63:i] := Convert_FP32_To_Int64_Truncate(a[l+31:l]) -ENDFOR -dst[MAX:512] := 0 + AVX512F + Miscellaneous + + + + + + + Scale the packed double-precision (64-bit) floating-point elements in "a" using values from "b", store the result in the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". + [round_note] + DEFINE SCALE(src1, src2) { + IF (src2 == NaN) + IF (src2 == SNaN) + RETURN QNAN(src2) + FI + ELSE IF (src1 == NaN) + IF (src1 == SNaN) + RETURN QNAN(src1) + FI + IF (src2 != INF) + RETURN QNAN(src1) + FI + ELSE + tmp_src2 := src2 + tmp_src1 := src1 + IF (IS_DENORMAL(src2) AND MXCSR.DAZ) + tmp_src2 := 0 + FI + IF (IS_DENORMAL(src1) AND MXCSR.DAZ) + tmp_src1 := 0 + FI + FI + dst[63:0] := tmp_src1[63:0] * POW(2.0, FLOOR(tmp_src2[63:0])) + RETURN dst[63:0] +} +IF k[0] + dst[63:0] := SCALE(a[63:0], b[63:0]) +ELSE + dst[63:0] := src[63:0] +FI +dst[127:64] := a[127:64] +dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point - Integer - AVX512DQ - Convert - - - - - Convert packed single-precision (32-bit) floating-point elements in "a" to packed 64-bit integers with truncation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - Pass __MM_FROUND_NO_EXC to "sae" to suppress all exceptions. - - -FOR j := 0 to 7 - i := j*64 - l := j*32 - IF k[j] - dst[i+63:i] := Convert_FP32_To_Int64_Truncate(a[l+31:l]) + AVX512F + Miscellaneous + + + + + + Scale the packed double-precision (64-bit) floating-point elements in "a" using values from "b", store the result in the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". + DEFINE SCALE(src1, src2) { + IF (src2 == NaN) + IF (src2 == SNaN) + RETURN QNAN(src2) + FI + ELSE IF (src1 == NaN) + IF (src1 == SNaN) + RETURN QNAN(src1) + FI + IF (src2 != INF) + RETURN QNAN(src1) + FI ELSE - dst[i+63:i] := src[i+63:i] + tmp_src2 := src2 + tmp_src1 := src1 + IF (IS_DENORMAL(src2) AND MXCSR.DAZ) + tmp_src2 := 0 + FI + IF (IS_DENORMAL(src1) AND MXCSR.DAZ) + tmp_src1 := 0 + FI FI -ENDFOR -dst[MAX:512] := 0 + dst[63:0] := tmp_src1[63:0] * POW(2.0, FLOOR(tmp_src2[63:0])) + RETURN dst[63:0] +} +IF k[0] + dst[63:0] := SCALE(a[63:0], b[63:0]) +ELSE + dst[63:0] := src[63:0] +FI +dst[127:64] := a[127:64] +dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point - Integer - AVX512DQ - Convert - - - - Convert packed single-precision (32-bit) floating-point elements in "a" to packed 64-bit integers with truncation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -FOR j := 0 to 7 - i := j*64 - l := j*32 - IF k[j] - dst[i+63:i] := Convert_FP32_To_Int64_Truncate(a[l+31:l]) + AVX512F + Miscellaneous + + + + + + Scale the packed double-precision (64-bit) floating-point elements in "a" using values from "b", store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". + [round_note] + DEFINE SCALE(src1, src2) { + IF (src2 == NaN) + IF (src2 == SNaN) + RETURN QNAN(src2) + FI + ELSE IF (src1 == NaN) + IF (src1 == SNaN) + RETURN QNAN(src1) + FI + IF (src2 != INF) + RETURN QNAN(src1) + FI ELSE - dst[i+63:i] := src[i+63:i] + tmp_src2 := src2 + tmp_src1 := src1 + IF (IS_DENORMAL(src2) AND MXCSR.DAZ) + tmp_src2 := 0 + FI + IF (IS_DENORMAL(src1) AND MXCSR.DAZ) + tmp_src1 := 0 + FI FI -ENDFOR -dst[MAX:512] := 0 + dst[63:0] := tmp_src1[63:0] * POW(2.0, FLOOR(tmp_src2[63:0])) + RETURN dst[63:0] +} +IF k[0] + dst[63:0] := SCALE(a[63:0], b[63:0]) +ELSE + dst[63:0] := 0 +FI +dst[127:64] := a[127:64] +dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point - Integer - AVX512DQ - Convert - - - - Convert packed single-precision (32-bit) floating-point elements in "a" to packed 64-bit integers with truncation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). Pass __MM_FROUND_NO_EXC to "sae" to suppress all exceptions. - - -FOR j := 0 to 7 - i := j*64 - l := j*32 - IF k[j] - dst[i+63:i] := Convert_FP32_To_Int64_Truncate(a[l+31:l]) + AVX512F + Miscellaneous + + + + + Scale the packed double-precision (64-bit) floating-point elements in "a" using values from "b", store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". + DEFINE SCALE(src1, src2) { + IF (src2 == NaN) + IF (src2 == SNaN) + RETURN QNAN(src2) + FI + ELSE IF (src1 == NaN) + IF (src1 == SNaN) + RETURN QNAN(src1) + FI + IF (src2 != INF) + RETURN QNAN(src1) + FI ELSE - dst[i+63:i] := 0 + tmp_src2 := src2 + tmp_src1 := src1 + IF (IS_DENORMAL(src2) AND MXCSR.DAZ) + tmp_src2 := 0 + FI + IF (IS_DENORMAL(src1) AND MXCSR.DAZ) + tmp_src1 := 0 + FI FI -ENDFOR -dst[MAX:512] := 0 + dst[63:0] := tmp_src1[63:0] * POW(2.0, FLOOR(tmp_src2[63:0])) + RETURN dst[63:0] +} +IF k[0] + dst[63:0] := SCALE(a[63:0], b[63:0]) +ELSE + dst[63:0] := 0 +FI +dst[127:64] := a[127:64] +dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point - Integer - AVX512DQ - Convert - - - Convert packed single-precision (32-bit) floating-point elements in "a" to packed 64-bit integers with truncation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - -FOR j := 0 to 7 - i := j*64 - l := j*32 - IF k[j] - dst[i+63:i] := Convert_FP32_To_Int64_Truncate(a[l+31:l]) + AVX512F + Miscellaneous + + + + + Scale the packed double-precision (64-bit) floating-point elements in "a" using values from "b", store the result in the lower element of "dst", and copy the upper element from "a" to the upper element of "dst". + [round_note] + DEFINE SCALE(src1, src2) { + IF (src2 == NaN) + IF (src2 == SNaN) + RETURN QNAN(src2) + FI + ELSE IF (src1 == NaN) + IF (src1 == SNaN) + RETURN QNAN(src1) + FI + IF (src2 != INF) + RETURN QNAN(src1) + FI ELSE - dst[i+63:i] := 0 + tmp_src2 := src2 + tmp_src1 := src1 + IF (IS_DENORMAL(src2) AND MXCSR.DAZ) + tmp_src2 := 0 + FI + IF (IS_DENORMAL(src1) AND MXCSR.DAZ) + tmp_src1 := 0 + FI FI -ENDFOR -dst[MAX:512] := 0 + dst[63:0] := tmp_src1[63:0] * POW(2.0, FLOOR(tmp_src2[63:0])) + RETURN dst[63:0] +} +dst[63:0] := SCALE(a[63:0], b[63:0]) +dst[127:64] := a[127:64] +dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point - Integer - AVX512VL - AVX512DQ - Convert - - Convert packed single-precision (32-bit) floating-point elements in "a" to packed 64-bit integers with truncation, and store the results in "dst". - -FOR j := 0 to 1 - i := j*64 - l := j*32 - dst[i+63:i] := Convert_FP32_To_Int64_Truncate(a[l+31:l]) -ENDFOR + AVX512F + Miscellaneous + + + + Scale the packed double-precision (64-bit) floating-point elements in "a" using values from "b", store the result in the lower element of "dst", and copy the upper element from "a" to the upper element of "dst". + DEFINE SCALE(src1, src2) { + IF (src2 == NaN) + IF (src2 == SNaN) + RETURN QNAN(src2) + FI + ELSE IF (src1 == NaN) + IF (src1 == SNaN) + RETURN QNAN(src1) + FI + IF (src2 != INF) + RETURN QNAN(src1) + FI + ELSE + tmp_src2 := src2 + tmp_src1 := src1 + IF (IS_DENORMAL(src2) AND MXCSR.DAZ) + tmp_src2 := 0 + FI + IF (IS_DENORMAL(src1) AND MXCSR.DAZ) + tmp_src1 := 0 + FI + FI + dst[63:0] := tmp_src1[63:0] * POW(2.0, FLOOR(tmp_src2[63:0])) + RETURN dst[63:0] +} +dst[63:0] := SCALE(a[63:0], b[63:0]) +dst[127:64] := a[127:64] dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point - Integer - AVX512VL - AVX512DQ - Convert - - - - Convert packed single-precision (32-bit) floating-point elements in "a" to packed 64-bit integers with truncation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -FOR j := 0 to 1 - i := j*64 - l := j*32 - IF k[j] - dst[i+63:i] := Convert_FP32_To_Int64_Truncate(a[l+31:l]) + AVX512F + Miscellaneous + + + + + + + Scale the packed single-precision (32-bit) floating-point elements in "a" using values from "b", store the result in the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". + [round_note] + DEFINE SCALE(src1, src2) { + IF (src2 == NaN) + IF (src2 == SNaN) + RETURN QNAN(src2) + FI + ELSE IF (src1 == NaN) + IF (src1 == SNaN) + RETURN QNAN(src1) + FI + IF (src2 != INF) + RETURN QNAN(src1) + FI ELSE - dst[i+63:i] := src[i+63:i] + tmp_src2 := src2 + tmp_src1 := src1 + IF (IS_DENORMAL(src2) AND MXCSR.DAZ) + tmp_src2 := 0 + FI + IF (IS_DENORMAL(src1) AND MXCSR.DAZ) + tmp_src1 := 0 + FI FI -ENDFOR + dst[31:0] := tmp_src1[31:0] * POW(2.0, FLOOR(tmp_src2[31:0])) + RETURN dst[63:0] +} +IF k[0] + dst[31:0] := SCALE(a[31:0], b[31:0]) +ELSE + dst[31:0] := src[31:0] +FI +dst[127:32] := a[127:32] dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point - Integer - AVX512VL - AVX512DQ - Convert - - - Convert packed single-precision (32-bit) floating-point elements in "a" to packed 64-bit integers with truncation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - -FOR j := 0 to 1 - i := j*64 - l := j*32 - IF k[j] - dst[i+63:i] := Convert_FP32_To_Int64_Truncate(a[l+31:l]) + AVX512F + Miscellaneous + + + + + + Scale the packed single-precision (32-bit) floating-point elements in "a" using values from "b", store the result in the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". + DEFINE SCALE(src1, src2) { + IF (src2 == NaN) + IF (src2 == SNaN) + RETURN QNAN(src2) + FI + ELSE IF (src1 == NaN) + IF (src1 == SNaN) + RETURN QNAN(src1) + FI + IF (src2 != INF) + RETURN QNAN(src1) + FI ELSE - dst[i+63:i] := 0 + tmp_src2 := src2 + tmp_src1 := src1 + IF (IS_DENORMAL(src2) AND MXCSR.DAZ) + tmp_src2 := 0 + FI + IF (IS_DENORMAL(src1) AND MXCSR.DAZ) + tmp_src1 := 0 + FI FI -ENDFOR + dst[31:0] := tmp_src1[31:0] * POW(2.0, FLOOR(tmp_src2[31:0])) + RETURN dst[63:0] +} +IF k[0] + dst[31:0] := SCALE(a[31:0], b[31:0]) +ELSE + dst[31:0] := src[31:0] +FI +dst[127:32] := a[127:32] dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point - Integer - AVX512VL AVX512F - Convert - - Convert packed single-precision (32-bit) floating-point elements in "a" to packed unsigned 32-bit integers with truncation, and store the results in "dst". - -FOR j := 0 to 7 - i := 32*j - dst[i+31:i] := Convert_FP32_To_UnsignedInt32_Truncate(a[i+31:i]) -ENDFOR -dst[MAX:256] := 0 + Miscellaneous + + + + + + Scale the packed single-precision (32-bit) floating-point elements in "a" using values from "b", store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". + [round_note] + DEFINE SCALE(src1, src2) { + IF (src2 == NaN) + IF (src2 == SNaN) + RETURN QNAN(src2) + FI + ELSE IF (src1 == NaN) + IF (src1 == SNaN) + RETURN QNAN(src1) + FI + IF (src2 != INF) + RETURN QNAN(src1) + FI + ELSE + tmp_src2 := src2 + tmp_src1 := src1 + IF (IS_DENORMAL(src2) AND MXCSR.DAZ) + tmp_src2 := 0 + FI + IF (IS_DENORMAL(src1) AND MXCSR.DAZ) + tmp_src1 := 0 + FI + FI + dst[31:0] := tmp_src1[31:0] * POW(2.0, FLOOR(tmp_src2[31:0])) + RETURN dst[63:0] +} +IF k[0] + dst[31:0] := SCALE(a[31:0], b[31:0]) +ELSE + dst[31:0] := 0 +FI +dst[127:32] := a[127:32] +dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point - Integer - AVX512VL AVX512F - Convert - - - - Convert packed double-precision (32-bit) floating-point elements in "a" to packed unsigned 32-bit integers with truncation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -FOR j := 0 to 7 - i := 32*j - IF k[j] - dst[i+31:i] := Convert_FP64_To_UnsignedInt32_Truncate(a[i+31:i]) + Miscellaneous + + + + + Scale the packed single-precision (32-bit) floating-point elements in "a" using values from "b", store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". + DEFINE SCALE(src1, src2) { + IF (src2 == NaN) + IF (src2 == SNaN) + RETURN QNAN(src2) + FI + ELSE IF (src1 == NaN) + IF (src1 == SNaN) + RETURN QNAN(src1) + FI + IF (src2 != INF) + RETURN QNAN(src1) + FI ELSE - dst[i+31:i] := src[i+31:i] + tmp_src2 := src2 + tmp_src1 := src1 + IF (IS_DENORMAL(src2) AND MXCSR.DAZ) + tmp_src2 := 0 + FI + IF (IS_DENORMAL(src1) AND MXCSR.DAZ) + tmp_src1 := 0 + FI FI -ENDFOR -dst[MAX:256] := 0 + dst[31:0] := tmp_src1[31:0] * POW(2.0, FLOOR(tmp_src2[31:0])) + RETURN dst[63:0] +} +IF k[0] + dst[31:0] := SCALE(a[31:0], b[31:0]) +ELSE + dst[31:0] := 0 +FI +dst[127:32] := a[127:32] +dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point - Integer - AVX512VL AVX512F - Convert - - - Convert packed double-precision (32-bit) floating-point elements in "a" to packed unsigned 32-bit integers with truncation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - -FOR j := 0 to 7 - i := 32*j - IF k[j] - dst[i+31:i] := Convert_FP64_To_UnsignedInt32_Truncate(a[i+31:i]) + Miscellaneous + + + + + Scale the packed single-precision (32-bit) floating-point elements in "a" using values from "b", store the result in the lower element of "dst", and copy the upper 3 packed elements from "a" to the upper elements of "dst". + [round_note] + DEFINE SCALE(src1, src2) { + IF (src2 == NaN) + IF (src2 == SNaN) + RETURN QNAN(src2) + FI + ELSE IF (src1 == NaN) + IF (src1 == SNaN) + RETURN QNAN(src1) + FI + IF (src2 != INF) + RETURN QNAN(src1) + FI ELSE - dst[i+31:i] := 0 + tmp_src2 := src2 + tmp_src1 := src1 + IF (IS_DENORMAL(src2) AND MXCSR.DAZ) + tmp_src2 := 0 + FI + IF (IS_DENORMAL(src1) AND MXCSR.DAZ) + tmp_src1 := 0 + FI FI -ENDFOR -dst[MAX:256] := 0 + dst[31:0] := tmp_src1[31:0] * POW(2.0, FLOOR(tmp_src2[31:0])) + RETURN dst[63:0] +} +dst[31:0] := SCALE(a[31:0], b[31:0]) +dst[127:32] := a[127:32] +dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point - Integer - AVX512VL AVX512F - Convert - - Convert packed single-precision (32-bit) floating-point elements in "a" to packed unsigned 32-bit integers with truncation, and store the results in "dst". - -FOR j := 0 to 3 - i := 32*j - dst[i+31:i] := Convert_FP32_To_UnsignedInt32_Truncate(a[i+31:i]) -ENDFOR + Miscellaneous + + + + Scale the packed single-precision (32-bit) floating-point elements in "a" using values from "b", store the result in the lower element of "dst", and copy the upper 3 packed elements from "a" to the upper elements of "dst". + DEFINE SCALE(src1, src2) { + IF (src2 == NaN) + IF (src2 == SNaN) + RETURN QNAN(src2) + FI + ELSE IF (src1 == NaN) + IF (src1 == SNaN) + RETURN QNAN(src1) + FI + IF (src2 != INF) + RETURN QNAN(src1) + FI + ELSE + tmp_src2 := src2 + tmp_src1 := src1 + IF (IS_DENORMAL(src2) AND MXCSR.DAZ) + tmp_src2 := 0 + FI + IF (IS_DENORMAL(src1) AND MXCSR.DAZ) + tmp_src1 := 0 + FI + FI + dst[31:0] := tmp_src1[31:0] * POW(2.0, FLOOR(tmp_src2[31:0])) + RETURN dst[63:0] +} +dst[31:0] := SCALE(a[31:0], b[31:0]) +dst[127:32] := a[127:32] dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point - Integer - AVX512VL AVX512F - Convert - - - - Convert packed double-precision (32-bit) floating-point elements in "a" to packed unsigned 32-bit integers with truncation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Store + + + + + + Scatter double-precision (64-bit) floating-point elements from "a" into memory using 32-bit indices. 64-bit elements are stored at addresses starting at "base_addr" and offset by each 32-bit element in "vindex" (each index is scaled by the factor in "scale"). "scale" should be 1, 2, 4 or 8. -FOR j := 0 to 3 - i := 32*j - IF k[j] - dst[i+31:i] := Convert_FP64_To_UnsignedInt32_Truncate(a[i+31:i]) - ELSE - dst[i+31:i] := src[i+31:i] - FI +FOR j := 0 to 7 + i := j*64 + m := j*32 + addr := base_addr + SignExtend64(vindex[m+31:m]) * ZeroExtend64(scale) * 8 + MEM[addr+63:addr] := a[i+63:i] ENDFOR -dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point - Integer - AVX512VL AVX512F - Convert - - - Convert packed double-precision (32-bit) floating-point elements in "a" to packed unsigned 32-bit integers with truncation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Store + + + + + + + Scatter double-precision (64-bit) floating-point elements from "a" into memory using 32-bit indices. 64-bit elements are stored at addresses starting at "base_addr" and offset by each 32-bit element in "vindex" (each index is scaled by the factor in "scale") subject to mask "k" (elements are not stored when the corresponding mask bit is not set). "scale" should be 1, 2, 4 or 8. -FOR j := 0 to 3 - i := 32*j +FOR j := 0 to 7 + i := j*64 + m := j*32 IF k[j] - dst[i+31:i] := Convert_FP64_To_UnsignedInt32_Truncate(a[i+31:i]) - ELSE - dst[i+31:i] := 0 + addr := base_addr + SignExtend64(vindex[m+31:m]) * ZeroExtend64(scale) * 8 + MEM[addr+63:addr] := a[i+63:i] FI ENDFOR -dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point - Integer - AVX512VL - AVX512DQ - Convert - - Convert packed single-precision (32-bit) floating-point elements in "a" to packed unsigned 64-bit integers with truncation, and store the results in "dst". + AVX512F + Store + + + + + + Scatter double-precision (64-bit) floating-point elements from "a" into memory using 64-bit indices. 64-bit elements are stored at addresses starting at "base_addr" and offset by each 64-bit element in "vindex" (each index is scaled by the factor in "scale"). "scale" should be 1, 2, 4 or 8. -FOR j := 0 to 3 +FOR j := 0 to 7 i := j*64 - l := j*32 - dst[i+63:i] := Convert_FP32_To_UnsignedInt64_Truncate(a[l+31:l]) + m := j*64 + addr := base_addr + vindex[m+63:m] * ZeroExtend64(scale) * 8 + MEM[addr+63:addr] := a[i+63:i] ENDFOR -dst[MAX:256] := 0 - +
immintrin.h
 - + Floating Point - Integer - AVX512VL - AVX512DQ - Convert - - - - Convert packed single-precision (32-bit) floating-point elements in "a" to packed unsigned 64-bit integers with truncation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + AVX512F + Store + + + + + + + Scatter double-precision (64-bit) floating-point elements from "a" into memory using 64-bit indices. 64-bit elements are stored at addresses starting at "base_addr" and offset by each 64-bit element in "vindex" (each index is scaled by the factor in "scale") subject to mask "k" (elements are not stored when the corresponding mask bit is not set). "scale" should be 1, 2, 4 or 8. -FOR j := 0 to 3 +FOR j := 0 to 7 i := j*64 - l := j*32 + m := j*64 IF k[j] - dst[i+63:i] := Convert_FP32_To_UnsignedInt64_Truncate(a[l+31:l]) - ELSE - dst[i+63:i] := src[i+63:i] + addr := base_addr + vindex[m+63:m] * ZeroExtend64(scale) * 8 + MEM[addr+63:addr] := a[i+63:i] FI ENDFOR -dst[MAX:256] := 0 - +
immintrin.h
 - + Floating Point - Integer - AVX512VL - AVX512DQ - Convert - - - Convert packed single-precision (32-bit) floating-point elements in "a" to packed unsigned 64-bit integers with truncation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - + AVX512F + Store + + + + + + Scatter single-precision (32-bit) floating-point elements from "a" into memory using 64-bit indices. 32-bit elements are stored at addresses starting at "base_addr" and offset by each 64-bit element in "vindex" (each index is scaled by the factor in "scale") subject to mask "k" (elements are not stored when the corresponding mask bit is not set). "scale" should be 1, 2, 4 or 8. -FOR j := 0 to 3 - i := j*64 - l := j*32 - IF k[j] - dst[i+63:i] := Convert_FP32_To_UnsignedInt64_Truncate(a[l+31:l]) - ELSE - dst[i+63:i] := 0 - FI +FOR j := 0 to 7 + i := j*32 + m := j*64 + addr := base_addr + vindex[m+63:m] * ZeroExtend64(scale) * 8 + MEM[addr+31:addr] := a[i+31:i] ENDFOR -dst[MAX:256] := 0 - +
immintrin.h
 - + Floating Point - Integer - AVX512DQ - Convert - - - Convert packed single-precision (32-bit) floating-point elements in "a" to packed unsigned 64-bit integers with truncation, and store the results in "dst". Pass __MM_FROUND_NO_EXC to "sae" to suppress all exceptions. + AVX512F + Store + + + + + + + Scatter single-precision (32-bit) floating-point elements from "a" into memory using 64-bit indices. 32-bit elements are stored at addresses starting at "base_addr" and offset by each 64-bit element in "vindex" (each index is scaled by the factor in "scale") subject to mask "k" (elements are not stored when the corresponding mask bit is not set). "scale" should be 1, 2, 4 or 8. FOR j := 0 to 7 - i := j*64 - l := j*32 - dst[i+63:i] := Convert_FP32_To_UnsignedInt64_Truncate(a[l+31:l]) + i := j*32 + m := j*64 + IF k[j] + addr := base_addr + vindex[m+63:m] * ZeroExtend64(scale) * 8 + MEM[addr+31:addr] := a[i+31:i] + FI ENDFOR -dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - Integer - AVX512DQ - Convert - - Convert packed single-precision (32-bit) floating-point elements in "a" to packed unsigned 64-bit integers with truncation, and store the results in "dst". - -FOR j := 0 to 7 - i := j*64 - l := j*32 - dst[i+63:i] := Convert_FP32_To_UnsignedInt64_Truncate(a[l+31:l]) + AVX512F + Swizzle + + + + + + + Shuffle 128-bits (composed of 4 single-precision (32-bit) floating-point elements) selected by "imm8" from "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +DEFINE SELECT4(src, control) { + CASE(control[1:0]) OF + 0: tmp[127:0] := src[127:0] + 1: tmp[127:0] := src[255:128] + 2: tmp[127:0] := src[383:256] + 3: tmp[127:0] := src[511:384] + ESAC + RETURN tmp[127:0] +} +tmp_dst[127:0] := SELECT4(a[511:0], imm8[1:0]) +tmp_dst[255:128] := SELECT4(a[511:0], imm8[3:2]) +tmp_dst[383:256] := SELECT4(b[511:0], imm8[5:4]) +tmp_dst[511:384] := SELECT4(b[511:0], imm8[7:6]) +FOR j := 0 to 15 + i := j*32 + IF k[j] + dst[i+31:i] := tmp_dst[i+31:i] + ELSE + dst[i+31:i] := src[i+31:i] + FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - Integer - AVX512DQ - Convert - - - - - Convert packed single-precision (32-bit) floating-point elements in "a" to packed unsigned 64-bit integers with truncation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - Pass __MM_FROUND_NO_EXC to "sae" to suppress all exceptions. - + AVX512F + Swizzle + + + + + + Shuffle 128-bits (composed of 4 single-precision (32-bit) floating-point elements) selected by "imm8" from "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := j*64 - l := j*32 +DEFINE SELECT4(src, control) { + CASE(control[1:0]) OF + 0: tmp[127:0] := src[127:0] + 1: tmp[127:0] := src[255:128] + 2: tmp[127:0] := src[383:256] + 3: tmp[127:0] := src[511:384] + ESAC + RETURN tmp[127:0] +} +tmp_dst[127:0] := SELECT4(a[511:0], imm8[1:0]) +tmp_dst[255:128] := SELECT4(a[511:0], imm8[3:2]) +tmp_dst[383:256] := SELECT4(b[511:0], imm8[5:4]) +tmp_dst[511:384] := SELECT4(b[511:0], imm8[7:6]) +FOR j := 0 to 15 + i := j*32 IF k[j] - dst[i+63:i] := Convert_FP32_To_UnsignedInt64_Truncate(a[l+31:l]) + dst[i+31:i] := tmp_dst[i+31:i] ELSE - dst[i+63:i] := src[i+63:i] + dst[i+31:i] := 0 FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - Integer - AVX512DQ - Convert - - - - Convert packed single-precision (32-bit) floating-point elements in "a" to packed unsigned 64-bit integers with truncation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -FOR j := 0 to 7 - i := j*64 - l := j*32 - IF k[j] - dst[i+63:i] := Convert_FP32_To_UnsignedInt64_Truncate(a[l+31:l]) - ELSE - dst[i+63:i] := src[i+63:i] - FI -ENDFOR + AVX512F + Swizzle + + + + + Shuffle 128-bits (composed of 4 single-precision (32-bit) floating-point elements) selected by "imm8" from "a" and "b", and store the results in "dst". + +DEFINE SELECT4(src, control) { + CASE(control[1:0]) OF + 0: tmp[127:0] := src[127:0] + 1: tmp[127:0] := src[255:128] + 2: tmp[127:0] := src[383:256] + 3: tmp[127:0] := src[511:384] + ESAC + RETURN tmp[127:0] +} +dst[127:0] := SELECT4(a[511:0], imm8[1:0]) +dst[255:128] := SELECT4(a[511:0], imm8[3:2]) +dst[383:256] := SELECT4(b[511:0], imm8[5:4]) +dst[511:384] := SELECT4(b[511:0], imm8[7:6]) dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - Integer - AVX512DQ - Convert - - - - Convert packed single-precision (32-bit) floating-point elements in "a" to packed unsigned 64-bit integers with truncation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). Pass __MM_FROUND_NO_EXC to "sae" to suppress all exceptions. - - + AVX512F + Swizzle + + + + + + + Shuffle 128-bits (composed of 2 double-precision (64-bit) floating-point elements) selected by "imm8" from "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +DEFINE SELECT4(src, control) { + CASE(control[1:0]) OF + 0: tmp[127:0] := src[127:0] + 1: tmp[127:0] := src[255:128] + 2: tmp[127:0] := src[383:256] + 3: tmp[127:0] := src[511:384] + ESAC + RETURN tmp[127:0] +} +tmp_dst[127:0] := SELECT4(a[511:0], imm8[1:0]) +tmp_dst[255:128] := SELECT4(a[511:0], imm8[3:2]) +tmp_dst[383:256] := SELECT4(b[511:0], imm8[5:4]) +tmp_dst[511:384] := SELECT4(b[511:0], imm8[7:6]) FOR j := 0 to 7 i := j*64 - l := j*32 IF k[j] - dst[i+63:i] := Convert_FP32_To_UnsignedInt64_Truncate(a[l+31:l]) + dst[i+63:i] := tmp_dst[i+63:i] ELSE - dst[i+63:i] := 0 + dst[i+63:i] := src[i+63:i] FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - Integer - AVX512DQ - Convert - - - Convert packed single-precision (32-bit) floating-point elements in "a" to packed unsigned 64-bit integers with truncation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - + AVX512F + Swizzle + + + + + + Shuffle 128-bits (composed of 2 double-precision (64-bit) floating-point elements) selected by "imm8" from "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). +DEFINE SELECT4(src, control) { + CASE(control[1:0]) OF + 0: tmp[127:0] := src[127:0] + 1: tmp[127:0] := src[255:128] + 2: tmp[127:0] := src[383:256] + 3: tmp[127:0] := src[511:384] + ESAC + RETURN tmp[127:0] +} +tmp_dst[127:0] := SELECT4(a[511:0], imm8[1:0]) +tmp_dst[255:128] := SELECT4(a[511:0], imm8[3:2]) +tmp_dst[383:256] := SELECT4(b[511:0], imm8[5:4]) +tmp_dst[511:384] := SELECT4(b[511:0], imm8[7:6]) FOR j := 0 to 7 i := j*64 - l := j*32 IF k[j] - dst[i+63:i] := Convert_FP32_To_UnsignedInt64_Truncate(a[l+31:l]) + dst[i+63:i] := tmp_dst[i+63:i] ELSE dst[i+63:i] := 0 FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - Integer - AVX512VL - AVX512DQ - Convert - - Convert packed single-precision (32-bit) floating-point elements in "a" to packed unsigned 64-bit integers with truncation, and store the results in "dst". + AVX512F + Swizzle + + + + + Shuffle 128-bits (composed of 2 double-precision (64-bit) floating-point elements) selected by "imm8" from "a" and "b", and store the results in "dst". -FOR j := 0 to 1 - i := j*64 - l := j*32 - dst[i+63:i] := Convert_FP32_To_UnsignedInt64_Truncate(a[l+31:l]) -ENDFOR -dst[MAX:128] := 0 +DEFINE SELECT4(src, control) { + CASE(control[1:0]) OF + 0: tmp[127:0] := src[127:0] + 1: tmp[127:0] := src[255:128] + 2: tmp[127:0] := src[383:256] + 3: tmp[127:0] := src[511:384] + ESAC + RETURN tmp[127:0] +} +dst[127:0] := SELECT4(a[511:0], imm8[1:0]) +dst[255:128] := SELECT4(a[511:0], imm8[3:2]) +dst[383:256] := SELECT4(b[511:0], imm8[5:4]) +dst[511:384] := SELECT4(b[511:0], imm8[7:6]) +dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point + Integer - AVX512VL - AVX512DQ - Convert - - - - Convert packed single-precision (32-bit) floating-point elements in "a" to packed unsigned 64-bit integers with truncation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -FOR j := 0 to 1 - i := j*64 - l := j*32 + AVX512F + Swizzle + + + + + + + Shuffle 128-bits (composed of 4 32-bit integers) selected by "imm8" from "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +DEFINE SELECT4(src, control) { + CASE(control[1:0]) OF + 0: tmp[127:0] := src[127:0] + 1: tmp[127:0] := src[255:128] + 2: tmp[127:0] := src[383:256] + 3: tmp[127:0] := src[511:384] + ESAC + RETURN tmp[127:0] +} +tmp_dst[127:0] := SELECT4(a[511:0], imm8[1:0]) +tmp_dst[255:128] := SELECT4(a[511:0], imm8[3:2]) +tmp_dst[383:256] := SELECT4(b[511:0], imm8[5:4]) +tmp_dst[511:384] := SELECT4(b[511:0], imm8[7:6]) +FOR j := 0 to 15 + i := j*32 IF k[j] - dst[i+63:i] := Convert_FP32_To_UnsignedInt64_Truncate(a[l+31:l]) + dst[i+31:i] := tmp_dst[i+31:i] ELSE - dst[i+63:i] := src[i+63:i] + dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point + Integer - AVX512VL - AVX512DQ - Convert - - - Convert packed single-precision (32-bit) floating-point elements in "a" to packed unsigned 64-bit integers with truncation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - + AVX512F + Swizzle + + + + + + Shuffle 128-bits (composed of 4 32-bit integers) selected by "imm8" from "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 1 - i := j*64 - l := j*32 +DEFINE SELECT4(src, control) { + CASE(control[1:0]) OF + 0: tmp[127:0] := src[127:0] + 1: tmp[127:0] := src[255:128] + 2: tmp[127:0] := src[383:256] + 3: tmp[127:0] := src[511:384] + ESAC + RETURN tmp[127:0] +} +tmp_dst[127:0] := SELECT4(a[511:0], imm8[1:0]) +tmp_dst[255:128] := SELECT4(a[511:0], imm8[3:2]) +tmp_dst[383:256] := SELECT4(b[511:0], imm8[5:4]) +tmp_dst[511:384] := SELECT4(b[511:0], imm8[7:6]) +FOR j := 0 to 15 + i := j*32 IF k[j] - dst[i+63:i] := Convert_FP32_To_UnsignedInt64_Truncate(a[l+31:l]) + dst[i+31:i] := tmp_dst[i+31:i] ELSE - dst[i+63:i] := 0 + dst[i+31:i] := 0 FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point + Integer - AVX512VL AVX512F - Convert - - Convert packed unsigned 32-bit integers in "a" to packed double-precision (64-bit) floating-point elements, and store the results in "dst". + Swizzle + + + + + Shuffle 128-bits (composed of 4 32-bit integers) selected by "imm8" from "a" and "b", and store the results in "dst". -FOR j := 0 to 3 - i := j*64 - l := j*32 - dst[i+63:i] := ConvertUnsignedIntegerTo_FP64(a[l+31:l]) -ENDFOR -dst[MAX:256] := 0 +DEFINE SELECT4(src, control) { + CASE(control[1:0]) OF + 0: tmp[127:0] := src[127:0] + 1: tmp[127:0] := src[255:128] + 2: tmp[127:0] := src[383:256] + 3: tmp[127:0] := src[511:384] + ESAC + RETURN tmp[127:0] +} +dst[127:0] := SELECT4(a[511:0], imm8[1:0]) +dst[255:128] := SELECT4(a[511:0], imm8[3:2]) +dst[383:256] := SELECT4(b[511:0], imm8[5:4]) +dst[511:384] := SELECT4(b[511:0], imm8[7:6]) +dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point + Integer - AVX512VL AVX512F - Convert - - - - Convert packed unsigned 32-bit integers in "a" to packed double-precision (64-bit) floating-point elements, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - - -FOR j := 0 to 3 + Swizzle + + + + + + + Shuffle 128-bits (composed of 2 64-bit integers) selected by "imm8" from "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +DEFINE SELECT4(src, control) { + CASE(control[1:0]) OF + 0: tmp[127:0] := src[127:0] + 1: tmp[127:0] := src[255:128] + 2: tmp[127:0] := src[383:256] + 3: tmp[127:0] := src[511:384] + ESAC + RETURN tmp[127:0] +} +tmp_dst[127:0] := SELECT4(a[511:0], imm8[1:0]) +tmp_dst[255:128] := SELECT4(a[511:0], imm8[3:2]) +tmp_dst[383:256] := SELECT4(b[511:0], imm8[5:4]) +tmp_dst[511:384] := SELECT4(b[511:0], imm8[7:6]) +FOR j := 0 to 7 i := j*64 - l := j*32 IF k[j] - dst[i+63:i] := ConvertUnsignedIntegerTo_FP64(a[l+31:l]) + dst[i+63:i] := tmp_dst[i+63:i] ELSE dst[i+63:i] := src[i+63:i] - FI + FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point + Integer - AVX512VL AVX512F - Convert - - - Convert packed unsigned 32-bit integers in "a" to packed double-precision (64-bit) floating-point elements, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Swizzle + + + + + + Shuffle 128-bits (composed of 2 64-bit integers) selected by "imm8" from "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 3 +DEFINE SELECT4(src, control) { + CASE(control[1:0]) OF + 0: tmp[127:0] := src[127:0] + 1: tmp[127:0] := src[255:128] + 2: tmp[127:0] := src[383:256] + 3: tmp[127:0] := src[511:384] + ESAC + RETURN tmp[127:0] +} +tmp_dst[127:0] := SELECT4(a[511:0], imm8[1:0]) +tmp_dst[255:128] := SELECT4(a[511:0], imm8[3:2]) +tmp_dst[383:256] := SELECT4(b[511:0], imm8[5:4]) +tmp_dst[511:384] := SELECT4(b[511:0], imm8[7:6]) +FOR j := 0 to 7 i := j*64 - l := j*32 IF k[j] - dst[i+63:i] := ConvertUnsignedIntegerTo_FP64(a[l+31:l]) + dst[i+63:i] := tmp_dst[i+63:i] ELSE dst[i+63:i] := 0 - FI + FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point + Integer - AVX512VL AVX512F - Convert - - Convert packed unsigned 32-bit integers in "a" to packed double-precision (64-bit) floating-point elements, and store the results in "dst". + Swizzle + + + + + Shuffle 128-bits (composed of 2 64-bit integers) selected by "imm8" from "a" and "b", and store the results in "dst". -FOR j := 0 to 1 - i := j*64 - l := j*32 - dst[i+63:i] := ConvertUnsignedIntegerTo_FP64(a[l+31:l]) -ENDFOR -dst[MAX:128] := 0 +DEFINE SELECT4(src, control) { + CASE(control[1:0]) OF + 0: tmp[127:0] := src[127:0] + 1: tmp[127:0] := src[255:128] + 2: tmp[127:0] := src[383:256] + 3: tmp[127:0] := src[511:384] + ESAC + RETURN tmp[127:0] +} +dst[127:0] := SELECT4(a[511:0], imm8[1:0]) +dst[255:128] := SELECT4(a[511:0], imm8[3:2]) +dst[383:256] := SELECT4(b[511:0], imm8[5:4]) +dst[511:384] := SELECT4(b[511:0], imm8[7:6]) +dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - Integer - AVX512VL AVX512F - Convert - - - - Convert packed unsigned 32-bit integers in "a" to packed double-precision (64-bit) floating-point elements, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - + Swizzle + + + + + + + Shuffle double-precision (64-bit) floating-point elements within 128-bit lanes using the control in "imm8", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 1 +tmp_dst[63:0] := (imm8[0] == 0) ? a[63:0] : a[127:64] +tmp_dst[127:64] := (imm8[1] == 0) ? b[63:0] : b[127:64] +tmp_dst[191:128] := (imm8[2] == 0) ? a[191:128] : a[255:192] +tmp_dst[255:192] := (imm8[3] == 0) ? b[191:128] : b[255:192] +tmp_dst[319:256] := (imm8[4] == 0) ? a[319:256] : a[383:320] +tmp_dst[383:320] := (imm8[5] == 0) ? b[319:256] : b[383:320] +tmp_dst[447:384] := (imm8[6] == 0) ? a[447:384] : a[511:448] +tmp_dst[511:448] := (imm8[7] == 0) ? b[447:384] : b[511:448] +FOR j := 0 to 7 i := j*64 - l := j*32 IF k[j] - dst[i+63:i] := ConvertUnsignedIntegerTo_FP64(a[l+31:l]) + dst[i+63:i] := tmp_dst[i+63:i] ELSE dst[i+63:i] := src[i+63:i] - FI + FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - Integer - AVX512VL AVX512F - Convert - - - Convert packed unsigned 32-bit integers in "a" to packed double-precision (64-bit) floating-point elements, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Swizzle + + + + + + Shuffle double-precision (64-bit) floating-point elements within 128-bit lanes using the control in "imm8", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 1 +tmp_dst[63:0] := (imm8[0] == 0) ? a[63:0] : a[127:64] +tmp_dst[127:64] := (imm8[1] == 0) ? b[63:0] : b[127:64] +tmp_dst[191:128] := (imm8[2] == 0) ? a[191:128] : a[255:192] +tmp_dst[255:192] := (imm8[3] == 0) ? b[191:128] : b[255:192] +tmp_dst[319:256] := (imm8[4] == 0) ? a[319:256] : a[383:320] +tmp_dst[383:320] := (imm8[5] == 0) ? b[319:256] : b[383:320] +tmp_dst[447:384] := (imm8[6] == 0) ? a[447:384] : a[511:448] +tmp_dst[511:448] := (imm8[7] == 0) ? b[447:384] : b[511:448] +FOR j := 0 to 7 i := j*64 - l := j*32 IF k[j] - dst[i+63:i] := ConvertUnsignedIntegerTo_FP64(a[l+31:l]) + dst[i+63:i] := tmp_dst[i+63:i] ELSE dst[i+63:i] := 0 - FI + FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - Integer - AVX512VL - AVX512DQ - Convert - - Convert packed unsigned 64-bit integers in "a" to packed double-precision (64-bit) floating-point elements, and store the results in "dst". + AVX512F + Swizzle + + + + + Shuffle double-precision (64-bit) floating-point elements within 128-bit lanes using the control in "imm8", and store the results in "dst". -FOR j := 0 to 3 - i := j*64 - dst[i+63:i] := ConvertUnsignedInt64_To_FP64(a[i+63:i]) -ENDFOR -dst[MAX:256] := 0 +dst[63:0] := (imm8[0] == 0) ? a[63:0] : a[127:64] +dst[127:64] := (imm8[1] == 0) ? b[63:0] : b[127:64] +dst[191:128] := (imm8[2] == 0) ? a[191:128] : a[255:192] +dst[255:192] := (imm8[3] == 0) ? b[191:128] : b[255:192] +dst[319:256] := (imm8[4] == 0) ? a[319:256] : a[383:320] +dst[383:320] := (imm8[5] == 0) ? b[319:256] : b[383:320] +dst[447:384] := (imm8[6] == 0) ? a[447:384] : a[511:448] +dst[511:448] := (imm8[7] == 0) ? b[447:384] : b[511:448] +dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - Integer - AVX512VL - AVX512DQ - Convert - - - - Convert packed unsigned 64-bit integers in "a" to packed double-precision (64-bit) floating-point elements, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - + AVX512F + Swizzle + + + + + + + Shuffle single-precision (32-bit) floating-point elements in "a" within 128-bit lanes using the control in "imm8", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*64 +DEFINE SELECT4(src, control) { + CASE(control[1:0]) OF + 0: tmp[31:0] := src[31:0] + 1: tmp[31:0] := src[63:32] + 2: tmp[31:0] := src[95:64] + 3: tmp[31:0] := src[127:96] + ESAC + RETURN tmp[31:0] +} +tmp_dst[31:0] := SELECT4(a[127:0], imm8[1:0]) +tmp_dst[63:32] := SELECT4(a[127:0], imm8[3:2]) +tmp_dst[95:64] := SELECT4(b[127:0], imm8[5:4]) +tmp_dst[127:96] := SELECT4(b[127:0], imm8[7:6]) +tmp_dst[159:128] := SELECT4(a[255:128], imm8[1:0]) +tmp_dst[191:160] := SELECT4(a[255:128], imm8[3:2]) +tmp_dst[223:192] := SELECT4(b[255:128], imm8[5:4]) +tmp_dst[255:224] := SELECT4(b[255:128], imm8[7:6]) +tmp_dst[287:256] := SELECT4(a[383:256], imm8[1:0]) +tmp_dst[319:288] := SELECT4(a[383:256], imm8[3:2]) +tmp_dst[351:320] := SELECT4(b[383:256], imm8[5:4]) +tmp_dst[383:352] := SELECT4(b[383:256], imm8[7:6]) +tmp_dst[415:384] := SELECT4(a[511:384], imm8[1:0]) +tmp_dst[447:416] := SELECT4(a[511:384], imm8[3:2]) +tmp_dst[479:448] := SELECT4(b[511:384], imm8[5:4]) +tmp_dst[511:480] := SELECT4(b[511:384], imm8[7:6]) +FOR j := 0 to 15 + i := j*32 IF k[j] - dst[i+63:i] := ConvertUnsignedInt64_To_FP64(a[i+63:i]) + dst[i+31:i] := tmp_dst[i+31:i] ELSE - dst[i+63:i] := src[i+63:i] + dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - Integer - AVX512VL - AVX512DQ - Convert - - - Convert packed unsigned 64-bit integers in "a" to packed double-precision (64-bit) floating-point elements, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + AVX512F + Swizzle + + + + + + Shuffle single-precision (32-bit) floating-point elements in "a" within 128-bit lanes using the control in "imm8", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*64 +DEFINE SELECT4(src, control) { + CASE(control[1:0]) OF + 0: tmp[31:0] := src[31:0] + 1: tmp[31:0] := src[63:32] + 2: tmp[31:0] := src[95:64] + 3: tmp[31:0] := src[127:96] + ESAC + RETURN tmp[31:0] +} +tmp_dst[31:0] := SELECT4(a[127:0], imm8[1:0]) +tmp_dst[63:32] := SELECT4(a[127:0], imm8[3:2]) +tmp_dst[95:64] := SELECT4(b[127:0], imm8[5:4]) +tmp_dst[127:96] := SELECT4(b[127:0], imm8[7:6]) +tmp_dst[159:128] := SELECT4(a[255:128], imm8[1:0]) +tmp_dst[191:160] := SELECT4(a[255:128], imm8[3:2]) +tmp_dst[223:192] := SELECT4(b[255:128], imm8[5:4]) +tmp_dst[255:224] := SELECT4(b[255:128], imm8[7:6]) +tmp_dst[287:256] := SELECT4(a[383:256], imm8[1:0]) +tmp_dst[319:288] := SELECT4(a[383:256], imm8[3:2]) +tmp_dst[351:320] := SELECT4(b[383:256], imm8[5:4]) +tmp_dst[383:352] := SELECT4(b[383:256], imm8[7:6]) +tmp_dst[415:384] := SELECT4(a[511:384], imm8[1:0]) +tmp_dst[447:416] := SELECT4(a[511:384], imm8[3:2]) +tmp_dst[479:448] := SELECT4(b[511:384], imm8[5:4]) +tmp_dst[511:480] := SELECT4(b[511:384], imm8[7:6]) +FOR j := 0 to 15 + i := j*32 IF k[j] - dst[i+63:i] := ConvertUnsignedInt64_To_FP64(a[i+63:i]) + dst[i+31:i] := tmp_dst[i+31:i] ELSE - dst[i+63:i] := 0 + dst[i+31:i] := 0 FI ENDFOR -dst[MAX:256] := 0 - - -
immintrin.h
- - - Floating Point - Integer - AVX512DQ - Convert - - - Convert packed unsigned 64-bit integers in "a" to packed double-precision (64-bit) floating-point elements, and store the results in "dst". - [round_note] - - -FOR j := 0 to 7 - i := j*64 - dst[i+63:i] := ConvertUnsignedInt64_To_FP64(a[i+63:i]) -ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - Integer - AVX512DQ - Convert - - Convert packed unsigned 64-bit integers in "a" to packed double-precision (64-bit) floating-point elements, and store the results in "dst". + AVX512F + Swizzle + + + + + Shuffle single-precision (32-bit) floating-point elements in "a" within 128-bit lanes using the control in "imm8", and store the results in "dst". -FOR j := 0 to 7 - i := j*64 - dst[i+63:i] := ConvertUnsignedInt64_To_FP64(a[i+63:i]) -ENDFOR +DEFINE SELECT4(src, control) { + CASE(control[1:0]) OF + 0: tmp[31:0] := src[31:0] + 1: tmp[31:0] := src[63:32] + 2: tmp[31:0] := src[95:64] + 3: tmp[31:0] := src[127:96] + ESAC + RETURN tmp[31:0] +} +dst[31:0] := SELECT4(a[127:0], imm8[1:0]) +dst[63:32] := SELECT4(a[127:0], imm8[3:2]) +dst[95:64] := SELECT4(b[127:0], imm8[5:4]) +dst[127:96] := SELECT4(b[127:0], imm8[7:6]) +dst[159:128] := SELECT4(a[255:128], imm8[1:0]) +dst[191:160] := SELECT4(a[255:128], imm8[3:2]) +dst[223:192] := SELECT4(b[255:128], imm8[5:4]) +dst[255:224] := SELECT4(b[255:128], imm8[7:6]) +dst[287:256] := SELECT4(a[383:256], imm8[1:0]) +dst[319:288] := SELECT4(a[383:256], imm8[3:2]) +dst[351:320] := SELECT4(b[383:256], imm8[5:4]) +dst[383:352] := SELECT4(b[383:256], imm8[7:6]) +dst[415:384] := SELECT4(a[511:384], imm8[1:0]) +dst[447:416] := SELECT4(a[511:384], imm8[3:2]) +dst[479:448] := SELECT4(b[511:384], imm8[5:4]) +dst[511:480] := SELECT4(b[511:384], imm8[7:6]) dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - Integer - AVX512DQ - Convert - - - - - Convert packed unsigned 64-bit integers in "a" to packed double-precision (64-bit) floating-point elements, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - [round_note] - + AVX512F + Elementary Math Functions + + + + + Compute the square root of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). FOR j := 0 to 7 i := j*64 IF k[j] - dst[i+63:i] := ConvertUnsignedInt64_To_FP64(a[i+63:i]) + dst[i+63:i] := SQRT(a[i+63:i]) ELSE dst[i+63:i] := src[i+63:i] FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - Integer - AVX512DQ - Convert - - - - Convert packed unsigned 64-bit integers in "a" to packed double-precision (64-bit) floating-point elements, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - + AVX512F + Elementary Math Functions + + + + + + Compute the square root of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + [round_note] FOR j := 0 to 7 i := j*64 IF k[j] - dst[i+63:i] := ConvertUnsignedInt64_To_FP64(a[i+63:i]) + dst[i+63:i] := SQRT(a[i+63:i]) ELSE dst[i+63:i] := src[i+63:i] FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - Integer - AVX512DQ - Convert - - - - Convert packed unsigned 64-bit integers in "a" to packed double-precision (64-bit) floating-point elements, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - [round_note] - + AVX512F + Elementary Math Functions + + + + Compute the square root of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). FOR j := 0 to 7 i := j*64 IF k[j] - dst[i+63:i] := ConvertUnsignedInt64_To_FP64(a[i+63:i]) + dst[i+63:i] := SQRT(a[i+63:i]) ELSE dst[i+63:i] := 0 FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - Integer - AVX512DQ - Convert - - - Convert packed unsigned 64-bit integers in "a" to packed double-precision (64-bit) floating-point elements, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + AVX512F + Elementary Math Functions + + + + + Compute the square root of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + [round_note]. FOR j := 0 to 7 i := j*64 IF k[j] - dst[i+63:i] := ConvertUnsignedInt64_To_FP64(a[i+63:i]) + dst[i+63:i] := SQRT(a[i+63:i]) ELSE dst[i+63:i] := 0 FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - Integer - AVX512VL - AVX512DQ - Convert - - Convert packed unsigned 64-bit integers in "a" to packed double-precision (64-bit) floating-point elements, and store the results in "dst". + AVX512F + Elementary Math Functions + + + Compute the square root of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst". -FOR j := 0 to 1 +FOR j := 0 to 7 i := j*64 - dst[i+63:i] := ConvertUnsignedInt64_To_FP64(a[i+63:i]) + dst[i+63:i] := SQRT(a[i+63:i]) ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - Integer - AVX512VL - AVX512DQ - Convert - - - - Convert packed unsigned 64-bit integers in "a" to packed double-precision (64-bit) floating-point elements, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - + AVX512F + Elementary Math Functions + + + + Compute the square root of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst". + [round_note]. -FOR j := 0 to 1 +FOR j := 0 to 7 i := j*64 - IF k[j] - dst[i+63:i] := ConvertUnsignedInt64_To_FP64(a[i+63:i]) - ELSE - dst[i+63:i] := src[i+63:i] - FI + dst[i+63:i] := SQRT(a[i+63:i]) ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - Integer - AVX512VL - AVX512DQ - Convert - - - Convert packed unsigned 64-bit integers in "a" to packed double-precision (64-bit) floating-point elements, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + AVX512F + Elementary Math Functions + + + + + Compute the square root of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 1 - i := j*64 +FOR j := 0 to 15 + i := j*32 IF k[j] - dst[i+63:i] := ConvertUnsignedInt64_To_FP64(a[i+63:i]) + dst[i+31:i] := SQRT(a[i+31:i]) ELSE - dst[i+63:i] := 0 + dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - Integer - AVX512VL - AVX512DQ - Convert - - Convert packed unsigned 64-bit integers in "a" to packed single-precision (32-bit) floating-point elements, and store the results in "dst". + AVX512F + Elementary Math Functions + + + + + + Compute the square root of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + [round_note] -FOR j := 0 to 3 - i := j*64 - l := j*32 - dst[l+31:l] := ConvertUnsignedInt64_To_FP32(a[i+63:i]) +FOR j := 0 to 15 + i := j*32 + IF k[j] + dst[i+31:i] := SQRT(a[i+31:i]) + ELSE + dst[i+31:i] := src[i+31:i] + FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - Integer - AVX512VL - AVX512DQ - Convert - - - - Convert packed unsigned 64-bit integers in "a" to packed single-precision (32-bit) floating-point elements, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - + AVX512F + Elementary Math Functions + + + + Compute the square root of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*64 - l := j*32 +FOR j := 0 to 15 + i := j*32 IF k[j] - dst[l+31:l] := ConvertUnsignedInt64_To_FP32(a[i+63:i]) + dst[i+31:i] := SQRT(a[i+31:i]) ELSE - dst[l+31:l] := src[l+31:l] + dst[i+31:i] := 0 FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - Integer - AVX512VL - AVX512DQ - Convert - - - Convert packed unsigned 64-bit integers in "a" to packed single-precision (32-bit) floating-point elements, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + AVX512F + Elementary Math Functions + + + + + Compute the square root of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + [round_note] -FOR j := 0 to 3 - i := j*64 - l := j*32 +FOR j := 0 to 15 + i := j*32 IF k[j] - dst[l+31:l] := ConvertUnsignedInt64_To_FP32(a[i+63:i]) + dst[i+31:i] := SQRT(a[i+31:i]) ELSE - dst[l+31:l] := 0 + dst[i+31:i] := 0 FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - Integer - AVX512DQ - Convert - - - Convert packed unsigned 64-bit integers in "a" to packed single-precision (32-bit) floating-point elements, and store the results in "dst". - [round_note] - + AVX512F + Elementary Math Functions + + + Compute the square root of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst". -FOR j := 0 to 7 - i := j*64 - l := j*32 - dst[l+31:l] := ConvertUnsignedInt64_To_FP32(a[i+63:i]) +FOR j := 0 to 15 + i := j*32 + dst[i+31:i] := SQRT(a[i+31:i]) ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - Integer - AVX512DQ - Convert - - Convert packed unsigned 64-bit integers in "a" to packed single-precision (32-bit) floating-point elements, and store the results in "dst". + AVX512F + Elementary Math Functions + + + + Compute the square root of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst". + [round_note]. -FOR j := 0 to 7 - i := j*64 - l := j*32 - dst[l+31:l] := ConvertUnsignedInt64_To_FP32(a[i+63:i]) +FOR j := 0 to 15 + i := j*32 + dst[i+31:i] := SQRT(a[i+31:i]) ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - Integer - AVX512DQ - Convert - - - - - Convert packed unsigned 64-bit integers in "a" to packed single-precision (32-bit) floating-point elements, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - [round_note] - + AVX512F + Elementary Math Functions + + + + + + + Compute the square root of the lower double-precision (64-bit) floating-point element in "b", store the result in the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". + [round_note] -FOR j := 0 to 7 - i := j*64 - l := j*32 - IF k[j] - dst[l+31:l] := ConvertUnsignedInt64_To_FP32(a[i+63:i]) - ELSE - dst[l+31:l] := src[l+31:l] - FI -ENDFOR -dst[MAX:256] := 0 +IF k[0] + dst[63:0] := SQRT(b[63:0]) +ELSE + dst[63:0] := src[63:0] +FI +dst[127:64] := a[127:64] +dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point - Integer - AVX512DQ - Convert - - - - Convert packed unsigned 64-bit integers in "a" to packed single-precision (32-bit) floating-point elements, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - + AVX512F + Elementary Math Functions + + + + + + Compute the square root of the lower double-precision (64-bit) floating-point element in "b", store the result in the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". -FOR j := 0 to 7 - i := j*64 - l := j*32 - IF k[j] - dst[l+31:l] := ConvertUnsignedInt64_To_FP32(a[i+63:i]) - ELSE - dst[l+31:l] := src[l+31:l] - FI -ENDFOR -dst[MAX:256] := 0 +IF k[0] + dst[63:0] := SQRT(b[63:0]) +ELSE + dst[63:0] := src[63:0] +FI +dst[127:64] := a[127:64] +dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point - Integer - AVX512DQ - Convert - - - - Convert packed unsigned 64-bit integers in "a" to packed single-precision (32-bit) floating-point elements, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - [round_note] - + AVX512F + Elementary Math Functions + + + + + + Compute the square root of the lower double-precision (64-bit) floating-point element in "b", store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". + [round_note] -FOR j := 0 to 7 - i := j*64 - l := j*32 - IF k[j] - dst[l+31:l] := ConvertUnsignedInt64_To_FP32(a[i+63:i]) - ELSE - dst[l+31:l] := 0 - FI -ENDFOR -dst[MAX:256] := 0 +IF k[0] + dst[63:0] := SQRT(b[63:0]) +ELSE + dst[63:0] := 0 +FI +dst[127:64] := a[127:64] +dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point - Integer - AVX512DQ - Convert - - - Convert packed unsigned 64-bit integers in "a" to packed single-precision (32-bit) floating-point elements, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + AVX512F + Elementary Math Functions + + + + + Compute the square root of the lower double-precision (64-bit) floating-point element in "b", store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". -FOR j := 0 to 7 - i := j*64 - l := j*32 - IF k[j] - dst[l+31:l] := ConvertUnsignedInt64_To_FP32(a[i+63:i]) - ELSE - dst[l+31:l] := 0 - FI -ENDFOR -dst[MAX:256] := 0 +IF k[0] + dst[63:0] := SQRT(b[63:0]) +ELSE + dst[63:0] := 0 +FI +dst[127:64] := a[127:64] +dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point - Integer - AVX512VL - AVX512DQ - Convert - - Convert packed unsigned 64-bit integers in "a" to packed single-precision (32-bit) floating-point elements, and store the results in "dst". + AVX512F + Elementary Math Functions + + + + + Compute the square root of the lower double-precision (64-bit) floating-point element in "b", store the result in the lower element of "dst", and copy the upper element from "a" to the upper element of "dst". + [round_note] -FOR j := 0 to 1 - i := j*64 - l := j*32 - dst[l+31:l] := ConvertUnsignedInt64_To_FP32(a[i+63:i]) -ENDFOR -dst[MAX:64] := 0 +dst[63:0] := SQRT(b[63:0]) +dst[127:64] := a[127:64] +dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point - Integer - AVX512VL - AVX512DQ - Convert - - - - Convert packed unsigned 64-bit integers in "a" to packed single-precision (32-bit) floating-point elements, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - + AVX512F + Elementary Math Functions + + + + + + + Compute the square root of the lower single-precision (32-bit) floating-point element in "b", store the result in the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". + [round_note] -FOR j := 0 to 1 - i := j*64 - l := j*32 - IF k[j] - dst[l+31:l] := ConvertUnsignedInt64_To_FP32(a[i+63:i]) - ELSE - dst[l+31:l] := src[l+31:l] - FI -ENDFOR -dst[MAX:64] := 0 +IF k[0] + dst[31:0] := SQRT(b[31:0]) +ELSE + dst[31:0] := src[31:0] +FI +dst[127:32] := a[127:32] +dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point - Integer - AVX512VL - AVX512DQ - Convert - - - Convert packed unsigned 64-bit integers in "a" to packed single-precision (32-bit) floating-point elements, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - -FOR j := 0 to 1 - i := j*64 - l := j*32 - IF k[j] - dst[l+31:l] := ConvertUnsignedInt64_To_FP32(a[i+63:i]) - ELSE - dst[l+31:l] := 0 - FI -ENDFOR -dst[MAX:64] := 0 - - -
immintrin.h
- - - Integer - AVX512VL - AVX512BW - Miscellaneous - - - - Compute the sum of absolute differences (SADs) of quadruplets of unsigned 8-bit integers in "a" compared to those in "b", and store the 16-bit results in "dst". - Four SADs are performed on four 8-bit quadruplets for each 64-bit lane. The first two SADs use the lower 8-bit quadruplet of the lane from "a", and the last two SADs use the uppper 8-bit quadruplet of the lane from "a". Quadruplets from "b" are selected from within 128-bit lanes according to the control in "imm8", and each SAD in each 64-bit lane uses the selected quadruplet at 8-bit offsets. - + AVX512F + Elementary Math Functions + + + + + + Compute the square root of the lower single-precision (32-bit) floating-point element in "b", store the result in the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". -FOR j := 0 to 1 - i := j*128 - tmp[i+31:i] := select(b[i+127:i], imm8[1:0]) - tmp[i+63:i+32] := select(b[i+127:i], imm8[3:2]) - tmp[i+95:i+64] := select(b[i+127:i], imm8[5:4]) - tmp[i+127:i+96] := select(b[i+127:i], imm8[7:6]) -ENDFOR - -FOR j := 0 to 3 - i := j*64 - dst[i+15:i] := ABS(a[i+7:i] - tmp[i+7:i]) + ABS(a[i+15:i+8] - tmp[i+15:i+8]) - + ABS(a[i+23:i+16] - tmp[i+23:i+16]) + ABS(a[i+31:i+24] - tmp[i+31:i+24]) - - dst[i+31:i+16] := ABS(a[i+7:i] - tmp[i+15:i+8]) + ABS(a[i+15:i+8] - tmp[i+23:i+16]) - + ABS(a[i+23:i+16] - tmp[i+31:i+24]) + ABS(a[i+31:i+24] - tmp[i+39:i+32]) - - dst[i+47:i+32] := ABS(a[i+39:i+32] - tmp[i+23:i+16]) + ABS(a[i+47:i+40] - tmp[i+31:i+24]) - + ABS(a[i+55:i+48] - tmp[i+39:i+32]) + ABS(a[i+63:i+56] - tmp[i+47:i+40]) - - dst[i+63:i+48] := ABS(a[i+39:i+32] - tmp[i+31:i+24]) + ABS(a[i+47:i+40] - tmp[i+39:i+32]) - + ABS(a[i+55:i+48] - tmp[i+47:i+40]) + ABS(a[i+63:i+56] - tmp[i+55:i+48]) -ENDFOR -dst[MAX:256] := 0 +IF k[0] + dst[31:0] := SQRT(b[31:0]) +ELSE + dst[31:0] := src[31:0] +FI +dst[127:32] := a[127:32] +dst[MAX:128] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512BW - Miscellaneous - - - - - - Compute the sum of absolute differences (SADs) of quadruplets of unsigned 8-bit integers in "a" compared to those in "b", and store the 16-bit results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - Four SADs are performed on four 8-bit quadruplets for each 64-bit lane. The first two SADs use the lower 8-bit quadruplet of the lane from "a", and the last two SADs use the uppper 8-bit quadruplet of the lane from "a". Quadruplets from "b" are selected from within 128-bit lanes according to the control in "imm8", and each SAD in each 64-bit lane uses the selected quadruplet at 8-bit offsets. - + + Floating Point + AVX512F + Elementary Math Functions + + + + + + Compute the square root of the lower single-precision (32-bit) floating-point element in "b", store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". + [round_note] -FOR j := 0 to 1 - i := j*128 - tmp[i+31:i] := select(b[i+127:i], imm8[1:0]) - tmp[i+63:i+32] := select(b[i+127:i], imm8[3:2]) - tmp[i+95:i+64] := select(b[i+127:i], imm8[5:4]) - tmp[i+127:i+96] := select(b[i+127:i], imm8[7:6]) -ENDFOR - -FOR j := 0 to 3 - i := j*64 - tmp_dst[i+15:i] := ABS(a[i+7:i] - tmp[i+7:i]) + ABS(a[i+15:i+8] - tmp[i+15:i+8]) - + ABS(a[i+23:i+16] - tmp[i+23:i+16]) + ABS(a[i+31:i+24] - tmp[i+31:i+24]) - - tmp_dst[i+31:i+16] := ABS(a[i+7:i] - tmp[i+15:i+8]) + ABS(a[i+15:i+8] - tmp[i+23:i+16]) - + ABS(a[i+23:i+16] - tmp[i+31:i+24]) + ABS(a[i+31:i+24] - tmp[i+39:i+32]) - - tmp_dst[i+47:i+32] := ABS(a[i+39:i+32] - tmp[i+23:i+16]) + ABS(a[i+47:i+40] - tmp[i+31:i+24]) - + ABS(a[i+55:i+48] - tmp[i+39:i+32]) + ABS(a[i+63:i+56] - tmp[i+47:i+40]) - - tmp_dst[i+63:i+48] := ABS(a[i+39:i+32] - tmp[i+31:i+24]) + ABS(a[i+47:i+40] - tmp[i+39:i+32]) - + ABS(a[i+55:i+48] - tmp[i+47:i+40]) + ABS(a[i+63:i+56] - tmp[i+55:i+48]) -ENDFOR - -FOR j := 0 to 15 - i := j*16 - IF k[j] - dst[i+15:i] := tmp_dst[i+15:i] - ELSE - dst[i+15:i] := src[i+15:i] - FI -ENDFOR -dst[MAX:256] := 0 +IF k[0] + dst[31:0] := SQRT(b[31:0]) +ELSE + dst[31:0] := 0 +FI +dst[127:32] := a[127:32] +dst[MAX:128] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512BW - Miscellaneous - - - - - Compute the sum of absolute differences (SADs) of quadruplets of unsigned 8-bit integers in "a" compared to those in "b", and store the 16-bit results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - Four SADs are performed on four 8-bit quadruplets for each 64-bit lane. The first two SADs use the lower 8-bit quadruplet of the lane from "a", and the last two SADs use the uppper 8-bit quadruplet of the lane from "a". Quadruplets from "b" are selected from within 128-bit lanes according to the control in "imm8", and each SAD in each 64-bit lane uses the selected quadruplet at 8-bit offsets. - + + Floating Point + AVX512F + Elementary Math Functions + + + + + Compute the square root of the lower single-precision (32-bit) floating-point element in "b", store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". -FOR j := 0 to 1 - i := j*128 - tmp[i+31:i] := select(b[i+127:i], imm8[1:0]) - tmp[i+63:i+32] := select(b[i+127:i], imm8[3:2]) - tmp[i+95:i+64] := select(b[i+127:i], imm8[5:4]) - tmp[i+127:i+96] := select(b[i+127:i], imm8[7:6]) -ENDFOR - -FOR j := 0 to 3 - i := j*64 - tmp_dst[i+15:i] := ABS(a[i+7:i] - tmp[i+7:i]) + ABS(a[i+15:i+8] - tmp[i+15:i+8]) - + ABS(a[i+23:i+16] - tmp[i+23:i+16]) + ABS(a[i+31:i+24] - tmp[i+31:i+24]) - - tmp_dst[i+31:i+16] := ABS(a[i+7:i] - tmp[i+15:i+8]) + ABS(a[i+15:i+8] - tmp[i+23:i+16]) - + ABS(a[i+23:i+16] - tmp[i+31:i+24]) + ABS(a[i+31:i+24] - tmp[i+39:i+32]) - - tmp_dst[i+47:i+32] := ABS(a[i+39:i+32] - tmp[i+23:i+16]) + ABS(a[i+47:i+40] - tmp[i+31:i+24]) - + ABS(a[i+55:i+48] - tmp[i+39:i+32]) + ABS(a[i+63:i+56] - tmp[i+47:i+40]) - - tmp_dst[i+63:i+48] := ABS(a[i+39:i+32] - tmp[i+31:i+24]) + ABS(a[i+47:i+40] - tmp[i+39:i+32]) - + ABS(a[i+55:i+48] - tmp[i+47:i+40]) + ABS(a[i+63:i+56] - tmp[i+55:i+48]) -ENDFOR - -FOR j := 0 to 15 - i := j*16 - IF k[j] - dst[i+15:i] := tmp_dst[i+15:i] - ELSE - dst[i+15:i] := 0 - FI -ENDFOR -dst[MAX:256] := 0 +IF k[0] + dst[31:0] := SQRT(b[31:0]) +ELSE + dst[31:0] := 0 +FI +dst[127:32] := a[127:32] +dst[MAX:128] := 0 - +
immintrin.h
 - - Integer - AVX512BW - Miscellaneous - - - - Compute the sum of absolute differences (SADs) of quadruplets of unsigned 8-bit integers in "a" compared to those in "b", and store the 16-bit results in "dst". - Four SADs are performed on four 8-bit quadruplets for each 64-bit lane. The first two SADs use the lower 8-bit quadruplet of the lane from "a", and the last two SADs use the uppper 8-bit quadruplet of the lane from "a". Quadruplets from "b" are selected from within 128-bit lanes according to the control in "imm8", and each SAD in each 64-bit lane uses the selected quadruplet at 8-bit offsets. - + + Floating Point + AVX512F + Elementary Math Functions + + + + + Compute the square root of the lower single-precision (32-bit) floating-point element in "b", store the result in the lower element of "dst", and copy the upper 3 packed elements from "a" to the upper elements of "dst". + [round_note] -FOR j := 0 to 3 - i := j*128 - tmp[i+31:i] := select(b[i+127:i], imm8[1:0]) - tmp[i+63:i+32] := select(b[i+127:i], imm8[3:2]) - tmp[i+95:i+64] := select(b[i+127:i], imm8[5:4]) - tmp[i+127:i+96] := select(b[i+127:i], imm8[7:6]) -ENDFOR - -FOR j := 0 to 7 - i := j*64 - dst[i+15:i] := ABS(a[i+7:i] - tmp[i+7:i]) + ABS(a[i+15:i+8] - tmp[i+15:i+8]) - + ABS(a[i+23:i+16] - tmp[i+23:i+16]) + ABS(a[i+31:i+24] - tmp[i+31:i+24]) - - dst[i+31:i+16] := ABS(a[i+7:i] - tmp[i+15:i+8]) + ABS(a[i+15:i+8] - tmp[i+23:i+16]) - + ABS(a[i+23:i+16] - tmp[i+31:i+24]) + ABS(a[i+31:i+24] - tmp[i+39:i+32]) - - dst[i+47:i+32] := ABS(a[i+39:i+32] - tmp[i+23:i+16]) + ABS(a[i+47:i+40] - tmp[i+31:i+24]) - + ABS(a[i+55:i+48] - tmp[i+39:i+32]) + ABS(a[i+63:i+56] - tmp[i+47:i+40]) - - dst[i+63:i+48] := ABS(a[i+39:i+32] - tmp[i+31:i+24]) + ABS(a[i+47:i+40] - tmp[i+39:i+32]) - + ABS(a[i+55:i+48] - tmp[i+47:i+40]) + ABS(a[i+63:i+56] - tmp[i+55:i+48]) -ENDFOR -dst[MAX:512] := 0 +dst[31:0] := SQRT(b[31:0]) +dst[127:32] := a[127:32] +dst[MAX:128] := 0 - +
immintrin.h
 - - Integer - AVX512BW - Miscellaneous - - - - - - Compute the sum of absolute differences (SADs) of quadruplets of unsigned 8-bit integers in "a" compared to those in "b", and store the 16-bit results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - Four SADs are performed on four 8-bit quadruplets for each 64-bit lane. The first two SADs use the lower 8-bit quadruplet of the lane from "a", and the last two SADs use the uppper 8-bit quadruplet of the lane from "a". Quadruplets from "b" are selected from within 128-bit lanes according to the control in "imm8", and each SAD in each 64-bit lane uses the selected quadruplet at 8-bit offsets. - - -FOR j := 0 to 3 - i := j*128 - tmp[i+31:i] := select(b[i+127:i], imm8[1:0]) - tmp[i+63:i+32] := select(b[i+127:i], imm8[3:2]) - tmp[i+95:i+64] := select(b[i+127:i], imm8[5:4]) - tmp[i+127:i+96] := select(b[i+127:i], imm8[7:6]) -ENDFOR - -FOR j := 0 to 7 - i := j*64 - tmp_dst[i+15:i] := ABS(a[i+7:i] - tmp[i+7:i]) + ABS(a[i+15:i+8] - tmp[i+15:i+8]) - + ABS(a[i+23:i+16] - tmp[i+23:i+16]) + ABS(a[i+31:i+24] - tmp[i+31:i+24]) - - tmp_dst[i+31:i+16] := ABS(a[i+7:i] - tmp[i+15:i+8]) + ABS(a[i+15:i+8] - tmp[i+23:i+16]) - + ABS(a[i+23:i+16] - tmp[i+31:i+24]) + ABS(a[i+31:i+24] - tmp[i+39:i+32]) - - tmp_dst[i+47:i+32] := ABS(a[i+39:i+32] - tmp[i+23:i+16]) + ABS(a[i+47:i+40] - tmp[i+31:i+24]) - + ABS(a[i+55:i+48] - tmp[i+39:i+32]) + ABS(a[i+63:i+56] - tmp[i+47:i+40]) - - tmp_dst[i+63:i+48] := ABS(a[i+39:i+32] - tmp[i+31:i+24]) + ABS(a[i+47:i+40] - tmp[i+39:i+32]) - + ABS(a[i+55:i+48] - tmp[i+47:i+40]) + ABS(a[i+63:i+56] - tmp[i+55:i+48]) -ENDFOR - -FOR j := 0 to 31 - i := j*16 + + Floating Point + AVX512F + Arithmetic + + + + + Subtract packed double-precision (64-bit) floating-point elements in "b" from packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + +FOR j := 0 to 7 + i := j*64 IF k[j] - dst[i+15:i] := tmp_dst[i+15:i] + dst[i+63:i] := a[i+63:i] - b[i+63:i] ELSE - dst[i+15:i] := src[i+15:i] + dst[i+63:i] := 0 FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - AVX512BW - Miscellaneous - - - - - Compute the sum of absolute differences (SADs) of quadruplets of unsigned 8-bit integers in "a" compared to those in "b", and store the 16-bit results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - Four SADs are performed on four 8-bit quadruplets for each 64-bit lane. The first two SADs use the lower 8-bit quadruplet of the lane from "a", and the last two SADs use the uppper 8-bit quadruplet of the lane from "a". Quadruplets from "b" are selected from within 128-bit lanes according to the control in "imm8", and each SAD in each 64-bit lane uses the selected quadruplet at 8-bit offsets. - + + Floating Point + AVX512F + Arithmetic + + + + + + Subtract packed double-precision (64-bit) floating-point elements in "b" from packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + [round_note] -FOR j := 0 to 3 - i := j*128 - tmp[i+31:i] := select(b[i+127:i], imm8[1:0]) - tmp[i+63:i+32] := select(b[i+127:i], imm8[3:2]) - tmp[i+95:i+64] := select(b[i+127:i], imm8[5:4]) - tmp[i+127:i+96] := select(b[i+127:i], imm8[7:6]) -ENDFOR - FOR j := 0 to 7 i := j*64 - tmp_dst[i+15:i] := ABS(a[i+7:i] - tmp[i+7:i]) + ABS(a[i+15:i+8] - tmp[i+15:i+8]) - + ABS(a[i+23:i+16] - tmp[i+23:i+16]) + ABS(a[i+31:i+24] - tmp[i+31:i+24]) - - tmp_dst[i+31:i+16] := ABS(a[i+7:i] - tmp[i+15:i+8]) + ABS(a[i+15:i+8] - tmp[i+23:i+16]) - + ABS(a[i+23:i+16] - tmp[i+31:i+24]) + ABS(a[i+31:i+24] - tmp[i+39:i+32]) - - tmp_dst[i+47:i+32] := ABS(a[i+39:i+32] - tmp[i+23:i+16]) + ABS(a[i+47:i+40] - tmp[i+31:i+24]) - + ABS(a[i+55:i+48] - tmp[i+39:i+32]) + ABS(a[i+63:i+56] - tmp[i+47:i+40]) - - tmp_dst[i+63:i+48] := ABS(a[i+39:i+32] - tmp[i+31:i+24]) + ABS(a[i+47:i+40] - tmp[i+39:i+32]) - + ABS(a[i+55:i+48] - tmp[i+47:i+40]) + ABS(a[i+63:i+56] - tmp[i+55:i+48]) -ENDFOR - -FOR j := 0 to 31 - i := j*16 IF k[j] - dst[i+15:i] := tmp_dst[i+15:i] + dst[i+63:i] := a[i+63:i] - b[i+63:i] ELSE - dst[i+15:i] := 0 + dst[i+63:i] := 0 FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512BW - Miscellaneous - - - - Compute the sum of absolute differences (SADs) of quadruplets of unsigned 8-bit integers in "a" compared to those in "b", and store the 16-bit results in "dst". - Four SADs are performed on four 8-bit quadruplets for each 64-bit lane. The first two SADs use the lower 8-bit quadruplet of the lane from "a", and the last two SADs use the uppper 8-bit quadruplet of the lane from "a". Quadruplets from "b" are selected according to the control in "imm8", and each SAD in each 64-bit lane uses the selected quadruplet at 8-bit offsets. - + + Floating Point + AVX512F + Arithmetic + + + + + Subtract packed single-precision (32-bit) floating-point elements in "b" from packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -tmp[31:0] := select(b[127:0], imm8[1:0]) -tmp[63:32] := select(b[127:0], imm8[3:2]) -tmp[95:64] := select(b[127:0], imm8[5:4]) -tmp[127:96] := select(b[127:0], imm8[7:6]) - -FOR j := 0 to 1 - i := j*64 - dst[i+15:i] := ABS(a[i+7:i] - tmp[i+7:i]) + ABS(a[i+15:i+8] - tmp[i+15:i+8]) - + ABS(a[i+23:i+16] - tmp[i+23:i+16]) + ABS(a[i+31:i+24] - tmp[i+31:i+24]) - - dst[i+31:i+16] := ABS(a[i+7:i] - tmp[i+15:i+8]) + ABS(a[i+15:i+8] - tmp[i+23:i+16]) - + ABS(a[i+23:i+16] - tmp[i+31:i+24]) + ABS(a[i+31:i+24] - tmp[i+39:i+32]) - - dst[i+47:i+32] := ABS(a[i+39:i+32] - tmp[i+23:i+16]) + ABS(a[i+47:i+40] - tmp[i+31:i+24]) - + ABS(a[i+55:i+48] - tmp[i+39:i+32]) + ABS(a[i+63:i+56] - tmp[i+47:i+40]) - - dst[i+63:i+48] := ABS(a[i+39:i+32] - tmp[i+31:i+24]) + ABS(a[i+47:i+40] - tmp[i+39:i+32]) - + ABS(a[i+55:i+48] - tmp[i+47:i+40]) + ABS(a[i+63:i+56] - tmp[i+55:i+48]) +FOR j := 0 to 15 + i := j*32 + IF k[j] + dst[i+31:i] := a[i+31:i] - b[i+31:i] + ELSE + dst[i+31:i] := 0 + FI ENDFOR - -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512BW - Miscellaneous - - - - - - Compute the sum of absolute differences (SADs) of quadruplets of unsigned 8-bit integers in "a" compared to those in "b", and store the 16-bit results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - Four SADs are performed on four 8-bit quadruplets for each 64-bit lane. The first two SADs use the lower 8-bit quadruplet of the lane from "a", and the last two SADs use the uppper 8-bit quadruplet of the lane from "a". Quadruplets from "b" are selected according to the control in "imm8", and each SAD in each 64-bit lane uses the selected quadruplet at 8-bit offsets. - + + Floating Point + AVX512F + Arithmetic + + + + + + Subtract packed single-precision (32-bit) floating-point elements in "b" from packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + [round_note] -tmp[31:0] := select(b[127:0], imm8[1:0]) -tmp[63:32] := select(b[127:0], imm8[3:2]) -tmp[95:64] := select(b[127:0], imm8[5:4]) -tmp[127:96] := select(b[127:0], imm8[7:6]) - -FOR j := 0 to 1 - i := j*64 - tmp_dst[i+15:i] := ABS(a[i+7:i] - tmp[i+7:i]) + ABS(a[i+15:i+8] - tmp[i+15:i+8]) - + ABS(a[i+23:i+16] - tmp[i+23:i+16]) + ABS(a[i+31:i+24] - tmp[i+31:i+24]) - - tmp_dst[i+31:i+16] := ABS(a[i+7:i] - tmp[i+15:i+8]) + ABS(a[i+15:i+8] - tmp[i+23:i+16]) - + ABS(a[i+23:i+16] - tmp[i+31:i+24]) + ABS(a[i+31:i+24] - tmp[i+39:i+32]) - - tmp_dst[i+47:i+32] := ABS(a[i+39:i+32] - tmp[i+23:i+16]) + ABS(a[i+47:i+40] - tmp[i+31:i+24]) - + ABS(a[i+55:i+48] - tmp[i+39:i+32]) + ABS(a[i+63:i+56] - tmp[i+47:i+40]) - - tmp_dst[i+63:i+48] := ABS(a[i+39:i+32] - tmp[i+31:i+24]) + ABS(a[i+47:i+40] - tmp[i+39:i+32]) - + ABS(a[i+55:i+48] - tmp[i+47:i+40]) + ABS(a[i+63:i+56] - tmp[i+55:i+48]) -ENDFOR - -FOR j := 0 to 7 - i := j*16 +FOR j := 0 to 15 + i := j*32 IF k[j] - dst[i+15:i] := tmp_dst[i+15:i] + dst[i+31:i] := a[i+31:i] - b[i+31:i] ELSE - dst[i+15:i] := src[i+15:i] - FI + dst[i+31:i] := 0 + FI ENDFOR +dst[MAX:512] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512F + Arithmetic + + + + + + + Subtract the lower double-precision (64-bit) floating-point element in "b" from the lower double-precision (64-bit) floating-point element in "a", store the result in the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". + [round_note] + +IF k[0] + dst[63:0] := a[63:0] - b[63:0] +ELSE + dst[63:0] := src[63:0] +FI +dst[127:64] := a[127:64] dst[MAX:128] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512BW - Miscellaneous - - - - - Compute the sum of absolute differences (SADs) of quadruplets of unsigned 8-bit integers in "a" compared to those in "b", and store the 16-bit results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - Four SADs are performed on four 8-bit quadruplets for each 64-bit lane. The first two SADs use the lower 8-bit quadruplet of the lane from "a", and the last two SADs use the uppper 8-bit quadruplet of the lane from "a". Quadruplets from "b" are selected according to the control in "imm8", and each SAD in each 64-bit lane uses the selected quadruplet at 8-bit offsets. - + + Floating Point + AVX512F + Arithmetic + + + + + + Subtract the lower double-precision (64-bit) floating-point element in "b" from the lower double-precision (64-bit) floating-point element in "a", store the result in the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". -tmp[31:0] := select(b[127:0], imm8[1:0]) -tmp[63:32] := select(b[127:0], imm8[3:2]) -tmp[95:64] := select(b[127:0], imm8[5:4]) -tmp[127:96] := select(b[127:0], imm8[7:6]) - -FOR j := 0 to 1 - i := j*64 - tmp_dst[i+15:i] := ABS(a[i+7:i] - tmp[i+7:i]) + ABS(a[i+15:i+8] - tmp[i+15:i+8]) - + ABS(a[i+23:i+16] - tmp[i+23:i+16]) + ABS(a[i+31:i+24] - tmp[i+31:i+24]) - - tmp_dst[i+31:i+16] := ABS(a[i+7:i] - tmp[i+15:i+8]) + ABS(a[i+15:i+8] - tmp[i+23:i+16]) - + ABS(a[i+23:i+16] - tmp[i+31:i+24]) + ABS(a[i+31:i+24] - tmp[i+39:i+32]) - - tmp_dst[i+47:i+32] := ABS(a[i+39:i+32] - tmp[i+23:i+16]) + ABS(a[i+47:i+40] - tmp[i+31:i+24]) - + ABS(a[i+55:i+48] - tmp[i+39:i+32]) + ABS(a[i+63:i+56] - tmp[i+47:i+40]) - - tmp_dst[i+63:i+48] := ABS(a[i+39:i+32] - tmp[i+31:i+24]) + ABS(a[i+47:i+40] - tmp[i+39:i+32]) - + ABS(a[i+55:i+48] - tmp[i+47:i+40]) + ABS(a[i+63:i+56] - tmp[i+55:i+48]) -ENDFOR - -FOR j := 0 to 7 - i := j*16 - IF k[j] - dst[i+15:i] := tmp_dst[i+15:i] - ELSE - dst[i+15:i] := 0 - FI -ENDFOR +IF k[0] + dst[63:0] := a[63:0] - b[63:0] +ELSE + dst[63:0] := src[63:0] +FI +dst[127:64] := a[127:64] dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point - AVX512VL AVX512F Arithmetic - - - - - Divide packed double-precision (64-bit) floating-point elements in "a" by packed elements in "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + + + + + Subtract the lower double-precision (64-bit) floating-point element in "b" from the lower double-precision (64-bit) floating-point element in "a", store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". + [round_note] -FOR j := 0 to 3 - i := 64*j - IF k[j] - dst[i+63:i] := a[i+63:i] / b[i+63:i] - ELSE - dst[i+63:i] := src[i+63:i] - FI -ENDFOR -dst[MAX:256] := 0 +IF k[0] + dst[63:0] := a[63:0] - b[63:0] +ELSE + dst[63:0] := 0 +FI +dst[127:64] := a[127:64] +dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point - AVX512VL AVX512F Arithmetic - - - - Divide packed double-precision (64-bit) floating-point elements in "a" by packed elements in "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + + + + Subtract the lower double-precision (64-bit) floating-point element in "b" from the lower double-precision (64-bit) floating-point element in "a", store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". -FOR j := 0 to 3 - i := 64*j - IF k[j] - dst[i+63:i] := a[i+63:i] / b[i+63:i] - ELSE - dst[i+63:i] := 0 - FI -ENDFOR -dst[MAX:256] := 0 +IF k[0] + dst[63:0] := a[63:0] - b[63:0] +ELSE + dst[63:0] := 0 +FI +dst[127:64] := a[127:64] +dst[MAX:128] := 0 - +
immintrin.h
 - + Floating Point - AVX512VL AVX512F Arithmetic - - - - - Divide packed double-precision (64-bit) floating-point elements in "a" by packed elements in "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + + + + Subtract the lower double-precision (64-bit) floating-point element in "b" from the lower double-precision (64-bit) floating-point element in "a", store the result in the lower element of "dst", and copy the upper element from "a" to the upper element of "dst". + [round_note] -FOR j := 0 to 1 - i := 64*j +dst[63:0] := a[63:0] - b[63:0] +dst[127:64] := a[127:64] +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512F + Arithmetic + + + + + + + Subtract the lower single-precision (32-bit) floating-point element in "b" from the lower single-precision (32-bit) floating-point element in "a", store the result in the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". + [round_note] + +IF k[0] + dst[31:0] := a[31:0] - b[31:0] +ELSE + dst[31:0] := src[31:0] +FI +dst[127:32] := a[127:32] +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512F + Arithmetic + + + + + + Subtract the lower single-precision (32-bit) floating-point element in "b" from the lower single-precision (32-bit) floating-point element in "a", store the result in the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". + +IF k[0] + dst[31:0] := a[31:0] - b[31:0] +ELSE + dst[31:0] := src[31:0] +FI +dst[127:32] := a[127:32] +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512F + Arithmetic + + + + + + Subtract the lower single-precision (32-bit) floating-point element in "b" from the lower single-precision (32-bit) floating-point element in "a", store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". + [round_note] + +IF k[0] + dst[31:0] := a[31:0] - b[31:0] +ELSE + dst[31:0] := 0 +FI +dst[127:32] := a[127:32] +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512F + Arithmetic + + + + + Subtract the lower single-precision (32-bit) floating-point element in "b" from the lower single-precision (32-bit) floating-point element in "a", store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". + +IF k[0] + dst[31:0] := a[31:0] - b[31:0] +ELSE + dst[31:0] := 0 +FI +dst[127:32] := a[127:32] +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512F + Arithmetic + + + + + Subtract the lower single-precision (32-bit) floating-point element in "b" from the lower single-precision (32-bit) floating-point element in "a", store the result in the lower element of "dst", and copy the upper 3 packed elements from "a" to the upper elements of "dst". + [round_note] + +dst[31:0] := a[31:0] - b[31:0] +dst[127:32] := a[127:32] +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512F + Swizzle + + + + + + Unpack and interleave double-precision (64-bit) floating-point elements from the high half of each 128-bit lane in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +DEFINE INTERLEAVE_HIGH_QWORDS(src1[127:0], src2[127:0]) { + dst[63:0] := src1[127:64] + dst[127:64] := src2[127:64] + RETURN dst[127:0] +} +tmp_dst[127:0] := INTERLEAVE_HIGH_QWORDS(a[127:0], b[127:0]) +tmp_dst[255:128] := INTERLEAVE_HIGH_QWORDS(a[255:128], b[255:128]) +tmp_dst[383:256] := INTERLEAVE_HIGH_QWORDS(a[383:256], b[383:256]) +tmp_dst[511:384] := INTERLEAVE_HIGH_QWORDS(a[511:384], b[511:384]) +FOR j := 0 to 7 + i := j*64 IF k[j] - dst[i+63:i] := a[i+63:i] / b[i+63:i] + dst[i+63:i] := tmp_dst[i+63:i] ELSE dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - AVX512VL AVX512F - Arithmetic - - - - Divide packed double-precision (64-bit) floating-point elements in "a" by packed elements in "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Swizzle + + + + + Unpack and interleave double-precision (64-bit) floating-point elements from the high half of each 128-bit lane in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 1 - i := 64*j +DEFINE INTERLEAVE_HIGH_QWORDS(src1[127:0], src2[127:0]) { + dst[63:0] := src1[127:64] + dst[127:64] := src2[127:64] + RETURN dst[127:0] +} +tmp_dst[127:0] := INTERLEAVE_HIGH_QWORDS(a[127:0], b[127:0]) +tmp_dst[255:128] := INTERLEAVE_HIGH_QWORDS(a[255:128], b[255:128]) +tmp_dst[383:256] := INTERLEAVE_HIGH_QWORDS(a[383:256], b[383:256]) +tmp_dst[511:384] := INTERLEAVE_HIGH_QWORDS(a[511:384], b[511:384]) +FOR j := 0 to 7 + i := j*64 IF k[j] - dst[i+63:i] := a[i+63:i] / b[i+63:i] + dst[i+63:i] := tmp_dst[i+63:i] ELSE dst[i+63:i] := 0 FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - AVX512VL AVX512F - Arithmetic - - - - - Divide packed single-precision (32-bit) floating-point elements in "a" by packed elements in "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Swizzle + + + + Unpack and interleave double-precision (64-bit) floating-point elements from the high half of each 128-bit lane in "a" and "b", and store the results in "dst". -FOR j := 0 to 7 - i := 32*j +DEFINE INTERLEAVE_HIGH_QWORDS(src1[127:0], src2[127:0]) { + dst[63:0] := src1[127:64] + dst[127:64] := src2[127:64] + RETURN dst[127:0] +} +dst[127:0] := INTERLEAVE_HIGH_QWORDS(a[127:0], b[127:0]) +dst[255:128] := INTERLEAVE_HIGH_QWORDS(a[255:128], b[255:128]) +dst[383:256] := INTERLEAVE_HIGH_QWORDS(a[383:256], b[383:256]) +dst[511:384] := INTERLEAVE_HIGH_QWORDS(a[511:384], b[511:384]) +dst[MAX:512] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512F + Swizzle + + + + + + Unpack and interleave single-precision (32-bit) floating-point elements from the high half of each 128-bit lane in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +DEFINE INTERLEAVE_HIGH_DWORDS(src1[127:0], src2[127:0]) { + dst[31:0] := src1[95:64] + dst[63:32] := src2[95:64] + dst[95:64] := src1[127:96] + dst[127:96] := src2[127:96] + RETURN dst[127:0] +} +tmp_dst[127:0] := INTERLEAVE_HIGH_DWORDS(a[127:0], b[127:0]) +tmp_dst[255:128] := INTERLEAVE_HIGH_DWORDS(a[255:128], b[255:128]) +tmp_dst[383:256] := INTERLEAVE_HIGH_DWORDS(a[383:256], b[383:256]) +tmp_dst[511:384] := INTERLEAVE_HIGH_DWORDS(a[511:384], b[511:384]) +FOR j := 0 to 15 + i := j*32 IF k[j] - dst[i+31:i] := a[i+31:i] / b[i+31:i] + dst[i+31:i] := tmp_dst[i+31:i] ELSE dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - AVX512VL AVX512F - Arithmetic - - - - Divide packed single-precision (32-bit) floating-point elements in "a" by packed elements in "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Swizzle + + + + + Unpack and interleave single-precision (32-bit) floating-point elements from the high half of each 128-bit lane in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := 32*j +DEFINE INTERLEAVE_HIGH_DWORDS(src1[127:0], src2[127:0]) { + dst[31:0] := src1[95:64] + dst[63:32] := src2[95:64] + dst[95:64] := src1[127:96] + dst[127:96] := src2[127:96] + RETURN dst[127:0] +} +tmp_dst[127:0] := INTERLEAVE_HIGH_DWORDS(a[127:0], b[127:0]) +tmp_dst[255:128] := INTERLEAVE_HIGH_DWORDS(a[255:128], b[255:128]) +tmp_dst[383:256] := INTERLEAVE_HIGH_DWORDS(a[383:256], b[383:256]) +tmp_dst[511:384] := INTERLEAVE_HIGH_DWORDS(a[511:384], b[511:384]) +FOR j := 0 to 15 + i := j*32 IF k[j] - dst[i+31:i] := a[i+31:i] / b[i+31:i] + dst[i+31:i] := tmp_dst[i+31:i] ELSE dst[i+31:i] := 0 FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - AVX512VL AVX512F - Arithmetic - - - - - Divide packed single-precision (32-bit) floating-point elements in "a" by packed elements in "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Swizzle + + + + Unpack and interleave single-precision (32-bit) floating-point elements from the high half of each 128-bit lane in "a" and "b", and store the results in "dst". -FOR j := 0 to 3 - i := 32*j +DEFINE INTERLEAVE_HIGH_DWORDS(src1[127:0], src2[127:0]) { + dst[31:0] := src1[95:64] + dst[63:32] := src2[95:64] + dst[95:64] := src1[127:96] + dst[127:96] := src2[127:96] + RETURN dst[127:0] +} +dst[127:0] := INTERLEAVE_HIGH_DWORDS(a[127:0], b[127:0]) +dst[255:128] := INTERLEAVE_HIGH_DWORDS(a[255:128], b[255:128]) +dst[383:256] := INTERLEAVE_HIGH_DWORDS(a[383:256], b[383:256]) +dst[511:384] := INTERLEAVE_HIGH_DWORDS(a[511:384], b[511:384]) +dst[MAX:512] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512F + Swizzle + + + + + + Unpack and interleave double-precision (64-bit) floating-point elements from the low half of each 128-bit lane in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +DEFINE INTERLEAVE_QWORDS(src1[127:0], src2[127:0]) { + dst[63:0] := src1[63:0] + dst[127:64] := src2[63:0] + RETURN dst[127:0] +} +tmp_dst[127:0] := INTERLEAVE_QWORDS(a[127:0], b[127:0]) +tmp_dst[255:128] := INTERLEAVE_QWORDS(a[255:128], b[255:128]) +tmp_dst[383:256] := INTERLEAVE_QWORDS(a[383:256], b[383:256]) +tmp_dst[511:384] := INTERLEAVE_QWORDS(a[511:384], b[511:384]) +FOR j := 0 to 7 + i := j*64 IF k[j] - dst[i+31:i] := a[i+31:i] / b[i+31:i] + dst[i+63:i] := tmp_dst[i+63:i] ELSE - dst[i+31:i] := src[i+31:i] + dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - AVX512VL AVX512F - Arithmetic - - - - Divide packed single-precision (32-bit) floating-point elements in "a" by packed elements in "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Swizzle + + + + + Unpack and interleave double-precision (64-bit) floating-point elements from the low half of each 128-bit lane in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := 32*j +DEFINE INTERLEAVE_QWORDS(src1[127:0], src2[127:0]) { + dst[63:0] := src1[63:0] + dst[127:64] := src2[63:0] + RETURN dst[127:0] +} +tmp_dst[127:0] := INTERLEAVE_QWORDS(a[127:0], b[127:0]) +tmp_dst[255:128] := INTERLEAVE_QWORDS(a[255:128], b[255:128]) +tmp_dst[383:256] := INTERLEAVE_QWORDS(a[383:256], b[383:256]) +tmp_dst[511:384] := INTERLEAVE_QWORDS(a[511:384], b[511:384]) +FOR j := 0 to 7 + i := j*64 IF k[j] - dst[i+31:i] := a[i+31:i] / b[i+31:i] + dst[i+63:i] := tmp_dst[i+63:i] ELSE - dst[i+31:i] := 0 + dst[i+63:i] := 0 FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - AVX512VL AVX512F - Miscellaneous - - - - Load contiguous active double-precision (64-bit) floating-point elements from "a" (those with their respective bit set in mask "k"), and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Swizzle + + + + Unpack and interleave double-precision (64-bit) floating-point elements from the low half of each 128-bit lane in "a" and "b", and store the results in "dst". -m := 0 -FOR j := 0 to 3 - i := j*64 +DEFINE INTERLEAVE_QWORDS(src1[127:0], src2[127:0]) { + dst[63:0] := src1[63:0] + dst[127:64] := src2[63:0] + RETURN dst[127:0] +} +dst[127:0] := INTERLEAVE_QWORDS(a[127:0], b[127:0]) +dst[255:128] := INTERLEAVE_QWORDS(a[255:128], b[255:128]) +dst[383:256] := INTERLEAVE_QWORDS(a[383:256], b[383:256]) +dst[511:384] := INTERLEAVE_QWORDS(a[511:384], b[511:384]) +dst[MAX:512] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512F + Swizzle + + + + + + Unpack and interleave single-precision (32-bit) floating-point elements from the low half of each 128-bit lane in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +DEFINE INTERLEAVE_DWORDS(src1[127:0], src2[127:0]) { + dst[31:0] := src1[31:0] + dst[63:32] := src2[31:0] + dst[95:64] := src1[63:32] + dst[127:96] := src2[63:32] + RETURN dst[127:0] +} +tmp_dst[127:0] := INTERLEAVE_DWORDS(a[127:0], b[127:0]) +tmp_dst[255:128] := INTERLEAVE_DWORDS(a[255:128], b[255:128]) +tmp_dst[383:256] := INTERLEAVE_DWORDS(a[383:256], b[383:256]) +tmp_dst[511:384] := INTERLEAVE_DWORDS(a[511:384], b[511:384]) +FOR j := 0 to 15 + i := j*32 IF k[j] - dst[i+63:i] := a[m+63:m] - m := m + 64 + dst[i+31:i] := tmp_dst[i+31:i] ELSE - dst[i+63:i] := src[i+63:i] + dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - AVX512VL AVX512F - Load - - - - Load contiguous active double-precision (64-bit) floating-point elements from unaligned memory at "mem_addr" (those with their respective bit set in mask "k"), and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Swizzle + + + + + Unpack and interleave single-precision (32-bit) floating-point elements from the low half of each 128-bit lane in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -m := 0 -FOR j := 0 to 3 - i := j*64 +DEFINE INTERLEAVE_DWORDS(src1[127:0], src2[127:0]) { + dst[31:0] := src1[31:0] + dst[63:32] := src2[31:0] + dst[95:64] := src1[63:32] + dst[127:96] := src2[63:32] + RETURN dst[127:0] +} +tmp_dst[127:0] := INTERLEAVE_DWORDS(a[127:0], b[127:0]) +tmp_dst[255:128] := INTERLEAVE_DWORDS(a[255:128], b[255:128]) +tmp_dst[383:256] := INTERLEAVE_DWORDS(a[383:256], b[383:256]) +tmp_dst[511:384] := INTERLEAVE_DWORDS(a[511:384], b[511:384]) +FOR j := 0 to 15 + i := j*32 IF k[j] - dst[i+63:i] := MEM[mem_addr+m+63:mem_addr+m] - m := m + 64 + dst[i+31:i] := tmp_dst[i+31:i] ELSE - dst[i+63:i] := src[i+63:i] + dst[i+31:i] := 0 FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512F + Swizzle + + + + Unpack and interleave single-precision (32-bit) floating-point elements from the low half of each 128-bit lane in "a" and "b", and store the results in "dst". + +DEFINE INTERLEAVE_DWORDS(src1[127:0], src2[127:0]) { + dst[31:0] := src1[31:0] + dst[63:32] := src2[31:0] + dst[95:64] := src1[63:32] + dst[127:96] := src2[63:32] + RETURN dst[127:0] +} +dst[127:0] := INTERLEAVE_DWORDS(a[127:0], b[127:0]) +dst[255:128] := INTERLEAVE_DWORDS(a[255:128], b[255:128]) +dst[383:256] := INTERLEAVE_DWORDS(a[383:256], b[383:256]) +dst[511:384] := INTERLEAVE_DWORDS(a[511:384], b[511:384]) +dst[MAX:512] := 0 - + +
immintrin.h
+ + + Floating Point + AVX512F + Cast + + + Cast vector of type __m128d to type __m512d; the upper 384 bits of the result are undefined. + This intrinsic is only used for compilation and does not generate any instructions, thus it has zero latency. +
immintrin.h
+ + + Floating Point + AVX512F + Cast + + + Cast vector of type __m256d to type __m512d; the upper 256 bits of the result are undefined. + This intrinsic is only used for compilation and does not generate any instructions, thus it has zero latency. +
immintrin.h
+ + + Floating Point + AVX512F + Cast + + + Cast vector of type __m512d to type __m128d. + This intrinsic is only used for compilation and does not generate any instructions, thus it has zero latency. +
immintrin.h
+ + + Floating Point + AVX512F + Cast + + + Cast vector of type __m512 to type __m128. + This intrinsic is only used for compilation and does not generate any instructions, thus it has zero latency. +
immintrin.h
+ + + Floating Point + AVX512F + Cast + + + Cast vector of type __m512d to type __m256d. + This intrinsic is only used for compilation and does not generate any instructions, thus it has zero latency.
immintrin.h
 - + Floating Point - AVX512VL AVX512F - Miscellaneous - - - Load contiguous active double-precision (64-bit) floating-point elements from "a" (those with their respective bit set in mask "k"), and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - -m := 0 -FOR j := 0 to 3 - i := j*64 - IF k[j] - dst[i+63:i] := a[m+63:m] - m := m + 64 - ELSE - dst[i+63:i] := 0 - FI -ENDFOR -dst[MAX:256] := 0 - - + Cast + + + Cast vector of type __m128 to type __m512; the upper 384 bits of the result are undefined. + This intrinsic is only used for compilation and does not generate any instructions, thus it has zero latency.
immintrin.h
 - + Floating Point - AVX512VL AVX512F - Load - - - Load contiguous active double-precision (64-bit) floating-point elements from unaligned memory at "mem_addr" (those with their respective bit set in mask "k"), and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - -m := 0 -FOR j := 0 to 3 - i := j*64 - IF k[j] - dst[i+63:i] := MEM[mem_addr+m+63:mem_addr+m] - m := m + 64 - ELSE - dst[i+63:i] := 0 - FI -ENDFOR -dst[MAX:256] := 0 - - + Cast + + + Cast vector of type __m256 to type __m512; the upper 256 bits of the result are undefined. + This intrinsic is only used for compilation and does not generate any instructions, thus it has zero latency.
immintrin.h
 - + Floating Point - AVX512VL AVX512F - Miscellaneous - - - - Load contiguous active double-precision (64-bit) floating-point elements from "a" (those with their respective bit set in mask "k"), and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -m := 0 -FOR j := 0 to 1 - i := j*64 - IF k[j] - dst[i+63:i] := a[m+63:m] - m := m + 64 - ELSE - dst[i+63:i] := src[i+63:i] - FI -ENDFOR -dst[MAX:128] := 0 - - + Cast + + + Cast vector of type __m512 to type __m256. + This intrinsic is only used for compilation and does not generate any instructions, thus it has zero latency.
immintrin.h
 - - Floating Point - AVX512VL + + Integer AVX512F - Load - - - - Load contiguous active double-precision (64-bit) floating-point elements from unaligned memory at "mem_addr" (those with their respective bit set in mask "k"), and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -m := 0 -FOR j := 0 to 1 - i := j*64 - IF k[j] - dst[i+63:i] := MEM[mem_addr+m+63:mem_addr+m] - m := m + 64 - ELSE - dst[i+63:i] := src[i+63:i] - FI -ENDFOR -dst[MAX:128] := 0 - - + Cast + + + Cast vector of type __m128i to type __m512i; the upper 384 bits of the result are undefined. + This intrinsic is only used for compilation and does not generate any instructions, thus it has zero latency.
immintrin.h
 - - Floating Point - AVX512VL + + Integer AVX512F - Miscellaneous - - - Load contiguous active double-precision (64-bit) floating-point elements from "a" (those with their respective bit set in mask "k"), and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - -m := 0 -FOR j := 0 to 1 - i := j*64 - IF k[j] - dst[i+63:i] := a[m+63:m] - m := m + 64 - ELSE - dst[i+63:i] := 0 - FI -ENDFOR -dst[MAX:128] := 0 - - + Cast + + + Cast vector of type __m256i to type __m512i; the upper 256 bits of the result are undefined. + This intrinsic is only used for compilation and does not generate any instructions, thus it has zero latency.
immintrin.h
 - - Floating Point - AVX512VL + + Integer AVX512F - Load - - - Load contiguous active double-precision (64-bit) floating-point elements from unaligned memory at "mem_addr" (those with their respective bit set in mask "k"), and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - -m := 0 -FOR j := 0 to 1 - i := j*64 - IF k[j] - dst[i+63:i] := MEM[mem_addr+m+63:mem_addr+m] - m := m + 64 - ELSE - dst[i+63:i] := 0 - FI -ENDFOR -dst[MAX:128] := 0 - - + Cast + + + Cast vector of type __m512i to type __m128i. + This intrinsic is only used for compilation and does not generate any instructions, thus it has zero latency. +
immintrin.h
+ + + Integer + AVX512F + Cast + + + Cast vector of type __m512i to type __m256i. + This intrinsic is only used for compilation and does not generate any instructions, thus it has zero latency.
immintrin.h
 - + Floating Point - AVX512VL AVX512F - Miscellaneous - - - - Load contiguous active single-precision (32-bit) floating-point elements from "a" (those with their respective bit set in mask "k"), and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -m := 0 -FOR j := 0 to 7 - i := j*32 - IF k[j] - dst[i+31:i] := a[m+31:m] - m := m + 32 - ELSE - dst[i+31:i] := src[i+31:i] - FI -ENDFOR -dst[MAX:256] := 0 - - + Cast + + + Cast vector of type __m128d to type __m512d; the upper 384 bits of the result are zeroed. This intrinsic is only used for compilation and does not generate any instructions, thus it has zero latency.
immintrin.h
 - + Floating Point - AVX512VL AVX512F - Load - - - - Load contiguous active single-precision (32-bit) floating-point elements from unaligned memory at "mem_addr" (those with their respective bit set in mask "k"), and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -m := 0 -FOR j := 0 to 7 - i := j*32 - IF k[j] - dst[i+31:i] := MEM[mem_addr+m+31:mem_addr+m] - m := m + 32 - ELSE - dst[i+31:i] := src[i+31:i] - FI -ENDFOR -dst[MAX:256] := 0 - - + Cast + + + Cast vector of type __m128 to type __m512; the upper 384 bits of the result are zeroed. This intrinsic is only used for compilation and does not generate any instructions, thus it has zero latency. +
immintrin.h
+ + + Integer + AVX512F + Cast + + + Cast vector of type __m128i to type __m512i; the upper 384 bits of the result are zeroed. This intrinsic is only used for compilation and does not generate any instructions, thus it has zero latency.
immintrin.h
 - + Floating Point - AVX512VL AVX512F - Miscellaneous - - - Load contiguous active single-precision (32-bit) floating-point elements from "a" (those with their respective bit set in mask "k"), and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - -m := 0 -FOR j := 0 to 7 - i := j*32 - IF k[j] - dst[i+31:i] := a[m+31:m] - m := m + 32 - ELSE - dst[i+31:i] := 0 - FI -ENDFOR -dst[MAX:256] := 0 - - + Cast + + + Cast vector of type __m256d to type __m512d; the upper 256 bits of the result are zeroed. This intrinsic is only used for compilation and does not generate any instructions, thus it has zero latency.
immintrin.h
 - + Floating Point - AVX512VL AVX512F - Load - - - Load contiguous active single-precision (32-bit) floating-point elements from unaligned memory at "mem_addr" (those with their respective bit set in mask "k"), and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - -m := 0 -FOR j := 0 to 7 - i := j*32 - IF k[j] - dst[i+31:i] := MEM[mem_addr+m+31:mem_addr+m] - m := m + 32 - ELSE - dst[i+31:i] := 0 - FI -ENDFOR -dst[MAX:256] := 0 - - + Cast + + + Cast vector of type __m256 to type __m512; the upper 256 bits of the result are zeroed. This intrinsic is only used for compilation and does not generate any instructions, thus it has zero latency. +
immintrin.h
+ + + Integer + AVX512F + Cast + + + Cast vector of type __m256i to type __m512i; the upper 256 bits of the result are zeroed. This intrinsic is only used for compilation and does not generate any instructions, thus it has zero latency.
immintrin.h
 - + Floating Point - AVX512VL AVX512F - Miscellaneous - - - - Load contiguous active single-precision (32-bit) floating-point elements from "a" (those with their respective bit set in mask "k"), and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Set + + + Broadcast double-precision (64-bit) floating-point value "a" to all elements of "dst". -m := 0 -FOR j := 0 to 3 - i := j*32 - IF k[j] - dst[i+31:i] := a[m+31:m] - m := m + 32 - ELSE - dst[i+31:i] := src[i+31:i] - FI +FOR j := 0 to 7 + i := j*64 + dst[i+63:i] := a[63:0] ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 -
immintrin.h
 - + Floating Point - AVX512VL AVX512F - Load - - - - Load contiguous active single-precision (32-bit) floating-point elements from unaligned memory at "mem_addr" (those with their respective bit set in mask "k"), and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Set + + + Broadcast single-precision (32-bit) floating-point value "a" to all elements of "dst". -m := 0 -FOR j := 0 to 3 +FOR j := 0 to 15 i := j*32 - IF k[j] - dst[i+31:i] := MEM[mem_addr+m+31:mem_addr+m] - m := m + 32 - ELSE - dst[i+31:i] := src[i+31:i] - FI + dst[i+31:i] := a[31:0] ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 -
immintrin.h
 - - Floating Point - AVX512VL + + Integer AVX512F - Miscellaneous - - - Load contiguous active single-precision (32-bit) floating-point elements from "a" (those with their respective bit set in mask "k"), and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Set + + + + + + Set packed 32-bit integers in "dst" with the repeated 4 element sequence. -m := 0 -FOR j := 0 to 3 - i := j*32 - IF k[j] - dst[i+31:i] := a[m+31:m] - m := m + 32 - ELSE - dst[i+31:i] := 0 - FI -ENDFOR -dst[MAX:128] := 0 +dst[31:0] := a +dst[63:32] := b +dst[95:64] := c +dst[127:96] := d +dst[159:128] := a +dst[191:160] := b +dst[223:192] := c +dst[255:224] := d +dst[287:256] := a +dst[319:288] := b +dst[351:320] := c +dst[383:352] := d +dst[415:384] := a +dst[447:416] := b +dst[479:448] := c +dst[511:480] := d +dst[MAX:512] := 0 -
immintrin.h
 - - Floating Point - AVX512VL + + Integer AVX512F - Load - - - Load contiguous active single-precision (32-bit) floating-point elements from unaligned memory at "mem_addr" (those with their respective bit set in mask "k"), and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Set + + + + + + Set packed 64-bit integers in "dst" with the repeated 4 element sequence. -m := 0 -FOR j := 0 to 3 - i := j*32 - IF k[j] - dst[i+31:i] := MEM[mem_addr+m+31:mem_addr+m] - m := m + 32 - ELSE - dst[i+31:i] := 0 - FI -ENDFOR -dst[MAX:128] := 0 +dst[63:0] := a +dst[127:64] := b +dst[191:128] := c +dst[255:192] := d +dst[319:256] := a +dst[383:320] := b +dst[447:384] := c +dst[511:448] := d +dst[MAX:512] := 0 -
immintrin.h
 - + Floating Point - AVX512VL AVX512F - Miscellaneous - - - Extract 128 bits (composed of 4 packed single-precision (32-bit) floating-point elements) from "a", selected with "imm8", and store the result in "dst". + Set + + + + + + Set packed double-precision (64-bit) floating-point elements in "dst" with the repeated 4 element sequence. -CASE imm8[7:0] of -0: dst[127:0] := a[127:0] -1: dst[127:0] := a[255:128] -ESAC -dst[MAX:128] := 0 +dst[63:0] := a +dst[127:64] := b +dst[191:128] := c +dst[255:192] := d +dst[319:256] := a +dst[383:320] := b +dst[447:384] := c +dst[511:448] := d +dst[MAX:512] := 0 -
immintrin.h
 - + Floating Point - AVX512VL AVX512F - Miscellaneous - - - - - Extract 128 bits (composed of 4 packed single-precision (32-bit) floating-point elements) from "a", selected with "imm8", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Set + + + + + + Set packed single-precision (32-bit) floating-point elements in "dst" with the repeated 4 element sequence. -CASE imm8[7:0] of -0: tmp[127:0] := a[127:0] -1: tmp[127:0] := a[255:128] -ESAC -FOR j := 0 to 3 - i := j*32 - IF k[j] - dst[i+31:i] := tmp[i+31:i] - ELSE - dst[i+31:i] := src[i+31:i] - FI -ENDFOR -dst[MAX:128] := 0 +dst[31:0] := a +dst[63:32] := b +dst[95:64] := c +dst[127:96] := d +dst[159:128] := a +dst[191:160] := b +dst[223:192] := c +dst[255:224] := d +dst[287:256] := a +dst[319:288] := b +dst[351:320] := c +dst[383:352] := d +dst[415:384] := a +dst[447:416] := b +dst[479:448] := c +dst[511:480] := d +dst[MAX:512] := 0 -
immintrin.h
 - - Floating Point - AVX512VL + + Integer AVX512F - Miscellaneous - - - - Extract 128 bits (composed of 4 packed single-precision (32-bit) floating-point elements) from "a", selected with "imm8", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Set + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + Set packed 8-bit integers in "dst" with the supplied values. -CASE imm8[7:0] of -0: tmp[127:0] := a[127:0] -1: tmp[127:0] := a[255:128] -ESAC -FOR j := 0 to 3 - i := j*32 - IF k[j] - dst[i+31:i] := tmp[i+31:i] - ELSE - dst[i+31:i] := 0 - FI -ENDFOR -dst[MAX:128] := 0 +dst[7:0] := e0 +dst[15:8] := e1 +dst[23:16] := e2 +dst[31:24] := e3 +dst[39:32] := e4 +dst[47:40] := e5 +dst[55:48] := e6 +dst[63:56] := e7 +dst[71:64] := e8 +dst[79:72] := e9 +dst[87:80] := e10 +dst[95:88] := e11 +dst[103:96] := e12 +dst[111:104] := e13 +dst[119:112] := e14 +dst[127:120] := e15 +dst[135:128] := e16 +dst[143:136] := e17 +dst[151:144] := e18 +dst[159:152] := e19 +dst[167:160] := e20 +dst[175:168] := e21 +dst[183:176] := e22 +dst[191:184] := e23 +dst[199:192] := e24 +dst[207:200] := e25 +dst[215:208] := e26 +dst[223:216] := e27 +dst[231:224] := e28 +dst[239:232] := e29 +dst[247:240] := e30 +dst[255:248] := e31 +dst[263:256] := e32 +dst[271:264] := e33 +dst[279:272] := e34 +dst[287:280] := e35 +dst[295:288] := e36 +dst[303:296] := e37 +dst[311:304] := e38 +dst[319:312] := e39 +dst[327:320] := e40 +dst[335:328] := e41 +dst[343:336] := e42 +dst[351:344] := e43 +dst[359:352] := e44 +dst[367:360] := e45 +dst[375:368] := e46 +dst[383:376] := e47 +dst[391:384] := e48 +dst[399:392] := e49 +dst[407:400] := e50 +dst[415:408] := e51 +dst[423:416] := e52 +dst[431:424] := e53 +dst[439:432] := e54 +dst[447:440] := e55 +dst[455:448] := e56 +dst[463:456] := e57 +dst[471:464] := e58 +dst[479:472] := e59 +dst[487:480] := e60 +dst[495:488] := e61 +dst[503:496] := e62 +dst[511:504] := e63 +dst[MAX:512] := 0 -
immintrin.h
 - - Floating Point - AVX512DQ - Miscellaneous - - - Extract 256 bits (composed of 8 packed single-precision (32-bit) floating-point elements) from "a", selected with "imm8", and store the result in "dst". + + Integer + AVX512F + Set + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + Set packed 16-bit integers in "dst" with the supplied values. -CASE imm8[7:0] of -0: dst[255:0] := a[255:0] -1: dst[255:0] := a[511:256] -ESAC -dst[MAX:256] := 0 +dst[15:0] := e0 +dst[31:16] := e1 +dst[47:32] := e2 +dst[63:48] := e3 +dst[79:64] := e4 +dst[95:80] := e5 +dst[111:96] := e6 +dst[127:112] := e7 +dst[143:128] := e8 +dst[159:144] := e9 +dst[175:160] := e10 +dst[191:176] := e11 +dst[207:192] := e12 +dst[223:208] := e13 +dst[239:224] := e14 +dst[255:240] := e15 +dst[271:256] := e16 +dst[287:272] := e17 +dst[303:288] := e18 +dst[319:304] := e19 +dst[335:320] := e20 +dst[351:336] := e21 +dst[367:352] := e22 +dst[383:368] := e23 +dst[399:384] := e24 +dst[415:400] := e25 +dst[431:416] := e26 +dst[447:432] := e27 +dst[463:448] := e28 +dst[479:464] := e29 +dst[495:480] := e30 +dst[511:496] := e31 +dst[MAX:512] := 0 -
immintrin.h
 - - Floating Point - AVX512DQ - Miscellaneous - - - - - Extract 256 bits (composed of 8 packed single-precision (32-bit) floating-point elements) from "a", selected with "imm8", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Integer + AVX512F + Set + + + + + + + + + + + + + + + + + + Set packed 32-bit integers in "dst" with the supplied values. -CASE imm8[7:0] of -0: tmp[255:0] := a[255:0] -1: tmp[255:0] := a[511:256] -ESAC - -FOR j := 0 to 7 - i := j*32 - IF k[j] - dst[i+31:i] := tmp[i+31:i] - ELSE - dst[i+31:i] := src[i+31:i] - FI -ENDFOR -dst[MAX:256] := 0 +dst[31:0] := e0 +dst[63:32] := e1 +dst[95:64] := e2 +dst[127:96] := e3 +dst[159:128] := e4 +dst[191:160] := e5 +dst[223:192] := e6 +dst[255:224] := e7 +dst[287:256] := e8 +dst[319:288] := e9 +dst[351:320] := e10 +dst[383:352] := e11 +dst[415:384] := e12 +dst[447:416] := e13 +dst[479:448] := e14 +dst[511:480] := e15 +dst[MAX:512] := 0 -
immintrin.h
 - - Floating Point - AVX512DQ - Miscellaneous - - - - Extract 256 bits (composed of 8 packed single-precision (32-bit) floating-point elements) from "a", selected with "imm8", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Integer + AVX512F + Set + + + + + + + + + + Set packed 64-bit integers in "dst" with the supplied values. -CASE imm8[7:0] of -0: tmp[255:0] := a[255:0] -1: tmp[255:0] := a[511:256] -ESAC - -FOR j := 0 to 7 - i := j*32 - IF k[j] - dst[i+31:i] := tmp[i+31:i] - ELSE - dst[i+31:i] := 0 - FI -ENDFOR -dst[MAX:256] := 0 +dst[63:0] := e0 +dst[127:64] := e1 +dst[191:128] := e2 +dst[255:192] := e3 +dst[319:256] := e4 +dst[383:320] := e5 +dst[447:384] := e6 +dst[511:448] := e7 +dst[MAX:512] := 0 -
immintrin.h
 - + Floating Point - AVX512VL - AVX512DQ - Miscellaneous - - - Extract 128 bits (composed of 2 packed double-precision (64-bit) floating-point elements) from "a", selected with "imm8", and store the result in "dst". + AVX512F + Set + + + + + + + + + + Set packed double-precision (64-bit) floating-point elements in "dst" with the supplied values. -CASE imm8[7:0] of -0: dst[127:0] := a[127:0] -1: dst[127:0] := a[255:128] -ESAC -dst[MAX:128] := 0 +dst[63:0] := e0 +dst[127:64] := e1 +dst[191:128] := e2 +dst[255:192] := e3 +dst[319:256] := e4 +dst[383:320] := e5 +dst[447:384] := e6 +dst[511:448] := e7 +dst[MAX:512] := 0 -
immintrin.h
 - + Floating Point - AVX512VL - AVX512DQ - Miscellaneous - - - - - Extract 128 bits (composed of 2 packed double-precision (64-bit) floating-point elements) from "a", selected with "imm8", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + AVX512F + Set + + + + + + + + + + + + + + + + + + Set packed single-precision (32-bit) floating-point elements in "dst" with the supplied values. -CASE imm8[7:0] of -0: tmp[127:0] := a[127:0] -1: tmp[127:0] := a[255:128] -ESAC - -FOR j := 0 to 1 - i := j*64 - IF k[j] - dst[i+63:i] := tmp[i+63:i] - ELSE - dst[i+63:i] := src[i+63:i] - FI -ENDFOR -dst[MAX:128] := 0 +dst[31:0] := e0 +dst[63:32] := e1 +dst[95:64] := e2 +dst[127:96] := e3 +dst[159:128] := e4 +dst[191:160] := e5 +dst[223:192] := e6 +dst[255:224] := e7 +dst[287:256] := e8 +dst[319:288] := e9 +dst[351:320] := e10 +dst[383:352] := e11 +dst[415:384] := e12 +dst[447:416] := e13 +dst[479:448] := e14 +dst[511:480] := e15 +dst[MAX:512] := 0 -
immintrin.h
 - - Floating Point - AVX512VL - AVX512DQ - Miscellaneous - - - - Extract 128 bits (composed of 2 packed double-precision (64-bit) floating-point elements) from "a", selected with "imm8", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Integer + AVX512F + Set + + + + + + Set packed 32-bit integers in "dst" with the repeated 4 element sequence in reverse order. -CASE imm8[7:0] of -0: tmp[127:0] := a[127:0] -1: tmp[127:0] := a[255:128] -ESAC - -FOR j := 0 to 1 - i := j*64 - IF k[j] - dst[i+63:i] := tmp[i+63:i] - ELSE - dst[i+63:i] := 0 - FI -ENDFOR -dst[MAX:128] := 0 +dst[31:0] := d +dst[63:32] := c +dst[95:64] := b +dst[127:96] := a +dst[159:128] := d +dst[191:160] := c +dst[223:192] := b +dst[255:224] := a +dst[287:256] := d +dst[319:288] := c +dst[351:320] := b +dst[383:352] := a +dst[415:384] := d +dst[447:416] := c +dst[479:448] := b +dst[511:480] := a +dst[MAX:512] := 0 -
immintrin.h
 - - Floating Point - AVX512DQ - Miscellaneous - - - Extract 128 bits (composed of 2 packed double-precision (64-bit) floating-point elements) from "a", selected with "imm8", and store the result in "dst". + + Integer + AVX512F + Set + + + + + + Set packed 64-bit integers in "dst" with the repeated 4 element sequence in reverse order. -CASE imm8[7:0] of -0: dst[127:0] := a[127:0] -1: dst[127:0] := a[255:128] -2: dst[127:0] := a[383:256] -3: dst[127:0] := a[511:384] -ESAC -dst[MAX:128] := 0 +dst[63:0] := d +dst[127:64] := c +dst[191:128] := b +dst[255:192] := a +dst[319:256] := d +dst[383:320] := c +dst[447:384] := b +dst[511:448] := a +dst[MAX:512] := 0 -
immintrin.h
 - + Floating Point - AVX512DQ - Miscellaneous - - - - - Extract 128 bits (composed of 2 packed double-precision (64-bit) floating-point elements) from "a", selected with "imm8", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + AVX512F + Set + + + + + + Set packed double-precision (64-bit) floating-point elements in "dst" with the repeated 4 element sequence in reverse order. -CASE imm8[7:0] of -0: tmp[127:0] := a[127:0] -1: tmp[127:0] := a[255:128] -2: tmp[127:0] := a[383:256] -3: tmp[127:0] := a[511:384] -ESAC - -FOR j := 0 to 1 - i := j*64 - IF k[j] - dst[i+63:i] := tmp[i+63:i] - ELSE - dst[i+63:i] := src[i+63:i] - FI -ENDFOR -dst[MAX:128] := 0 +dst[63:0] := d +dst[127:64] := c +dst[191:128] := b +dst[255:192] := a +dst[319:256] := d +dst[383:320] := c +dst[447:384] := b +dst[511:448] := a +dst[MAX:512] := 0 -
immintrin.h
 - + Floating Point - AVX512DQ - Miscellaneous - - - - Extract 128 bits (composed of 2 packed double-precision (64-bit) floating-point elements) from "a", selected with "imm8", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + AVX512F + Set + + + + + + Set packed single-precision (32-bit) floating-point elements in "dst" with the repeated 4 element sequence in reverse order. -CASE imm8[7:0] of -0: tmp[127:0] := a[127:0] -1: tmp[127:0] := a[255:128] -2: tmp[127:0] := a[383:256] -3: tmp[127:0] := a[511:384] -ESAC - -FOR j := 0 to 1 - i := j*64 - IF k[j] - dst[i+63:i] := tmp[i+63:i] - ELSE - dst[i+63:i] := 0 - FI -ENDFOR -dst[MAX:128] := 0 +dst[31:0] := d +dst[63:32] := c +dst[95:64] := b +dst[127:96] := a +dst[159:128] := d +dst[191:160] := c +dst[223:192] := b +dst[255:224] := a +dst[287:256] := d +dst[319:288] := c +dst[351:320] := b +dst[383:352] := a +dst[415:384] := d +dst[447:416] := c +dst[479:448] := b +dst[511:480] := a +dst[MAX:512] := 0 -
immintrin.h
 - + Integer - AVX512VL AVX512F - Miscellaneous - - - Extract 128 bits (composed of 4 packed 32-bit integers) from "a", selected with "imm8", and store the result in "dst". + Set + + + + + + + + + + + + + + + + + + Set packed 32-bit integers in "dst" with the supplied values in reverse order. -CASE imm8[7:0] of -0: dst[127:0] := a[127:0] -1: dst[127:0] := a[255:128] -ESAC -dst[MAX:128] := 0 +dst[31:0] := e15 +dst[63:32] := e14 +dst[95:64] := e13 +dst[127:96] := e12 +dst[159:128] := e11 +dst[191:160] := e10 +dst[223:192] := e9 +dst[255:224] := e8 +dst[287:256] := e7 +dst[319:288] := e6 +dst[351:320] := e5 +dst[383:352] := e4 +dst[415:384] := e3 +dst[447:416] := e2 +dst[479:448] := e1 +dst[511:480] := e0 +dst[MAX:512] := 0 -
immintrin.h
 - + Integer - AVX512VL AVX512F - Miscellaneous - - - - - Extract 128 bits (composed of 4 packed 32-bit integers) from "a", selected with "imm8", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Set + + + + + + + + + + Set packed 64-bit integers in "dst" with the supplied values in reverse order. -CASE imm8[7:0] of -0: dst[127:0] := a[127:0] -1: dst[127:0] := a[255:128] -ESAC -FOR j := 0 to 3 - i := j*32 - IF k[j] - dst[i+31:i] := tmp[i+31:i] - ELSE - dst[i+31:i] := src[i+31:i] - FI -ENDFOR -dst[MAX:128] := 0 +dst[63:0] := e7 +dst[127:64] := e6 +dst[191:128] := e5 +dst[255:192] := e4 +dst[319:256] := e3 +dst[383:320] := e2 +dst[447:384] := e1 +dst[511:448] := e0 +dst[MAX:512] := 0 -
immintrin.h
 - - Integer - AVX512VL + + Floating Point AVX512F - Miscellaneous - - - - Extract 128 bits (composed of 4 packed 32-bit integers) from "a", selected with "imm8", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Set + + + + + + + + + + Set packed double-precision (64-bit) floating-point elements in "dst" with the supplied values in reverse order. -CASE imm8[7:0] of -0: dst[127:0] := a[127:0] -1: dst[127:0] := a[255:128] -ESAC -FOR j := 0 to 3 - i := j*32 - IF k[j] - dst[i+31:i] := tmp[i+31:i] - ELSE - dst[i+31:i] := 0 - FI -ENDFOR -dst[MAX:128] := 0 +dst[63:0] := e7 +dst[127:64] := e6 +dst[191:128] := e5 +dst[255:192] := e4 +dst[319:256] := e3 +dst[383:320] := e2 +dst[447:384] := e1 +dst[511:448] := e0 +dst[MAX:512] := 0 -
immintrin.h
 - - Integer - AVX512DQ - Miscellaneous - - - Extract 256 bits (composed of 8 packed 32-bit integers) from "a", selected with "imm8", and store the result in "dst". + + Floating Point + AVX512F + Set + + + + + + + + + + + + + + + + + + Set packed single-precision (32-bit) floating-point elements in "dst" with the supplied values in reverse order. -CASE imm8[7:0] of -0: dst[255:0] := a[255:0] -1: dst[255:0] := a[511:256] -ESAC -dst[MAX:256] := 0 +dst[31:0] := e15 +dst[63:32] := e14 +dst[95:64] := e13 +dst[127:96] := e12 +dst[159:128] := e11 +dst[191:160] := e10 +dst[223:192] := e9 +dst[255:224] := e8 +dst[287:256] := e7 +dst[319:288] := e6 +dst[351:320] := e5 +dst[383:352] := e4 +dst[415:384] := e3 +dst[447:416] := e2 +dst[479:448] := e1 +dst[511:480] := e0 +dst[MAX:512] := 0 -
immintrin.h
 - - Integer - AVX512DQ - Miscellaneous - - - - - Extract 256 bits (composed of 8 packed 32-bit integers) from "a", selected with "imm8", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + AVX512F + Set + + + Return vector of type __m512 with all elements set to zero. -CASE imm8[7:0] of -0: tmp[255:0] := a[255:0] -1: tmp[255:0] := a[511:256] -ESAC - -FOR j := 0 to 7 - i := j*32 - IF k[j] - dst[i+31:i] := tmp[i+31:i] - ELSE - dst[i+31:i] := src[i+31:i] - FI -ENDFOR -dst[MAX:256] := 0 +dst[MAX:0] := 0 - +
immintrin.h
 - + Integer - AVX512DQ - Miscellaneous - - - - Extract 256 bits (composed of 8 packed 32-bit integers) from "a", selected with "imm8", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + AVX512F + Set + + Return vector of type __m512i with all elements set to zero. -CASE imm8[7:0] of -0: tmp[255:0] := a[255:0] -1: tmp[255:0] := a[511:256] -ESAC - -FOR j := 0 to 7 - i := j*32 - IF k[j] - dst[i+31:i] := tmp[i+31:i] - ELSE - dst[i+31:i] := 0 - FI -ENDFOR -dst[MAX:256] := 0 +dst[MAX:0] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512DQ - Miscellaneous - - - Extract 128 bits (composed of 2 packed 64-bit integers) from "a", selected with "imm8", and store the result in "dst". + + Floating Point + AVX512F + Set + + Return vector of type __m512d with all elements set to zero. -CASE imm8[7:0] of -0: dst[127:0] := a[127:0] -1: dst[127:0] := a[255:128] -ESAC -dst[MAX:128] := 0 +dst[MAX:0] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512DQ - Miscellaneous - - - - - Extract 128 bits (composed of 2 packed 64-bit integers) from "a", selected with "imm8", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Floating Point + AVX512F + Set + + Return vector of type __m512 with all elements set to zero. -CASE imm8[7:0] of -0: tmp[127:0] := a[127:0] -1: tmp[127:0] := a[255:128] -ESAC - -FOR j := 0 to 1 - i := j*64 - IF k[j] - dst[i+63:i] := tmp[i+63:i] - ELSE - dst[i+63:i] := src[i+63:i] - FI -ENDFOR -dst[MAX:128] := 0 +dst[MAX:0] := 0 - +
immintrin.h
 - + Integer - AVX512VL - AVX512DQ - Miscellaneous - - - - Extract 128 bits (composed of 2 packed 64-bit integers) from "a", selected with "imm8", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + AVX512F + Set + + Return vector of type __m512i with all elements set to zero. -CASE imm8[7:0] of -0: tmp[127:0] := a[127:0] -1: tmp[127:0] := a[255:128] -ESAC - -FOR j := 0 to 1 - i := j*64 - IF k[j] - dst[i+63:i] := tmp[i+63:i] - ELSE - dst[i+63:i] := 0 - FI -ENDFOR -dst[MAX:128] := 0 +dst[MAX:0] := 0 - +
immintrin.h
 - - Integer - AVX512DQ - Miscellaneous - - - Extract 128 bits (composed of 2 packed 64-bit integers) from "a", selected with "imm8", and store the result in "dst". - -CASE imm8[7:0] of -0: dst[127:0] := a[127:0] -1: dst[127:0] := a[255:128] -2: dst[127:0] := a[383:256] -3: dst[127:0] := a[511:384] -ESAC -dst[MAX:128] := 0 - - + + AVX512F + General Support + + + Return vector of type __m512 with undefined elements.
immintrin.h
 - + Integer - AVX512DQ - Miscellaneous - - - - - Extract 128 bits (composed of 2 packed 64-bit integers) from "a", selected with "imm8", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -CASE imm8[7:0] of -0: tmp[127:0] := a[127:0] -1: tmp[127:0] := a[255:128] -2: tmp[127:0] := a[383:256] -3: tmp[127:0] := a[511:384] -ESAC - -FOR j := 0 to 1 - i := j*64 - IF k[j] - dst[i+63:i] := tmp[i+63:i] - ELSE - dst[i+63:i] := src[i+63:i] - FI -ENDFOR -dst[MAX:128] := 0 - - + AVX512F + General Support + + Return vector of type __m512i with undefined elements.
immintrin.h
 - - Integer - AVX512DQ - Miscellaneous - - - - Extract 128 bits (composed of 2 packed 64-bit integers) from "a", selected with "imm8", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - -CASE imm8[7:0] of -0: tmp[127:0] := a[127:0] -1: tmp[127:0] := a[255:128] -2: tmp[127:0] := a[383:256] -3: tmp[127:0] := a[511:384] -ESAC - -FOR j := 0 to 1 - i := j*64 - IF k[j] - dst[i+63:i] := tmp[i+63:i] - ELSE - dst[i+63:i] := 0 - FI -ENDFOR -dst[MAX:128] := 0 - - + + Floating Point + AVX512F + General Support + + Return vector of type __m512d with undefined elements.
immintrin.h
 - + Floating Point - AVX512VL AVX512F - Miscellaneous - - - - - Fix up packed double-precision (64-bit) floating-point elements in "a" and "b" using packed 64-bit integers in "c", and store the results in "dst". "imm8" is used to set the required flags reporting. - -enum TOKEN_TYPE { - QNAN_TOKEN := 0, - SNAN_TOKEN := 1, - ZERO_VALUE_TOKEN := 2, - ONE_VALUE_TOKEN := 3, - NEG_INF_TOKEN := 4, - POS_INF_TOKEN := 5, - NEG_VALUE_TOKEN := 6, - POS_VALUE_TOKEN := 7 -} -FIXUPIMMPD(src1[63:0], src2[63:0], src3[63:0], imm8[7:0]){ - tsrc[63:0] := ((src2[62:52] == 0) AND (MXCSR.DAZ == 1)) ? 0.0 : src2[63:0] - CASE(tsrc[63:0] of TOKEN_TYPE) - QNAN_TOKEN:j := 0 - SNAN_TOKEN:j := 1 - ZERO_VALUE_TOKEN: j := 2 - ONE_VALUE_TOKEN: j := 3 - NEG_INF_TOKEN: j := 4 - POS_INF_TOKEN: j := 5 - NEG_VALUE_TOKEN: j := 6 - POS_VALUE_TOKEN: j := 7 - ESAC - - token_response[3:0] := src3[3+4*j:4*j] - - CASE(token_response[3:0]) of - 0 : dest[63:0] := src1[63:0] - 1 : dest[63:0] := tsrc[63:0] - 2 : dest[63:0] := QNaN(tsrc[63:0]) - 3 : dest[63:0] := QNAN_Indefinite - 4 : dest[63:0] := -INF - 5 : dest[63:0] := +INF - 6 : dest[63:0] := tsrc.sign? –INF : +INF - 7 : dest[63:0] := -0 - 8 : dest[63:0] := +0 - 9 : dest[63:0] := -1 - 10: dest[63:0] := +1 - 11: dest[63:0] := 1⁄2 - 12: dest[63:0] := 90.0 - 13: dest[63:0] := PI/2 - 14: dest[63:0] := MAX_FLOAT - 15: dest[63:0] := -MAX_FLOAT - ESAC - - CASE(tsrc[31:0] of TOKEN_TYPE) - ZERO_VALUE_TOKEN: if imm8[0] then set #ZE - ZERO_VALUE_TOKEN: if imm8[1] then set #IE - ONE_VALUE_TOKEN: if imm8[2] then set #ZE - ONE_VALUE_TOKEN: if imm8[3] then set #IE - SNAN_TOKEN: if imm8[4] then set #IE - NEG_INF_TOKEN: if imm8[5] then set #IE - NEG_VALUE_TOKEN: if imm8[6] then set #IE - POS_INF_TOKEN: if imm8[7] then set #IE - ESAC - RETURN dest[63:0] -} - -FOR j := 0 to 3 + General Support + + Return vector of type __m512 with undefined elements. +
immintrin.h
+ + + Floating Point + AVX512F + Trigonometry + + + Compute the inverse cosine of packed double-precision (64-bit) floating-point elements in "a" expressed in radians, and store the results in "dst". + +FOR j := 0 to 7 i := j*64 - dst[i+63:i] := FIXUPIMMPD(a[i+63:i], b[i+63:i], c[i+63:i], imm8[7:0]) + dst[i+63:i] := ACOS(a[i+63:i]) ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 -
immintrin.h
 - + Floating Point - AVX512VL AVX512F - Miscellaneous - - - - - - Fix up packed double-precision (64-bit) floating-point elements in "a" and "b" using packed 64-bit integers in "c", and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). "imm8" is used to set the required flags reporting. + Trigonometry + + + + + Compute the inverse cosine of packed double-precision (64-bit) floating-point elements in "a" expressed in radians, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -enum TOKEN_TYPE { - QNAN_TOKEN := 0, - SNAN_TOKEN := 1, - ZERO_VALUE_TOKEN := 2, - ONE_VALUE_TOKEN := 3, - NEG_INF_TOKEN := 4, - POS_INF_TOKEN := 5, - NEG_VALUE_TOKEN := 6, - POS_VALUE_TOKEN := 7 -} -FIXUPIMMPD(src1[63:0], src2[63:0], src3[63:0], imm8[7:0]){ - tsrc[63:0] := ((src2[62:52] == 0) AND (MXCSR.DAZ == 1)) ? 0.0 : src2[63:0] - CASE(tsrc[63:0] of TOKEN_TYPE) - QNAN_TOKEN:j := 0 - SNAN_TOKEN:j := 1 - ZERO_VALUE_TOKEN: j := 2 - ONE_VALUE_TOKEN: j := 3 - NEG_INF_TOKEN: j := 4 - POS_INF_TOKEN: j := 5 - NEG_VALUE_TOKEN: j := 6 - POS_VALUE_TOKEN: j := 7 - ESAC - - token_response[3:0] := src3[3+4*j:4*j] - - CASE(token_response[3:0]) of - 0 : dest[63:0] := src1[63:0] - 1 : dest[63:0] := tsrc[63:0] - 2 : dest[63:0] := QNaN(tsrc[63:0]) - 3 : dest[63:0] := QNAN_Indefinite - 4 : dest[63:0] := -INF - 5 : dest[63:0] := +INF - 6 : dest[63:0] := tsrc.sign? –INF : +INF - 7 : dest[63:0] := -0 - 8 : dest[63:0] := +0 - 9 : dest[63:0] := -1 - 10: dest[63:0] := +1 - 11: dest[63:0] := 1⁄2 - 12: dest[63:0] := 90.0 - 13: dest[63:0] := PI/2 - 14: dest[63:0] := MAX_FLOAT - 15: dest[63:0] := -MAX_FLOAT - ESAC - - CASE(tsrc[31:0] of TOKEN_TYPE) - ZERO_VALUE_TOKEN: if imm8[0] then set #ZE - ZERO_VALUE_TOKEN: if imm8[1] then set #IE - ONE_VALUE_TOKEN: if imm8[2] then set #ZE - ONE_VALUE_TOKEN: if imm8[3] then set #IE - SNAN_TOKEN: if imm8[4] then set #IE - NEG_INF_TOKEN: if imm8[5] then set #IE - NEG_VALUE_TOKEN: if imm8[6] then set #IE - POS_INF_TOKEN: if imm8[7] then set #IE - ESAC - RETURN dest[63:0] -} - -FOR j := 0 to 3 +FOR j := 0 to 7 i := j*64 IF k[j] - dst[i+63:i] := FIXUPIMMPD(a[i+63:i], b[i+63:i], c[i+63:i], imm8[7:0]) + dst[i+63:i] := ACOS(a[i+63:i]) ELSE - dst[i+63:i] := a[i+63:i] + dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 -
immintrin.h
 - + Floating Point - AVX512VL AVX512F - Miscellaneous - - - - - - Fix up packed double-precision (64-bit) floating-point elements in "a" and "b" using packed 64-bit integers in "c", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). "imm8" is used to set the required flags reporting. + Trigonometry + + + Compute the inverse cosine of packed single-precision (32-bit) floating-point elements in "a" expressed in radians, and store the results in "dst". -enum TOKEN_TYPE { - QNAN_TOKEN := 0, - SNAN_TOKEN := 1, - ZERO_VALUE_TOKEN := 2, - ONE_VALUE_TOKEN := 3, - NEG_INF_TOKEN := 4, - POS_INF_TOKEN := 5, - NEG_VALUE_TOKEN := 6, - POS_VALUE_TOKEN := 7 -} -FIXUPIMMPD(src1[63:0], src2[63:0], src3[63:0], imm8[7:0]){ - tsrc[63:0] := ((src2[62:52] == 0) AND (MXCSR.DAZ == 1)) ? 0.0 : src2[63:0] - CASE(tsrc[63:0] of TOKEN_TYPE) - QNAN_TOKEN:j := 0 - SNAN_TOKEN:j := 1 - ZERO_VALUE_TOKEN: j := 2 - ONE_VALUE_TOKEN: j := 3 - NEG_INF_TOKEN: j := 4 - POS_INF_TOKEN: j := 5 - NEG_VALUE_TOKEN: j := 6 - POS_VALUE_TOKEN: j := 7 - ESAC - - token_response[3:0] := src3[3+4*j:4*j] - - CASE(token_response[3:0]) of - 0 : dest[63:0] := src1[63:0] - 1 : dest[63:0] := tsrc[63:0] - 2 : dest[63:0] := QNaN(tsrc[63:0]) - 3 : dest[63:0] := QNAN_Indefinite - 4 : dest[63:0] := -INF - 5 : dest[63:0] := +INF - 6 : dest[63:0] := tsrc.sign? –INF : +INF - 7 : dest[63:0] := -0 - 8 : dest[63:0] := +0 - 9 : dest[63:0] := -1 - 10: dest[63:0] := +1 - 11: dest[63:0] := 1⁄2 - 12: dest[63:0] := 90.0 - 13: dest[63:0] := PI/2 - 14: dest[63:0] := MAX_FLOAT - 15: dest[63:0] := -MAX_FLOAT - ESAC - - CASE(tsrc[31:0] of TOKEN_TYPE) - ZERO_VALUE_TOKEN: if imm8[0] then set #ZE - ZERO_VALUE_TOKEN: if imm8[1] then set #IE - ONE_VALUE_TOKEN: if imm8[2] then set #ZE - ONE_VALUE_TOKEN: if imm8[3] then set #IE - SNAN_TOKEN: if imm8[4] then set #IE - NEG_INF_TOKEN: if imm8[5] then set #IE - NEG_VALUE_TOKEN: if imm8[6] then set #IE - POS_INF_TOKEN: if imm8[7] then set #IE - ESAC - RETURN dest[63:0] -} - -FOR j := 0 to 3 - i := j*64 +FOR j := 0 to 15 + i := j*32 + dst[i+31:i] := ACOS(a[i+31:i]) +ENDFOR +dst[MAX:512] := 0 + +
immintrin.h
+ + + Floating Point + AVX512F + Trigonometry + + + + + Compute the inverse cosine of packed single-precision (32-bit) floating-point elements in "a" expressed in radians, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +FOR j := 0 to 15 + i := j*32 IF k[j] - dst[i+63:i] := FIXUPIMMPD(a[i+63:i], b[i+63:i], c[i+63:i], imm8[7:0]) + dst[i+31:i] := ACOS(a[i+31:i]) ELSE - dst[i+63:i] := 0 + dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 -
immintrin.h
 - + Floating Point - AVX512VL AVX512F - Miscellaneous - - - - - Fix up packed double-precision (64-bit) floating-point elements in "a" and "b" using packed 64-bit integers in "c", and store the results in "dst". "imm8" is used to set the required flags reporting. + Trigonometry + + + Compute the inverse hyperbolic cosine of packed double-precision (64-bit) floating-point elements in "a" expressed in radians, and store the results in "dst". -enum TOKEN_TYPE { - QNAN_TOKEN := 0, - SNAN_TOKEN := 1, - ZERO_VALUE_TOKEN := 2, - ONE_VALUE_TOKEN := 3, - NEG_INF_TOKEN := 4, - POS_INF_TOKEN := 5, - NEG_VALUE_TOKEN := 6, - POS_VALUE_TOKEN := 7 -} -FIXUPIMMPD(src1[63:0], src2[63:0], src3[63:0], imm8[7:0]){ - tsrc[63:0] := ((src2[62:52] == 0) AND (MXCSR.DAZ == 1)) ? 0.0 : src2[63:0] - CASE(tsrc[63:0] of TOKEN_TYPE) - QNAN_TOKEN:j := 0 - SNAN_TOKEN:j := 1 - ZERO_VALUE_TOKEN: j := 2 - ONE_VALUE_TOKEN: j := 3 - NEG_INF_TOKEN: j := 4 - POS_INF_TOKEN: j := 5 - NEG_VALUE_TOKEN: j := 6 - POS_VALUE_TOKEN: j := 7 - ESAC - - token_response[3:0] := src3[3+4*j:4*j] - - CASE(token_response[3:0]) of - 0 : dest[63:0] := src1[63:0] - 1 : dest[63:0] := tsrc[63:0] - 2 : dest[63:0] := QNaN(tsrc[63:0]) - 3 : dest[63:0] := QNAN_Indefinite - 4 : dest[63:0] := -INF - 5 : dest[63:0] := +INF - 6 : dest[63:0] := tsrc.sign? –INF : +INF - 7 : dest[63:0] := -0 - 8 : dest[63:0] := +0 - 9 : dest[63:0] := -1 - 10: dest[63:0] := +1 - 11: dest[63:0] := 1⁄2 - 12: dest[63:0] := 90.0 - 13: dest[63:0] := PI/2 - 14: dest[63:0] := MAX_FLOAT - 15: dest[63:0] := -MAX_FLOAT - ESAC - - CASE(tsrc[31:0] of TOKEN_TYPE) - ZERO_VALUE_TOKEN: if imm8[0] then set #ZE - ZERO_VALUE_TOKEN: if imm8[1] then set #IE - ONE_VALUE_TOKEN: if imm8[2] then set #ZE - ONE_VALUE_TOKEN: if imm8[3] then set #IE - SNAN_TOKEN: if imm8[4] then set #IE - NEG_INF_TOKEN: if imm8[5] then set #IE - NEG_VALUE_TOKEN: if imm8[6] then set #IE - POS_INF_TOKEN: if imm8[7] then set #IE - ESAC - RETURN dest[63:0] -} - -FOR j := 0 to 1 +FOR j := 0 to 7 i := j*64 - dst[i+63:i] := FIXUPIMMPD(a[i+63:i], b[i+63:i], c[i+63:i], imm8[7:0]) + dst[i+63:i] := ACOSH(a[i+63:i]) ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 -
immintrin.h
 - + Floating Point - AVX512VL AVX512F - Miscellaneous - - - - - - Fix up packed double-precision (64-bit) floating-point elements in "a" and "b" using packed 64-bit integers in "c", and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). "imm8" is used to set the required flags reporting. + Trigonometry + + + + + Compute the inverse hyperbolic cosine of packed double-precision (64-bit) floating-point elements in "a" expressed in radians, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -enum TOKEN_TYPE { - QNAN_TOKEN := 0, - SNAN_TOKEN := 1, - ZERO_VALUE_TOKEN := 2, - ONE_VALUE_TOKEN := 3, - NEG_INF_TOKEN := 4, - POS_INF_TOKEN := 5, - NEG_VALUE_TOKEN := 6, - POS_VALUE_TOKEN := 7 -} -FIXUPIMMPD(src1[63:0], src2[63:0], src3[63:0], imm8[7:0]){ - tsrc[63:0] := ((src2[62:52] == 0) AND (MXCSR.DAZ == 1)) ? 0.0 : src2[63:0] - CASE(tsrc[63:0] of TOKEN_TYPE) - QNAN_TOKEN:j := 0 - SNAN_TOKEN:j := 1 - ZERO_VALUE_TOKEN: j := 2 - ONE_VALUE_TOKEN: j := 3 - NEG_INF_TOKEN: j := 4 - POS_INF_TOKEN: j := 5 - NEG_VALUE_TOKEN: j := 6 - POS_VALUE_TOKEN: j := 7 - ESAC - - token_response[3:0] := src3[3+4*j:4*j] - - CASE(token_response[3:0]) of - 0 : dest[63:0] := src1[63:0] - 1 : dest[63:0] := tsrc[63:0] - 2 : dest[63:0] := QNaN(tsrc[63:0]) - 3 : dest[63:0] := QNAN_Indefinite - 4 : dest[63:0] := -INF - 5 : dest[63:0] := +INF - 6 : dest[63:0] := tsrc.sign? –INF : +INF - 7 : dest[63:0] := -0 - 8 : dest[63:0] := +0 - 9 : dest[63:0] := -1 - 10: dest[63:0] := +1 - 11: dest[63:0] := 1⁄2 - 12: dest[63:0] := 90.0 - 13: dest[63:0] := PI/2 - 14: dest[63:0] := MAX_FLOAT - 15: dest[63:0] := -MAX_FLOAT - ESAC - - CASE(tsrc[31:0] of TOKEN_TYPE) - ZERO_VALUE_TOKEN: if imm8[0] then set #ZE - ZERO_VALUE_TOKEN: if imm8[1] then set #IE - ONE_VALUE_TOKEN: if imm8[2] then set #ZE - ONE_VALUE_TOKEN: if imm8[3] then set #IE - SNAN_TOKEN: if imm8[4] then set #IE - NEG_INF_TOKEN: if imm8[5] then set #IE - NEG_VALUE_TOKEN: if imm8[6] then set #IE - POS_INF_TOKEN: if imm8[7] then set #IE - ESAC - RETURN dest[63:0] -} - -FOR j := 0 to 1 +FOR j := 0 to 7 i := j*64 IF k[j] - dst[i+63:i] := FIXUPIMMPD(a[i+63:i], b[i+63:i], c[i+63:i], imm8[7:0]) + dst[i+63:i] := ACOSH(a[i+63:i]) ELSE - dst[i+63:i] := a[i+63:i] + dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 -
immintrin.h
 - + Floating Point - AVX512VL AVX512F - Miscellaneous - - - - - - Fix up packed double-precision (64-bit) floating-point elements in "a" and "b" using packed 64-bit integers in "c", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). "imm8" is used to set the required flags reporting. + Trigonometry + + + Compute the inverse hyperbolic cosine of packed single-precision (32-bit) floating-point elements in "a" expressed in radians, and store the results in "dst". -enum TOKEN_TYPE { - QNAN_TOKEN := 0, - SNAN_TOKEN := 1, - ZERO_VALUE_TOKEN := 2, - ONE_VALUE_TOKEN := 3, - NEG_INF_TOKEN := 4, - POS_INF_TOKEN := 5, - NEG_VALUE_TOKEN := 6, - POS_VALUE_TOKEN := 7 -} -FIXUPIMMPD(src1[63:0], src2[63:0], src3[63:0], imm8[7:0]){ - tsrc[63:0] := ((src2[62:52] == 0) AND (MXCSR.DAZ == 1)) ? 0.0 : src2[63:0] - CASE(tsrc[63:0] of TOKEN_TYPE) - QNAN_TOKEN:j := 0 - SNAN_TOKEN:j := 1 - ZERO_VALUE_TOKEN: j := 2 - ONE_VALUE_TOKEN: j := 3 - NEG_INF_TOKEN: j := 4 - POS_INF_TOKEN: j := 5 - NEG_VALUE_TOKEN: j := 6 - POS_VALUE_TOKEN: j := 7 - ESAC - - token_response[3:0] := src3[3+4*j:4*j] - - CASE(token_response[3:0]) of - 0 : dest[63:0] := src1[63:0] - 1 : dest[63:0] := tsrc[63:0] - 2 : dest[63:0] := QNaN(tsrc[63:0]) - 3 : dest[63:0] := QNAN_Indefinite - 4 : dest[63:0] := -INF - 5 : dest[63:0] := +INF - 6 : dest[63:0] := tsrc.sign? –INF : +INF - 7 : dest[63:0] := -0 - 8 : dest[63:0] := +0 - 9 : dest[63:0] := -1 - 10: dest[63:0] := +1 - 11: dest[63:0] := 1⁄2 - 12: dest[63:0] := 90.0 - 13: dest[63:0] := PI/2 - 14: dest[63:0] := MAX_FLOAT - 15: dest[63:0] := -MAX_FLOAT - ESAC - - CASE(tsrc[31:0] of TOKEN_TYPE) - ZERO_VALUE_TOKEN: if imm8[0] then set #ZE - ZERO_VALUE_TOKEN: if imm8[1] then set #IE - ONE_VALUE_TOKEN: if imm8[2] then set #ZE - ONE_VALUE_TOKEN: if imm8[3] then set #IE - SNAN_TOKEN: if imm8[4] then set #IE - NEG_INF_TOKEN: if imm8[5] then set #IE - NEG_VALUE_TOKEN: if imm8[6] then set #IE - POS_INF_TOKEN: if imm8[7] then set #IE - ESAC - RETURN dest[63:0] -} - -FOR j := 0 to 1 - i := j*64 +FOR j := 0 to 15 + i := j*32 + dst[i+31:i] := ACOSH(a[i+31:i]) +ENDFOR +dst[MAX:512] := 0 + +
immintrin.h
+ + + Floating Point + AVX512F + Trigonometry + + + + + Compute the inverse hyperbolic cosine of packed single-precision (32-bit) floating-point elements in "a" expressed in radians, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +FOR j := 0 to 15 + i := j*32 IF k[j] - dst[i+63:i] := FIXUPIMMPD(a[i+63:i], b[i+63:i], c[i+63:i], imm8[7:0]) + dst[i+31:i] := ACOSH(a[i+31:i]) ELSE - dst[i+63:i] := 0 + dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 -
immintrin.h
 - + Floating Point - AVX512VL AVX512F - Miscellaneous - - - - - Fix up packed single-precision (32-bit) floating-point elements in "a" and "b" using packed 32-bit integers in "c", and store the results in "dst". "imm8" is used to set the required flags reporting. + Trigonometry + + + Compute the inverse sine of packed double-precision (64-bit) floating-point elements in "a" expressed in radians, and store the results in "dst". -enum TOKEN_TYPE { - QNAN_TOKEN := 0, - SNAN_TOKEN L= 1, - ZERO_VALUE_TOKEN := 2, - ONE_VALUE_TOKEN := 3, - NEG_INF_TOKEN := 4, - POS_INF_TOKEN := 5, - NEG_VALUE_TOKEN := 6, - POS_VALUE_TOKEN := 7 -} -FIXUPIMMPD(src1[31:0], src2[31:0], src3[31:0], imm8[7:0]){ - tsrc[31:0] := ((src2[30:23] == 0) AND (MXCSR.DAZ == 1)) ? 0.0 : src2[31:0] - CASE(tsrc[31:0] of TOKEN_TYPE) - QNAN_TOKEN:j := 0 - SNAN_TOKEN:j := 1 - ZERO_VALUE_TOKEN: j := 2 - ONE_VALUE_TOKEN: j := 3 - NEG_INF_TOKEN: j := 4 - POS_INF_TOKEN: j := 5 - NEG_VALUE_TOKEN: j := 6 - POS_VALUE_TOKEN: j := 7 - ESAC - - token_response[3:0] := src3[3+4*j:4*j] - - CASE(token_response[3:0]) of - 0 : dest[31:0] := src1[31:0] - 1 : dest[31:0] := tsrc[31:0] - 2 : dest[31:0] := QNaN(tsrc[31:0]) - 3 : dest[31:0] := QNAN_Indefinite - 4 : dest[31:0] := -INF - 5 : dest[31:0] := +INF - 6 : dest[31:0] := tsrc.sign? –INF : +INF - 7 : dest[31:0] := -0 - 8 : dest[31:0] := +0 - 9 : dest[31:0] := -1 - 10: dest[31:0] := +1 - 11: dest[31:0] := 1⁄2 - 12: dest[31:0] := 90.0 - 13: dest[31:0] := PI/2 - 14: dest[31:0] := MAX_FLOAT - 15: dest[31:0] := -MAX_FLOAT - ESAC - - CASE(tsrc[31:0] of TOKEN_TYPE) - ZERO_VALUE_TOKEN: if imm8[0] then set #ZE - ZERO_VALUE_TOKEN: if imm8[1] then set #IE - ONE_VALUE_TOKEN: if imm8[2] then set #ZE - ONE_VALUE_TOKEN: if imm8[3] then set #IE - SNAN_TOKEN: if imm8[4] then set #IE - NEG_INF_TOKEN: if imm8[5] then set #IE - NEG_VALUE_TOKEN: if imm8[6] then set #IE - POS_INF_TOKEN: if imm8[7] then set #IE - ESAC - RETURN dest[31:0] -} - FOR j := 0 to 7 - i := j*32 - dst[i+31:i] := FIXUPIMMPD(a[i+31:i], b[i+31:i], c[i+31:i], imm8[7:0]) + i := j*64 + dst[i+63:i] := ASIN(a[i+63:i]) ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 -
immintrin.h
 - + Floating Point - AVX512VL AVX512F - Miscellaneous - - - - - - Fix up packed single-precision (32-bit) floating-point elements in "a" and "b" using packed 32-bit integers in "c", and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). "imm8" is used to set the required flags reporting. + Trigonometry + + + + + Compute the inverse sine of packed double-precision (64-bit) floating-point elements in "a" expressed in radians, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -enum TOKEN_TYPE { - QNAN_TOKEN := 0, - SNAN_TOKEN L= 1, - ZERO_VALUE_TOKEN := 2, - ONE_VALUE_TOKEN := 3, - NEG_INF_TOKEN := 4, - POS_INF_TOKEN := 5, - NEG_VALUE_TOKEN := 6, - POS_VALUE_TOKEN := 7 -} -FIXUPIMMPD(src1[31:0], src2[31:0], src3[31:0], imm8[7:0]){ - tsrc[31:0] := ((src2[30:23] == 0) AND (MXCSR.DAZ == 1)) ? 0.0 : src2[31:0] - CASE(tsrc[31:0] of TOKEN_TYPE) - QNAN_TOKEN:j := 0 - SNAN_TOKEN:j := 1 - ZERO_VALUE_TOKEN: j := 2 - ONE_VALUE_TOKEN: j := 3 - NEG_INF_TOKEN: j := 4 - POS_INF_TOKEN: j := 5 - NEG_VALUE_TOKEN: j := 6 - POS_VALUE_TOKEN: j := 7 - ESAC - - token_response[3:0] := src3[3+4*j:4*j] - - CASE(token_response[3:0]) of - 0 : dest[31:0] := src1[31:0] - 1 : dest[31:0] := tsrc[31:0] - 2 : dest[31:0] := QNaN(tsrc[31:0]) - 3 : dest[31:0] := QNAN_Indefinite - 4 : dest[31:0] := -INF - 5 : dest[31:0] := +INF - 6 : dest[31:0] := tsrc.sign? –INF : +INF - 7 : dest[31:0] := -0 - 8 : dest[31:0] := +0 - 9 : dest[31:0] := -1 - 10: dest[31:0] := +1 - 11: dest[31:0] := 1⁄2 - 12: dest[31:0] := 90.0 - 13: dest[31:0] := PI/2 - 14: dest[31:0] := MAX_FLOAT - 15: dest[31:0] := -MAX_FLOAT - ESAC - - CASE(tsrc[31:0] of TOKEN_TYPE) - ZERO_VALUE_TOKEN: if imm8[0] then set #ZE - ZERO_VALUE_TOKEN: if imm8[1] then set #IE - ONE_VALUE_TOKEN: if imm8[2] then set #ZE - ONE_VALUE_TOKEN: if imm8[3] then set #IE - SNAN_TOKEN: if imm8[4] then set #IE - NEG_INF_TOKEN: if imm8[5] then set #IE - NEG_VALUE_TOKEN: if imm8[6] then set #IE - POS_INF_TOKEN: if imm8[7] then set #IE - ESAC - RETURN dest[31:0] -} - FOR j := 0 to 7 - i := j*32 + i := j*64 IF k[j] - dst[i+31:i] := FIXUPIMMPD(a[i+31:i], b[i+31:i], c[i+31:i], imm8[7:0]) + dst[i+63:i] := ASIN(a[i+63:i]) ELSE - dst[i+31:i] := a[i+31:i] + dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 -
immintrin.h
 - + Floating Point - AVX512VL AVX512F - Miscellaneous - - - - - - Fix up packed single-precision (32-bit) floating-point elements in "a" and "b" using packed 32-bit integers in "c", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). "imm8" is used to set the required flags reporting. + Trigonometry + + + Compute the inverse sine of packed single-precision (32-bit) floating-point elements in "a" expressed in radians, and store the results in "dst". -enum TOKEN_TYPE { - QNAN_TOKEN := 0, - SNAN_TOKEN L= 1, - ZERO_VALUE_TOKEN := 2, - ONE_VALUE_TOKEN := 3, - NEG_INF_TOKEN := 4, - POS_INF_TOKEN := 5, - NEG_VALUE_TOKEN := 6, - POS_VALUE_TOKEN := 7 -} -FIXUPIMMPD(src1[31:0], src2[31:0], src3[31:0], imm8[7:0]){ - tsrc[31:0] := ((src2[30:23] == 0) AND (MXCSR.DAZ == 1)) ? 0.0 : src2[31:0] - CASE(tsrc[31:0] of TOKEN_TYPE) - QNAN_TOKEN:j := 0 - SNAN_TOKEN:j := 1 - ZERO_VALUE_TOKEN: j := 2 - ONE_VALUE_TOKEN: j := 3 - NEG_INF_TOKEN: j := 4 - POS_INF_TOKEN: j := 5 - NEG_VALUE_TOKEN: j := 6 - POS_VALUE_TOKEN: j := 7 - ESAC - - token_response[3:0] := src3[3+4*j:4*j] - - CASE(token_response[3:0]) of - 0 : dest[31:0] := src1[31:0] - 1 : dest[31:0] := tsrc[31:0] - 2 : dest[31:0] := QNaN(tsrc[31:0]) - 3 : dest[31:0] := QNAN_Indefinite - 4 : dest[31:0] := -INF - 5 : dest[31:0] := +INF - 6 : dest[31:0] := tsrc.sign? –INF : +INF - 7 : dest[31:0] := -0 - 8 : dest[31:0] := +0 - 9 : dest[31:0] := -1 - 10: dest[31:0] := +1 - 11: dest[31:0] := 1⁄2 - 12: dest[31:0] := 90.0 - 13: dest[31:0] := PI/2 - 14: dest[31:0] := MAX_FLOAT - 15: dest[31:0] := -MAX_FLOAT - ESAC - - CASE(tsrc[31:0] of TOKEN_TYPE) - ZERO_VALUE_TOKEN: if imm8[0] then set #ZE - ZERO_VALUE_TOKEN: if imm8[1] then set #IE - ONE_VALUE_TOKEN: if imm8[2] then set #ZE - ONE_VALUE_TOKEN: if imm8[3] then set #IE - SNAN_TOKEN: if imm8[4] then set #IE - NEG_INF_TOKEN: if imm8[5] then set #IE - NEG_VALUE_TOKEN: if imm8[6] then set #IE - POS_INF_TOKEN: if imm8[7] then set #IE - ESAC - RETURN dest[31:0] -} - -FOR j := 0 to 7 +FOR j := 0 to 15 + i := j*32 + dst[i+31:i] := ASIN(a[i+31:i]) +ENDFOR +dst[MAX:512] := 0 + +
immintrin.h
+ + + Floating Point + AVX512F + Trigonometry + + + + + Compute the inverse sine of packed single-precision (32-bit) floating-point elements in "a" expressed in radians, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +FOR j := 0 to 15 i := j*32 IF k[j] - dst[i+31:i] := FIXUPIMMPD(a[i+31:i], b[i+31:i], c[i+31:i], imm8[7:0]) + dst[i+31:i] := ASIN(a[i+31:i]) ELSE - dst[i+31:i] := 0 + dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 -
immintrin.h
 - + Floating Point - AVX512VL AVX512F - Miscellaneous - - - - - Fix up packed single-precision (32-bit) floating-point elements in "a" and "b" using packed 32-bit integers in "c", and store the results in "dst". "imm8" is used to set the required flags reporting. + Trigonometry + + + Compute the inverse hyperbolic sine of packed double-precision (64-bit) floating-point elements in "a" expressed in radians, and store the results in "dst". -enum TOKEN_TYPE { - QNAN_TOKEN := 0, - SNAN_TOKEN L= 1, - ZERO_VALUE_TOKEN := 2, - ONE_VALUE_TOKEN := 3, - NEG_INF_TOKEN := 4, - POS_INF_TOKEN := 5, - NEG_VALUE_TOKEN := 6, - POS_VALUE_TOKEN := 7 -} -FIXUPIMMPD(src1[31:0], src2[31:0], src3[31:0], imm8[7:0]){ - tsrc[31:0] := ((src2[30:23] == 0) AND (MXCSR.DAZ == 1)) ? 0.0 : src2[31:0] - CASE(tsrc[31:0] of TOKEN_TYPE) - QNAN_TOKEN:j := 0 - SNAN_TOKEN:j := 1 - ZERO_VALUE_TOKEN: j := 2 - ONE_VALUE_TOKEN: j := 3 - NEG_INF_TOKEN: j := 4 - POS_INF_TOKEN: j := 5 - NEG_VALUE_TOKEN: j := 6 - POS_VALUE_TOKEN: j := 7 - ESAC - - token_response[3:0] := src3[3+4*j:4*j] - - CASE(token_response[3:0]) of - 0 : dest[31:0] := src1[31:0] - 1 : dest[31:0] := tsrc[31:0] - 2 : dest[31:0] := QNaN(tsrc[31:0]) - 3 : dest[31:0] := QNAN_Indefinite - 4 : dest[31:0] := -INF - 5 : dest[31:0] := +INF - 6 : dest[31:0] := tsrc.sign? –INF : +INF - 7 : dest[31:0] := -0 - 8 : dest[31:0] := +0 - 9 : dest[31:0] := -1 - 10: dest[31:0] := +1 - 11: dest[31:0] := 1⁄2 - 12: dest[31:0] := 90.0 - 13: dest[31:0] := PI/2 - 14: dest[31:0] := MAX_FLOAT - 15: dest[31:0] := -MAX_FLOAT - ESAC - - CASE(tsrc[31:0] of TOKEN_TYPE) - ZERO_VALUE_TOKEN: if imm8[0] then set #ZE - ZERO_VALUE_TOKEN: if imm8[1] then set #IE - ONE_VALUE_TOKEN: if imm8[2] then set #ZE - ONE_VALUE_TOKEN: if imm8[3] then set #IE - SNAN_TOKEN: if imm8[4] then set #IE - NEG_INF_TOKEN: if imm8[5] then set #IE - NEG_VALUE_TOKEN: if imm8[6] then set #IE - POS_INF_TOKEN: if imm8[7] then set #IE - ESAC - RETURN dest[31:0] -} - -FOR j := 0 to 3 - i := j*32 - dst[i+31:i] := FIXUPIMMPD(a[i+31:i], b[i+31:i], c[i+31:i], imm8[7:0]) +FOR j := 0 to 7 + i := j*64 + dst[i+63:i] := ASINH(a[i+63:i]) ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 -
immintrin.h
 - + Floating Point - AVX512VL AVX512F - Miscellaneous - - - - - - Fix up packed single-precision (32-bit) floating-point elements in "a" and "b" using packed 32-bit integers in "c", and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). "imm8" is used to set the required flags reporting. + Trigonometry + + + + + Compute the inverse hyperbolic sine of packed double-precision (64-bit) floating-point elements in "a" expressed in radians, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -enum TOKEN_TYPE { - QNAN_TOKEN := 0, - SNAN_TOKEN L= 1, - ZERO_VALUE_TOKEN := 2, - ONE_VALUE_TOKEN := 3, - NEG_INF_TOKEN := 4, - POS_INF_TOKEN := 5, - NEG_VALUE_TOKEN := 6, - POS_VALUE_TOKEN := 7 -} -FIXUPIMMPD(src1[31:0], src2[31:0], src3[31:0], imm8[7:0]){ - tsrc[31:0] := ((src2[30:23] == 0) AND (MXCSR.DAZ == 1)) ? 0.0 : src2[31:0] - CASE(tsrc[31:0] of TOKEN_TYPE) - QNAN_TOKEN:j := 0 - SNAN_TOKEN:j := 1 - ZERO_VALUE_TOKEN: j := 2 - ONE_VALUE_TOKEN: j := 3 - NEG_INF_TOKEN: j := 4 - POS_INF_TOKEN: j := 5 - NEG_VALUE_TOKEN: j := 6 - POS_VALUE_TOKEN: j := 7 - ESAC - - token_response[3:0] := src3[3+4*j:4*j] - - CASE(token_response[3:0]) of - 0 : dest[31:0] := src1[31:0] - 1 : dest[31:0] := tsrc[31:0] - 2 : dest[31:0] := QNaN(tsrc[31:0]) - 3 : dest[31:0] := QNAN_Indefinite - 4 : dest[31:0] := -INF - 5 : dest[31:0] := +INF - 6 : dest[31:0] := tsrc.sign? –INF : +INF - 7 : dest[31:0] := -0 - 8 : dest[31:0] := +0 - 9 : dest[31:0] := -1 - 10: dest[31:0] := +1 - 11: dest[31:0] := 1⁄2 - 12: dest[31:0] := 90.0 - 13: dest[31:0] := PI/2 - 14: dest[31:0] := MAX_FLOAT - 15: dest[31:0] := -MAX_FLOAT - ESAC - - CASE(tsrc[31:0] of TOKEN_TYPE) - ZERO_VALUE_TOKEN: if imm8[0] then set #ZE - ZERO_VALUE_TOKEN: if imm8[1] then set #IE - ONE_VALUE_TOKEN: if imm8[2] then set #ZE - ONE_VALUE_TOKEN: if imm8[3] then set #IE - SNAN_TOKEN: if imm8[4] then set #IE - NEG_INF_TOKEN: if imm8[5] then set #IE - NEG_VALUE_TOKEN: if imm8[6] then set #IE - POS_INF_TOKEN: if imm8[7] then set #IE - ESAC - RETURN dest[31:0] -} - -FOR j := 0 to 3 - i := j*32 +FOR j := 0 to 7 + i := j*64 IF k[j] - dst[i+31:i] := FIXUPIMMPD(a[i+31:i], b[i+31:i], c[i+31:i], imm8[7:0]) + dst[i+63:i] := ASINH(a[i+63:i]) ELSE - dst[i+31:i] := a[i+31:i] + dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 -
immintrin.h
 - + Floating Point - AVX512VL AVX512F - Miscellaneous - - - - - - Fix up packed single-precision (32-bit) floating-point elements in "a" and "b" using packed 32-bit integers in "c", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). "imm8" is used to set the required flags reporting. + Trigonometry + + + Compute the inverse hyperbolic sine of packed single-precision (32-bit) floating-point elements in "a" expressed in radians, and store the results in "dst". -enum TOKEN_TYPE { - QNAN_TOKEN := 0, - SNAN_TOKEN L= 1, - ZERO_VALUE_TOKEN := 2, - ONE_VALUE_TOKEN := 3, - NEG_INF_TOKEN := 4, - POS_INF_TOKEN := 5, - NEG_VALUE_TOKEN := 6, - POS_VALUE_TOKEN := 7 -} -FIXUPIMMPD(src1[31:0], src2[31:0], src3[31:0], imm8[7:0]){ - tsrc[31:0] := ((src2[30:23] == 0) AND (MXCSR.DAZ == 1)) ? 0.0 : src2[31:0] - CASE(tsrc[31:0] of TOKEN_TYPE) - QNAN_TOKEN:j := 0 - SNAN_TOKEN:j := 1 - ZERO_VALUE_TOKEN: j := 2 - ONE_VALUE_TOKEN: j := 3 - NEG_INF_TOKEN: j := 4 - POS_INF_TOKEN: j := 5 - NEG_VALUE_TOKEN: j := 6 - POS_VALUE_TOKEN: j := 7 - ESAC - - token_response[3:0] := src3[3+4*j:4*j] - - CASE(token_response[3:0]) of - 0 : dest[31:0] := src1[31:0] - 1 : dest[31:0] := tsrc[31:0] - 2 : dest[31:0] := QNaN(tsrc[31:0]) - 3 : dest[31:0] := QNAN_Indefinite - 4 : dest[31:0] := -INF - 5 : dest[31:0] := +INF - 6 : dest[31:0] := tsrc.sign? –INF : +INF - 7 : dest[31:0] := -0 - 8 : dest[31:0] := +0 - 9 : dest[31:0] := -1 - 10: dest[31:0] := +1 - 11: dest[31:0] := 1⁄2 - 12: dest[31:0] := 90.0 - 13: dest[31:0] := PI/2 - 14: dest[31:0] := MAX_FLOAT - 15: dest[31:0] := -MAX_FLOAT - ESAC - - CASE(tsrc[31:0] of TOKEN_TYPE) - ZERO_VALUE_TOKEN: if imm8[0] then set #ZE - ZERO_VALUE_TOKEN: if imm8[1] then set #IE - ONE_VALUE_TOKEN: if imm8[2] then set #ZE - ONE_VALUE_TOKEN: if imm8[3] then set #IE - SNAN_TOKEN: if imm8[4] then set #IE - NEG_INF_TOKEN: if imm8[5] then set #IE - NEG_VALUE_TOKEN: if imm8[6] then set #IE - POS_INF_TOKEN: if imm8[7] then set #IE - ESAC - RETURN dest[31:0] -} - -FOR j := 0 to 3 +FOR j := 0 to 15 i := j*32 - IF k[j] - dst[i+31:i] := FIXUPIMMPD(a[i+31:i], b[i+31:i], c[i+31:i], imm8[7:0]) - ELSE - dst[i+31:i] := 0 - FI + dst[i+31:i] := ASINH(a[i+31:i]) ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 -
immintrin.h
 - + Floating Point - AVX512VL AVX512F - Arithmetic - - - - - Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", add the intermediate result to packed elements in "c", and store the results in "dst" using writemask "k" (elements are copied from "c" when the corresponding mask bit is not set). - + Trigonometry + + + + + Compute the inverse hyperbolic sine of packed single-precision (32-bit) floating-point elements in "a" expressed in radians, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*64 +FOR j := 0 to 15 + i := j*32 IF k[j] - dst[i+63:i] := (a[i+63:i] * b[i+63:i]) + c[i+63:i] + dst[i+31:i] := ASINH(a[i+31:i]) ELSE - dst[i+63:i] := c[i+63:i] + dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - - -
immintrin.h
 - + Floating Point - AVX512VL AVX512F - Arithmetic - - - - - Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", add the intermediate result to packed elements in "c", and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). - + Trigonometry + + + + Compute the inverse tangent of packed double-precision (64-bit) floating-point elements in "a" divided by packed elements in "b", and store the results in "dst" expressed in radians. -FOR j := 0 to 3 +FOR j := 0 to 7 i := j*64 - IF k[j] - dst[i+63:i] := (a[i+63:i] * b[i+63:i]) + c[i+63:i] - ELSE - dst[i+63:i] := a[i+63:i] - FI + dst[i+63:i] := ATAN2(a[i+63:i], b[i+63:i]) ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - - -
immintrin.h
 - + Floating Point - AVX512VL AVX512F - Arithmetic - - - - - Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", add the intermediate result to packed elements in "c", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - + Trigonometry + + + + + + Compute the inverse tangent of packed double-precision (64-bit) floating-point elements in "a" divided by packed elements in "b", and store the results in "dst" expressed in radians using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 3 +FOR j := 0 to 7 i := j*64 IF k[j] - dst[i+63:i] := (a[i+63:i] * b[i+63:i]) + c[i+63:i] + dst[i+63:i] := ATAN2(a[i+63:i], b[i+63:i]) ELSE - dst[i+63:i] := 0 + dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - - -
immintrin.h
 - + Floating Point - AVX512VL AVX512F - Arithmetic - - - - - Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", add the intermediate result to packed elements in "c", and store the results in "dst" using writemask "k" (elements are copied from "c" when the corresponding mask bit is not set). - + Trigonometry + + + + Compute the inverse tangent of packed single-precision (32-bit) floating-point elements in "a" divided by packed elements in "b", and store the results in "dst" expressed in radians. -FOR j := 0 to 1 - i := j*64 +FOR j := 0 to 15 + i := j*32 + dst[i+31:i] := ATAN2(a[i+31:i], b[i+31:i]) +ENDFOR +dst[MAX:512] := 0 + +
immintrin.h
+ + + Floating Point + AVX512F + Trigonometry + + + + + + Compute the inverse tangent of packed single-precision (32-bit) floating-point elements in "a" divided by packed elements in "b", and store the results in "dst" expressed in radians using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +FOR j := 0 to 15 + i := j*32 IF k[j] - dst[i+63:i] := (a[i+63:i] * b[i+63:i]) + c[i+63:i] + dst[i+31:i] := ATAN2(a[i+31:i], b[i+31:i]) ELSE - dst[i+63:i] := c[i+63:i] + dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - - -
immintrin.h
 - + Floating Point - AVX512VL AVX512F - Arithmetic - - - - - Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", add the intermediate result to packed elements in "c", and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). - + Trigonometry + + + Compute the inverse tangent of packed double-precision (64-bit) floating-point elements in "a" and store the results in "dst" expressed in radians. -FOR j := 0 to 1 +FOR j := 0 to 7 i := j*64 - IF k[j] - dst[i+63:i] := (a[i+63:i] * b[i+63:i]) + c[i+63:i] - ELSE - dst[i+63:i] := a[i+63:i] - FI + dst[i+63:i] := ATAN(a[i+63:i]) ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - - -
immintrin.h
 - + Floating Point - AVX512VL AVX512F - Arithmetic - - - - - Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", add the intermediate result to packed elements in "c", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - + Trigonometry + + + + + Compute the inverse tangent of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst" expressed in radians using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 1 +FOR j := 0 to 7 i := j*64 IF k[j] - dst[i+63:i] := (a[i+63:i] * b[i+63:i]) + c[i+63:i] + dst[i+63:i] := ATAN(a[i+63:i]) ELSE - dst[i+63:i] := 0 + dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - - -
immintrin.h
 - + Floating Point - AVX512VL AVX512F - Arithmetic - - - - - Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", add the intermediate result to packed elements in "c", and store the results in "dst" using writemask "k" (elements are copied from "c" when the corresponding mask bit is not set). - + Trigonometry + + + Compute the inverse tangent of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst" expressed in radians. -FOR j := 0 to 7 +FOR j := 0 to 15 i := j*32 - IF k[j] - dst[i+31:i] := (a[i+31:i] * b[i+31:i]) + c[i+31:i] - ELSE - dst[i+31:i] := c[i+31:i] - FI + dst[i+31:i] := ATAN(a[i+31:i]) ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - - -
immintrin.h
 - + Floating Point - AVX512VL AVX512F - Arithmetic - - - - - Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", add the intermediate result to packed elements in "c", and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). - + Trigonometry + + + + + Compute the inverse tangent of packed single-precision (32-bit) floating-point elements in "a" expressed in radians, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 7 +FOR j := 0 to 15 i := j*32 IF k[j] - dst[i+31:i] := (a[i+31:i] * b[i+31:i]) + c[i+31:i] + dst[i+31:i] := ATAN(a[i+31:i]) ELSE - dst[i+31:i] := a[i+31:i] + dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - - -
immintrin.h
 - + Floating Point - AVX512VL AVX512F - Arithmetic - - - - - Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", add the intermediate result to packed elements in "c", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - + Trigonometry + + + Compute the inverse hyperbolic tangent of packed double-precision (64-bit) floating-point elements in "a" and store the results in "dst" expressed in radians. FOR j := 0 to 7 - i := j*32 + i := j*64 + dst[i+63:i] := ATANH(a[i+63:i]) +ENDFOR +dst[MAX:512] := 0 + +
immintrin.h
+ + + Floating Point + AVX512F + Trigonometry + + + + + Compute the inverse hyperbolic tangent of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst" expressed in radians using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +FOR j := 0 to 7 + i := j*64 IF k[j] - dst[i+31:i] := (a[i+31:i] * b[i+31:i]) + c[i+31:i] + dst[i+63:i] := ATANH(a[i+63:i]) ELSE - dst[i+31:i] := 0 + dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - - -
immintrin.h
 - + Floating Point - AVX512VL AVX512F - Arithmetic - - - - - Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", add the intermediate result to packed elements in "c", and store the results in "dst" using writemask "k" (elements are copied from "c" when the corresponding mask bit is not set). + Trigonometry + + + Compute the inverse hyperblic tangent of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst" expressed in radians. -FOR j := 0 to 3 +FOR j := 0 to 15 + i := j*32 + dst[i+31:i] := ATANH(a[i+31:i]) +ENDFOR +dst[MAX:512] := 0 + +
immintrin.h
+ + + Floating Point + AVX512F + Trigonometry + + + + + Compute the inverse hyperbolic tangent of packed single-precision (32-bit) floating-point elements in "a" expressed in radians, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +FOR j := 0 to 15 i := j*32 IF k[j] - dst[i+31:i] := (a[i+31:i] * b[i+31:i]) + c[i+31:i] + dst[i+31:i] := ATANH(a[i+31:i]) ELSE - dst[i+31:i] := c[i+31:i] + dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - - -
immintrin.h
 - + Floating Point - AVX512VL AVX512F - Arithmetic - - - - - Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", add the intermediate result to packed elements in "c", and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). - - -FOR j := 0 to 3 - i := j*32 + Elementary Math Functions + + + Compute the cube root of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst". + FOR j := 0 to 7 + i := j*64 + dst[i+63:i] := CubeRoot(a[i+63:i]) +ENDFOR +dst[MAX:512] := 0 + +
immintrin.h
+ + + Floating Point + AVX512F + Elementary Math Functions + + + + + Compute the cube root of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + FOR j := 0 to 7 + i := j*64 IF k[j] - dst[i+31:i] := (a[i+31:i] * b[i+31:i]) + c[i+31:i] + dst[i+63:i] := CubeRoot(a[i+63:i]) ELSE - dst[i+31:i] := a[i+31:i] + dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - - -
immintrin.h
 - + Floating Point - AVX512VL AVX512F - Arithmetic - - - - - Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", add the intermediate result to packed elements in "c", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - - -FOR j := 0 to 3 + Elementary Math Functions + + + Compute the cube root of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst". + FOR j := 0 to 15 + i := j*32 + dst[i+31:i] := CubeRoot(a[i+31:i]) +ENDFOR +dst[MAX:512] := 0 + +
immintrin.h
+ + + Floating Point + AVX512F + Elementary Math Functions + + + + + Compute the cube root of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + FOR j := 0 to 15 i := j*32 IF k[j] - dst[i+31:i] := (a[i+31:i] * b[i+31:i]) + c[i+31:i] + dst[i+31:i] := CubeRoot(a[i+31:i]) ELSE - dst[i+31:i] := 0 + dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - - -
immintrin.h
 - + Floating Point - AVX512VL AVX512F - Arithmetic - - - - - Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", alternatively add and subtract packed elements in "c" to/from the intermediate result, and store the results in "dst" using writemask "k" (elements are copied from "c" when the corresponding mask bit is not set). - -FOR j := 0 to 3 + Probability/Statistics + + + Compute the cumulative distribution function of packed double-precision (64-bit) floating-point elements in "a" using the normal distribution, and store the results in "dst". + FOR j := 0 to 7 + i := j*64 + dst[i+63:i] := CDFNormal(a[i+63:i]) +ENDFOR +dst[MAX:512] := 0 + +
immintrin.h
+ + + Floating Point + AVX512F + Probability/Statistics + + + + + Compute the cumulative distribution function of packed double-precision (64-bit) floating-point elements in "a" using the normal distribution, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + FOR j := 0 to 7 i := j*64 IF k[j] - IF (j is even) - dst[i+63:i] := (a[i+63:i] * b[i+63:i]) - c[i+63:i] - ELSE - dst[i+63:i] := (a[i+63:i] * b[i+63:i]) + c[i+63:i] - FI + dst[i+63:i] := CDFNormal(a[i+63:i]) ELSE - dst[i+63:i] := c[i+63:i] + dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - - -
immintrin.h
 - + Floating Point - AVX512VL AVX512F - Arithmetic - - - - - Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", alternatively add and subtract packed elements in "c" to/from the intermediate result, and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). - -FOR j := 0 to 3 - i := j*64 + Probability/Statistics + + + Compute the cumulative distribution function of packed single-precision (32-bit) floating-point elements in "a" using the normal distribution, and store the results in "dst". + FOR j := 0 to 15 + i := j*32 + dst[i+31:i] := CDFNormal(a[i+31:i]) +ENDFOR +dst[MAX:512] := 0 + +
immintrin.h
+ + + Floating Point + AVX512F + Probability/Statistics + + + + + Compute the cumulative distribution function of packed single-precision (32-bit) floating-point elements in "a" using the normal distribution, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + FOR j := 0 to 15 + i := j*32 IF k[j] - IF (j is even) - dst[i+63:i] := (a[i+63:i] * b[i+63:i]) - c[i+63:i] - ELSE - dst[i+63:i] := (a[i+63:i] * b[i+63:i]) + c[i+63:i] - FI + dst[i+31:i] := CDFNormal(a[i+31:i]) ELSE - dst[i+63:i] := a[i+63:i] + dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - - -
immintrin.h
 - + Floating Point - AVX512VL AVX512F - Arithmetic - - - - - Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", alternatively add and subtract packed elements in "c" to/from the intermediate result, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - - -FOR j := 0 to 3 + Probability/Statistics + + + Compute the inverse cumulative distribution function of packed double-precision (64-bit) floating-point elements in "a" using the normal distribution, and store the results in "dst". + FOR j := 0 to 7 + i := j*64 + dst[i+63:i] := InverseCDFNormal(a[i+63:i]) +ENDFOR +dst[MAX:512] := 0 + +
immintrin.h
+ + + Floating Point + AVX512F + Probability/Statistics + + + + + Compute the inverse cumulative distribution function of packed double-precision (64-bit) floating-point elements in "a" using the normal distribution, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + FOR j := 0 to 7 i := j*64 IF k[j] - IF (j is even) - dst[i+63:i] := (a[i+63:i] * b[i+63:i]) - c[i+63:i] - ELSE - dst[i+63:i] := (a[i+63:i] * b[i+63:i]) + c[i+63:i] - FI + dst[i+63:i] := InverseCDFNormal(a[i+63:i]) ELSE - dst[i+63:i] := 0 + dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - - -
immintrin.h
 - + Floating Point - AVX512VL AVX512F - Arithmetic - - - - - Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", alternatively add and subtract packed elements in "c" to/from the intermediate result, and store the results in "dst" using writemask "k" (elements are copied from "c" when the corresponding mask bit is not set). - -FOR j := 0 to 1 - i := j*64 + Probability/Statistics + + + Compute the inverse cumulative distribution function of packed single-precision (32-bit) floating-point elements in "a" using the normal distribution, and store the results in "dst". + FOR j := 0 to 15 + i := j*32 + dst[i+31:i] := InverseCDFNormal(a[i+31:i]) +ENDFOR +dst[MAX:512] := 0 + +
immintrin.h
+ + + Floating Point + AVX512F + Probability/Statistics + + + + + Compute the inverse cumulative distribution function of packed single-precision (32-bit) floating-point elements in "a" using the normal distribution, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + FOR j := 0 to 15 + i := j*32 IF k[j] - IF (j is even) - dst[i+63:i] := (a[i+63:i] * b[i+63:i]) - c[i+63:i] - ELSE - dst[i+63:i] := (a[i+63:i] * b[i+63:i]) + c[i+63:i] - FI + dst[i+31:i] := InverseCDFNormal(a[i+31:i]) ELSE - dst[i+63:i] := c[i+63:i] + dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - - -
immintrin.h
 - + Floating Point - AVX512VL AVX512F - Arithmetic - - - - - Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", alternatively add and subtract packed elements in "c" to/from the intermediate result, and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). + Special Math Functions + + + Round the packed double-precision (64-bit) floating-point elements in "a" up to an integer value, and store the results as packed double-precision floating-point elements in "dst". -FOR j := 0 to 1 +FOR j := 0 to 7 i := j*64 - IF k[j] - IF (j is even) - dst[i+63:i] := (a[i+63:i] * b[i+63:i]) - c[i+63:i] - ELSE - dst[i+63:i] := (a[i+63:i] * b[i+63:i]) + c[i+63:i] - FI - ELSE - dst[i+63:i] := a[i+63:i] - FI + dst[i+63:i] := CEIL(a[i+63:i]) ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - - -
immintrin.h
 - + Floating Point - AVX512VL AVX512F - Arithmetic - - - - - Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", alternatively add and subtract packed elements in "c" to/from the intermediate result, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - + Special Math Functions + + + + + Round the packed double-precision (64-bit) floating-point elements in "a" up to an integer value, and store the results as packed double-precision floating-point elements in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 1 +FOR j := 0 to 7 i := j*64 IF k[j] - IF (j is even) - dst[i+63:i] := (a[i+63:i] * b[i+63:i]) - c[i+63:i] - ELSE - dst[i+63:i] := (a[i+63:i] * b[i+63:i]) + c[i+63:i] - FI + dst[i+63:i] := CEIL(a[i+63:i]) ELSE - dst[i+63:i] := 0 + dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - - -
immintrin.h
 - + Floating Point - AVX512VL AVX512F - Arithmetic - - - - - Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", alternatively add and subtract packed elements in "c" to/from the intermediate result, and store the results in "dst" using writemask "k" (elements are copied from "c" when the corresponding mask bit is not set). + Special Math Functions + + + Round the packed single-precision (32-bit) floating-point elements in "a" up to an integer value, and store the results as packed single-precision floating-point elements in "dst". -FOR j := 0 to 7 +FOR j := 0 to 15 + i := j*32 + dst[i+31:i] := CEIL(a[i+31:i]) +ENDFOR +dst[MAX:512] := 0 + +
immintrin.h
+ + + Floating Point + AVX512F + Special Math Functions + + + + + Round the packed single-precision (32-bit) floating-point elements in "a" up to an integer value, and store the results as packed single-precision floating-point elements in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +FOR j := 0 to 15 i := j*32 IF k[j] - IF (j is even) - dst[i+31:i] := (a[i+31:i] * b[i+31:i]) - c[i+31:i] - ELSE - dst[i+31:i] := (a[i+31:i] * b[i+31:i]) + c[i+31:i] - FI + dst[i+31:i] := CEIL(a[i+31:i]) ELSE - dst[i+31:i] := c[i+31:i] + dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - - -
immintrin.h
 - + Floating Point - AVX512VL AVX512F - Arithmetic - - - - - Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", alternatively add and subtract packed elements in "c" to/from the intermediate result, and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). + Trigonometry + + + Compute the cosine of packed double-precision (64-bit) floating-point elements in "a" expressed in radians, and store the results in "dst". FOR j := 0 to 7 - i := j*32 + i := j*64 + dst[i+63:i] := COS(a[i+63:i]) +ENDFOR +dst[MAX:512] := 0 + +
immintrin.h
+ + + Floating Point + AVX512F + Trigonometry + + + + + Compute the cosine of packed double-precision (64-bit) floating-point elements in "a" expressed in radians, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +FOR j := 0 to 7 + i := j*64 IF k[j] - IF (j is even) - dst[i+31:i] := (a[i+31:i] * b[i+31:i]) - c[i+31:i] - ELSE - dst[i+31:i] := (a[i+31:i] * b[i+31:i]) + c[i+31:i] - FI + dst[i+63:i] := COS(a[i+63:i]) ELSE - dst[i+31:i] := a[i+31:i] + dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - - -
immintrin.h
 - + Floating Point - AVX512VL AVX512F - Arithmetic - - - - - Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", alternatively add and subtract packed elements in "c" to/from the intermediate result, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - + Trigonometry + + + Compute the cosine of packed single-precision (32-bit) floating-point elements in "a" expressed in radians, and store the results in "dst". -FOR j := 0 to 7 +FOR j := 0 to 15 + i := j*32 + dst[i+31:i] := COS(a[i+31:i]) +ENDFOR +dst[MAX:512] := 0 + +
immintrin.h
+ + + Floating Point + AVX512F + Trigonometry + + + + + Compute the cosine of packed single-precision (32-bit) floating-point elements in "a" expressed in radians, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +FOR j := 0 to 15 i := j*32 IF k[j] - IF (j is even) - dst[i+31:i] := (a[i+31:i] * b[i+31:i]) - c[i+31:i] - ELSE - dst[i+31:i] := (a[i+31:i] * b[i+31:i]) + c[i+31:i] - FI + dst[i+31:i] := COS(a[i+31:i]) ELSE - dst[i+31:i] := 0 + dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 + +
immintrin.h
+ + + Floating Point + AVX512F + Trigonometry + + + Compute the cosine of packed double-precision (64-bit) floating-point elements in "a" expressed in degrees, and store the results in "dst". + FOR j := 0 to 7 + i := j*64 + dst[i+63:i] := COSD(a[i+63:i]) +ENDFOR +dst[MAX:512] := 0 - - -
immintrin.h
 - + Floating Point - AVX512VL AVX512F - Arithmetic - - - - - Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", alternatively add and subtract packed elements in "c" to/from the intermediate result, and store the results in "dst" using writemask "k" (elements are copied from "c" when the corresponding mask bit is not set). - -FOR j := 0 to 3 - i := j*32 + Trigonometry + + + + + Compute the cosine of packed double-precision (64-bit) floating-point elements in "a" expressed in degrees, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + FOR j := 0 to 7 + i := j*64 IF k[j] - IF (j is even) - dst[i+31:i] := (a[i+31:i] * b[i+31:i]) - c[i+31:i] - ELSE - dst[i+31:i] := (a[i+31:i] * b[i+31:i]) + c[i+31:i] - FI + dst[i+63:i] := COSD(a[i+63:i]) ELSE - dst[i+31:i] := c[i+31:i] + dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - - -
immintrin.h
 - + Floating Point - AVX512VL AVX512F - Arithmetic - - - - - Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", alternatively add and subtract packed elements in "c" to/from the intermediate result, and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). - -FOR j := 0 to 3 + Trigonometry + + + Compute the cosine of packed single-precision (32-bit) floating-point elements in "a" expressed in degrees, and store the results in "dst". + FOR j := 0 to 15 i := j*32 - IF k[j] - IF (j is even) - dst[i+31:i] := (a[i+31:i] * b[i+31:i]) - c[i+31:i] - ELSE - dst[i+31:i] := (a[i+31:i] * b[i+31:i]) + c[i+31:i] - FI - ELSE - dst[i+31:i] := a[i+31:i] - FI + dst[i+31:i] := COSD(a[i+31:i]) ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - - -
immintrin.h
 - + Floating Point - AVX512VL AVX512F - Arithmetic - - - - - Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", alternatively add and subtract packed elements in "c" to/from the intermediate result, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - - -FOR j := 0 to 3 + Trigonometry + + + + + Compute the cosine of packed single-precision (32-bit) floating-point elements in "a" expressed in degrees, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + FOR j := 0 to 15 i := j*32 IF k[j] - IF (j is even) - dst[i+31:i] := (a[i+31:i] * b[i+31:i]) - c[i+31:i] - ELSE - dst[i+31:i] := (a[i+31:i] * b[i+31:i]) + c[i+31:i] - FI + dst[i+31:i] := COSD(a[i+31:i]) ELSE - dst[i+31:i] := 0 + dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - - -
immintrin.h
 - + Floating Point - AVX512VL AVX512F - Arithmetic - - - - - Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", subtract packed elements in "c" from the intermediate result, and store the results in "dst" using writemask "k" (elements are copied from "c" when the corresponding mask bit is not set). + Trigonometry + + + Compute the hyperbolic cosine of packed double-precision (64-bit) floating-point elements in "a" expressed in radians, and store the results in "dst". -FOR j := 0 to 3 +FOR j := 0 to 7 i := j*64 - IF k[j] - dst[i+63:i] := (a[i+63:i] * b[i+63:i]) - c[i+63:i] - ELSE - dst[i+63:i] := c[i+63:i] - FI + dst[i+63:i] := COSH(a[i+63:i]) ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - - -
immintrin.h
 - + Floating Point - AVX512VL AVX512F - Arithmetic - - - - - Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", subtract packed elements in "c" from the intermediate result, and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). + Trigonometry + + + + + Compute the hyperbolic cosine of packed double-precision (64-bit) floating-point elements in "a" expressed in radians, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 3 +FOR j := 0 to 7 i := j*64 IF k[j] - dst[i+63:i] := (a[i+63:i] * b[i+63:i]) - c[i+63:i] + dst[i+63:i] := COSH(a[i+63:i]) ELSE - dst[i+63:i] := a[i+63:i] + dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - - -
immintrin.h
 - + Floating Point - AVX512VL AVX512F - Arithmetic - - - - - Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", subtract packed elements in "c" from the intermediate result, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - + Trigonometry + + + Compute the hyperbolic cosine of packed single-precision (32-bit) floating-point elements in "a" expressed in radians, and store the results in "dst". -FOR j := 0 to 3 - i := j*64 - IF k[j] - dst[i+63:i] := (a[i+63:i] * b[i+63:i]) - c[i+63:i] - ELSE - dst[i+63:i] := 0 - FI +FOR j := 0 to 15 + i := j*32 + dst[i+31:i] := COSH(a[i+31:i]) ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - - -
immintrin.h
 - + Floating Point - AVX512VL AVX512F - Arithmetic - - - - - Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", subtract packed elements in "c" from the intermediate result, and store the results in "dst" using writemask "k" (elements are copied from "c" when the corresponding mask bit is not set). + Trigonometry + + + + + Compute the hyperbolic cosine of packed single-precision (32-bit) floating-point elements in "a" expressed in radians, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 1 - i := j*64 +FOR j := 0 to 15 + i := j*32 IF k[j] - dst[i+63:i] := (a[i+63:i] * b[i+63:i]) - c[i+63:i] + dst[i+31:i] := COSH(a[i+31:i]) ELSE - dst[i+63:i] := c[i+63:i] + dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - - -
immintrin.h
 - + Floating Point - AVX512VL AVX512F - Arithmetic - - - - - Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", subtract packed elements in "c" from the intermediate result, and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). - -FOR j := 0 to 1 + Probability/Statistics + + + Compute the error function of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst". + FOR j := 0 to 7 i := j*64 - IF k[j] - dst[i+63:i] := (a[i+63:i] * b[i+63:i]) - c[i+63:i] - ELSE - dst[i+63:i] := a[i+63:i] - FI + dst[i+63:i] := ERF(a[i+63:i]) ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - - -
immintrin.h
 - + Floating Point - AVX512VL AVX512F - Arithmetic - - - - - Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", subtract packed elements in "c" from the intermediate result, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - - -FOR j := 0 to 1 + Probability/Statistics + + + + + Compute the error function of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + FOR j := 0 to 7 i := j*64 IF k[j] - dst[i+63:i] := (a[i+63:i] * b[i+63:i]) - c[i+63:i] + dst[i+63:i] := ERF(a[i+63:i]) ELSE - dst[i+63:i] := 0 + dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - - -
immintrin.h
 - + Floating Point - AVX512VL AVX512F - Arithmetic - - - - - Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", subtract packed elements in "c" from the intermediate result, and store the results in "dst" using writemask "k" (elements are copied from "c" when the corresponding mask bit is not set). - -FOR j := 0 to 7 - i := j*32 - IF k[j] - dst[i+31:i] := (a[i+31:i] * b[i+31:i]) - c[i+31:i] - ELSE - dst[i+31:i] := c[i+31:i] - FI + Probability/Statistics + + + Compute the complementary error function of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst". + FOR j := 0 to 7 + i := j*64 + dst[i+63:i] := 1.0 - ERF(a[i+63:i]) ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - - -
immintrin.h
 - + Floating Point - AVX512VL AVX512F - Arithmetic - - - - - Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", subtract packed elements in "c" from the intermediate result, and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). - -FOR j := 0 to 7 - i := j*32 + Probability/Statistics + + + + + Compute the complementary error function of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + FOR j := 0 to 7 + i := j*64 IF k[j] - dst[i+31:i] := (a[i+31:i] * b[i+31:i]) - c[i+31:i] + dst[i+63:i] := 1.0 - ERF(a[i+63:i]) ELSE - dst[i+31:i] := a[i+31:i] + dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - - -
immintrin.h
 - + Floating Point - AVX512VL AVX512F - Arithmetic - - - - - Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", subtract packed elements in "c" from the intermediate result, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - - -FOR j := 0 to 7 + Probability/Statistics + + + Compute the error function of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst". + FOR j := 0 to 15 i := j*32 - IF k[j] - dst[i+31:i] := (a[i+31:i] * b[i+31:i]) - c[i+31:i] - ELSE - dst[i+31:i] := 0 - FI + dst[i+31:i] := ERF(a[i+31:i]) ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - - -
immintrin.h
 - + Floating Point - AVX512VL AVX512F - Arithmetic - - - - - Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", subtract packed elements in "c" from the intermediate result, and store the results in "dst" using writemask "k" (elements are copied from "c" when the corresponding mask bit is not set). - -FOR j := 0 to 3 + Probability/Statistics + + + + + Compute the error function of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + FOR j := 0 to 15 i := j*32 IF k[j] - dst[i+31:i] := (a[i+31:i] * b[i+31:i]) - c[i+31:i] + dst[i+31:i] := ERF(a[i+31:i]) ELSE - dst[i+31:i] := c[i+31:i] + dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - - -
immintrin.h
 - + Floating Point - AVX512VL AVX512F - Arithmetic - - - - - Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", subtract packed elements in "c" from the intermediate result, and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). - -FOR j := 0 to 3 + Probability/Statistics + + + Compute the complementary error function of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst". + FOR j := 0 to 15 i := j*32 - IF k[j] - dst[i+31:i] := (a[i+31:i] * b[i+31:i]) - c[i+31:i] - ELSE - dst[i+31:i] := a[i+31:i] - FI + dst[i+63:i] := 1.0 - ERF(a[i+31:i]) ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - - -
immintrin.h
 - + Floating Point - AVX512VL AVX512F - Arithmetic - - - - - Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", subtract packed elements in "c" from the intermediate result, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - - -FOR j := 0 to 3 + Probability/Statistics + + + + + Compute the complementary error function of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + FOR j := 0 to 15 i := j*32 IF k[j] - dst[i+31:i] := (a[i+31:i] * b[i+31:i]) - c[i+31:i] + dst[i+63:i] := 1.0 - ERF(a[i+31:i]) ELSE - dst[i+31:i] := 0 + dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - - -
immintrin.h
 - + Floating Point - AVX512VL AVX512F - Arithmetic - - - - - Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", alternatively subtract and add packed elements in "c" from/to the intermediate result, and store the results in "dst" using writemask "k" (elements are copied from "c" when the corresponding mask bit is not set). - -FOR j := 0 to 3 + Probability/Statistics + + + Compute the inverse error function of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst". + FOR j := 0 to 7 i := j*64 - IF k[j] - IF (j is even) - dst[i+63:i] := (a[i+63:i] * b[i+63:i]) + c[i+63:i] - ELSE - dst[i+63:i] := (a[i+63:i] * b[i+63:i]) - c[i+63:i] - FI - ELSE - dst[i+63:i] := c[i+63:i] - FI + dst[i+63:i] := 1.0 / ERF(a[i+63:i]) ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - - -
immintrin.h
 - + Floating Point - AVX512VL AVX512F - Arithmetic - - - - - Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", alternatively subtract and add packed elements in "c" from/to the intermediate result, and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). - -FOR j := 0 to 3 + Probability/Statistics + + + + + Compute the inverse error function of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + FOR j := 0 to 7 i := j*64 IF k[j] - IF (j is even) - dst[i+63:i] := (a[i+63:i] * b[i+63:i]) + c[i+63:i] - ELSE - dst[i+63:i] := (a[i+63:i] * b[i+63:i]) - c[i+63:i] - FI + dst[i+63:i] := 1.0 / ERF(a[i+63:i]) ELSE - dst[i+63:i] := a[i+63:i] + dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - - -
immintrin.h
 - + Floating Point - AVX512VL AVX512F - Arithmetic - - - - - Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", alternatively subtract and add packed elements in "c" from/to the intermediate result, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - - -FOR j := 0 to 3 - i := j*64 - IF k[j] - IF (j is even) - dst[i+63:i] := (a[i+63:i] * b[i+63:i]) + c[i+63:i] - ELSE - dst[i+63:i] := (a[i+63:i] * b[i+63:i]) - c[i+63:i] - FI - ELSE - dst[i+63:i] := 0 - FI + Probability/Statistics + + + Compute the inverse error function of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst". + FOR j := 0 to 15 + i := j*32 + dst[i+63:i] := 1.0 / ERF(a[i+31:i]) ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - - -
immintrin.h
 - + Floating Point - AVX512VL AVX512F - Arithmetic - - - - - Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", alternatively subtract and add packed elements in "c" from/to the intermediate result, and store the results in "dst" using writemask "k" (elements are copied from "c" when the corresponding mask bit is not set). - -FOR j := 0 to 1 - i := j*64 + Probability/Statistics + + + + + Compute the inverse error function of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + FOR j := 0 to 15 + i := j*32 IF k[j] - IF (j is even) - dst[i+63:i] := (a[i+63:i] * b[i+63:i]) + c[i+63:i] - ELSE - dst[i+63:i] := (a[i+63:i] * b[i+63:i]) - c[i+63:i] - FI + dst[i+63:i] := 1.0 / ERF(a[i+31:i]) ELSE - dst[i+63:i] := c[i+63:i] + dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - - -
immintrin.h
 - + Floating Point - AVX512VL AVX512F - Arithmetic - - - - - Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", alternatively subtract and add packed elements in "c" from/to the intermediate result, and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). - -FOR j := 0 to 1 + Probability/Statistics + + + Compute the inverse complementary error function of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst". + FOR j := 0 to 7 i := j*64 - IF k[j] - IF (j is even) - dst[i+63:i] := (a[i+63:i] * b[i+63:i]) + c[i+63:i] - ELSE - dst[i+63:i] := (a[i+63:i] * b[i+63:i]) - c[i+63:i] - FI - ELSE - dst[i+63:i] := a[i+63:i] - FI + dst[i+63:i] := 1.0 / (1.0 - ERF(a[i+63:i])) ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - - -
immintrin.h
 - + Floating Point - AVX512VL AVX512F - Arithmetic - - - - - Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", alternatively subtract and add packed elements in "c" from/to the intermediate result, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - - -FOR j := 0 to 1 + Probability/Statistics + + + + + Compute the inverse complementary error function of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + FOR j := 0 to 7 i := j*64 IF k[j] - IF (j is even) - dst[i+63:i] := (a[i+63:i] * b[i+63:i]) + c[i+63:i] - ELSE - dst[i+63:i] := (a[i+63:i] * b[i+63:i]) - c[i+63:i] - FI + dst[i+63:i] := 1.0 / (1.0 - ERF(a[i+63:i])) ELSE - dst[i+63:i] := 0 + dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - - -
immintrin.h
 - + Floating Point - AVX512VL AVX512F - Arithmetic - - - - - Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", alternatively subtract and add packed elements in "c" from/to the intermediate result, and store the results in "dst" using writemask "k" (elements are copied from "c" when the corresponding mask bit is not set). - -FOR j := 0 to 7 + Probability/Statistics + + + Compute the inverse complementary error function of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst". + FOR j := 0 to 15 i := j*32 - IF k[j] - IF (j is even) - dst[i+31:i] := (a[i+31:i] * b[i+31:i]) + c[i+31:i] - ELSE - dst[i+31:i] := (a[i+31:i] * b[i+31:i]) - c[i+31:i] - FI - ELSE - dst[i+31:i] := c[i+31:i] - FI + dst[i+63:i] := 1.0 / (1.0 - ERF(a[i+31:i])) ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - - -
immintrin.h
 - + Floating Point - AVX512VL AVX512F - Arithmetic - - - - - Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", alternatively subtract and add packed elements in "c" from/to the intermediate result, and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). - -FOR j := 0 to 7 + Probability/Statistics + + + + + Compute the inverse complementary error function of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + FOR j := 0 to 15 i := j*32 IF k[j] - IF (j is even) - dst[i+31:i] := (a[i+31:i] * b[i+31:i]) + c[i+31:i] - ELSE - dst[i+31:i] := (a[i+31:i] * b[i+31:i]) - c[i+31:i] - FI + dst[i+63:i] := 1.0 / (1.0 - ERF(a[i+31:i])) ELSE - dst[i+31:i] := a[i+31:i] + dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - - -
immintrin.h
 - + Floating Point - AVX512VL AVX512F - Arithmetic - - - - - Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", alternatively subtract and add packed elements in "c" from/to the intermediate result, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - + Elementary Math Functions + + + Compute the exponential value of 10 raised to the power of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst". FOR j := 0 to 7 - i := j*32 - IF k[j] - IF (j is even) - dst[i+31:i] := (a[i+31:i] * b[i+31:i]) + c[i+31:i] - ELSE - dst[i+31:i] := (a[i+31:i] * b[i+31:i]) - c[i+31:i] - FI - ELSE - dst[i+31:i] := 0 - FI + i := j*64 + dst[i+63:i] := POW(10.0, a[i+63:i]) ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - - -
immintrin.h
 - + Floating Point - AVX512VL AVX512F - Arithmetic - - - - - Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", alternatively subtract and add packed elements in "c" from/to the intermediate result, and store the results in "dst" using writemask "k" (elements are copied from "c" when the corresponding mask bit is not set). + Elementary Math Functions + + + + + Compute the exponential value of 10 raised to the power of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*32 +FOR j := 0 to 7 + i := j*64 IF k[j] - IF (j is even) - dst[i+31:i] := (a[i+31:i] * b[i+31:i]) + c[i+31:i] - ELSE - dst[i+31:i] := (a[i+31:i] * b[i+31:i]) - c[i+31:i] - FI + dst[i+63:i] := POW(10.0, a[i+63:i]) ELSE - dst[i+31:i] := c[i+31:i] + dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - - -
immintrin.h
 - + Floating Point - AVX512VL AVX512F - Arithmetic - - - - - Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", alternatively subtract and add packed elements in "c" from/to the intermediate result, and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). + Elementary Math Functions + + + Compute the exponential value of 10 raised to the power of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst". -FOR j := 0 to 3 +FOR j := 0 to 15 i := j*32 - IF k[j] - IF (j is even) - dst[i+31:i] := (a[i+31:i] * b[i+31:i]) + c[i+31:i] - ELSE - dst[i+31:i] := (a[i+31:i] * b[i+31:i]) - c[i+31:i] - FI - ELSE - dst[i+31:i] := a[i+31:i] - FI + dst[i+31:i] := POW(FP32(10.0), a[i+31:i]) ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - - -
immintrin.h
 - + Floating Point - AVX512VL AVX512F - Arithmetic - - - - - Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", alternatively subtract and add packed elements in "c" from/to the intermediate result, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - + Elementary Math Functions + + + + + Compute the exponential value of 10 raised to the power of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 3 +FOR j := 0 to 15 i := j*32 IF k[j] - IF (j is even) - dst[i+31:i] := (a[i+31:i] * b[i+31:i]) + c[i+31:i] - ELSE - dst[i+31:i] := (a[i+31:i] * b[i+31:i]) - c[i+31:i] - FI + dst[i+31:i] := POW(FP32(10.0), a[i+31:i]) ELSE - dst[i+31:i] := 0 + dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - - -
immintrin.h
 - + Floating Point - AVX512VL AVX512F - Arithmetic - - - - - Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", add the negated intermediate result to packed elements in "c", and store the results in "dst" using writemask "k" (elements are copied from "c" when the corresponding mask bit is not set). + Elementary Math Functions + + + Compute the exponential value of 2 raised to the power of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst". -FOR j := 0 to 3 +FOR j := 0 to 7 i := j*64 - IF k[j] - dst[i+63:i] := -(a[i+63:i] * b[i+63:i]) + c[i+63:i] - ELSE - dst[i+63:i] := c[i+63:i] - FI -ENDFOR -dst[MAX:256] := 0 + dst[i+63:i] := POW(2.0, a[i+63:i]) +ENDFOR +dst[MAX:512] := 0 - - -
immintrin.h
 - + Floating Point - AVX512VL AVX512F - Arithmetic - - - - - Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", add the negated intermediate result to packed elements in "c", and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). + Elementary Math Functions + + + + + Compute the exponential value of 2 raised to the power of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 3 +FOR j := 0 to 7 i := j*64 IF k[j] - dst[i+63:i] := -(a[i+63:i] * b[i+63:i]) + c[i+63:i] + dst[i+63:i] := POW(2.0, a[i+63:i]) ELSE - dst[i+63:i] := a[i+63:i] + dst[i+63:i] := src[i+63:i] FI -ENDFOR -dst[MAX:256] := 0 +ENDFOR +dst[MAX:512] := 0 - - -
immintrin.h
 - + Floating Point - AVX512VL AVX512F - Arithmetic - - - - - Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", add the negated intermediate result to packed elements in "c", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Elementary Math Functions + + + Compute the exponential value of 2 raised to the power of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst". -FOR j := 0 to 3 - i := j*64 - IF k[j] - dst[i+63:i] := -(a[i+63:i] * b[i+63:i]) + c[i+63:i] - ELSE - dst[i+63:i] := 0 - FI -ENDFOR -dst[MAX:256] := 0 +FOR j := 0 to 15 + i := j*32 + dst[i+31:i] := POW(FP32(2.0), a[i+31:i]) +ENDFOR +dst[MAX:512] := 0 - - -
immintrin.h
 - + Floating Point - AVX512VL AVX512F - Arithmetic - - - - - Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", add the negated intermediate result to packed elements in "c", and store the results in "dst" using writemask "k" (elements are copied from "c" when the corresponding mask bit is not set). + Elementary Math Functions + + + + + Compute the exponential value of 2 raised to the power of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 1 - i := j*64 +FOR j := 0 to 15 + i := j*32 IF k[j] - dst[i+63:i] := -(a[i+63:i] * b[i+63:i]) + c[i+63:i] + dst[i+31:i] := POW(FP32(2.0), a[i+31:i]) ELSE - dst[i+63:i] := c[i+63:i] + dst[i+31:i] := src[i+31:i] FI -ENDFOR -dst[MAX:128] := 0 +ENDFOR +dst[MAX:512] := 0 - - -
immintrin.h
 - + Floating Point - AVX512VL AVX512F - Arithmetic - - - - - Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", add the negated intermediate result to packed elements in "c", and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). + Elementary Math Functions + + + Compute the exponential value of "e" raised to the power of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst". -FOR j := 0 to 1 +FOR j := 0 to 7 i := j*64 - IF k[j] - dst[i+63:i] := -(a[i+63:i] * b[i+63:i]) + c[i+63:i] - ELSE - dst[i+63:i] := a[i+63:i] - FI -ENDFOR -dst[MAX:128] := 0 + dst[i+63:i] := POW(e, a[i+63:i]) +ENDFOR +dst[MAX:512] := 0 - - -
immintrin.h
 - + Floating Point - AVX512VL AVX512F - Arithmetic - - - - - Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", add the negated intermediate result to packed elements in "c", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Elementary Math Functions + + + + + Compute the exponential value of "e" raised to the power of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 1 +FOR j := 0 to 7 i := j*64 IF k[j] - dst[i+63:i] := -(a[i+63:i] * b[i+63:i]) + c[i+63:i] + dst[i+63:i] := POW(e, a[i+63:i]) ELSE - dst[i+63:i] := 0 + dst[i+63:i] := src[i+63:i] FI -ENDFOR -dst[MAX:128] := 0 +ENDFOR +dst[MAX:512] := 0 - - -
immintrin.h
 - + Floating Point - AVX512VL AVX512F - Arithmetic - - - - - Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", add the negated intermediate result to packed elements in "c", and store the results in "dst" using writemask "k" (elements are copied from "c" when the corresponding mask bit is not set). + Elementary Math Functions + + + Compute the exponential value of "e" raised to the power of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst". -FOR j := 0 to 7 - i := j*32 - IF k[j] - dst[i+31:i] := -(a[i+31:i] * b[i+31:i]) + c[i+31:i] - ELSE - dst[i+31:i] := c[i+31:i] - FI -ENDFOR -dst[MAX:256] := 0 +FOR j := 0 to 15 + i := j*32 + dst[i+31:i] := POW(FP32(e), a[i+31:i]) +ENDFOR +dst[MAX:512] := 0 - - -
immintrin.h
 - + Floating Point - AVX512VL AVX512F - Arithmetic - - - - - Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", add the negated intermediate result to packed elements in "c", and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). + Elementary Math Functions + + + + + Compute the exponential value of "e" raised to the power of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 7 +FOR j := 0 to 15 i := j*32 IF k[j] - dst[i+31:i] := -(a[i+31:i] * b[i+31:i]) + c[i+31:i] + dst[i+31:i] := POW(FP32(e), a[i+31:i]) ELSE - dst[i+31:i] := a[i+31:i] + dst[i+31:i] := src[i+31:i] FI -ENDFOR -dst[MAX:256] := 0 +ENDFOR +dst[MAX:512] := 0 - - -
immintrin.h
 - + Floating Point - AVX512VL AVX512F - Arithmetic - - - - - Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", add the negated intermediate result to packed elements in "c", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Elementary Math Functions + + + Compute the exponential value of "e" raised to the power of packed double-precision (64-bit) floating-point elements in "a", subtract one from each element, and store the results in "dst". FOR j := 0 to 7 - i := j*32 - IF k[j] - dst[i+31:i] := -(a[i+31:i] * b[i+31:i]) + c[i+31:i] - ELSE - dst[i+31:i] := 0 - FI -ENDFOR -dst[MAX:256] := 0 + i := j*64 + dst[i+63:i] := POW(e, a[i+63:i]) - 1.0 +ENDFOR +dst[MAX:512] := 0 - - -
immintrin.h
 - + Floating Point - AVX512VL AVX512F - Arithmetic - - - - - Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", add the negated intermediate result to packed elements in "c", and store the results in "dst" using writemask "k" (elements are copied from "c" when the corresponding mask bit is not set). + Elementary Math Functions + + + + + Compute the exponential value of "e" raised to the power of packed double-precision (64-bit) floating-point elements in "a", subtract one from each element, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*32 +FOR j := 0 to 7 + i := j*64 IF k[j] - dst[i+31:i] := -(a[i+31:i] * b[i+31:i]) + c[i+31:i] + dst[i+63:i] := POW(e, a[i+63:i]) - 1.0 ELSE - dst[i+31:i] := c[i+31:i] + dst[i+63:i] := src[i+63:i] FI -ENDFOR -dst[MAX:128] := 0 +ENDFOR +dst[MAX:512] := 0 - - -
immintrin.h
 - + Floating Point - AVX512VL AVX512F - Arithmetic - - - - - Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", add the negated intermediate result to packed elements in "c", and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). + Elementary Math Functions + + + Compute the exponential value of "e" raised to the power of packed single-precision (32-bit) floating-point elements in "a", subtract one from each element, and store the results in "dst". -FOR j := 0 to 3 +FOR j := 0 to 15 i := j*32 - IF k[j] - dst[i+31:i] := -(a[i+31:i] * b[i+31:i]) + c[i+31:i] - ELSE - dst[i+31:i] := a[i+31:i] - FI -ENDFOR -dst[MAX:128] := 0 + dst[i+31:i] := POW(FP32(e), a[i+31:i]) - 1.0 +ENDFOR +dst[MAX:512] := 0 - - -
immintrin.h
 - + Floating Point - AVX512VL AVX512F - Arithmetic - - - - - Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", add the negated intermediate result to packed elements in "c", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Elementary Math Functions + + + + + Compute the exponential value of "e" raised to the power of packed single-precision (32-bit) floating-point elements in "a", subtract one from each element, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 3 +FOR j := 0 to 15 i := j*32 IF k[j] - dst[i+31:i] := -(a[i+31:i] * b[i+31:i]) + c[i+31:i] + dst[i+31:i] := POW(FP32(e), a[i+31:i]) - 1.0 ELSE - dst[i+31:i] := 0 + dst[i+31:i] := src[i+31:i] FI -ENDFOR -dst[MAX:128] := 0 +ENDFOR +dst[MAX:512] := 0 - - -
immintrin.h
 - + Floating Point - AVX512VL AVX512F - Arithmetic - - - - - Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", subtract packed elements in "c" from the negated intermediate result, and store the results in "dst" using writemask "k" (elements are copied from "c" when the corresponding mask bit is not set). + Special Math Functions + + + Round the packed double-precision (64-bit) floating-point elements in "a" down to an integer value, and store the results as packed double-precision floating-point elements in "dst". -FOR j := 0 to 3 +FOR j := 0 to 7 i := j*64 - IF k[j] - dst[i+63:i] := -(a[i+63:i] * b[i+63:i]) - c[i+63:i] - ELSE - dst[i+63:i] := c[i+63:i] - FI -ENDFOR -dst[MAX:256] := 0 + dst[i+63:i] := FLOOR(a[i+63:i]) +ENDFOR +dst[MAX:512] := 0 - - -
immintrin.h
 - + Floating Point - AVX512VL AVX512F - Arithmetic - - - - - Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", subtract packed elements in "c" from the negated intermediate result, and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). + Special Math Functions + + + + + Round the packed double-precision (64-bit) floating-point elements in "a" down to an integer value, and store the results as packed double-precision floating-point elements in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 3 +FOR j := 0 to 7 i := j*64 IF k[j] - dst[i+63:i] := -(a[i+63:i] * b[i+63:i]) - c[i+63:i] + dst[i+63:i] := FLOOR(a[i+63:i]) ELSE - dst[i+63:i] := a[i+63:i] + dst[i+63:i] := src[i+63:i] FI -ENDFOR -dst[MAX:256] := 0 +ENDFOR +dst[MAX:512] := 0 - - -
immintrin.h
 - + Floating Point - AVX512VL AVX512F - Arithmetic - - - - - Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", subtract packed elements in "c" from the negated intermediate result, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Special Math Functions + + + Round the packed single-precision (32-bit) floating-point elements in "a" down to an integer value, and store the results as packed single-precision floating-point elements in "dst". -FOR j := 0 to 3 - i := j*64 - IF k[j] - dst[i+63:i] := -(a[i+63:i] * b[i+63:i]) - c[i+63:i] - ELSE - dst[i+63:i] := 0 - FI -ENDFOR -dst[MAX:256] := 0 +FOR j := 0 to 15 + i := j*32 + dst[i+31:i] := FLOOR(a[i+31:i]) +ENDFOR +dst[MAX:512] := 0 - - -
immintrin.h
 - + Floating Point - AVX512VL AVX512F - Arithmetic - - - - - Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", subtract packed elements in "c" from the negated intermediate result, and store the results in "dst" using writemask "k" (elements are copied from "c" when the corresponding mask bit is not set). + Special Math Functions + + + + + Round the packed single-precision (32-bit) floating-point elements in "a" down to an integer value, and store the results as packed single-precision floating-point elements in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 1 - i := j*64 +FOR j := 0 to 15 + i := j*32 IF k[j] - dst[i+63:i] := -(a[i+63:i] * b[i+63:i]) - c[i+63:i] + dst[i+31:i] := FLOOR(a[i+31:i]) ELSE - dst[i+63:i] := c[i+63:i] + dst[i+31:i] := src[i+31:i] FI -ENDFOR -dst[MAX:128] := 0 +ENDFOR +dst[MAX:512] := 0 - - -
immintrin.h
 - + Floating Point - AVX512VL AVX512F - Arithmetic - - - - - Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", subtract packed elements in "c" from the negated intermediate result, and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). + Elementary Math Functions + + + + Compute the length of the hypotenous of a right triangle, with the lengths of the other two sides of the triangle stored as packed double-precision (64-bit) floating-point elements in "a" and "b", and store the results in "dst". -FOR j := 0 to 1 +FOR j := 0 to 7 i := j*64 - IF k[j] - dst[i+63:i] := -(a[i+63:i] * b[i+63:i]) - c[i+63:i] - ELSE - dst[i+63:i] := a[i+63:i] - FI -ENDFOR -dst[MAX:128] := 0 + dst[i+63:i] := SQRT(POW(a[i+63:i], 2.0) + POW(b[i+63:i], 2.0)) +ENDFOR +dst[MAX:512] := 0 - - -
immintrin.h
 - + Floating Point - AVX512VL AVX512F - Arithmetic - - - - - Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", subtract packed elements in "c" from the negated intermediate result, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Elementary Math Functions + + + + + + Compute the length of the hypotenous of a right triangle, with the lengths of the other two sides of the triangle stored as packed double-precision (64-bit) floating-point elements in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 1 +FOR j := 0 to 7 i := j*64 IF k[j] - dst[i+63:i] := -(a[i+63:i] * b[i+63:i]) - c[i+63:i] + dst[i+63:i] := SQRT(POW(a[i+63:i], 2.0) + POW(b[i+63:i], 2.0)) ELSE - dst[i+63:i] := 0 + dst[i+63:i] := src[i+63:i] FI -ENDFOR -dst[MAX:128] := 0 +ENDFOR +dst[MAX:512] := 0 - - -
immintrin.h
 - + Floating Point - AVX512VL AVX512F - Arithmetic - - - - - Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", subtract packed elements in "c" from the negated intermediate result, and store the results in "dst" using writemask "k" (elements are copied from "c" when the corresponding mask bit is not set). + Elementary Math Functions + + + + Compute the length of the hypotenous of a right triangle, with the lengths of the other two sides of the triangle stored as packed single-precision (32-bit) floating-point elements in "a" and "b", and store the results in "dst". -FOR j := 0 to 7 +FOR j := 0 to 15 i := j*32 - IF k[j] - dst[i+31:i] := -(a[i+31:i] * b[i+31:i]) - c[i+31:i] - ELSE - dst[i+31:i] := c[i+31:i] - FI -ENDFOR -dst[MAX:256] := 0 + dst[i+31:i] := SQRT(POW(a[i+31:i], 2.0) + POW(b[i+31:i], 2.0)) +ENDFOR +dst[MAX:512] := 0 - - -
immintrin.h
 - + Floating Point - AVX512VL AVX512F - Arithmetic - - - - - Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", subtract packed elements in "c" from the negated intermediate result, and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). + Elementary Math Functions + + + + + + Compute the length of the hypotenous of a right triangle, with the lengths of the other two sides of the triangle stored as packed single-precision (32-bit) floating-point elements in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 7 +FOR j := 0 to 15 i := j*32 IF k[j] - dst[i+31:i] := -(a[i+31:i] * b[i+31:i]) - c[i+31:i] + dst[i+31:i] := SQRT(POW(a[i+31:i], 2.0) + POW(b[i+31:i], 2.0)) ELSE - dst[i+31:i] := a[i+31:i] + dst[i+31:i] := src[i+31:i] FI -ENDFOR -dst[MAX:256] := 0 +ENDFOR +dst[MAX:512] := 0 - - -
immintrin.h
 - - Floating Point - AVX512VL + + Integer AVX512F Arithmetic - - - - - Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", subtract packed elements in "c" from the negated intermediate result, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + + + Divide packed signed 32-bit integers in "a" by packed elements in "b", and store the truncated results in "dst". -FOR j := 0 to 7 - i := j*32 - IF k[j] - dst[i+31:i] := -(a[i+31:i] * b[i+31:i]) - c[i+31:i] - ELSE - dst[i+31:i] := 0 +FOR j := 0 to 15 + i := 32*j + IF b[i+31:i] == 0 + #DE FI -ENDFOR -dst[MAX:256] := 0 + dst[i+31:i] := Truncate32(a[i+31:i] / b[i+31:i]) +ENDFOR +dst[MAX:512] := 0 - - -
immintrin.h
 - - Floating Point - AVX512VL + + Integer AVX512F Arithmetic - - - - - Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", subtract packed elements in "c" from the negated intermediate result, and store the results in "dst" using writemask "k" (elements are copied from "c" when the corresponding mask bit is not set). + + + + + + Divide packed signed 32-bit integers in "a" by packed elements in "b", and store the truncated results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*32 +FOR j := 0 to 15 + i := 32*j IF k[j] - dst[i+31:i] := -(a[i+31:i] * b[i+31:i]) - c[i+31:i] + IF b[i+31:i] == 0 + #DE + FI + dst[i+31:i] := Truncate32(a[i+31:i] / b[i+31:i]) ELSE - dst[i+31:i] := c[i+31:i] + dst[i+31:i] := src[i+31:i] FI -ENDFOR -dst[MAX:128] := 0 +ENDFOR +dst[MAX:512] := 0 - - -
immintrin.h
 - - Floating Point - AVX512VL + + Integer AVX512F Arithmetic - - - - - Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", subtract packed elements in "c" from the negated intermediate result, and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). + + + + Divide packed signed 8-bit integers in "a" by packed elements in "b", and store the truncated results in "dst". -FOR j := 0 to 3 - i := j*32 - IF k[j] - dst[i+31:i] := -(a[i+31:i] * b[i+31:i]) - c[i+31:i] - ELSE - dst[i+31:i] := a[i+31:i] +FOR j := 0 to 63 + i := 8*j + IF b[i+7:i] == 0 + #DE FI -ENDFOR -dst[MAX:128] := 0 + dst[i+7:i] := Truncate8(a[i+7:i] / b[i+7:i]) +ENDFOR +dst[MAX:512] := 0 - - -
immintrin.h
 - - Floating Point - AVX512VL + + Integer AVX512F Arithmetic - - - - - Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", subtract packed elements in "c" from the negated intermediate result, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + + + Divide packed signed 16-bit integers in "a" by packed elements in "b", and store the truncated results in "dst". -FOR j := 0 to 3 - i := j*32 - IF k[j] - dst[i+31:i] := -(a[i+31:i] * b[i+31:i]) - c[i+31:i] - ELSE - dst[i+31:i] := 0 +FOR j := 0 to 31 + i := 16*j + IF b[i+15:i] == 0 + #DE FI -ENDFOR -dst[MAX:128] := 0 - - - - -
immintrin.h
- - - Floating Point - Mask - AVX512VL - AVX512DQ - Miscellaneous - - - Test packed double-precision (64-bit) floating-point elements in "a" for special categories specified by "imm8", and store the results in mask vector "k". - [fpclass_note] - - -FOR j := 0 to 3 - i := j*64 - k[j] := CheckFPClass_FP64(a[i+63:i], imm8[7:0]) + dst[i+15:i] := Truncate16(a[i+15:i] / b[i+15:i]) ENDFOR -k[MAX:4] := 0 +dst[MAX:512] := 0 -
immintrin.h
 - - Floating Point - Mask - AVX512VL - AVX512DQ - Miscellaneous - - - - Test packed double-precision (64-bit) floating-point elements in "a" for special categories specified by "imm8", and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). - [fpclass_note] - + + Integer + AVX512F + Arithmetic + + + + Divide packed signed 64-bit integers in "a" by packed elements in "b", and store the truncated results in "dst". -FOR j := 0 to 3 - i := j*64 - IF k1[j] - k[j] := CheckFPClass_FP64(a[i+63:i], imm8[7:0]) - ELSE - k[j] := 0 +FOR j := 0 to 7 + i := 64*j + IF b[i+63:i] == 0 + #DE FI + dst[i+63:i] := Truncate64(a[i+63:i] / b[i+63:i]) ENDFOR -k[MAX:4] := 0 +dst[MAX:512] := 0 -
immintrin.h
 - + Floating Point - Mask - AVX512DQ - Miscellaneous - - - Test packed double-precision (64-bit) floating-point elements in "a" for special categories specified by "imm8", and store the results in mask vector "k". - [fpclass_note] - - -FOR j := 0 to 7 + AVX512F + Elementary Math Functions + + + Compute the inverse square root of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst". + FOR j := 0 to 7 i := j*64 - k[j] := CheckFPClass_FP64(a[i+63:i], imm8[7:0]) + dst[i+63:i] := InvSQRT(a[i+63:i]) ENDFOR -k[MAX:8] := 0 +dst[MAX:512] := 0 -
immintrin.h
 - + Floating Point - Mask - AVX512DQ - Miscellaneous - - - - Test packed double-precision (64-bit) floating-point elements in "a" for special categories specified by "imm8", and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). - [fpclass_note] - - -FOR j := 0 to 7 + AVX512F + Elementary Math Functions + + + + + Compute the inverse square root of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + FOR j := 0 to 7 i := j*64 - IF k1[j] - k[j] := CheckFPClass_FP64(a[i+63:i], imm8[7:0]) + IF k[j] + dst[i+63:i] := InvSQRT(a[i+63:i]) ELSE - k[j] := 0 + dst[i+63:i] := src[i+63:i] FI ENDFOR -k[MAX:8] := 0 +dst[MAX:512] := 0 -
immintrin.h
 - + Floating Point - Mask - AVX512VL - AVX512DQ - Miscellaneous - - - Test packed double-precision (64-bit) floating-point elements in "a" for special categories specified by "imm8", and store the results in mask vector "k". - [fpclass_note] - - -FOR j := 0 to 1 - i := j*64 - k[j] := CheckFPClass_FP64(a[i+63:i], imm8[7:0]) + AVX512F + Elementary Math Functions + + + Compute the inverse square root of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst". + FOR j := 0 to 15 + i := j*32 + dst[i+31:i] := InvSQRT(a[i+31:i]) ENDFOR -k[MAX:2] := 0 +dst[MAX:512] := 0 -
immintrin.h
 - + Floating Point - Mask - AVX512VL - AVX512DQ - Miscellaneous - - - - Test packed double-precision (64-bit) floating-point elements in "a" for special categories specified by "imm8", and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). - [fpclass_note] - - -FOR j := 0 to 1 - i := j*64 - IF k1[j] - k[j] := CheckFPClass_FP64(a[i+63:i], imm8[7:0]) + AVX512F + Elementary Math Functions + + + + + Compute the inverse square root of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + FOR j := 0 to 15 + i := j*32 + IF k[j] + dst[i+31:i] := InvSQRT(a[i+31:i]) ELSE - k[j] := 0 + dst[i+31:i] := src[i+31:i] FI ENDFOR -k[MAX:2] := 0 +dst[MAX:512] := 0 -
immintrin.h
 - - Floating Point - Mask - AVX512VL - AVX512DQ - Miscellaneous - - - Test packed single-precision (32-bit) floating-point elements in "a" for special categories specified by "imm8", and store the results in mask vector "k". - [fpclass_note] - - -FOR j := 0 to 7 - i := j*32 - k[j] := CheckFPClass_FP32(a[i+31:i], imm8[7:0]) + + Integer + AVX512F + Arithmetic + + + + Divide packed 32-bit integers in "a" by packed elements in "b", and store the remainders as packed 32-bit integers in "dst". + FOR j := 0 to 15 + i := 32*j + dst[i+31:i] := REMAINDER(a[i+31:i] / b[i+31:i]) ENDFOR -k[MAX:8] := 0 +dst[MAX:512] := 0 -
immintrin.h
 - - Floating Point - Mask - AVX512VL - AVX512DQ - Miscellaneous - - - - Test packed single-precision (32-bit) floating-point elements in "a" for special categories specified by "imm8", and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). - [fpclass_note] - - -FOR j := 0 to 7 - i := j*32 - IF k1[j] - k[j] := CheckFPClass_FP32(a[i+31:i], imm8[7:0]) + + Integer + AVX512F + Arithmetic + + + + + + Divide packed 32-bit integers in "a" by packed elements in "b", and store the remainders as packed 32-bit integers in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + FOR j := 0 to 15 + i := 32*j + IF k[j] + dst[i+31:i] := REMAINDER(a[i+31:i] / b[i+31:i]) ELSE - k[j] := 0 + dst[i+31:i] := src[i+31:i] FI ENDFOR -k[MAX:8] := 0 +dst[MAX:512] := 0 -
immintrin.h
 - - Floating Point - Mask - AVX512DQ - Miscellaneous - - - Test packed single-precision (32-bit) floating-point elements in "a" for special categories specified by "imm8", and store the results in mask vector "k". - [fpclass_note] - - -FOR j := 0 to 15 - i := j*32 - k[j] := CheckFPClass_FP32(a[i+31:i], imm8[7:0]) + + Integer + AVX512F + Arithmetic + + + + Divide packed 8-bit integers in "a" by packed elements in "b", and store the remainders as packed 32-bit integers in "dst". + FOR j := 0 to 63 + i := 8*j + dst[i+7:i] := REMAINDER(a[i+7:i] / b[i+7:i]) ENDFOR -k[MAX:16] := 0 +dst[MAX:512] := 0 -
immintrin.h
 - - Floating Point - Mask - AVX512DQ - Miscellaneous - - - - Test packed single-precision (32-bit) floating-point elements in "a" for special categories specified by "imm8", and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). - [fpclass_note] - - -FOR j := 0 to 15 - i := j*32 - IF k1[j] - k[j] := CheckFPClass_FP32(a[i+31:i], imm8[7:0]) - ELSE - k[j] := 0 - FI + + Integer + AVX512F + Arithmetic + + + + Divide packed 16-bit integers in "a" by packed elements in "b", and store the remainders as packed 32-bit integers in "dst". + FOR j := 0 to 31 + i := 16*j + dst[i+15:i] := REMAINDER(a[i+15:i] / b[i+15:i]) ENDFOR -k[MAX:16] := 0 +dst[MAX:512] := 0 -
immintrin.h
 - - Floating Point - Mask - AVX512VL - AVX512DQ - Miscellaneous - - - Test packed single-precision (32-bit) floating-point elements in "a" for special categories specified by "imm8", and store the results in mask vector "k". - [fpclass_note] - - -FOR j := 0 to 3 - i := j*32 - k[j] := CheckFPClass_FP32(a[i+31:i], imm8[7:0]) + + Integer + AVX512F + Arithmetic + + + + Divide packed 64-bit integers in "a" by packed elements in "b", and store the remainders as packed 32-bit integers in "dst". + FOR j := 0 to 7 + i := 64*j + dst[i+63:i] := REMAINDER(a[i+63:i] / b[i+63:i]) ENDFOR -k[MAX:4] := 0 +dst[MAX:512] := 0 -
immintrin.h
 - + Floating Point - Mask - AVX512VL - AVX512DQ - Miscellaneous - - - - Test packed single-precision (32-bit) floating-point elements in "a" for special categories specified by "imm8", and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). - [fpclass_note] - + AVX512F + Elementary Math Functions + + + Compute the base-10 logarithm of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst". -FOR j := 0 to 3 - i := j*32 - IF k1[j] - k[j] := CheckFPClass_FP32(a[i+31:i], imm8[7:0]) - ELSE - k[j] := 0 - FI +FOR j := 0 to 7 + i := j*64 + dst[i+63:i] := LOG(a[i+63:i]) / LOG(10.0) ENDFOR -k[MAX:4] := 0 +dst[MAX:512] := 0 -
immintrin.h
 - + Floating Point - Mask - AVX512DQ - Miscellaneous - - - Test the lower double-precision (64-bit) floating-point element in "a" for special categories specified by "imm8", and store the result in mask vector "k". - [fpclass_note] - + AVX512F + Elementary Math Functions + + + + + Compute the base-10 logarithm of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -k[0] := CheckFPClass_FP64(a[63:0], imm8[7:0]) -k[MAX:1] := 0 +FOR j := 0 to 7 + i := j*64 + IF k[j] + dst[i+63:i] := LOG(a[i+63:i]) / LOG(10.0) + ELSE + dst[i+63:i] := src[i+63:i] + FI +ENDFOR +dst[MAX:512] := 0 -
immintrin.h
 - + Floating Point - Mask - AVX512DQ - Miscellaneous - - - - Test the lower double-precision (64-bit) floating-point element in "a" for special categories specified by "imm8", and store the result in mask vector "k" using zeromask "k1" (the element is zeroed out when mask bit 0 is not set). - [fpclass_note] - + AVX512F + Elementary Math Functions + + + Compute the base-10 logarithm of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst". -IF k1[0] - k[0] := CheckFPClass_FP64(a[63:0], imm8[7:0]) -ELSE - k[0] := 0 -FI -k[MAX:1] := 0 +FOR j := 0 to 15 + i := j*32 + dst[i+31:i] := LOG(a[i+31:i]) / LOG(10.0) +ENDFOR +dst[MAX:512] := 0 -
immintrin.h
 - + Floating Point - Mask - AVX512DQ - Miscellaneous - - - Test the lower single-precision (32-bit) floating-point element in "a" for special categories specified by "imm8", and store the result in mask vector "k. - [fpclass_note] - + AVX512F + Elementary Math Functions + + + + + Compute the base-10 logarithm of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -k[0] := CheckFPClass_FP32(a[31:0], imm8[7:0]) -k[MAX:1] := 0 +FOR j := 0 to 15 + i := j*32 + IF k[j] + dst[i+31:i] := LOG(a[i+31:i]) / LOG(10.0) + ELSE + dst[i+31:i] := src[i+31:i] + FI +ENDFOR +dst[MAX:512] := 0 -
immintrin.h
 - + Floating Point - Mask - AVX512DQ - Miscellaneous - - - - Test the lower single-precision (32-bit) floating-point element in "a" for special categories specified by "imm8", and store the result in mask vector "k" using zeromask "k1" (the element is zeroed out when mask bit 0 is not set). - [fpclass_note] - + AVX512F + Elementary Math Functions + + + Compute the natural logarithm of one plus packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst". -IF k1[0] - k[0] := CheckFPClass_FP32(a[31:0], imm8[7:0]) -ELSE - k[0] := 0 -FI -k[MAX:1] := 0 +FOR j := 0 to 7 + i := j*64 + dst[i+63:i] := LOG(1.0 + a[i+63:i]) +ENDFOR +dst[MAX:512] := 0 -
immintrin.h
 - + Floating Point - AVX512VL AVX512F - Load - - - - - - - Gather double-precision (64-bit) floating-point elements from memory using 32-bit indices. 64-bit elements are loaded from addresses starting at "base_addr" and offset by each 32-bit element in "vindex" (each index is scaled by the factor in "scale"). Gathered elements are merged into "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). "scale" should be 1, 2, 4 or 8. - + Elementary Math Functions + + + + + Compute the natural logarithm of one plus packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 3 +FOR j := 0 to 7 i := j*64 - m := j*32 IF k[j] - dst[i+63:i] := MEM[base_addr + SignExtend(vindex[m+31:m])*scale] - k[j] := 0 + dst[i+63:i] := LOG(1.0 + a[i+63:i]) ELSE dst[i+63:i] := src[i+63:i] FI ENDFOR -k[MAX:4] := 0 -dst[MAX:256] := 0 +dst[MAX:512] := 0 -
immintrin.h
 - + Floating Point - AVX512VL AVX512F - Load - - - - - - - Gather double-precision (64-bit) floating-point elements from memory using 32-bit indices. 64-bit elements are loaded from addresses starting at "base_addr" and offset by each 32-bit element in "vindex" (each index is scaled by the factor in "scale"). Gathered elements are merged into "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). "scale" should be 1, 2, 4 or 8. - + Elementary Math Functions + + + Compute the natural logarithm of one plus packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst". -FOR j := 0 to 1 - i := j*64 - m := j*32 - IF k[j] - dst[i+63:i] := MEM[base_addr + SignExtend(vindex[m+31:m])*scale] - k[j] := 0 - ELSE - dst[i+63:i] := src[i+63:i] - FI +FOR j := 0 to 15 + i := j*32 + dst[i+31:i] := LOG(1.0 + a[i+31:i]) ENDFOR -k[MAX:2] := 0 -dst[MAX:128] := 0 +dst[MAX:512] := 0 -
immintrin.h
 - + Floating Point - AVX512VL AVX512F - Load - - - - - - - Gather single-precision (32-bit) floating-point elements from memory using 32-bit indices. 32-bit elements are loaded from addresses starting at "base_addr" and offset by each 32-bit element in "vindex" (each index is scaled by the factor in "scale"). Gathered elements are merged into "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). "scale" should be 1, 2, 4 or 8. - + Elementary Math Functions + + + + + Compute the natural logarithm of one plus packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 7 +FOR j := 0 to 15 i := j*32 IF k[j] - dst[i+31:i] := MEM[base_addr + SignExtend(vindex[i+31:i])*scale] - k[j] := 0 + dst[i+31:i] := LOG(1.0 + a[i+31:i]) ELSE dst[i+31:i] := src[i+31:i] FI ENDFOR -k[MAX:8] := 0 -dst[MAX:256] := 0 +dst[MAX:512] := 0 -
immintrin.h
 - + Floating Point - AVX512VL AVX512F - Load - - - - - - - Gather single-precision (32-bit) floating-point elements from memory using 32-bit indices. 32-bit elements are loaded from addresses starting at "base_addr" and offset by each 32-bit element in "vindex" (each index is scaled by the factor in "scale"). Gathered elements are merged into "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). "scale" should be 1, 2, 4 or 8. - + Elementary Math Functions + + + Compute the base-2 logarithm of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst". -FOR j := 0 to 3 - i := j*32 - IF k[j] - dst[i+31:i] := MEM[base_addr + SignExtend(vindex[i+31:i])*scale] - k[j] := 0 - ELSE - dst[i+31:i] := src[i+31:i] - FI +FOR j := 0 to 7 + i := j*64 + dst[i+63:i] := LOG(a[i+63:i]) / LOG(2.0) ENDFOR -k[MAX:4] := 0 -dst[MAX:128] := 0 +dst[MAX:512] := 0 -
immintrin.h
 - + Floating Point - AVX512VL AVX512F - Load - - - - - - - Gather double-precision (64-bit) floating-point elements from memory using 64-bit indices. 64-bit elements are loaded from addresses starting at "base_addr" and offset by each 64-bit element in "vindex" (each index is scaled by the factor in "scale"). Gathered elements are merged into "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). "scale" should be 1, 2, 4 or 8. - + Elementary Math Functions + + + + + Compute the base-2 logarithm of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 3 +FOR j := 0 to 7 i := j*64 IF k[j] - dst[i+63:i] := MEM[base_addr + SignExtend(vindex[i+63:i])*scale] - k[j] := 0 + dst[i+63:i] := LOG(a[i+63:i]) / LOG(2.0) ELSE dst[i+63:i] := src[i+63:i] FI ENDFOR -k[MAX:4] := 0 -dst[MAX:256] := 0 +dst[MAX:512] := 0 -
immintrin.h
 - + Floating Point - AVX512VL AVX512F - Load - - - - - - - Gather double-precision (64-bit) floating-point elements from memory using 64-bit indices. 64-bit elements are loaded from addresses starting at "base_addr" and offset by each 64-bit element in "vindex" (each index is scaled by the factor in "scale"). Gathered elements are merged into "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). "scale" should be 1, 2, 4 or 8. - + Elementary Math Functions + + + Compute the natural logarithm of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst". -FOR j := 0 to 1 +FOR j := 0 to 7 + i := j*64 + dst[i+63:i] := LOG(a[i+63:i]) +ENDFOR +dst[MAX:512] := 0 + +
immintrin.h
+ + + Floating Point + AVX512F + Elementary Math Functions + + + + + Compute the natural logarithm of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +FOR j := 0 to 7 i := j*64 IF k[j] - dst[i+63:i] := MEM[base_addr + SignExtend(vindex[i+63:i])*scale] - k[j] := 0 + dst[i+63:i] := LOG(a[i+63:i]) ELSE dst[i+63:i] := src[i+63:i] FI ENDFOR -k[MAX:2] := 0 -dst[MAX:128] := 0 +dst[MAX:512] := 0 -
immintrin.h
 - + Floating Point - AVX512VL AVX512F - Load - - - - - - - Gather single-precision (32-bit) floating-point elements from memory using 64-bit indices. 32-bit elements are loaded from addresses starting at "base_addr" and offset by each 64-bit element in "vindex" (each index is scaled by the factor in "scale"). Gathered elements are merged into "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). "scale" should be 1, 2, 4 or 8. - + Elementary Math Functions + + + Compute the natural logarithm of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst". -FOR j := 0 to 3 +FOR j := 0 to 15 i := j*32 - m := j*64 - IF k[j] - dst[i+31:i] := MEM[base_addr + SignExtend(vindex[i+63:i])*scale] - k[j] := 0 - ELSE - dst[i+31:i] := src[i+31:i] - FI + dst[i+31:i] := LOG(a[i+31:i]) ENDFOR -k[MAX:4] := 0 -dst[MAX:128] := 0 +dst[MAX:512] := 0 -
immintrin.h
 - + Floating Point - AVX512VL AVX512F - Load - - - - - - - Gather single-precision (32-bit) floating-point elements from memory using 64-bit indices. 32-bit elements are loaded from addresses starting at "base_addr" and offset by each 64-bit element in "vindex" (each index is scaled by the factor in "scale"). Gathered elements are merged into "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). "scale" should be 1, 2, 4 or 8. - + Elementary Math Functions + + + + + Compute the natural logarithm of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 1 +FOR j := 0 to 15 i := j*32 - m := j*64 IF k[j] - dst[i+31:i] := MEM[base_addr + SignExtend(vindex[i+63:i])*scale] - k[j] := 0 + dst[i+31:i] := LOG(a[i+31:i]) ELSE dst[i+31:i] := src[i+31:i] FI ENDFOR -k[MAX:2] := 0 -dst[MAX:64] := 0 +dst[MAX:512] := 0 -
immintrin.h
 - + Floating Point - AVX512VL AVX512F - Miscellaneous - - Convert the exponent of each packed double-precision (64-bit) floating-point element in "a" to a double-precision (64-bit) floating-point number representing the integer exponent, and store the results in "dst". This intrinsic essentially calculates "floor(log2(x))" for each element. - -FOR j := 0 to 3 + Elementary Math Functions + + + Convert the exponent of each packed double-precision (64-bit) floating-point element in "a" to a double-precision floating-point number representing the integer exponent, and store the results in "dst". This intrinsic essentially calculates "floor(log2(x))" for each element. + FOR j := 0 to 7 i := j*64 dst[i+63:i] := ConvertExpFP64(a[i+63:i]) ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 -
immintrin.h
 - + Floating Point - AVX512VL AVX512F - Miscellaneous - - - - Convert the exponent of each packed double-precision (64-bit) floating-point element in "a" to a double-precision (64-bit) floating-point number representing the integer exponent, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). This intrinsic essentially calculates "floor(log2(x))" for each element. - -FOR j := 0 to 3 + Elementary Math Functions + + + + + Convert the exponent of each packed double-precision (64-bit) floating-point element in "a" to a double-precision floating-point number representing the integer exponent, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). This intrinsic essentially calculates "floor(log2(x))" for each element. + FOR j := 0 to 7 i := j*64 IF k[j] dst[i+63:i] := ConvertExpFP64(a[i+63:i]) @@ -87358,47504 +103624,44212 @@ FOR j := 0 to 3 dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 -
immintrin.h
 - + Floating Point - AVX512VL AVX512F - Miscellaneous - - - Convert the exponent of each packed double-precision (64-bit) floating-point element in "a" to a double-precision (64-bit) floating-point number representing the integer exponent, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). This intrinsic essentially calculates "floor(log2(x))" for each element. - -FOR j := 0 to 3 - i := j*64 - IF k[j] - dst[i+63:i] := ConvertExpFP64(a[i+63:i]) - ELSE - dst[i+63:i] := 0 - FI + Elementary Math Functions + + + Convert the exponent of each packed single-precision (32-bit) floating-point element in "a" to a single-precision floating-point number representing the integer exponent, and store the results in "dst". This intrinsic essentially calculates "floor(log2(x))" for each element. + FOR j := 0 to 15 + i := j*32 + dst[i+31:i] := ConvertExpFP32(a[i+31:i]) ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 -
immintrin.h
 - + Floating Point - AVX512VL AVX512F - Miscellaneous - - Convert the exponent of each packed double-precision (64-bit) floating-point element in "a" to a double-precision (64-bit) floating-point number representing the integer exponent, and store the results in "dst". This intrinsic essentially calculates "floor(log2(x))" for each element. - -FOR j := 0 to 1 - i := j*64 - dst[i+63:i] := ConvertExpFP64(a[i+63:i]) + Elementary Math Functions + + + + + Convert the exponent of each packed single-precision (32-bit) floating-point element in "a" to a single-precision floating-point number representing the integer exponent, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). This intrinsic essentially calculates "floor(log2(x))" for each element. + FOR j := 0 to 15 + i := j*32 + IF k[j] + dst[i+31:i] := ConvertExpFP32(a[i+31:i]) + ELSE + dst[i+31:i] := src[i+31:i] + FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 -
immintrin.h
 - + Floating Point - AVX512VL AVX512F - Miscellaneous - - - - Convert the exponent of each packed double-precision (64-bit) floating-point element in "a" to a double-precision (64-bit) floating-point number representing the integer exponent, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). This intrinsic essentially calculates "floor(log2(x))" for each element. - -FOR j := 0 to 1 + Special Math Functions + + + Rounds each packed double-precision (64-bit) floating-point element in "a" to the nearest integer value and stores the results as packed double-precision floating-point elements in "dst". + FOR j := 0 to 7 i := j*64 - IF k[j] - dst[i+63:i] := ConvertExpFP64(a[i+63:i]) - ELSE - dst[i+63:i] := src[i+63:i] - FI + dst[i+63:i] := NearbyInt(a[i+63:i]) ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 -
immintrin.h
 - + Floating Point - AVX512VL AVX512F - Miscellaneous - - - Convert the exponent of each packed double-precision (64-bit) floating-point element in "a" to a double-precision (64-bit) floating-point number representing the integer exponent, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). This intrinsic essentially calculates "floor(log2(x))" for each element. - -FOR j := 0 to 1 + Special Math Functions + + + + + Rounds each packed double-precision (64-bit) floating-point element in "a" to the nearest integer value and stores the results as packed double-precision floating-point elements in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + FOR j := 0 to 7 i := j*64 IF k[j] - dst[i+63:i] := ConvertExpFP64(a[i+63:i]) + dst[i+63:i] := NearbyInt(a[i+63:i]) ELSE - dst[i+63:i] := 0 + dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 -
immintrin.h
 - + Floating Point - AVX512VL AVX512F - Miscellaneous - - Convert the exponent of each packed single-precision (32-bit) floating-point element in "a" to a single-precision (32-bit) floating-point number representing the integer exponent, and store the results in "dst". This intrinsic essentially calculates "floor(log2(x))" for each element. - -FOR j := 0 to 7 + Special Math Functions + + + Rounds each packed single-precision (32-bit) floating-point element in "a" to the nearest integer value and stores the results as packed double-precision floating-point elements in "dst". + FOR j := 0 to 15 i := j*32 - dst[i+31:i] := ConvertExpFP32(a[i+31:i]) + dst[i+31:i] := NearbyInt(a[i+31:i]) ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 -
immintrin.h
 - + Floating Point - AVX512VL AVX512F - Miscellaneous - - - - Convert the exponent of each packed single-precision (32-bit) floating-point element in "a" to a single-precision (32-bit) floating-point number representing the integer exponent, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). This intrinsic essentially calculates "floor(log2(x))" for each element. - -FOR j := 0 to 7 + Special Math Functions + + + + + Rounds each packed single-precision (32-bit) floating-point element in "a" to the nearest integer value and stores the results as packed double-precision floating-point elements in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + FOR j := 0 to 15 i := j*32 IF k[j] - dst[i+31:i] := ConvertExpFP32(a[i+31:i]) + dst[i+31:i] := NearbyInt(a[i+31:i]) ELSE dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 -
immintrin.h
 - + Floating Point - AVX512VL AVX512F - Miscellaneous - - - Convert the exponent of each packed single-precision (32-bit) floating-point element in "a" to a single-precision (32-bit) floating-point number representing the integer exponent, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). This intrinsic essentially calculates "floor(log2(x))" for each element. + Elementary Math Functions + + + + Compute the exponential value of packed double-precision (64-bit) floating-point elements in "a" raised by packed elements in "b", and store the results in "dst". FOR j := 0 to 7 - i := j*32 - IF k[j] - dst[i+31:i] := ConvertExpFP32(a[i+31:i]) - ELSE - dst[i+31:i] := 0 - FI + i := j*64 + dst[i+63:i] := POW(a[i+63:i], b[i+63:i]) ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 -
immintrin.h
 - + Floating Point - AVX512VL AVX512F - Miscellaneous - - Convert the exponent of each packed single-precision (32-bit) floating-point element in "a" to a single-precision (32-bit) floating-point number representing the integer exponent, and store the results in "dst". This intrinsic essentially calculates "floor(log2(x))" for each element. + Elementary Math Functions + + + + + + Compute the exponential value of packed double-precision (64-bit) floating-point elements in "a" raised by packed elements in "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*32 - dst[i+31:i] := ConvertExpFP32(a[i+31:i]) +FOR j := 0 to 7 + i := j*64 + IF k[j] + dst[i+63:i] := POW(a[i+63:i], b[i+63:i]) + ELSE + dst[i+63:i] := src[i+63:i] + FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 -
immintrin.h
 - + Floating Point - AVX512VL AVX512F - Miscellaneous - - - - Convert the exponent of each packed single-precision (32-bit) floating-point element in "a" to a single-precision (32-bit) floating-point number representing the integer exponent, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). This intrinsic essentially calculates "floor(log2(x))" for each element. + Elementary Math Functions + + + + Compute the exponential value of packed single-precision (32-bit) floating-point elements in "a" raised by packed elements in "b", and store the results in "dst". -FOR j := 0 to 3 +FOR j := 0 to 15 i := j*32 - IF k[j] - dst[i+31:i] := ConvertExpFP32(a[i+31:i]) - ELSE - dst[i+31:i] := src[i+31:i] - FI + dst[i+31:i] := POW(a[i+31:i], b[i+31:i]) ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 -
immintrin.h
 - + Floating Point - AVX512VL AVX512F - Miscellaneous - - - Convert the exponent of each packed single-precision (32-bit) floating-point element in "a" to a single-precision (32-bit) floating-point number representing the integer exponent, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). This intrinsic essentially calculates "floor(log2(x))" for each element. + Elementary Math Functions + + + + + + Compute the exponential value of packed single-precision (32-bit) floating-point elements in "a" raised by packed elements in "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 3 +FOR j := 0 to 15 i := j*32 IF k[j] - dst[i+31:i] := ConvertExpFP32(a[i+31:i]) + dst[i+31:i] := POW(a[i+31:i], b[i+31:i]) ELSE - dst[i+31:i] := 0 + dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 -
immintrin.h
 - + Floating Point - AVX512VL AVX512F - Miscellaneous - - - - Normalize the mantissas of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst". This intrinsic essentially calculates "±(2^k)*|x.significand|", where "k" depends on the interval range defined by "interv" and the sign depends on "sc" and the source sign. - [getmant_note] + Elementary Math Functions + + + Computes the reciprocal of packed double-precision (64-bit) floating-point elements in "a", storing the results in "dst". -FOR j := 0 to 3 +FOR j := 0 to 7 i := j*64 - dst[i+63:i] := GetNormalizedMantissa(a[i+63:i], sc, interv) + dst[i+63:i] := (1.0 / a[i+63:i]) ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 -
immintrin.h
 - + Floating Point - AVX512VL AVX512F - Miscellaneous - - - - - - Normalize the mantissas of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). This intrinsic essentially calculates "±(2^k)*|x.significand|", where "k" depends on the interval range defined by "interv" and the sign depends on "sc" and the source sign. - [getmant_note] + Elementary Math Functions + + + + + Computes the reciprocal of packed double-precision (64-bit) floating-point elements in "a", storing the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 3 +FOR j := 0 to 7 i := j*64 IF k[j] - dst[i+63:i] := GetNormalizedMantissa(a[i+63:i], sc, interv) + dst[i+63:i] := (1.0 / a[i+63:i]) ELSE dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 -
immintrin.h
 - + Floating Point - AVX512VL AVX512F - Miscellaneous - - - - - Normalize the mantissas of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). This intrinsic essentially calculates "±(2^k)*|x.significand|", where "k" depends on the interval range defined by "interv" and the sign depends on "sc" and the source sign. - [getmant_note] + Elementary Math Functions + + + Computes the reciprocal of packed single-precision (32-bit) floating-point elements in "a", storing the results in "dst". -FOR j := 0 to 3 - i := j*64 +FOR j := 0 to 15 + i := j*32 + dst[i+31:i] := (1.0 / a[i+31:i]) +ENDFOR +dst[MAX:512] := 0 + +
immintrin.h
+ + + Floating Point + AVX512F + Elementary Math Functions + + + + + Computes the reciprocal of packed single-precision (32-bit) floating-point elements in "a", storing the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +FOR j := 0 to 15 + i := j*32 IF k[j] - dst[i+63:i] := GetNormalizedMantissa(a[i+63:i], sc, interv) + dst[i+31:i] := (1.0 / a[i+31:i]) ELSE - dst[i+63:i] := 0 + dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 -
immintrin.h
 - + Floating Point - AVX512VL AVX512F - Miscellaneous - - - - Normalize the mantissas of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst". This intrinsic essentially calculates "±(2^k)*|x.significand|", where "k" depends on the interval range defined by "interv" and the sign depends on "sc" and the source sign. - [getmant_note] - -FOR j := 0 to 1 + Special Math Functions + + + Rounds the packed double-precision (64-bit) floating-point elements in "a" to the nearest even integer value and stores the results in "dst". + FOR j := 0 to 7 i := j*64 - dst[i+63:i] := GetNormalizedMantissa(a[i+63:i], sc, interv) + dst[i+63:i] := RoundToNearestEven(a[i+63:i]) ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 -
immintrin.h
 - + Floating Point - AVX512VL AVX512F - Miscellaneous - - - - - - Normalize the mantissas of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). This intrinsic essentially calculates "±(2^k)*|x.significand|", where "k" depends on the interval range defined by "interv" and the sign depends on "sc" and the source sign. - [getmant_note] - -FOR j := 0 to 1 + Special Math Functions + + + + + Rounds the packed double-precision (64-bit) floating-point elements in "a" to the nearest even integer value and stores the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + FOR j := 0 to 7 i := j*64 IF k[j] - dst[i+63:i] := GetNormalizedMantissa(a[i+63:i], sc, interv) + dst[i+63:i] := RoundToNearestEven(a[i+63:i]) ELSE dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 -
immintrin.h
 - + Floating Point - AVX512VL AVX512F - Miscellaneous - - - - - Normalize the mantissas of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). This intrinsic essentially calculates "±(2^k)*|x.significand|", where "k" depends on the interval range defined by "interv" and the sign depends on "sc" and the source sign. - [getmant_note] - -FOR j := 0 to 1 - i := j*64 + Special Math Functions + + + Rounds the packed single-precision (32-bit) floating-point elements in "a" to the nearest even integer value and stores the results in "dst". + FOR j := 0 to 15 + i := j*32 + dst[i+31:i] := RoundToNearestEven(a[i+31:i]) +ENDFOR +dst[MAX:512] := 0 + +
immintrin.h
+ + + Floating Point + AVX512F + Special Math Functions + + + + + Rounds the packed single-precision (32-bit) floating-point elements in "a" to the nearest even integer value and stores the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + FOR j := 0 to 15 + i := j*32 IF k[j] - dst[i+63:i] := GetNormalizedMantissa(a[i+63:i], sc, interv) + dst[i+31:i] := RoundToNearestEven(a[i+31:i]) ELSE - dst[i+63:i] := 0 + dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 -
immintrin.h
 - + Floating Point - AVX512VL AVX512F - Miscellaneous - - - - Normalize the mantissas of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst". This intrinsic essentially calculates "±(2^k)*|x.significand|", where "k" depends on the interval range defined by "interv" and the sign depends on "sc" and the source sign. - [getmant_note] + Special Math Functions + + + Round the packed double-precision (64-bit) floating-point elements in "a" to the nearest integer value, and store the results as packed double-precision floating-point elements in "dst". FOR j := 0 to 7 - i := j*32 - dst[i+31:i] := GetNormalizedMantissa(a[i+31:i], sc, interv) + i := j*64 + dst[i+63:i] := ROUND(a[i+63:i]) ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 -
immintrin.h
 - + Floating Point - AVX512VL AVX512F - Miscellaneous - - - - - - Normalize the mantissas of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). This intrinsic essentially calculates "±(2^k)*|x.significand|", where "k" depends on the interval range defined by "interv" and the sign depends on "sc" and the source sign. - [getmant_note] + Special Math Functions + + + + + Round the packed double-precision (64-bit) floating-point elements in "a" to the nearest integer value, and store the results as packed double-precision floating-point elements in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + [round_note] FOR j := 0 to 7 - i := j*32 + i := j*64 IF k[j] - dst[i+31:i] := GetNormalizedMantissa(a[i+31:i], sc, interv) + dst[i+63:i] := ROUND(a[i+63:i]) ELSE - dst[i+31:i] := src[i+31:i] + dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 -
immintrin.h
 - + Floating Point - AVX512VL AVX512F - Miscellaneous - - - - - Normalize the mantissas of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). This intrinsic essentially calculates "±(2^k)*|x.significand|", where "k" depends on the interval range defined by "interv" and the sign depends on "sc" and the source sign. - [getmant_note] + Trigonometry + + + Compute the sine of packed double-precision (64-bit) floating-point elements in "a" expressed in radians, and store the results in "dst". FOR j := 0 to 7 - i := j*32 + i := j*64 + dst[i+63:i] := SIN(a[i+63:i]) +ENDFOR +dst[MAX:512] := 0 + +
immintrin.h
+ + + Floating Point + AVX512F + Trigonometry + + + + + Compute the sine of packed double-precision (64-bit) floating-point elements in "a" expressed in radians, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +FOR j := 0 to 7 + i := j*64 IF k[j] - dst[i+31:i] := GetNormalizedMantissa(a[i+31:i], sc, interv) + dst[i+63:i] := SIN(a[i+63:i]) ELSE - dst[i+31:i] := 0 + dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 -
immintrin.h
 - + Floating Point - AVX512VL AVX512F - Miscellaneous - - - - Normalize the mantissas of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst". This intrinsic essentially calculates "±(2^k)*|x.significand|", where "k" depends on the interval range defined by "interv" and the sign depends on "sc" and the source sign. - [getmant_note] + Trigonometry + + + Compute the sine of packed single-precision (32-bit) floating-point elements in "a" expressed in radians, and store the results in "dst". -FOR j := 0 to 3 +FOR j := 0 to 15 i := j*32 - dst[i+31:i] := GetNormalizedMantissa(a[i+31:i], sc, interv) + dst[i+31:i] := SIN(a[i+31:i]) ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 -
immintrin.h
 - + Floating Point - AVX512VL AVX512F - Miscellaneous - - - - - - Normalize the mantissas of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). This intrinsic essentially calculates "±(2^k)*|x.significand|", where "k" depends on the interval range defined by "interv" and the sign depends on "sc" and the source sign. - [getmant_note] + Trigonometry + + + + + Compute the sine of packed single-precision (32-bit) floating-point elements in "a" expressed in radians, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 3 +FOR j := 0 to 15 i := j*32 IF k[j] - dst[i+31:i] := GetNormalizedMantissa(a[i+31:i], sc, interv) + dst[i+31:i] := SIN(a[i+31:i]) ELSE dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 -
immintrin.h
 - + Floating Point - AVX512VL AVX512F - Miscellaneous - - - - - Normalize the mantissas of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). This intrinsic essentially calculates "±(2^k)*|x.significand|", where "k" depends on the interval range defined by "interv" and the sign depends on "sc" and the source sign. - [getmant_note] + Trigonometry + + + Compute the hyperbolic sine of packed double-precision (64-bit) floating-point elements in "a" expressed in radians, and store the results in "dst". -FOR j := 0 to 3 - i := j*32 - IF k[j] - dst[i+31:i] := GetNormalizedMantissa(a[i+31:i], sc, interv) - ELSE - dst[i+31:i] := 0 - FI +FOR j := 0 to 7 + i := j*64 + dst[i+63:i] := SINH(a[i+63:i]) ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 -
immintrin.h
 - + Floating Point - AVX512VL AVX512F - Miscellaneous - - - - Copy "a" to "dst", then insert 128 bits (composed of 4 packed single-precision (32-bit) floating-point elements) from "b" into "dst" at the location specified by "imm8". + Trigonometry + + + + + Compute the hyperbolic sine of packed double-precision (64-bit) floating-point elements in "a" expressed in radians, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -dst[255:0] := a[255:0] -CASE (imm8[1:0]) of -0: dst[127:0] := b[127:0] -1: dst[255:128] := b[127:0] -ESAC -dst[MAX:256] := 0 +FOR j := 0 to 7 + i := j*64 + IF k[j] + dst[i+63:i] := SINH(a[i+63:i]) + ELSE + dst[i+63:i] := src[i+63:i] + FI +ENDFOR +dst[MAX:512] := 0 -
immintrin.h
 - + Floating Point - AVX512VL AVX512F - Miscellaneous - - - - - - Copy "a" to "tmp", then insert 128 bits (composed of 4 packed single-precision (32-bit) floating-point elements) from "b" into "tmp" at the location specified by "imm8". Store "tmp" to "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - + Trigonometry + + + Compute the hyperbolic sine of packed single-precision (32-bit) floating-point elements in "a" expressed in radians, and store the results in "dst". -tmp[255:0] := a[255:0] -CASE (imm8[1:0]) of -0: tmp[127:0] := b[127:0] -1: tmp[255:128] := b[127:0] -ESAC -FOR j := 0 to 7 +FOR j := 0 to 15 i := j*32 - IF k[j] - dst[i+31:i] := tmp[i+31:i] - ELSE - dst[i+31:i] := src[i+31:i] - FI + dst[i+31:i] := SINH(a[i+31:i]) ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 -
immintrin.h
 - + Floating Point - AVX512VL AVX512F - Miscellaneous - - - - - Copy "a" to "tmp", then insert 128 bits (composed of 4 packed single-precision (32-bit) floating-point elements) from "b" into "tmp" at the location specified by "imm8". Store "tmp" to "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - + Trigonometry + + + + + Compute the hyperbolic sine of packed single-precision (32-bit) floating-point elements in "a" expressed in radians, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -tmp[255:0] := a[255:0] -CASE (imm8[1:0]) of -0: tmp[127:0] := b[127:0] -1: tmp[255:128] := b[127:0] -ESAC -FOR j := 0 to 7 +FOR j := 0 to 15 i := j*32 IF k[j] - dst[i+31:i] := tmp[i+31:i] + dst[i+31:i] := SINH(a[i+31:i]) ELSE - dst[i+31:i] := 0 + dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 -
immintrin.h
 - + Floating Point - AVX512DQ - Miscellaneous - - - - Copy "a" to "dst", then insert 256 bits (composed of 8 packed single-precision (32-bit) floating-point elements) from "b" into "dst" at the location specified by "imm8". - - -dst[511:0] := a[511:0] -CASE (imm8[7:0]) OF -0: dst[255:0] := b[255:0] -1: dst[511:256] := b[255:0] -ESAC + AVX512F + Trigonometry + + + Compute the sine of packed double-precision (64-bit) floating-point elements in "a" expressed in degrees, and store the results in "dst". + FOR j := 0 to 7 + i := j*64 + dst[i+63:i] := SIND(a[i+63:i]) +ENDFOR dst[MAX:512] := 0 -
immintrin.h
 - + Floating Point - AVX512DQ - Miscellaneous - - - - - - Copy "a" to "tmp", then insert 256 bits (composed of 8 packed single-precision (32-bit) floating-point elements) from "b" into "tmp" at the location specified by "imm8". Store "tmp" to "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - - -tmp[511:0] := a[511:0] -CASE (imm8[7:0]) OF -0: tmp[255:0] := b[255:0] -1: tmp[511:256] := b[255:0] -ESAC -FOR j := 0 to 15 - i := j*32 + AVX512F + Trigonometry + + + + + Compute the sine of packed double-precision (64-bit) floating-point elements in "a" expressed in degrees, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + FOR j := 0 to 7 + i := j*64 IF k[j] - dst[i+31:i] := tmp[i+31:i] + dst[i+63:i] := SIND(a[i+63:i]) ELSE - dst[i+31:i] := src[i+31:i] + dst[i+63:i] := src[i+63:i] FI ENDFOR dst[MAX:512] := 0 -
immintrin.h
 - + Floating Point - AVX512DQ - Miscellaneous - - - - - Copy "a" to "tmp", then insert 256 bits (composed of 8 packed single-precision (32-bit) floating-point elements) from "b" into "tmp" at the location specified by "imm8". Store "tmp" to "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - - -tmp[511:0] := a[511:0] -CASE (imm8[7:0]) OF -0: tmp[255:0] := b[255:0] -1: tmp[511:256] := b[255:0] -ESAC -FOR j := 0 to 15 + AVX512F + Trigonometry + + + Compute the sine of packed single-precision (32-bit) floating-point elements in "a" expressed in degrees, and store the results in "dst". + FOR j := 0 to 15 + i := j*32 + dst[i+31:i] := SIND(a[i+31:i]) +ENDFOR +dst[MAX:512] := 0 + +
immintrin.h
+ + + Floating Point + AVX512F + Trigonometry + + + + + Compute the sine of packed single-precision (32-bit) floating-point elements in "a" expressed in degrees, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + FOR j := 0 to 15 i := j*32 IF k[j] - dst[i+31:i] := tmp[i+31:i] + dst[i+31:i] := SIND(a[i+31:i]) ELSE - dst[i+31:i] := 0 + dst[i+31:i] := src[i+31:i] FI ENDFOR dst[MAX:512] := 0 -
immintrin.h
 - + Floating Point - AVX512VL - AVX512DQ - Miscellaneous - - - - Copy "a" to "dst", then insert 128 bits (composed of 2 packed double-precision (64-bit) floating-point elements) from "b" into "dst" at the location specified by "imm8". + AVX512F + Trigonometry + + + Compute the tangent of packed double-precision (64-bit) floating-point elements in "a" expressed in radians, and store the results in "dst". -dst[255:0] := a[255:0] -CASE imm8[7:0] of -0: dst[127:0] := b[127:0] -1: dst[255:128] := b[127:0] -ESAC -dst[MAX:256] := 0 +FOR j := 0 to 7 + i := j*64 + dst[i+63:i] := TAN(a[i+63:i]) +ENDFOR +dst[MAX:512] := 0 -
immintrin.h
 - + Floating Point - AVX512VL - AVX512DQ - Miscellaneous - - - - - - Copy "a" to "tmp", then insert 128 bits (composed of 2 packed double-precision (64-bit) floating-point elements) from "b" into "tmp" at the location specified by "imm8". Store "tmp" to "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - + AVX512F + Trigonometry + + + + + Compute the tangent of packed double-precision (64-bit) floating-point elements in "a" expressed in radians, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -tmp[255:0] := a[255:0] -CASE (imm8[1:0]) of -0: tmp[127:0] := b[127:0] -1: tmp[255:128] := b[127:0] -ESAC -FOR j := 0 to 3 +FOR j := 0 to 7 i := j*64 IF k[j] - dst[i+63:i] := tmp[i+63:i] + dst[i+63:i] := TAN(a[i+63:i]) ELSE dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 -
immintrin.h
 - + Floating Point - AVX512VL - AVX512DQ - Miscellaneous - - - - - Copy "a" to "tmp", then insert 128 bits (composed of 2 packed double-precision (64-bit) floating-point elements) from "b" into "tmp" at the location specified by "imm8". Store "tmp" to "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - + AVX512F + Trigonometry + + + Compute the tangent of packed single-precision (32-bit) floating-point elements in "a" expressed in radians, and store the results in "dst". -tmp[255:0] := a[255:0] -CASE (imm8[1:0]) of -0: tmp[127:0] := b[127:0] -1: tmp[255:128] := b[127:0] -ESAC -FOR j := 0 to 3 - i := j*64 - IF k[j] - dst[i+63:i] := tmp[i+63:i] - ELSE - dst[i+63:i] := 0 - FI +FOR j := 0 to 15 + i := j*32 + dst[i+31:i] := TAN(a[i+31:i]) ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 -
immintrin.h
 - + Floating Point - AVX512DQ - Miscellaneous - - - - Copy "a" to "dst", then insert 128 bits (composed of 2 packed double-precision (64-bit) floating-point elements) from "b" into "dst" at the location specified by "imm8". + AVX512F + Trigonometry + + + + + Compute the tangent of packed single-precision (32-bit) floating-point elements in "a" expressed in radians, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -dst[511:0] := a[511:0] -CASE imm8[7:0] of -0: dst[127:0] := b[127:0] -1: dst[255:128] := b[127:0] -2: dst[383:256] := b[127:0] -3: dst[511:384] := b[127:0] -ESAC +FOR j := 0 to 15 + i := j*32 + IF k[j] + dst[i+31:i] := TAN(a[i+31:i]) + ELSE + dst[i+31:i] := src[i+31:i] + FI +ENDFOR dst[MAX:512] := 0 -
immintrin.h
 - + Floating Point - AVX512DQ - Miscellaneous - - - - - - Copy "a" to "tmp", then insert 128 bits (composed of 2 packed double-precision (64-bit) floating-point elements) from "b" into "tmp" at the location specified by "imm8". Store "tmp" to "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - - -tmp[511:0] := a[511:0] -CASE (imm8[1:0]) of -0: tmp[127:0] := b[127:0] -1: tmp[255:128] := b[127:0] -2: tmp[383:256] := b[127:0] -3: tmp[511:384] := b[127:0] -ESAC -FOR j := 0 to 7 + AVX512F + Trigonometry + + + Compute the tangent of packed double-precision (64-bit) floating-point elements in "a" expressed in degrees, and store the results in "dst". + FOR j := 0 to 7 i := j*64 - IF k[j] - dst[i+63:i] := tmp[i+63:i] - ELSE - dst[i+63:i] := src[i+63:i] - FI + dst[i+63:i] := TAND(a[i+63:i]) ENDFOR dst[MAX:512] := 0 -
immintrin.h
 - + Floating Point - AVX512DQ - Miscellaneous - - - - - Copy "a" to "tmp", then insert 128 bits (composed of 2 packed double-precision (64-bit) floating-point elements) from "b" into "tmp" at the location specified by "imm8". Store "tmp" to "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - - -tmp[511:0] := a[511:0] -CASE (imm8[1:0]) of -0: tmp[127:0] := b[127:0] -1: tmp[255:128] := b[127:0] -2: tmp[383:256] := b[127:0] -3: tmp[511:384] := b[127:0] -ESAC -FOR j := 0 to 7 + AVX512F + Trigonometry + + + + + Compute the tangent of packed double-precision (64-bit) floating-point elements in "a" expressed in degrees, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + FOR j := 0 to 7 i := j*64 IF k[j] - dst[i+63:i] := tmp[i+63:i] + dst[i+63:i] := TAND(a[i+63:i]) ELSE - dst[i+63:i] := 0 + dst[i+63:i] := src[i+63:i] FI ENDFOR dst[MAX:512] := 0 -
immintrin.h
 - - AVX512VL + + Floating Point AVX512F - Miscellaneous - - - - Copy "a" to "dst", then insert 128 bits (composed of 4 packed 32-bit integers) from "b" into "dst" at the location specified by "imm8". - -dst[255:0] := a[255:0] -CASE (imm8[1:0]) of -0: dst[127:0] := b[127:0] -1: dst[255:128] := b[127:0] -ESAC -dst[MAX:256] := 0 + Trigonometry + + + Compute the tangent of packed single-precision (32-bit) floating-point elements in "a" expressed in degrees, and store the results in "dst". + FOR j := 0 to 15 + i := j*32 + dst[i+31:i] := TAND(a[i+31:i]) +ENDFOR +dst[MAX:512] := 0 -
immintrin.h
 - - AVX512VL + + Floating Point AVX512F - Miscellaneous - - - - - - Copy "a" to "tmp", then insert 128 bits (composed of 4 packed 32-bit integers) from "b" into "tmp" at the location specified by "imm8". Store "tmp" to "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - - -tmp[255:0] := a[255:0] -CASE (imm8[1:0]) of -0: tmp[127:0] := b[127:0] -1: tmp[255:128] := b[127:0] -ESAC -FOR j := 0 to 7 + Trigonometry + + + + + Compute the tangent of packed single-precision (32-bit) floating-point elements in "a" expressed in degrees, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + FOR j := 0 to 15 i := j*32 IF k[j] - dst[i+31:i] := tmp[i+31:i] + dst[i+31:i] := TAND(a[i+31:i]) ELSE dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 -
immintrin.h
 - - AVX512VL + + Floating Point AVX512F - Miscellaneous - - - - - Copy "a" to "tmp", then insert 128 bits (composed of 4 packed 32-bit integers) from "b" into "tmp" at the location specified by "imm8". Store "tmp" to "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - + Trigonometry + + + Compute the hyperbolic tangent of packed double-precision (64-bit) floating-point elements in "a" expressed in radians, and store the results in "dst". -tmp[255:0] := a[255:0] -CASE (imm8[1:0]) of -0: tmp[127:0] := b[127:0] -1: tmp[255:128] := b[127:0] -ESAC FOR j := 0 to 7 - i := j*32 - IF k[j] - dst[i+31:i] := tmp[i+31:i] - ELSE - dst[i+31:i] := 0 - FI + i := j*64 + dst[i+63:i] := TANH(a[i+63:i]) ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 -
immintrin.h
 - - AVX512DQ - Miscellaneous - - - - Copy "a" to "dst", then insert 256 bits (composed of 8 packed 32-bit integers) from "b" into "dst" at the location specified by "imm8". + + Floating Point + AVX512F + Trigonometry + + + + + Compute the hyperbolic tangent of packed double-precision (64-bit) floating-point elements in "a" expressed in radians, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -dst[511:0] := a[511:0] -CASE imm8[7:0] of -0: dst[255:0] := b[255:0] -1: dst[511:256] := b[255:0] -ESAC +FOR j := 0 to 7 + i := j*64 + IF k[j] + dst[i+63:i] := TANH(a[i+63:i]) + ELSE + dst[i+63:i] := src[i+63:i] + FI +ENDFOR dst[MAX:512] := 0 -
immintrin.h
 - - AVX512DQ - Miscellaneous - - - - - - Copy "a" to "tmp", then insert 256 bits (composed of 8 packed 32-bit integers) from "b" into "tmp" at the location specified by "imm8". Store "tmp" to "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - + + Floating Point + AVX512F + Trigonometry + + + Compute the hyperbolic tangent of packed single-precision (32-bit) floating-point elements in "a" expressed in radians, and store the results in "dst". -tmp[511:0] := a[511:0] -CASE (imm8[7:0]) OF -0: tmp[255:0] := b[255:0] -1: tmp[511:256] := b[255:0] -ESAC FOR j := 0 to 15 i := j*32 - IF k[j] - dst[i+31:i] := tmp[i+31:i] - ELSE - dst[i+31:i] := src[i+31:i] - FI + dst[i+31:i] := TANH(a[i+31:i]) ENDFOR dst[MAX:512] := 0 -
immintrin.h
 - - AVX512DQ - Miscellaneous - - - - - Copy "a" to "tmp", then insert 256 bits (composed of 8 packed 32-bit integers) from "b" into "tmp" at the location specified by "imm8". Store "tmp" to "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - + + Floating Point + AVX512F + Trigonometry + + + + + Compute the hyperbolic tangent of packed single-precision (32-bit) floating-point elements in "a" expressed in radians, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -tmp[511:0] := a[511:0] -CASE (imm8[7:0]) OF -0: tmp[255:0] := b[255:0] -1: tmp[511:256] := b[255:0] -ESAC FOR j := 0 to 15 i := j*32 IF k[j] - dst[i+31:i] := tmp[i+31:i] + dst[i+31:i] := TANH(a[i+31:i]) ELSE - dst[i+31:i] := 0 + dst[i+31:i] := src[i+31:i] FI ENDFOR dst[MAX:512] := 0 -
immintrin.h
 - - AVX512VL - AVX512DQ - Miscellaneous - - - - Copy "a" to "dst", then insert 128 bits (composed of 2 packed 64-bit integers) from "b" into "dst" at the location specified by "imm8". - -dst[255:0] := a[255:0] -CASE imm8[7:0] of -0: dst[127:0] := b[127:0] -1: dst[255:128] := b[127:0] -ESAC -dst[MAX:256] := 0 + + Floating Point + AVX512F + Special Math Functions + + + Truncate the packed double-precision (64-bit) floating-point elements in "a", and store the results as packed double-precision floating-point elements in "dst". + FOR j := 0 to 7 + i := j*64 + dst[i+63:i] := TRUNCATE(a[i+63:i]) +ENDFOR +dst[MAX:512] := 0 -
immintrin.h
 - - AVX512VL - AVX512DQ - Miscellaneous - - - - - - Copy "a" to "tmp", then insert 128 bits (composed of 2 packed 64-bit integers) from "b" into "tmp" at the location specified by "imm8". Store "tmp" to "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - - -tmp[255:0] := a[255:0] -CASE (imm8[1:0]) of -0: tmp[127:0] := b[127:0] -1: tmp[255:128] := b[127:0] -ESAC -FOR j := 0 to 3 + + Floating Point + AVX512F + Special Math Functions + + + + + Truncate the packed double-precision (64-bit) floating-point elements in "a", and store the results as packed double-precision floating-point elements in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + FOR j := 0 to 7 i := j*64 IF k[j] - dst[i+63:i] := tmp[i+63:i] + dst[i+63:i] := TRUNCATE(a[i+63:i]) ELSE dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 -
immintrin.h
 - - AVX512VL - AVX512DQ - Miscellaneous - - - - - Copy "a" to "tmp", then insert 128 bits (composed of 2 packed 64-bit integers) from "b" into "tmp" at the location specified by "imm8". Store "tmp" to "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - - -tmp[255:0] := a[255:0] -CASE (imm8[1:0]) of -0: tmp[127:0] := b[127:0] -1: tmp[255:128] := b[127:0] -ESAC -FOR j := 0 to 3 - i := j*64 + + Floating Point + AVX512F + Special Math Functions + + + Truncate the packed single-precision (32-bit) floating-point elements in "a", and store the results as packed single-precision floating-point elements in "dst". + FOR j := 0 to 15 + i := j*32 + dst[i+31:i] := TRUNCATE(a[i+31:i]) +ENDFOR +dst[MAX:512] := 0 + +
immintrin.h
+ + + Floating Point + AVX512F + Special Math Functions + + + + + Truncate the packed single-precision (32-bit) floating-point elements in "a", and store the results as packed single-precision floating-point elements in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + FOR j := 0 to 15 + i := j*32 IF k[j] - dst[i+63:i] := tmp[i+63:i] + dst[i+31:i] := TRUNCATE(a[i+31:i]) ELSE - dst[i+63:i] := 0 + dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 -
immintrin.h
 - - AVX512DQ - Miscellaneous - - - - Copy "a" to "dst", then insert 128 bits (composed of 2 packed 64-bit integers) from "b" into "dst" at the location specified by "imm8". + + Integer + AVX512F + Arithmetic + + + + Divide packed unsigned 32-bit integers in "a" by packed elements in "b", and store the truncated results in "dst". -dst[511:0] := a[511:0] -CASE imm8[7:0] of -0: dst[127:0] := b[127:0] -1: dst[255:128] := b[127:0] -2: dst[383:256] := b[127:0] -3: dst[511:384] := b[127:0] -ESAC +FOR j := 0 to 15 + i := 32*j + IF b[i+31:i] == 0 + #DE + FI + dst[i+31:i] := Truncate32(a[i+31:i] / b[i+31:i]) +ENDFOR dst[MAX:512] := 0 -
immintrin.h
 - - AVX512DQ - Miscellaneous - - - - - - Copy "a" to "tmp", then insert 128 bits (composed of 2 packed 64-bit integers) from "b" into "tmp" at the location specified by "imm8". Store "tmp" to "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - + + Integer + AVX512F + Arithmetic + + + + + + Divide packed unsigned 32-bit integers in "a" by packed elements in "b", and store the truncated results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -tmp[511:0] := a[511:0] -CASE (imm8[1:0]) of -0: tmp[127:0] := b[127:0] -1: tmp[255:128] := b[127:0] -2: tmp[383:256] := b[127:0] -3: tmp[511:384] := b[127:0] -ESAC -FOR j := 0 to 7 - i := j*64 +FOR j := 0 to 15 + i := 32*j IF k[j] - dst[i+63:i] := tmp[i+63:i] + IF b[i+31:i] == 0 + #DE + FI + dst[i+31:i] := Truncate32(a[i+31:i] / b[i+31:i]) ELSE - dst[i+63:i] := src[i+63:i] + dst[i+31:i] := src[i+31:i] FI ENDFOR dst[MAX:512] := 0 -
immintrin.h
 - - AVX512DQ - Miscellaneous - - - - - Copy "a" to "tmp", then insert 128 bits (composed of 2 packed 64-bit integers) from "b" into "tmp" at the location specified by "imm8". Store "tmp" to "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - + + Integer + AVX512F + Arithmetic + + + + Divide packed unsigned 8-bit integers in "a" by packed elements in "b", and store the truncated results in "dst". -tmp[511:0] := a[511:0] -CASE (imm8[1:0]) of -0: tmp[127:0] := b[127:0] -1: tmp[255:128] := b[127:0] -2: tmp[383:256] := b[127:0] -3: tmp[511:384] := b[127:0] -ESAC -FOR j := 0 to 7 - i := j*64 - IF k[j] - dst[i+63:i] := tmp[i+63:i] - ELSE - dst[i+63:i] := 0 +FOR j := 0 to 63 + i := 8*j + IF b[i+7:i] == 0 + #DE FI + dst[i+7:i] := Truncate8(a[i+7:i] / b[i+7:i]) ENDFOR dst[MAX:512] := 0 -
immintrin.h
 - - Floating Point - AVX512VL + + Integer AVX512F Arithmetic - - - - - Compare packed double-precision (64-bit) floating-point elements in "a" and "b", and store packed maximum values in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + + + Divide packed unsigned 16-bit integers in "a" by packed elements in "b", and store the truncated results in "dst". -FOR j := 0 to 3 - i := j*64 - IF k[j] - dst[i+63:i] := MAX(a[i+63:i], b[i+63:i]) - ELSE - dst[i+63:i] := src[i+63:i] +FOR j := 0 to 31 + i := 16*j + IF b[i+15:i] == 0 + #DE FI + dst[i+15:i] := Truncate16(a[i+15:i] / b[i+15:i]) ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 -
immintrin.h
 - - Floating Point - AVX512VL + + Integer AVX512F Arithmetic - - - - Compare packed double-precision (64-bit) floating-point elements in "a" and "b", and store packed maximum values in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - + + + + Divide packed unsigned 64-bit integers in "a" by packed elements in "b", and store the truncated results in "dst". -FOR j := 0 to 3 - i := j*64 - IF k[j] - dst[i+63:i] := MAX(a[i+63:i], b[i+63:i]) - ELSE - dst[i+63:i] := 0 +FOR j := 0 to 7 + i := 64*j + IF b[i+63:i] == 0 + #DE FI + dst[i+63:i] := Truncate64(a[i+63:i] / b[i+63:i]) ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 -
immintrin.h
 - - Floating Point - AVX512VL + + Integer AVX512F Arithmetic - - - - - Compare packed double-precision (64-bit) floating-point elements in "a" and "b", and store packed maximum values in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -FOR j := 0 to 1 - i := j*64 - IF k[j] - dst[i+63:i] := MAX(a[i+63:i], b[i+63:i]) - ELSE - dst[i+63:i] := src[i+63:i] - FI + + + + Divide packed unsigned 32-bit integers in "a" by packed elements in "b", and store the remainders as packed unsigned 32-bit integers in "dst". + FOR j := 0 to 15 + i := 32*j + dst[i+31:i] := REMAINDER(a[i+31:i] / b[i+31:i]) ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 -
immintrin.h
 - - Floating Point - AVX512VL + + Integer AVX512F Arithmetic - - - - Compare packed double-precision (64-bit) floating-point elements in "a" and "b", and store packed maximum values in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - - -FOR j := 0 to 1 - i := j*64 + + + + + + Divide packed unsigned 32-bit integers in "a" by packed elements in "b", and store the remainders as packed unsigned 32-bit integers in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + FOR j := 0 to 15 + i := 32*j IF k[j] - dst[i+63:i] := MAX(a[i+63:i], b[i+63:i]) + dst[i+31:i] := REMAINDER(a[i+31:i] / b[i+31:i]) ELSE - dst[i+63:i] := 0 + dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 -
immintrin.h
 - - Floating Point - AVX512VL + + Integer AVX512F Arithmetic - - - - - Compare packed single-precision (32-bit) floating-point elements in "a" and "b", and store packed maximum values in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -FOR j := 0 to 7 - i := j*32 - IF k[j] - dst[i+31:i] := MAX(a[i+31:i], b[i+31:i]) - ELSE - dst[i+31:i] := src[i+31:i] - FI + + + + Divide packed unsigned 8-bit integers in "a" by packed elements in "b", and store the remainders as packed unsigned 32-bit integers in "dst". + FOR j := 0 to 63 + i := 8*j + dst[i+7:i] := REMAINDER(a[i+7:i] / b[i+7:i]) ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 -
immintrin.h
 - - Floating Point - AVX512VL + + Integer AVX512F Arithmetic - - - - Compare packed single-precision (32-bit) floating-point elements in "a" and "b", and store packed maximum values in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - - -FOR j := 0 to 7 - i := j*32 - IF k[j] - dst[i+31:i] := MAX(a[i+31:i], b[i+31:i]) - ELSE - dst[i+31:i] := 0 - FI + + + + Divide packed unsigned 16-bit integers in "a" by packed elements in "b", and store the remainders as packed unsigned 32-bit integers in "dst". + FOR j := 0 to 31 + i := 16*j + dst[i+15:i] := REMAINDER(a[i+15:i] / b[i+15:i]) ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 -
immintrin.h
 - - Floating Point - AVX512VL + + Integer AVX512F Arithmetic - - - - - Compare packed single-precision (32-bit) floating-point elements in "a" and "b", and store packed maximum values in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -FOR j := 0 to 3 - i := j*32 - IF k[j] - dst[i+31:i] := MAX(a[i+31:i], b[i+31:i]) - ELSE - dst[i+31:i] := src[i+31:i] - FI + + + + Divide packed unsigned 64-bit integers in "a" by packed elements in "b", and store the remainders as packed unsigned 32-bit integers in "dst". + FOR j := 0 to 7 + i := 64*j + dst[i+63:i] := REMAINDER(a[i+63:i] / b[i+63:i]) ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 -
immintrin.h
 - - Floating Point - AVX512VL + + Mask AVX512F - Arithmetic - - - - Compare packed single-precision (32-bit) floating-point elements in "a" and "b", and store packed maximum values in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - - -FOR j := 0 to 3 - i := j*32 - IF k[j] - dst[i+31:i] := MAX(a[i+31:i], b[i+31:i]) - ELSE - dst[i+31:i] := 0 - FI -ENDFOR -dst[MAX:128] := 0 + Mask + + + + Performs bitwise OR between "k1" and "k2", storing the result in "dst". ZF flag is set if "dst" is 0. + dst[15:0] := k1[15:0] | k2[15:0] +IF dst == 0 + SetZF() +FI - +
immintrin.h
 - - Floating Point - AVX512VL + + Mask AVX512F - Arithmetic - - - - - Compare packed double-precision (64-bit) floating-point elements in "a" and "b", and store packed minimum values in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Mask + + + + Performs bitwise OR between "k1" and "k2", storing the result in "dst". CF flag is set if "dst" consists of all 1's. + dst[15:0] := k1[15:0] | k2[15:0] +IF PopCount(dst[15:0]) == 16 + SetCF() +FI + + +
immintrin.h
+ + + AVX512F + Mask + + + Converts bit mask "k1" into an integer value, storing the results in "dst". -FOR j := 0 to 3 - i := j*64 - IF k[j] - dst[i+63:i] := MIN(a[i+63:i], b[i+63:i]) - ELSE - dst[i+63:i] := src[i+63:i] - FI -ENDFOR -dst[MAX:256] := 0 +dst := ZeroExtend32(k1) - +
immintrin.h
 - - Floating Point - AVX512VL + AVX512F - Arithmetic - - - - Compare packed double-precision (64-bit) floating-point elements in "a" and "b", and store packed minimum values in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - + Mask + + + Converts integer "mask" into bitmask, storing the result in "dst". -FOR j := 0 to 3 - i := j*64 - IF k[j] - dst[i+63:i] := MIN(a[i+63:i], b[i+63:i]) - ELSE - dst[i+63:i] := 0 - FI -ENDFOR -dst[MAX:256] := 0 +dst := mask[15:0] - +
immintrin.h
 - - Floating Point - AVX512VL + + Integer AVX512F - Arithmetic - - - - - Compare packed double-precision (64-bit) floating-point elements in "a" and "b", and store packed minimum values in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Store + + + + Multiplies elements in packed 64-bit integer vectors "a" and "b" together, storing the lower 64 bits of the result in "dst". -FOR j := 0 to 1 +FOR j := 0 to 7 i := j*64 - IF k[j] - dst[i+63:i] := MIN(a[i+63:i], b[i+63:i]) - ELSE - dst[i+63:i] := src[i+63:i] - FI + dst[i+63:i] := a[i+63:i] * b[i+63:i] ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 -
immintrin.h
 - - Floating Point - AVX512VL + + Integer AVX512F - Arithmetic - - - - Compare packed double-precision (64-bit) floating-point elements in "a" and "b", and store packed minimum values in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - + Store + + + + + + Multiplies elements in packed 64-bit integer vectors "a" and "b" together, storing the lower 64 bits of the result in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 1 +FOR j := 0 to 7 i := j*64 IF k[j] - dst[i+63:i] := MIN(a[i+63:i], b[i+63:i]) + dst[i+63:i] := a[i+63:i] * b[i+63:i] ELSE - dst[i+63:i] := 0 + dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 -
immintrin.h
 - + Floating Point - AVX512VL AVX512F - Arithmetic - - - - - Compare packed single-precision (32-bit) floating-point elements in "a" and "b", and store packed minimum values in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Trigonometry + + + + Compute the sine and cosine of packed double-precision (64-bit) floating-point elements in "a" expressed in radians, store the sine in "dst", and store the cosine into memory at "mem_addr". FOR j := 0 to 7 - i := j*32 - IF k[j] - dst[i+31:i] := MIN(a[i+31:i], b[i+31:i]) - ELSE - dst[i+31:i] := src[i+31:i] - FI + i := j*64 + dst[i+63:i] := SIN(a[i+63:i]) + MEM[mem_addr+i+63:mem_addr+i] := COS(a[i+63:i]) ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 +cos_res[MAX:512] := 0 -
immintrin.h
 - + Floating Point - AVX512VL AVX512F - Arithmetic - - - - Compare packed single-precision (32-bit) floating-point elements in "a" and "b", and store packed minimum values in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - + Trigonometry + + + + + + + Compute the sine and cosine of packed double-precision (64-bit) floating-point elements in "a" expressed in radians, store the sine in "dst", store the cosine into memory at "mem_addr". Elements are written to their respective locations using writemask "k" (elements are copied from "sin_src" or "cos_src" when the corresponding mask bit is not set). FOR j := 0 to 7 - i := j*32 + i := j*64 IF k[j] - dst[i+31:i] := MIN(a[i+31:i], b[i+31:i]) + dst[i+63:i] := SIN(a[i+63:i]) + MEM[mem_addr+i+63:mem_addr+i] := COS(a[i+63:i]) ELSE - dst[i+31:i] := 0 + dst[i+63:i] := sin_src[i+63:i] + MEM[mem_addr+i+63:mem_addr+i] := cos_src[i+63:i] FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 +cos_res[MAX:512] := 0 -
immintrin.h
 - + Floating Point - AVX512VL AVX512F - Arithmetic - - - - - Compare packed single-precision (32-bit) floating-point elements in "a" and "b", and store packed minimum values in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Trigonometry + + + + Compute the sine and cosine of packed single-precision (32-bit) floating-point elements in "a" expressed in radians, store the sine in "dst", and store the cosine into memory at "mem_addr". -FOR j := 0 to 3 +FOR j := 0 to 15 i := j*32 - IF k[j] - dst[i+31:i] := MIN(a[i+31:i], b[i+31:i]) - ELSE - dst[i+31:i] := src[i+31:i] - FI + dst[i+31:i] := SIN(a[i+31:i]) + MEM[mem_addr+i+31:mem_addr+i] := COS(a[i+31:i]) ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 +cos_res[MAX:512] := 0 -
immintrin.h
 - + Floating Point - AVX512VL AVX512F - Arithmetic - - - - Compare packed single-precision (32-bit) floating-point elements in "a" and "b", and store packed minimum values in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - + Trigonometry + + + + + + + Compute the sine and cosine of packed single-precision (32-bit) floating-point elements in "a" expressed in radians, store the sine in "dst", store the cosine into memory at "mem_addr". Elements are written to their respective locations using writemask "k" (elements are copied from "sin_src" or "cos_src" when the corresponding mask bit is not set). -FOR j := 0 to 3 +FOR j := 0 to 15 i := j*32 IF k[j] - dst[i+31:i] := MIN(a[i+31:i], b[i+31:i]) + dst[i+31:i] := SIN(a[i+31:i]) + MEM[mem_addr+i+31:mem_addr+i] := COS(a[i+31:i]) ELSE - dst[i+31:i] := 0 + dst[i+31:i] := sin_src[i+31:i] + MEM[mem_addr+i+31:mem_addr+i] := cos_src[i+31:i] FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 +cos_res[MAX:512] := 0 -
immintrin.h
 - + Floating Point - AVX512VL AVX512F - Load - - - - Load packed double-precision (64-bit) floating-point elements from memory into "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). "mem_addr" must be aligned on a 32-byte boundary or a general-protection exception may be generated. + Convert + + + Copy the lower single-precision (32-bit) floating-point element of "a" to "dst". -FOR j := 0 to 3 - i := j*64 - IF k[j] - dst[i+63:i] := MEM[mem_addr+i+63:mem_addr+i] - ELSE - dst[i+63:i] := src[i+63:i] - FI -ENDFOR -dst[MAX:256] := 0 +dst[31:0] := a[31:0] - +
immintrin.h
 - + Floating Point - AVX512VL AVX512F - Move - - - - Move packed double-precision (64-bit) floating-point elements from "a" to "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Convert + + + Copy the lower double-precision (64-bit) floating-point element of "a" to "dst". -FOR j := 0 to 3 - i := j*64 - IF k[j] - dst[i+63:i] := a[i+63:i] - ELSE - dst[i+63:i] := src[i+63:i] - FI -ENDFOR -dst[MAX:256] := 0 +dst[63:0] := a[63:0] - +
immintrin.h
 - - Floating Point - AVX512VL + + Integer AVX512F - Store - - - - Store packed double-precision (64-bit) floating-point elements from "a" into memory using writemask "k". - "mem_addr" must be aligned on a 32-byte boundary or a general-protection exception may be generated. + Convert + + + Copy the lower 32-bit integer in "a" to "dst". -FOR j := 0 to 3 - i := j*64 - IF k[j] - MEM[mem_addr+i+63:mem_addr+i] := a[i+63:i] - FI -ENDFOR +dst[31:0] := a[31:0] - +
immintrin.h
 - + Floating Point - AVX512VL - AVX512F - Load - - - Load packed double-precision (64-bit) floating-point elements from memory into "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). "mem_addr" must be aligned on a 32-byte boundary or a general-protection exception may be generated. + AVX512F/KNCNI + Arithmetic + + + + Add packed double-precision (64-bit) floating-point elements in "a" and "b", and store the results in "dst". -FOR j := 0 to 3 +FOR j := 0 to 7 i := j*64 - IF k[j] - dst[i+63:i] := MEM[mem_addr+i+63:mem_addr+i] - ELSE - dst[i+63:i] := 0 - FI + dst[i+63:i] := a[i+63:i] + b[i+63:i] ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - AVX512VL - AVX512F - Move - - - Move packed double-precision (64-bit) floating-point elements from "a" into "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + AVX512F/KNCNI + Arithmetic + + + + + Add packed double-precision (64-bit) floating-point elements in "a" and "b", and store the results in "dst". + [round_note] -FOR j := 0 to 3 +FOR j := 0 to 7 i := j*64 - IF k[j] - dst[i+63:i] := a[i+63:i] - ELSE - dst[i+63:i] := 0 - FI + dst[i+63:i] := a[i+63:i] + b[i+63:i] ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - AVX512VL - AVX512F - Load - - - - Load packed double-precision (64-bit) floating-point elements from memory into "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). "mem_addr" must be aligned on a 16-byte boundary or a general-protection exception may be generated. + AVX512F/KNCNI + Arithmetic + + + + + + Add packed double-precision (64-bit) floating-point elements in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 1 +FOR j := 0 to 7 i := j*64 IF k[j] - dst[i+63:i] := MEM[mem_addr+i+63:mem_addr+i] + dst[i+63:i] := a[i+63:i] + b[i+63:i] ELSE dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - AVX512VL - AVX512F - Move - - - - Move packed double-precision (64-bit) floating-point elements from "a" to "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + AVX512F/KNCNI + Arithmetic + + + + + + + Add packed double-precision (64-bit) floating-point elements in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + [round_note] -FOR j := 0 to 1 +FOR j := 0 to 7 i := j*64 IF k[j] - dst[i+63:i] := a[i+63:i] + dst[i+63:i] := a[i+63:i] + b[i+63:i] ELSE dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - AVX512VL - AVX512F - Store - - - - Store packed double-precision (64-bit) floating-point elements from "a" into memory using writemask "k". - "mem_addr" must be aligned on a 16-byte boundary or a general-protection exception may be generated. + AVX512F/KNCNI + Arithmetic + + + + Add packed single-precision (32-bit) floating-point elements in "a" and "b", and store the results in "dst". -FOR j := 0 to 1 - i := j*64 - IF k[j] - MEM[mem_addr+i+63:mem_addr+i] := a[i+63:i] - FI +FOR j := 0 to 15 + i := j*32 + dst[i+31:i] := a[i+31:i] + b[i+31:i] ENDFOR +dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - AVX512VL - AVX512F - Load - - - Load packed double-precision (64-bit) floating-point elements from memory into "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). "mem_addr" must be aligned on a 16-byte boundary or a general-protection exception may be generated. + AVX512F/KNCNI + Arithmetic + + + + + Add packed single-precision (32-bit) floating-point elements in "a" and "b", and store the results in "dst". + [round_note] -FOR j := 0 to 1 - i := j*64 - IF k[j] - dst[i+63:i] := MEM[mem_addr+i+63:mem_addr+i] - ELSE - dst[i+63:i] := 0 - FI +FOR j := 0 to 15 + i := j*32 + dst[i+31:i] := a[i+31:i] + b[i+31:i] ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - AVX512VL - AVX512F - Move - - - Move packed double-precision (64-bit) floating-point elements from "a" into "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + AVX512F/KNCNI + Arithmetic + + + + + + Add packed single-precision (32-bit) floating-point elements in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 1 - i := j*64 +FOR j := 0 to 15 + i := j*32 IF k[j] - dst[i+63:i] := a[i+63:i] + dst[i+31:i] := a[i+31:i] + b[i+31:i] ELSE - dst[i+63:i] := 0 + dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - AVX512VL - AVX512F - Load - - - - Load packed single-precision (32-bit) floating-point elements from memory into "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). "mem_addr" must be aligned on a 32-byte boundary or a general-protection exception may be generated. + AVX512F/KNCNI + Arithmetic + + + + + + + Add packed single-precision (32-bit) floating-point elements in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + [round_note] -FOR j := 0 to 7 +FOR j := 0 to 15 i := j*32 IF k[j] - dst[i+31:i] := MEM[mem_addr+i+31:mem_addr+i] + dst[i+31:i] := a[i+31:i] + b[i+31:i] ELSE dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point - AVX512VL - AVX512F - Move - - - - Move packed single-precision (32-bit) floating-point elements from "a" to "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Integer + AVX512F/KNCNI + Miscellaneous + + + + + Concatenate "a" and "b" into a 128-byte immediate result, shift the result right by "imm8" 32-bit elements, and store the low 64 bytes (16 elements) in "dst". -FOR j := 0 to 7 +temp[1023:512] := a[511:0] +temp[511:0] := b[511:0] +temp[1023:0] := temp[1023:0] >> (32*imm8[3:0]) +dst[511:0] := temp[511:0] +dst[MAX:512] := 0 + + +
immintrin.h
+ + + Integer + AVX512F/KNCNI + Miscellaneous + + + + + + + Concatenate "a" and "b" into a 128-byte immediate result, shift the result right by "imm8" 32-bit elements, and store the low 64 bytes (16 elements) in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +temp[1023:512] := a[511:0] +temp[511:0] := b[511:0] +temp[1023:0] := temp[1023:0] >> (32*imm8[3:0]) +FOR j := 0 to 15 i := j*32 IF k[j] - dst[i+31:i] := a[i+31:i] + dst[i+31:i] := temp[i+31:i] ELSE dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - AVX512VL - AVX512F - Store - - - - Store packed single-precision (32-bit) floating-point elements from "a" into memory using writemask "k". - "mem_addr" must be aligned on a 32-byte boundary or a general-protection exception may be generated. + AVX512F/KNCNI + Swizzle + + + + + Blend packed double-precision (64-bit) floating-point elements from "a" and "b" using control mask "k", and store the results in "dst". FOR j := 0 to 7 - i := j*32 + i := j*64 IF k[j] - MEM[mem_addr+i+31:mem_addr+i] := a[i+31:i] + dst[i+63:i] := b[i+63:i] + ELSE + dst[i+63:i] := a[i+63:i] FI ENDFOR +dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - AVX512VL - AVX512F - Load - - - Load packed single-precision (32-bit) floating-point elements from memory into "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). "mem_addr" must be aligned on a 32-byte boundary or a general-protection exception may be generated. + AVX512F/KNCNI + Swizzle + + + + + Blend packed single-precision (32-bit) floating-point elements from "a" and "b" using control mask "k", and store the results in "dst". -FOR j := 0 to 7 +FOR j := 0 to 15 i := j*32 IF k[j] - dst[i+31:i] := MEM[mem_addr+i+31:mem_addr+i] + dst[i+31:i] := b[i+31:i] ELSE - dst[i+31:i] := 0 + dst[i+31:i] := a[i+31:i] FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - AVX512VL - AVX512F - Move - - - Move packed single-precision (32-bit) floating-point elements from "a" into "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - + Mask + AVX512F/KNCNI + Compare + + + + + Compare packed double-precision (64-bit) floating-point elements in "a" and "b" based on the comparison operand specified by "imm8", and store the results in mask vector "k". + CASE (imm8[4:0]) OF +0: OP := _CMP_EQ_OQ +1: OP := _CMP_LT_OS +2: OP := _CMP_LE_OS +3: OP := _CMP_UNORD_Q +4: OP := _CMP_NEQ_UQ +5: OP := _CMP_NLT_US +6: OP := _CMP_NLE_US +7: OP := _CMP_ORD_Q +8: OP := _CMP_EQ_UQ +9: OP := _CMP_NGE_US +10: OP := _CMP_NGT_US +11: OP := _CMP_FALSE_OQ +12: OP := _CMP_NEQ_OQ +13: OP := _CMP_GE_OS +14: OP := _CMP_GT_OS +15: OP := _CMP_TRUE_UQ +16: OP := _CMP_EQ_OS +17: OP := _CMP_LT_OQ +18: OP := _CMP_LE_OQ +19: OP := _CMP_UNORD_S +20: OP := _CMP_NEQ_US +21: OP := _CMP_NLT_UQ +22: OP := _CMP_NLE_UQ +23: OP := _CMP_ORD_S +24: OP := _CMP_EQ_US +25: OP := _CMP_NGE_UQ +26: OP := _CMP_NGT_UQ +27: OP := _CMP_FALSE_OS +28: OP := _CMP_NEQ_OS +29: OP := _CMP_GE_OQ +30: OP := _CMP_GT_OQ +31: OP := _CMP_TRUE_US +ESAC FOR j := 0 to 7 - i := j*32 - IF k[j] - dst[i+31:i] := a[i+31:i] - ELSE - dst[i+31:i] := 0 - FI + i := j*64 + k[j] := (a[i+63:i] OP b[i+63:i]) ? 1 : 0 ENDFOR -dst[MAX:256] := 0 +k[MAX:8] := 0 - +
immintrin.h
 - + Floating Point - AVX512VL - AVX512F - Load - - - - Load packed single-precision (32-bit) floating-point elements from memory into "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). "mem_addr" must be aligned on a 16-byte boundary or a general-protection exception may be generated. - -FOR j := 0 to 3 - i := j*32 - IF k[j] - dst[i+31:i] := MEM[mem_addr+i+31:mem_addr+i] - ELSE - dst[i+31:i] := src[i+31:i] - FI + Mask + AVX512F/KNCNI + Compare + + + + + + Compare packed double-precision (64-bit) floating-point elements in "a" and "b" based on the comparison operand specified by "imm8", and store the results in mask vector "k". [sae_note] + CASE (imm8[4:0]) OF +0: OP := _CMP_EQ_OQ +1: OP := _CMP_LT_OS +2: OP := _CMP_LE_OS +3: OP := _CMP_UNORD_Q +4: OP := _CMP_NEQ_UQ +5: OP := _CMP_NLT_US +6: OP := _CMP_NLE_US +7: OP := _CMP_ORD_Q +8: OP := _CMP_EQ_UQ +9: OP := _CMP_NGE_US +10: OP := _CMP_NGT_US +11: OP := _CMP_FALSE_OQ +12: OP := _CMP_NEQ_OQ +13: OP := _CMP_GE_OS +14: OP := _CMP_GT_OS +15: OP := _CMP_TRUE_UQ +16: OP := _CMP_EQ_OS +17: OP := _CMP_LT_OQ +18: OP := _CMP_LE_OQ +19: OP := _CMP_UNORD_S +20: OP := _CMP_NEQ_US +21: OP := _CMP_NLT_UQ +22: OP := _CMP_NLE_UQ +23: OP := _CMP_ORD_S +24: OP := _CMP_EQ_US +25: OP := _CMP_NGE_UQ +26: OP := _CMP_NGT_UQ +27: OP := _CMP_FALSE_OS +28: OP := _CMP_NEQ_OS +29: OP := _CMP_GE_OQ +30: OP := _CMP_GT_OQ +31: OP := _CMP_TRUE_US +ESAC +FOR j := 0 to 7 + i := j*64 + k[j] := (a[i+63:i] OP b[i+63:i]) ? 1 : 0 ENDFOR -dst[MAX:128] := 0 +k[MAX:8] := 0 - +
immintrin.h
 - + Floating Point - AVX512VL - AVX512F - Move - - - - Move packed single-precision (32-bit) floating-point elements from "a" to "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Mask + AVX512F/KNCNI + Compare + + + + Compare packed double-precision (64-bit) floating-point elements in "a" and "b" for equality, and store the results in mask vector "k". -FOR j := 0 to 3 - i := j*32 - IF k[j] - dst[i+31:i] := a[i+31:i] - ELSE - dst[i+31:i] := src[i+31:i] - FI -ENDFOR -dst[MAX:128] := 0 +FOR j := 0 to 7 + i := j*64 + k[j] := (a[i+63:i] == b[i+63:i]) ? 1 : 0 +ENDFOR +k[MAX:8] := 0 - +
immintrin.h
 - + Floating Point - AVX512VL - AVX512F - Store - - - - Store packed single-precision (32-bit) floating-point elements from "a" into memory using writemask "k". - "mem_addr" must be aligned on a 16-byte boundary or a general-protection exception may be generated. + Mask + AVX512F/KNCNI + Compare + + + + Compare packed double-precision (64-bit) floating-point elements in "a" and "b" for less-than-or-equal, and store the results in mask vector "k". -FOR j := 0 to 3 - i := j*32 - IF k[j] - MEM[mem_addr+i+31:mem_addr+i] := a[i+31:i] - FI +FOR j := 0 to 7 + i := j*64 + k[j] := (a[i+63:i] <= b[i+63:i]) ? 1 : 0 ENDFOR +k[MAX:8] := 0 - +
immintrin.h
 - + Floating Point - AVX512VL - AVX512F - Load - - - Load packed single-precision (32-bit) floating-point elements from memory into "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). "mem_addr" must be aligned on a 16-byte boundary or a general-protection exception may be generated. + Mask + AVX512F/KNCNI + Compare + + + + Compare packed double-precision (64-bit) floating-point elements in "a" and "b" for less-than, and store the results in mask vector "k". -FOR j := 0 to 3 - i := j*32 - IF k[j] - dst[i+31:i] := MEM[mem_addr+i+31:mem_addr+i] - ELSE - dst[i+31:i] := 0 - FI +FOR j := 0 to 7 + i := j*64 + k[j] := (a[i+63:i] < b[i+63:i]) ? 1 : 0 ENDFOR -dst[MAX:128] := 0 +k[MAX:8] := 0 - +
immintrin.h
 - + Floating Point - AVX512VL - AVX512F - Move - - - Move packed single-precision (32-bit) floating-point elements from "a" into "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Mask + AVX512F/KNCNI + Compare + + + + Compare packed double-precision (64-bit) floating-point elements in "a" and "b" for not-equal, and store the results in mask vector "k". -FOR j := 0 to 3 - i := j*32 - IF k[j] - dst[i+31:i] := a[i+31:i] - ELSE - dst[i+31:i] := 0 - FI +FOR j := 0 to 7 + i := j*64 + k[j] := (a[i+63:i] != b[i+63:i]) ? 1 : 0 ENDFOR -dst[MAX:128] := 0 +k[MAX:8] := 0 - +
immintrin.h
 - + Floating Point - AVX512VL - AVX512F - Move - - - - Duplicate even-indexed double-precision (64-bit) floating-point elements from "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Mask + AVX512F/KNCNI + Compare + + + + Compare packed double-precision (64-bit) floating-point elements in "a" and "b" for not-less-than-or-equal, and store the results in mask vector "k". -tmp[63:0] := a[63:0] -tmp[127:64] := a[63:0] -tmp[191:128] := a[191:128] -tmp[255:192] := a[191:128] -FOR j := 0 to 3 +FOR j := 0 to 7 i := j*64 - IF k[j] - dst[i+63:i] := tmp[i+63:i] - ELSE - dst[i+63:i] := src[i+63:i] - FI + k[j] := (!(a[i+63:i] <= b[i+63:i])) ? 1 : 0 ENDFOR -dst[MAX:256] := 0 +k[MAX:8] := 0 - +
immintrin.h
 - + Floating Point - AVX512VL - AVX512F - Move - - - Duplicate even-indexed double-precision (64-bit) floating-point elements from "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Mask + AVX512F/KNCNI + Compare + + + + Compare packed double-precision (64-bit) floating-point elements in "a" and "b" for not-less-than, and store the results in mask vector "k". -tmp[63:0] := a[63:0] -tmp[127:64] := a[63:0] -tmp[191:128] := a[191:128] -tmp[255:192] := a[191:128] -FOR j := 0 to 3 +FOR j := 0 to 7 i := j*64 - IF k[j] - dst[i+63:i] := tmp[i+63:i] - ELSE - dst[i+63:i] := 0 - FI + k[j] := (!(a[i+63:i] < b[i+63:i])) ? 1 : 0 ENDFOR -dst[MAX:256] := 0 +k[MAX:8] := 0 - +
immintrin.h
 - + Floating Point - AVX512VL - AVX512F - Move - - - - Duplicate even-indexed double-precision (64-bit) floating-point elements from "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -tmp[63:0] := a[63:0] -tmp[127:64] := a[63:0] -FOR j := 0 to 1 + Mask + AVX512F/KNCNI + Compare + + + + Compare packed double-precision (64-bit) floating-point elements in "a" and "b" to see if neither is NaN, and store the results in mask vector "k". + FOR j := 0 to 7 i := j*64 - IF k[j] - dst[i+63:i] := tmp[i+63:i] - ELSE - dst[i+63:i] := src[i+63:i] - FI + k[j] := (a[i+63:i] != NaN AND b[i+63:i] != NaN) ? 1 : 0 ENDFOR -dst[MAX:128] := 0 +k[MAX:8] := 0 - +
immintrin.h
 - + Floating Point - AVX512VL - AVX512F - Move - - - Duplicate even-indexed double-precision (64-bit) floating-point elements from "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - -tmp[63:0] := a[63:0] -tmp[127:64] := a[63:0] -FOR j := 0 to 1 + Mask + AVX512F/KNCNI + Compare + + + + Compare packed double-precision (64-bit) floating-point elements in "a" and "b" to see if either is NaN, and store the results in mask vector "k". + FOR j := 0 to 7 i := j*64 - IF k[j] - dst[i+63:i] := tmp[i+63:i] - ELSE - dst[i+63:i] := 0 - FI + k[j] := (a[i+63:i] == NaN OR b[i+63:i] == NaN) ? 1 : 0 ENDFOR -dst[MAX:128] := 0 +k[MAX:8] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Load - - - - Load packed 32-bit integers from memory into "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - "mem_addr" must be aligned on a 32-byte boundary or a general-protection exception may be generated. - + + Floating Point + Mask + AVX512F/KNCNI + Compare + + + + + + Compare packed double-precision (64-bit) floating-point elements in "a" and "b" based on the comparison operand specified by "imm8", and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). + CASE (imm8[4:0]) OF +0: OP := _CMP_EQ_OQ +1: OP := _CMP_LT_OS +2: OP := _CMP_LE_OS +3: OP := _CMP_UNORD_Q +4: OP := _CMP_NEQ_UQ +5: OP := _CMP_NLT_US +6: OP := _CMP_NLE_US +7: OP := _CMP_ORD_Q +8: OP := _CMP_EQ_UQ +9: OP := _CMP_NGE_US +10: OP := _CMP_NGT_US +11: OP := _CMP_FALSE_OQ +12: OP := _CMP_NEQ_OQ +13: OP := _CMP_GE_OS +14: OP := _CMP_GT_OS +15: OP := _CMP_TRUE_UQ +16: OP := _CMP_EQ_OS +17: OP := _CMP_LT_OQ +18: OP := _CMP_LE_OQ +19: OP := _CMP_UNORD_S +20: OP := _CMP_NEQ_US +21: OP := _CMP_NLT_UQ +22: OP := _CMP_NLE_UQ +23: OP := _CMP_ORD_S +24: OP := _CMP_EQ_US +25: OP := _CMP_NGE_UQ +26: OP := _CMP_NGT_UQ +27: OP := _CMP_FALSE_OS +28: OP := _CMP_NEQ_OS +29: OP := _CMP_GE_OQ +30: OP := _CMP_GT_OQ +31: OP := _CMP_TRUE_US +ESAC FOR j := 0 to 7 - i := j*32 - IF k[j] - dst[i+31:i] := MEM[mem_addr+i+31:mem_addr+i] + i := j*64 + IF k1[j] + k[j] := ( a[i+63:i] OP b[i+63:i] ) ? 1 : 0 ELSE - dst[i+31:i] := src[i+31:i] + k[j] := 0 FI ENDFOR -dst[MAX:256] := 0 +k[MAX:8] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Move - - - - Move packed 32-bit integers from "a" to "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - + + Floating Point + Mask + AVX512F/KNCNI + Compare + + + + + + + Compare packed double-precision (64-bit) floating-point elements in "a" and "b" based on the comparison operand specified by "imm8", and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). [sae_note] + CASE (imm8[4:0]) OF +0: OP := _CMP_EQ_OQ +1: OP := _CMP_LT_OS +2: OP := _CMP_LE_OS +3: OP := _CMP_UNORD_Q +4: OP := _CMP_NEQ_UQ +5: OP := _CMP_NLT_US +6: OP := _CMP_NLE_US +7: OP := _CMP_ORD_Q +8: OP := _CMP_EQ_UQ +9: OP := _CMP_NGE_US +10: OP := _CMP_NGT_US +11: OP := _CMP_FALSE_OQ +12: OP := _CMP_NEQ_OQ +13: OP := _CMP_GE_OS +14: OP := _CMP_GT_OS +15: OP := _CMP_TRUE_UQ +16: OP := _CMP_EQ_OS +17: OP := _CMP_LT_OQ +18: OP := _CMP_LE_OQ +19: OP := _CMP_UNORD_S +20: OP := _CMP_NEQ_US +21: OP := _CMP_NLT_UQ +22: OP := _CMP_NLE_UQ +23: OP := _CMP_ORD_S +24: OP := _CMP_EQ_US +25: OP := _CMP_NGE_UQ +26: OP := _CMP_NGT_UQ +27: OP := _CMP_FALSE_OS +28: OP := _CMP_NEQ_OS +29: OP := _CMP_GE_OQ +30: OP := _CMP_GT_OQ +31: OP := _CMP_TRUE_US +ESAC FOR j := 0 to 7 - i := j*32 - IF k[j] - dst[i+31:i] := a[i+31:i] - ELSE - dst[i+31:i] := src[i+31:i] + i := j*64 + IF k1[j] + k[j] := ( a[i+63:i] OP b[i+63:i] ) ? 1 : 0 + ELSE + k[j] := 0 FI ENDFOR -dst[MAX:256] := 0 +k[MAX:8] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Store - - - - Store packed 32-bit integers from "a" into memory using writemask "k". - "mem_addr" must be aligned on a 32-byte boundary or a general-protection exception may be generated. - + + Floating Point + Mask + AVX512F/KNCNI + Compare + + + + + Compare packed double-precision (64-bit) floating-point elements in "a" and "b" for equality, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). FOR j := 0 to 7 - i := j*32 - IF k[j] - MEM[mem_addr+i+31:mem_addr+i] := a[i+31:i] + i := j*64 + IF k1[j] + k[j] := (a[i+63:i] == b[i+63:i]) ? 1 : 0 + ELSE + k[j] := 0 FI -ENDFOR +ENDFOR +k[MAX:8] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Load - - - Load packed 32-bit integers from memory into "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - "mem_addr" must be aligned on a 32-byte boundary or a general-protection exception may be generated. + + Floating Point + Mask + AVX512F/KNCNI + Compare + + + + + Compare packed double-precision (64-bit) floating-point elements in "a" and "b" for less-than-or-equal, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). FOR j := 0 to 7 - i := j*32 - IF k[j] - dst[i+31:i] := MEM[mem_addr+i+31:mem_addr+i] - ELSE - dst[i+31:i] := 0 + i := j*64 + IF k1[j] + k[j] := (a[i+63:i] <= b[i+63:i]) ? 1 : 0 + ELSE + k[j] := 0 FI ENDFOR -dst[MAX:256] := 0 +k[MAX:8] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Move - - - Move packed 32-bit integers from "a" into "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Floating Point + Mask + AVX512F/KNCNI + Compare + + + + + Compare packed double-precision (64-bit) floating-point elements in "a" and "b" for less-than, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). FOR j := 0 to 7 - i := j*32 - IF k[j] - dst[i+31:i] := a[i+31:i] - ELSE - dst[i+31:i] := 0 + i := j*64 + IF k1[j] + k[j] := (a[i+63:i] < b[i+63:i]) ? 1 : 0 + ELSE + k[j] := 0 FI ENDFOR -dst[MAX:256] := 0 +k[MAX:8] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Load - - - - Load packed 32-bit integers from memory into "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - "mem_addr" must be aligned on a 16-byte boundary or a general-protection exception may be generated. + + Floating Point + Mask + AVX512F/KNCNI + Compare + + + + + Compare packed double-precision (64-bit) floating-point elements in "a" and "b" for not-equal, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*32 - IF k[j] - dst[i+31:i] := MEM[mem_addr+i+31:mem_addr+i] - ELSE - dst[i+31:i] := src[i+31:i] +FOR j := 0 to 7 + i := j*64 + IF k1[j] + k[j] := (a[i+63:i] != b[i+63:i]) ? 1 : 0 + ELSE + k[j] := 0 FI ENDFOR -dst[MAX:128] := 0 +k[MAX:8] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Move - - - - Move packed 32-bit integers from "a" to "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Floating Point + Mask + AVX512F/KNCNI + Compare + + + + + Compare packed double-precision (64-bit) floating-point elements in "a" and "b" for not-less-than-or-equal, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*32 - IF k[j] - dst[i+31:i] := a[i+31:i] - ELSE - dst[i+31:i] := src[i+31:i] +FOR j := 0 to 7 + i := j*64 + IF k1[j] + k[j] := (!(a[i+63:i] <= b[i+63:i])) ? 1 : 0 + ELSE + k[j] := 0 FI ENDFOR -dst[MAX:128] := 0 +k[MAX:8] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Store - - - - Store packed 32-bit integers from "a" into memory using writemask "k". - "mem_addr" must be aligned on a 16-byte boundary or a general-protection exception may be generated. - + + Floating Point + Mask + AVX512F/KNCNI + Compare + + + + + Compare packed double-precision (64-bit) floating-point elements in "a" and "b" for not-less-than, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*32 - IF k[j] - MEM[mem_addr+i+31:mem_addr+i] := a[i+31:i] +FOR j := 0 to 7 + i := j*64 + IF k1[j] + k[j] := (!(a[i+63:i] < b[i+63:i])) ? 1 : 0 + ELSE + k[j] := 0 FI ENDFOR +k[MAX:8] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Load - - - Load packed 32-bit integers from memory into "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - "mem_addr" must be aligned on a 16-byte boundary or a general-protection exception may be generated. - -FOR j := 0 to 3 - i := j*32 - IF k[j] - dst[i+31:i] := MEM[mem_addr+i+31:mem_addr+i] - ELSE - dst[i+31:i] := 0 + + Floating Point + Mask + AVX512F/KNCNI + Compare + + + + + Compare packed double-precision (64-bit) floating-point elements in "a" and "b" to see if neither is NaN, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). + FOR j := 0 to 7 + i := j*64 + IF k1[j] + k[j] := (a[i+63:i] != NaN AND b[i+63:i] != NaN) ? 1 : 0 + ELSE + k[j] := 0 FI ENDFOR -dst[MAX:128] := 0 +k[MAX:8] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Move - - - Move packed 32-bit integers from "a" into "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - -FOR j := 0 to 3 - i := j*32 - IF k[j] - dst[i+31:i] := a[i+31:i] - ELSE - dst[i+31:i] := 0 + + Floating Point + Mask + AVX512F/KNCNI + Compare + + + + + Compare packed double-precision (64-bit) floating-point elements in "a" and "b" to see if either is NaN, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). + FOR j := 0 to 7 + i := j*64 + IF k1[j] + k[j] := (a[i+63:i] == NaN OR b[i+63:i] == NaN) ? 1 : 0 + ELSE + k[j] := 0 FI ENDFOR -dst[MAX:128] := 0 +k[MAX:8] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Load - - - - Load packed 64-bit integers from memory into "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - "mem_addr" must be aligned on a 32-byte boundary or a general-protection exception may be generated. - -FOR j := 0 to 3 - i := j*64 - IF k[j] - dst[i+63:i] := MEM[mem_addr+i+63:mem_addr+i] - ELSE - dst[i+63:i] := src[i+63:i] - FI + + Floating Point + Mask + AVX512F/KNCNI + Compare + + + + + Compare packed single-precision (32-bit) floating-point elements in "a" and "b" based on the comparison operand specified by "imm8", and store the results in mask vector "k". + CASE (imm8[4:0]) OF +0: OP := _CMP_EQ_OQ +1: OP := _CMP_LT_OS +2: OP := _CMP_LE_OS +3: OP := _CMP_UNORD_Q +4: OP := _CMP_NEQ_UQ +5: OP := _CMP_NLT_US +6: OP := _CMP_NLE_US +7: OP := _CMP_ORD_Q +8: OP := _CMP_EQ_UQ +9: OP := _CMP_NGE_US +10: OP := _CMP_NGT_US +11: OP := _CMP_FALSE_OQ +12: OP := _CMP_NEQ_OQ +13: OP := _CMP_GE_OS +14: OP := _CMP_GT_OS +15: OP := _CMP_TRUE_UQ +16: OP := _CMP_EQ_OS +17: OP := _CMP_LT_OQ +18: OP := _CMP_LE_OQ +19: OP := _CMP_UNORD_S +20: OP := _CMP_NEQ_US +21: OP := _CMP_NLT_UQ +22: OP := _CMP_NLE_UQ +23: OP := _CMP_ORD_S +24: OP := _CMP_EQ_US +25: OP := _CMP_NGE_UQ +26: OP := _CMP_NGT_UQ +27: OP := _CMP_FALSE_OS +28: OP := _CMP_NEQ_OS +29: OP := _CMP_GE_OQ +30: OP := _CMP_GT_OQ +31: OP := _CMP_TRUE_US +ESAC +FOR j := 0 to 15 + i := j*32 + k[j] := (a[i+31:i] OP b[i+31:i]) ? 1 : 0 ENDFOR -dst[MAX:256] := 0 +k[MAX:16] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Move - - - - Move packed 64-bit integers from "a" to "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -FOR j := 0 to 3 - i := j*64 - IF k[j] - dst[i+63:i] := a[i+63:i] - ELSE - dst[i+63:i] := src[i+63:i] - FI + + Floating Point + Mask + AVX512F/KNCNI + Compare + + + + + + Compare packed single-precision (32-bit) floating-point elements in "a" and "b" based on the comparison operand specified by "imm8", and store the results in mask vector "k". [sae_note] + CASE (imm8[4:0]) OF +0: OP := _CMP_EQ_OQ +1: OP := _CMP_LT_OS +2: OP := _CMP_LE_OS +3: OP := _CMP_UNORD_Q +4: OP := _CMP_NEQ_UQ +5: OP := _CMP_NLT_US +6: OP := _CMP_NLE_US +7: OP := _CMP_ORD_Q +8: OP := _CMP_EQ_UQ +9: OP := _CMP_NGE_US +10: OP := _CMP_NGT_US +11: OP := _CMP_FALSE_OQ +12: OP := _CMP_NEQ_OQ +13: OP := _CMP_GE_OS +14: OP := _CMP_GT_OS +15: OP := _CMP_TRUE_UQ +16: OP := _CMP_EQ_OS +17: OP := _CMP_LT_OQ +18: OP := _CMP_LE_OQ +19: OP := _CMP_UNORD_S +20: OP := _CMP_NEQ_US +21: OP := _CMP_NLT_UQ +22: OP := _CMP_NLE_UQ +23: OP := _CMP_ORD_S +24: OP := _CMP_EQ_US +25: OP := _CMP_NGE_UQ +26: OP := _CMP_NGT_UQ +27: OP := _CMP_FALSE_OS +28: OP := _CMP_NEQ_OS +29: OP := _CMP_GE_OQ +30: OP := _CMP_GT_OQ +31: OP := _CMP_TRUE_US +ESAC +FOR j := 0 to 15 + i := j*32 + k[j] := (a[i+31:i] OP b[i+31:i]) ? 1 : 0 ENDFOR -dst[MAX:256] := 0 +k[MAX:16] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Store - - - - Store packed 64-bit integers from "a" into memory using writemask "k". - "mem_addr" must be aligned on a 32-byte boundary or a general-protection exception may be generated. - + + Floating Point + Mask + AVX512F/KNCNI + Compare + + + + Compare packed single-precision (32-bit) floating-point elements in "a" and "b" for equality, and store the results in mask vector "k". -FOR j := 0 to 3 - i := j*64 - IF k[j] - MEM[mem_addr+i+63:mem_addr+i] := a[i+63:i] - FI -ENDFOR +FOR j := 0 to 15 + i := j*32 + k[j] := (a[i+31:i] == b[i+31:i]) ? 1 : 0 +ENDFOR +k[MAX:16] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Load - - - Load packed 64-bit integers from memory into "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - "mem_addr" must be aligned on a 32-byte boundary or a general-protection exception may be generated. - + + Floating Point + Mask + AVX512F/KNCNI + Compare + + + + Compare packed single-precision (32-bit) floating-point elements in "a" and "b" for less-than-or-equal, and store the results in mask vector "k". -FOR j := 0 to 3 - i := j*64 - IF k[j] - dst[i+63:i] := MEM[mem_addr+i+63:mem_addr+i] - ELSE - dst[i+63:i] := 0 - FI +FOR j := 0 to 15 + i := j*32 + k[j] := (a[i+31:i] <= b[i+31:i]) ? 1 : 0 ENDFOR -dst[MAX:256] := 0 +k[MAX:16] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Move - - - Move packed 64-bit integers from "a" into "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Floating Point + Mask + AVX512F/KNCNI + Compare + + + + Compare packed single-precision (32-bit) floating-point elements in "a" and "b" for less-than, and store the results in mask vector "k". -FOR j := 0 to 3 - i := j*64 - IF k[j] - dst[i+63:i] := a[i+63:i] - ELSE - dst[i+63:i] := 0 - FI +FOR j := 0 to 15 + i := j*32 + k[j] := (a[i+31:i] < b[i+31:i]) ? 1 : 0 ENDFOR -dst[MAX:256] := 0 +k[MAX:16] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Load - - - - Load packed 64-bit integers from memory into "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - "mem_addr" must be aligned on a 16-byte boundary or a general-protection exception may be generated. + + Floating Point + Mask + AVX512F/KNCNI + Compare + + + + Compare packed single-precision (32-bit) floating-point elements in "a" and "b" for not-equal, and store the results in mask vector "k". -FOR j := 0 to 1 - i := j*64 - IF k[j] - dst[i+63:i] := MEM[mem_addr+i+63:mem_addr+i] - ELSE - dst[i+63:i] := src[i+63:i] - FI +FOR j := 0 to 15 + i := j*32 + k[j] := (a[i+31:i] != b[i+31:i]) ? 1 : 0 ENDFOR -dst[MAX:128] := 0 +k[MAX:16] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Move - - - - Move packed 64-bit integers from "a" to "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Floating Point + Mask + AVX512F/KNCNI + Compare + + + + Compare packed single-precision (32-bit) floating-point elements in "a" and "b" for not-less-than-or-equal, and store the results in mask vector "k". -FOR j := 0 to 1 - i := j*64 - IF k[j] - dst[i+63:i] := a[i+63:i] - ELSE - dst[i+63:i] := src[i+63:i] - FI +FOR j := 0 to 15 + i := j*32 + k[j] := (!(a[i+31:i] <= b[i+31:i])) ? 1 : 0 ENDFOR -dst[MAX:128] := 0 +k[MAX:16] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Store - - - - Store packed 64-bit integers from "a" into memory using writemask "k". - "mem_addr" must be aligned on a 16-byte boundary or a general-protection exception may be generated. - + + Floating Point + Mask + AVX512F/KNCNI + Compare + + + + Compare packed single-precision (32-bit) floating-point elements in "a" and "b" for not-less-than, and store the results in mask vector "k". -FOR j := 0 to 1 - i := j*64 - IF k[j] - MEM[mem_addr+i+63:mem_addr+i] := a[i+63:i] - FI +FOR j := 0 to 15 + i := j*32 + k[j] := (!(a[i+31:i] < b[i+31:i])) ? 1 : 0 ENDFOR +k[MAX:16] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Load - - - Load packed 64-bit integers from memory into "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - "mem_addr" must be aligned on a 16-byte boundary or a general-protection exception may be generated. - - -FOR j := 0 to 1 - i := j*64 - IF k[j] - dst[i+63:i] := MEM[mem_addr+i+63:mem_addr+i] - ELSE - dst[i+63:i] := 0 - FI + + Floating Point + Mask + AVX512F/KNCNI + Compare + + + + Compare packed single-precision (32-bit) floating-point elements in "a" and "b" to see if neither is NaN, and store the results in mask vector "k". + FOR j := 0 to 15 + i := j*32 + k[j] := ((a[i+31:i] != NaN) AND (b[i+31:i] != NaN)) ? 1 : 0 ENDFOR -dst[MAX:128] := 0 +k[MAX:16] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Move - - - Move packed 64-bit integers from "a" into "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - -FOR j := 0 to 1 - i := j*64 - IF k[j] - dst[i+63:i] := a[i+63:i] - ELSE - dst[i+63:i] := 0 + + Floating Point + Mask + AVX512F/KNCNI + Compare + + + + Compare packed single-precision (32-bit) floating-point elements in "a" and "b" to see if either is NaN, and store the results in mask vector "k". + FOR j := 0 to 15 + i := j*32 + k[j] := ((a[i+31:i] == NaN) OR (b[i+31:i] == NaN)) ? 1 : 0 +ENDFOR +k[MAX:16] := 0 + + +
immintrin.h
+ + + Floating Point + Mask + AVX512F/KNCNI + Compare + + + + + + Compare packed single-precision (32-bit) floating-point elements in "a" and "b" based on the comparison operand specified by "imm8", and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). + CASE (imm8[4:0]) OF +0: OP := _CMP_EQ_OQ +1: OP := _CMP_LT_OS +2: OP := _CMP_LE_OS +3: OP := _CMP_UNORD_Q +4: OP := _CMP_NEQ_UQ +5: OP := _CMP_NLT_US +6: OP := _CMP_NLE_US +7: OP := _CMP_ORD_Q +8: OP := _CMP_EQ_UQ +9: OP := _CMP_NGE_US +10: OP := _CMP_NGT_US +11: OP := _CMP_FALSE_OQ +12: OP := _CMP_NEQ_OQ +13: OP := _CMP_GE_OS +14: OP := _CMP_GT_OS +15: OP := _CMP_TRUE_UQ +16: OP := _CMP_EQ_OS +17: OP := _CMP_LT_OQ +18: OP := _CMP_LE_OQ +19: OP := _CMP_UNORD_S +20: OP := _CMP_NEQ_US +21: OP := _CMP_NLT_UQ +22: OP := _CMP_NLE_UQ +23: OP := _CMP_ORD_S +24: OP := _CMP_EQ_US +25: OP := _CMP_NGE_UQ +26: OP := _CMP_NGT_UQ +27: OP := _CMP_FALSE_OS +28: OP := _CMP_NEQ_OS +29: OP := _CMP_GE_OQ +30: OP := _CMP_GT_OQ +31: OP := _CMP_TRUE_US +ESAC +FOR j := 0 to 15 + i := j*32 + IF k1[j] + k[j] := ( a[i+31:i] OP b[i+31:i] ) ? 1 : 0 + ELSE + k[j] := 0 FI ENDFOR -dst[MAX:128] := 0 +k[MAX:16] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512BW - Load - - - - Load packed 16-bit integers from memory into "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - "mem_addr" does not need to be aligned on any particular boundary. - - + + Floating Point + Mask + AVX512F/KNCNI + Compare + + + + + + + Compare packed single-precision (32-bit) floating-point elements in "a" and "b" based on the comparison operand specified by "imm8", and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). [sae_note] + CASE (imm8[4:0]) OF +0: OP := _CMP_EQ_OQ +1: OP := _CMP_LT_OS +2: OP := _CMP_LE_OS +3: OP := _CMP_UNORD_Q +4: OP := _CMP_NEQ_UQ +5: OP := _CMP_NLT_US +6: OP := _CMP_NLE_US +7: OP := _CMP_ORD_Q +8: OP := _CMP_EQ_UQ +9: OP := _CMP_NGE_US +10: OP := _CMP_NGT_US +11: OP := _CMP_FALSE_OQ +12: OP := _CMP_NEQ_OQ +13: OP := _CMP_GE_OS +14: OP := _CMP_GT_OS +15: OP := _CMP_TRUE_UQ +16: OP := _CMP_EQ_OS +17: OP := _CMP_LT_OQ +18: OP := _CMP_LE_OQ +19: OP := _CMP_UNORD_S +20: OP := _CMP_NEQ_US +21: OP := _CMP_NLT_UQ +22: OP := _CMP_NLE_UQ +23: OP := _CMP_ORD_S +24: OP := _CMP_EQ_US +25: OP := _CMP_NGE_UQ +26: OP := _CMP_NGT_UQ +27: OP := _CMP_FALSE_OS +28: OP := _CMP_NEQ_OS +29: OP := _CMP_GE_OQ +30: OP := _CMP_GT_OQ +31: OP := _CMP_TRUE_US +ESAC FOR j := 0 to 15 - i := j*16 - IF k[j] - dst[i+15:i] := MEM[mem_addr+i+15:mem_addr+i] - ELSE - dst[i+15:i] := src[i+15:i] + i := j*32 + IF k1[j] + k[j] := ( a[i+31:i] OP b[i+31:i] ) ? 1 : 0 + ELSE + k[j] := 0 FI ENDFOR -dst[MAX:256] := 0 +k[MAX:16] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512BW - Move - - - - Move packed 16-bit integers from "a" into "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Floating Point + Mask + AVX512F/KNCNI + Compare + + + + + Compare packed single-precision (32-bit) floating-point elements in "a" and "b" for equality, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). FOR j := 0 to 15 - i := j*16 - IF k[j] - dst[i+15:i] := a[i+15:i] - ELSE - dst[i+15:i] := src[i+15:i] + i := j*32 + IF k1[j] + k[j] := (a[i+31:i] == b[i+31:i]) ? 1 : 0 + ELSE + k[j] := 0 FI -ENDFOR -dst[MAX:256] := 0 +ENDFOR +k[MAX:16] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512BW - Store - - - - Store packed 16-bit integers from "a" into memory using writemask "k". - "mem_addr" does not need to be aligned on any particular boundary. - + + Floating Point + Mask + AVX512F/KNCNI + Compare + + + + + Compare packed single-precision (32-bit) floating-point elements in "a" and "b" for less-than-or-equal, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). FOR j := 0 to 15 - i := j*16 - IF k[j] - MEM[mem_addr+i+15:mem_addr+i] := a[i+15:i] + i := j*32 + IF k1[j] + k[j] := (a[i+31:i] <= b[i+31:i]) ? 1 : 0 + ELSE + k[j] := 0 FI ENDFOR -dst[MAX:256] := 0 +k[MAX:16] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512BW - Load - - - Load packed 16-bit integers from memory into "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - "mem_addr" does not need to be aligned on any particular boundary. - + + Floating Point + Mask + AVX512F/KNCNI + Compare + + + + + Compare packed single-precision (32-bit) floating-point elements in "a" and "b" for less-than, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). FOR j := 0 to 15 - i := j*16 - IF k[j] - dst[i+15:i] := MEM[mem_addr+i+15:mem_addr+i] - ELSE - dst[i+15:i] := 0 + i := j*32 + IF k1[j] + k[j] := (a[i+31:i] < b[i+31:i]) ? 1 : 0 + ELSE + k[j] := 0 FI ENDFOR -dst[MAX:256] := 0 +k[MAX:16] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512BW - Move - - - Move packed 16-bit integers from "a" into "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - + + Floating Point + Mask + AVX512F/KNCNI + Compare + + + + + Compare packed single-precision (32-bit) floating-point elements in "a" and "b" for not-equal, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). FOR j := 0 to 15 - i := j*16 - IF k[j] - dst[i+15:i] := a[i+15:i] - ELSE - dst[i+15:i] := 0 + i := j*32 + IF k1[j] + k[j] := (a[i+31:i] != b[i+31:i]) ? 1 : 0 + ELSE + k[j] := 0 FI ENDFOR -dst[MAX:256] := 0 +k[MAX:16] := 0 - +
immintrin.h
 - - Integer - AVX512BW - Load - - - - Load packed 16-bit integers from memory into "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - "mem_addr" does not need to be aligned on any particular boundary. - + + Floating Point + Mask + AVX512F/KNCNI + Compare + + + + + Compare packed single-precision (32-bit) floating-point elements in "a" and "b" for not-less-than-or-equal, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 31 - i := j*16 - IF k[j] - dst[i+15:i] := MEM[mem_addr+i+15:mem_addr+i] - ELSE - dst[i+15:i] := src[i+15:i] +FOR j := 0 to 15 + i := j*32 + IF k1[j] + k[j] := (!(a[i+31:i] <= b[i+31:i])) ? 1 : 0 + ELSE + k[j] := 0 FI ENDFOR -dst[MAX:512] := 0 +k[MAX:16] := 0 - +
immintrin.h
 - - Integer - AVX512BW - Move - - - - Move packed 16-bit integers from "a" into "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Floating Point + Mask + AVX512F/KNCNI + Compare + + + + + Compare packed single-precision (32-bit) floating-point elements in "a" and "b" for not-less-than, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 31 - i := j*16 - IF k[j] - dst[i+15:i] := a[i+15:i] - ELSE - dst[i+15:i] := src[i+15:i] +FOR j := 0 to 15 + i := j*32 + IF k1[j] + k[j] := (!(a[i+31:i] < b[i+31:i])) ? 1 : 0 + ELSE + k[j] := 0 FI ENDFOR -dst[MAX:512] := 0 +k[MAX:16] := 0 - +
immintrin.h
 - - Integer - AVX512BW - Store - - - - Store packed 16-bit integers from "a" into memory using writemask "k". - "mem_addr" does not need to be aligned on any particular boundary. - - -FOR j := 0 to 31 - i := j*16 - IF k[j] - MEM[mem_addr+i+15:mem_addr+i] := a[i+15:i] + + Floating Point + Mask + AVX512F/KNCNI + Compare + + + + + Compare packed single-precision (32-bit) floating-point elements in "a" and "b" to see if neither is NaN, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). + FOR j := 0 to 15 + i := j*32 + IF k1[j] + k[j] := ((a[i+31:i] != NaN) AND (b[i+31:i] != NaN)) ? 1 : 0 + ELSE + k[j] := 0 FI ENDFOR -dst[MAX:512] := 0 +k[MAX:16] := 0 - +
immintrin.h
 - - Integer - AVX512BW - Load - - - Load packed 16-bit integers from memory into "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - "mem_addr" does not need to be aligned on any particular boundary. - - -FOR j := 0 to 31 - i := j*16 - IF k[j] - dst[i+15:i] := MEM[mem_addr+i+15:mem_addr+i] - ELSE - dst[i+15:i] := 0 + + Floating Point + Mask + AVX512F/KNCNI + Compare + + + + + Compare packed single-precision (32-bit) floating-point elements in "a" and "b" to see if either is NaN, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). + FOR j := 0 to 15 + i := j*32 + IF k1[j] + k[j] := ((a[i+31:i] == NaN) OR (b[i+31:i] == NaN)) ? 1 : 0 + ELSE + k[j] := 0 FI ENDFOR -dst[MAX:512] := 0 +k[MAX:16] := 0 - +
immintrin.h
 - - Integer - AVX512BW - Move - - - Move packed 16-bit integers from "a" into "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - + + Floating Point + AVX512F/KNCNI + Arithmetic + + + + + Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", add the intermediate result to packed elements in "c", and store the results in "dst". -FOR j := 0 to 31 - i := j*16 - IF k[j] - dst[i+15:i] := a[i+15:i] - ELSE - dst[i+15:i] := 0 - FI +FOR j := 0 to 7 + i := j*64 + dst[i+63:i] := (a[i+63:i] * b[i+63:i]) + c[i+63:i] ENDFOR dst[MAX:512] := 0 - + + +
immintrin.h
 - - Integer - AVX512VL - AVX512BW - Load - - - - Load packed 16-bit integers from memory into "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - "mem_addr" does not need to be aligned on any particular boundary. - + + Floating Point + AVX512F/KNCNI + Arithmetic + + + + + + Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", add the intermediate result to packed elements in "c", and store the results in "dst". + [round_note] FOR j := 0 to 7 - i := j*16 - IF k[j] - dst[i+15:i] := MEM[mem_addr+i+15:mem_addr+i] - ELSE - dst[i+15:i] := src[i+15:i] - FI + i := j*64 + dst[i+63:i] := (a[i+63:i] * b[i+63:i]) + c[i+63:i] ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - + + +
immintrin.h
 - - Integer - AVX512VL - AVX512BW - Move - - - - Move packed 16-bit integers from "a" into "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Floating Point + AVX512F/KNCNI + Arithmetic + + + + + + Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", add the intermediate result to packed elements in "c", and store the results in "dst" using writemask "k" (elements are copied from "c" when the corresponding mask bit is not set). FOR j := 0 to 7 - i := j*16 + i := j*64 IF k[j] - dst[i+15:i] := a[i+15:i] + dst[i+63:i] := (a[i+63:i] * b[i+63:i]) + c[i+63:i] ELSE - dst[i+15:i] := src[i+15:i] + dst[i+63:i] := c[i+63:i] FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - + + +
immintrin.h
 - - Integer - AVX512VL - AVX512BW - Store - - - - Store packed 16-bit integers from "a" into memory using writemask "k". - "mem_addr" does not need to be aligned on any particular boundary. - + + Floating Point + AVX512F/KNCNI + Arithmetic + + + + + + + Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", add the intermediate result to packed elements in "c", and store the results in "dst" using writemask "k" (elements are copied from "c" when the corresponding mask bit is not set). + [round_note] FOR j := 0 to 7 - i := j*16 + i := j*64 IF k[j] - MEM[mem_addr+i+15:mem_addr+i] := a[i+15:i] + dst[i+63:i] := (a[i+63:i] * b[i+63:i]) + c[i+63:i] + ELSE + dst[i+63:i] := c[i+63:i] FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - + + +
immintrin.h
 - - Integer - AVX512VL - AVX512BW - Load - - - Load packed 16-bit integers from memory into "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - "mem_addr" does not need to be aligned on any particular boundary. - + + Floating Point + AVX512F/KNCNI + Arithmetic + + + + + + Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", add the intermediate result to packed elements in "c", and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). FOR j := 0 to 7 - i := j*16 + i := j*64 IF k[j] - dst[i+15:i] := MEM[mem_addr+i+15:mem_addr+i] + dst[i+63:i] := (a[i+63:i] * b[i+63:i]) + c[i+63:i] ELSE - dst[i+15:i] := 0 + dst[i+63:i] := a[i+63:i] FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - + + +
immintrin.h
 - - Integer - AVX512VL - AVX512BW - Move - - - Move packed 16-bit integers from "a" into "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - + + Floating Point + AVX512F/KNCNI + Arithmetic + + + + + + + Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", add the intermediate result to packed elements in "c", and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). + [round_note] FOR j := 0 to 7 - i := j*16 + i := j*64 IF k[j] - dst[i+15:i] := a[i+15:i] + dst[i+63:i] := (a[i+63:i] * b[i+63:i]) + c[i+63:i] ELSE - dst[i+15:i] := 0 + dst[i+63:i] := a[i+63:i] FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - + + +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Load - - - - Load packed 32-bit integers from memory into "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - "mem_addr" does not need to be aligned on any particular boundary. - + + Floating Point + AVX512F/KNCNI + Arithmetic + + + + + Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", add the intermediate result to packed elements in "c", and store the results in "dst". -FOR j := 0 to 7 +FOR j := 0 to 15 i := j*32 - IF k[j] - dst[i+31:i] := MEM[mem_addr+i+31:mem_addr+i] - ELSE - dst[i+31:i] := src[i+31:i] - FI + dst[i+31:i] := (a[i+31:i] * b[i+31:i]) + c[i+31:i] ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - + + +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Store - - - - Store packed 32-bit integers from "a" into memory using writemask "k". - "mem_addr" does not need to be aligned on any particular boundary. - + + Floating Point + AVX512F/KNCNI + Arithmetic + + + + + + Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", add the intermediate result to packed elements in "c", and store the results in "dst". + [round_note] -FOR j := 0 to 7 +FOR j := 0 to 15 i := j*32 - IF k[j] - MEM[mem_addr+i+31:mem_addr+i] := a[i+31:i] - FI + dst[i+31:i] := (a[i+31:i] * b[i+31:i]) + c[i+31:i] ENDFOR +dst[MAX:512] := 0 - + + +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Load - - - Load packed 32-bit integers from memory into "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - "mem_addr" does not need to be aligned on any particular boundary. - + + Floating Point + AVX512F/KNCNI + Arithmetic + + + + + + Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", add the intermediate result to packed elements in "c", and store the results in "dst" using writemask "k" (elements are copied from "c" when the corresponding mask bit is not set). -FOR j := 0 to 7 +FOR j := 0 to 15 i := j*32 IF k[j] - dst[i+31:i] := MEM[mem_addr+i+31:mem_addr+i] + dst[i+31:i] := (a[i+31:i] * b[i+31:i]) + c[i+31:i] ELSE - dst[i+31:i] := 0 + dst[i+31:i] := c[i+31:i] FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - + + +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Load - - - - Load packed 32-bit integers from memory into "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - "mem_addr" does not need to be aligned on any particular boundary. - + + Floating Point + AVX512F/KNCNI + Arithmetic + + + + + + + Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", add the intermediate result to packed elements in "c", and store the results in "dst" using writemask "k" (elements are copied from "c" when the corresponding mask bit is not set). + [round_note] -FOR j := 0 to 3 +FOR j := 0 to 15 i := j*32 IF k[j] - dst[i+31:i] := MEM[mem_addr+i+31:mem_addr+i] + dst[i+31:i] := (a[i+31:i] * b[i+31:i]) + c[i+31:i] ELSE - dst[i+31:i] := src[i+31:i] + dst[i+31:i] := c[i+31:i] FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - + + +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Store - - - - Store packed 32-bit integers from "a" into memory using writemask "k". - "mem_addr" does not need to be aligned on any particular boundary. - + + Floating Point + AVX512F/KNCNI + Arithmetic + + + + + + Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", add the intermediate result to packed elements in "c", and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). -FOR j := 0 to 3 +FOR j := 0 to 15 i := j*32 IF k[j] - MEM[mem_addr+i+31:mem_addr+i] := a[i+31:i] + dst[i+31:i] := (a[i+31:i] * b[i+31:i]) + c[i+31:i] + ELSE + dst[i+31:i] := a[i+31:i] FI ENDFOR +dst[MAX:512] := 0 - + + +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Load - - - Load packed 32-bit integers from memory into "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - "mem_addr" does not need to be aligned on any particular boundary. - + + Floating Point + AVX512F/KNCNI + Arithmetic + + + + + + + Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", add the intermediate result to packed elements in "c", and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). + [round_note] -FOR j := 0 to 3 +FOR j := 0 to 15 i := j*32 IF k[j] - dst[i+31:i] := MEM[mem_addr+i+31:mem_addr+i] + dst[i+31:i] := (a[i+31:i] * b[i+31:i]) + c[i+31:i] ELSE - dst[i+31:i] := 0 + dst[i+31:i] := a[i+31:i] FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - + + +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Load - - - - Load packed 64-bit integers from memory into "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - "mem_addr" does not need to be aligned on any particular boundary. - + + Floating Point + AVX512F/KNCNI + Arithmetic + + + + + Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", subtract packed elements in "c" from the intermediate result, and store the results in "dst". -FOR j := 0 to 3 +FOR j := 0 to 7 i := j*64 - IF k[j] - dst[i+63:i] := MEM[mem_addr+i+63:mem_addr+i] - ELSE - dst[i+63:i] := src[i+63:i] - FI + dst[i+63:i] := (a[i+63:i] * b[i+63:i]) - c[i+63:i] ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - + + +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Store - - - - Store packed 64-bit integers from "a" into memory using writemask "k". - "mem_addr" does not need to be aligned on any particular boundary. - + + Floating Point + AVX512F/KNCNI + Arithmetic + + + + + + Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", subtract packed elements in "c" from the intermediate result, and store the results in "dst". + [round_note] -FOR j := 0 to 3 +FOR j := 0 to 7 i := j*64 - IF k[j] - MEM[mem_addr+i+63:mem_addr+i] := a[i+63:i] - FI + dst[i+63:i] := (a[i+63:i] * b[i+63:i]) - c[i+63:i] ENDFOR +dst[MAX:512] := 0 - + + +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Load - - - Load packed 64-bit integers from memory into "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - "mem_addr" does not need to be aligned on any particular boundary. - + + Floating Point + AVX512F/KNCNI + Arithmetic + + + + + + Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", subtract packed elements in "c" from the intermediate result, and store the results in "dst" using writemask "k" (elements are copied from "c" when the corresponding mask bit is not set). -FOR j := 0 to 3 +FOR j := 0 to 7 i := j*64 IF k[j] - dst[i+63:i] := MEM[mem_addr+i+63:mem_addr+i] + dst[i+63:i] := (a[i+63:i] * b[i+63:i]) - c[i+63:i] ELSE - dst[i+63:i] := 0 + dst[i+63:i] := c[i+63:i] FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - + + +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Load - - - - Load packed 64-bit integers from memory into "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - "mem_addr" does not need to be aligned on any particular boundary. - + + Floating Point + AVX512F/KNCNI + Arithmetic + + + + + + + Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", subtract packed elements in "c" from the intermediate result, and store the results in "dst" using writemask "k" (elements are copied from "c" when the corresponding mask bit is not set). [round_note] -FOR j := 0 to 1 +FOR j := 0 to 7 i := j*64 IF k[j] - dst[i+63:i] := MEM[mem_addr+i+63:mem_addr+i] + dst[i+63:i] := (a[i+63:i] * b[i+63:i]) - c[i+63:i] ELSE - dst[i+63:i] := src[i+63:i] + dst[i+63:i] := c[i+63:i] FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - + + +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Store - - - - Store packed 64-bit integers from "a" into memory using writemask "k". - "mem_addr" does not need to be aligned on any particular boundary. - + + Floating Point + AVX512F/KNCNI + Arithmetic + + + + + + Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", subtract packed elements in "c" from the intermediate result, and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). -FOR j := 0 to 1 +FOR j := 0 to 7 i := j*64 IF k[j] - MEM[mem_addr+i+63:mem_addr+i] := a[i+63:i] + dst[i+63:i] := (a[i+63:i] * b[i+63:i]) - c[i+63:i] + ELSE + dst[i+63:i] := a[i+63:i] FI ENDFOR +dst[MAX:512] := 0 - + + +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Load - - - Load packed 64-bit integers from memory into "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - "mem_addr" does not need to be aligned on any particular boundary. - + + Floating Point + AVX512F/KNCNI + Arithmetic + + + + + + + Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", subtract packed elements in "c" from the intermediate result, and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). [round_note] -FOR j := 0 to 1 +FOR j := 0 to 7 i := j*64 IF k[j] - dst[i+63:i] := MEM[mem_addr+i+63:mem_addr+i] + dst[i+63:i] := (a[i+63:i] * b[i+63:i]) - c[i+63:i] ELSE - dst[i+63:i] := 0 + dst[i+63:i] := a[i+63:i] FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - + + +
immintrin.h
 - - Integer - AVX512VL - AVX512BW - Load - - - - Load packed 8-bit integers from memory into "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - "mem_addr" does not need to be aligned on any particular boundary. - + + Floating Point + AVX512F/KNCNI + Arithmetic + + + + + Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", subtract packed elements in "c" from the intermediate result, and store the results in "dst". -FOR j := 0 to 31 - i := j*8 +FOR j := 0 to 15 + i := j*32 + dst[i+31:i] := (a[i+31:i] * b[i+31:i]) - c[i+31:i] +ENDFOR +dst[MAX:512] := 0 + + + + +
immintrin.h
+ + + Floating Point + AVX512F/KNCNI + Arithmetic + + + + + + Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", subtract packed elements in "c" from the intermediate result, and store the results in "dst". + [round_note] + +FOR j := 0 to 15 + i := j*32 + dst[i+31:i] := (a[i+31:i] * b[i+31:i]) - c[i+31:i] +ENDFOR +dst[MAX:512] := 0 + + + + +
immintrin.h
+ + + Floating Point + AVX512F/KNCNI + Arithmetic + + + + + + Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", subtract packed elements in "c" from the intermediate result, and store the results in "dst" using writemask "k" (elements are copied from "c" when the corresponding mask bit is not set). + +FOR j := 0 to 15 + i := j*32 IF k[j] - dst[i+7:i] := MEM[mem_addr+i+7:mem_addr+i] + dst[i+31:i] := (a[i+31:i] * b[i+31:i]) - c[i+31:i] ELSE - dst[i+7:i] := src[i+7:i] + dst[i+31:i] := c[i+31:i] FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - + + +
immintrin.h
 - - Integer - AVX512VL - AVX512BW - Move - - - - Move packed 8-bit integers from "a" into "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Floating Point + AVX512F/KNCNI + Arithmetic + + + + + + + Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", subtract packed elements in "c" from the intermediate result, and store the results in "dst" using writemask "k" (elements are copied from "c" when the corresponding mask bit is not set). [round_note] -FOR j := 0 to 31 - i := j*8 +FOR j := 0 to 15 + i := j*32 IF k[j] - dst[i+7:i] := a[i+7:i] + dst[i+31:i] := (a[i+31:i] * b[i+31:i]) - c[i+31:i] ELSE - dst[i+7:i] := src[i+7:i] + dst[i+31:i] := c[i+31:i] FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - + + +
immintrin.h
 - - Integer - AVX512VL - AVX512BW - Store - - - - Store packed 8-bit integers from "a" into memory using writemask "k". - "mem_addr" does not need to be aligned on any particular boundary. - + + Floating Point + AVX512F/KNCNI + Arithmetic + + + + + + Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", subtract packed elements in "c" from the intermediate result, and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). -FOR j := 0 to 31 - i := j*8 +FOR j := 0 to 15 + i := j*32 IF k[j] - MEM[mem_addr+i+7:mem_addr+i] := a[i+7:i] + dst[i+31:i] := (a[i+31:i] * b[i+31:i]) - c[i+31:i] + ELSE + dst[i+31:i] := a[i+31:i] FI ENDFOR +dst[MAX:512] := 0 - + + +
immintrin.h
 - - Integer - AVX512VL - AVX512BW - Load - - - Load packed 8-bit integers from memory into "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - "mem_addr" does not need to be aligned on any particular boundary. - + + Floating Point + AVX512F/KNCNI + Arithmetic + + + + + + + Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", subtract packed elements in "c" from the intermediate result, and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). [round_note] -FOR j := 0 to 31 - i := j*8 +FOR j := 0 to 15 + i := j*32 IF k[j] - dst[i+7:i] := MEM[mem_addr+i+7:mem_addr+i] + dst[i+31:i] := (a[i+31:i] * b[i+31:i]) - c[i+31:i] ELSE - dst[i+7:i] := 0 + dst[i+31:i] := a[i+31:i] FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - + + +
immintrin.h
 - - Integer - AVX512VL - AVX512BW - Move - - - Move packed 8-bit integers from "a" into "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - + + Floating Point + AVX512F/KNCNI + Arithmetic + + + + + Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", add the negated intermediate result to packed elements in "c", and store the results in "dst". -FOR j := 0 to 31 - i := j*8 +FOR j := 0 to 7 + i := j*64 + dst[i+63:i] := -(a[i+63:i] * b[i+63:i]) + c[i+63:i] +ENDFOR +dst[MAX:512] := 0 + + + + +
immintrin.h
+ + + Floating Point + AVX512F/KNCNI + Arithmetic + + + + + + Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", add the negated intermediate result to packed elements in "c", and store the results in "dst". + [round_note] + +FOR j := 0 to 7 + i := j*64 + dst[i+63:i] := -(a[i+63:i] * b[i+63:i]) + c[i+63:i] +ENDFOR +dst[MAX:512] := 0 + + + + +
immintrin.h
+ + + Floating Point + AVX512F/KNCNI + Arithmetic + + + + + + Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", add the negated intermediate result to packed elements in "c", and store the results in "dst" using writemask "k" (elements are copied from "c" when the corresponding mask bit is not set). + +FOR j := 0 to 7 + i := j*64 IF k[j] - dst[i+7:i] := a[i+7:i] + dst[i+63:i] := -(a[i+63:i] * b[i+63:i]) + c[i+63:i] ELSE - dst[i+7:i] := 0 + dst[i+63:i] := c[i+63:i] FI -ENDFOR -dst[MAX:256] := 0 +ENDFOR +dst[MAX:512] := 0 - + + +
immintrin.h
 - - Integer - AVX512BW - Load - - - - Load packed 8-bit integers from memory into "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - "mem_addr" does not need to be aligned on any particular boundary. - + + Floating Point + AVX512F/KNCNI + Arithmetic + + + + + + + Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", add the negated intermediate result to packed elements in "c", and store the results in "dst" using writemask "k" (elements are copied from "c" when the corresponding mask bit is not set). [round_note] -FOR j := 0 to 63 - i := j*8 +FOR j := 0 to 7 + i := j*64 IF k[j] - dst[i+7:i] := MEM[mem_addr+i+7:mem_addr+i] + dst[i+63:i] := -(a[i+63:i] * b[i+63:i]) + c[i+63:i] ELSE - dst[i+7:i] := src[i+7:i] + dst[i+63:i] := c[i+63:i] FI -ENDFOR +ENDFOR dst[MAX:512] := 0 - + + +
immintrin.h
 - - Integer - AVX512BW - Move - - - - Move packed 8-bit integers from "a" into "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Floating Point + AVX512F/KNCNI + Arithmetic + + + + + + Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", add the negated intermediate result to packed elements in "c", and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). -FOR j := 0 to 63 - i := j*8 +FOR j := 0 to 7 + i := j*64 IF k[j] - dst[i+7:i] := a[i+7:i] + dst[i+63:i] := -(a[i+63:i] * b[i+63:i]) + c[i+63:i] ELSE - dst[i+7:i] := src[i+7:i] + dst[i+63:i] := a[i+63:i] FI -ENDFOR +ENDFOR dst[MAX:512] := 0 - + + +
immintrin.h
 - - Integer - AVX512BW - Store - - - - Store packed 8-bit integers from "a" into memory using writemask "k". - "mem_addr" does not need to be aligned on any particular boundary. - + + Floating Point + AVX512F/KNCNI + Arithmetic + + + + + + + Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", add the negated intermediate result to packed elements in "c", and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). [round_note] -FOR j := 0 to 63 - i := j*8 +FOR j := 0 to 7 + i := j*64 IF k[j] - MEM[mem_addr+i+7:mem_addr+i] := a[i+7:i] + dst[i+63:i] := -(a[i+63:i] * b[i+63:i]) + c[i+63:i] + ELSE + dst[i+63:i] := a[i+63:i] FI -ENDFOR +ENDFOR +dst[MAX:512] := 0 - + + +
immintrin.h
 - - Integer - AVX512BW - Load - - - Load packed 8-bit integers from memory into "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - "mem_addr" does not need to be aligned on any particular boundary. - + + Floating Point + AVX512F/KNCNI + Arithmetic + + + + + Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", add the negated intermediate result to packed elements in "c", and store the results in "dst". -FOR j := 0 to 63 - i := j*8 +FOR j := 0 to 15 + i := j*32 + dst[i+31:i] := -(a[i+31:i] * b[i+31:i]) + c[i+31:i] +ENDFOR +dst[MAX:512] := 0 + + + + +
immintrin.h
+ + + Floating Point + AVX512F/KNCNI + Arithmetic + + + + + + Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", add the negated intermediate result to packed elements in "c", and store the results in "dst". + [round_note] + +FOR j := 0 to 15 + i := j*32 + dst[i+31:i] := -(a[i+31:i] * b[i+31:i]) + c[i+31:i] +ENDFOR +dst[MAX:512] := 0 + + + + +
immintrin.h
+ + + Floating Point + AVX512F/KNCNI + Arithmetic + + + + + + Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", add the negated intermediate result to packed elements in "c", and store the results in "dst" using writemask "k" (elements are copied from "c" when the corresponding mask bit is not set). + +FOR j := 0 to 15 + i := j*32 IF k[j] - dst[i+7:i] := MEM[mem_addr+i+7:mem_addr+i] + dst[i+31:i] := -(a[i+31:i] * b[i+31:i]) + c[i+31:i] ELSE - dst[i+7:i] := 0 + dst[i+31:i] := c[i+31:i] FI -ENDFOR +ENDFOR dst[MAX:512] := 0 - + + +
immintrin.h
 - - Integer - AVX512BW - Move - - - Move packed 8-bit integers from "a" into "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - + + Floating Point + AVX512F/KNCNI + Arithmetic + + + + + + + Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", add the negated intermediate result to packed elements in "c", and store the results in "dst" using writemask "k" (elements are copied from "c" when the corresponding mask bit is not set). [round_note] -FOR j := 0 to 63 - i := j*8 +FOR j := 0 to 15 + i := j*32 IF k[j] - dst[i+7:i] := a[i+7:i] + dst[i+31:i] := -(a[i+31:i] * b[i+31:i]) + c[i+31:i] ELSE - dst[i+7:i] := 0 + dst[i+31:i] := c[i+31:i] FI -ENDFOR +ENDFOR dst[MAX:512] := 0 - + + +
immintrin.h
 - - Integer - AVX512VL - AVX512BW - Load - - - - Load packed 8-bit integers from memory into "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - "mem_addr" does not need to be aligned on any particular boundary. - + + Floating Point + AVX512F/KNCNI + Arithmetic + + + + + + Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", add the negated intermediate result to packed elements in "c", and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). FOR j := 0 to 15 - i := j*8 + i := j*32 IF k[j] - dst[i+7:i] := MEM[mem_addr+i+7:mem_addr+i] + dst[i+31:i] := -(a[i+31:i] * b[i+31:i]) + c[i+31:i] ELSE - dst[i+7:i] := src[i+7:i] + dst[i+31:i] := a[i+31:i] FI -ENDFOR -dst[MAX:128] := 0 +ENDFOR +dst[MAX:512] := 0 - + + +
immintrin.h
 - - Integer - AVX512VL - AVX512BW - Move - - - - Move packed 8-bit integers from "a" into "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Floating Point + AVX512F/KNCNI + Arithmetic + + + + + + + Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", add the negated intermediate result to packed elements in "c", and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). [round_note] FOR j := 0 to 15 - i := j*8 + i := j*32 IF k[j] - dst[i+7:i] := a[i+7:i] + dst[i+31:i] := -(a[i+31:i] * b[i+31:i]) + c[i+31:i] ELSE - dst[i+7:i] := src[i+7:i] + dst[i+31:i] := a[i+31:i] FI -ENDFOR -dst[MAX:128] := 0 +ENDFOR +dst[MAX:512] := 0 - + + +
immintrin.h
 - - Integer - AVX512VL - AVX512BW - Store - - - - Store packed 8-bit integers from "a" into memory using writemask "k". - "mem_addr" does not need to be aligned on any particular boundary. - + + Floating Point + AVX512F/KNCNI + Arithmetic + + + + + Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", subtract packed elements in "c" from the negated intermediate result, and store the results in "dst". -FOR j := 0 to 15 - i := j*8 - IF k[j] - MEM[mem_addr+i+7:mem_addr+i] := a[i+7:i] - FI -ENDFOR +FOR j := 0 to 7 + i := j*64 + dst[i+63:i] := -(a[i+63:i] * b[i+63:i]) - c[i+63:i] +ENDFOR +dst[MAX:512] := 0 - + + +
immintrin.h
 - - Integer - AVX512VL - AVX512BW - Load - - - Load packed 8-bit integers from memory into "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - "mem_addr" does not need to be aligned on any particular boundary. - + + Floating Point + AVX512F/KNCNI + Arithmetic + + + + + + Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", subtract packed elements in "c" from the negated intermediate result, and store the results in "dst". + [round_note] -FOR j := 0 to 15 - i := j*8 - IF k[j] - dst[i+7:i] := MEM[mem_addr+i+7:mem_addr+i] - ELSE - dst[i+7:i] := 0 - FI -ENDFOR -dst[MAX:128] := 0 +FOR j := 0 to 7 + i := j*64 + dst[i+63:i] := -(a[i+63:i] * b[i+63:i]) - c[i+63:i] +ENDFOR +dst[MAX:512] := 0 - + + +
immintrin.h
 - - Integer - AVX512VL - AVX512BW - Move - - - Move packed 8-bit integers from "a" into "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - + + Floating Point + AVX512F/KNCNI + Arithmetic + + + + + + Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", subtract packed elements in "c" from the negated intermediate result, and store the results in "dst" using writemask "k" (elements are copied from "c" when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*8 +FOR j := 0 to 7 + i := j*64 IF k[j] - dst[i+7:i] := a[i+7:i] + dst[i+63:i] := -(a[i+63:i] * b[i+63:i]) - c[i+63:i] ELSE - dst[i+7:i] := 0 + dst[i+63:i] := c[i+63:i] FI -ENDFOR -dst[MAX:128] := 0 +ENDFOR +dst[MAX:512] := 0 - + + +
immintrin.h
 - + Floating Point - AVX512VL - AVX512F - Move - - - - Duplicate odd-indexed single-precision (32-bit) floating-point elements from "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + AVX512F/KNCNI + Arithmetic + + + + + + + Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", subtract packed elements in "c" from the negated intermediate result, and store the results in "dst" using writemask "k" (elements are copied from "c" when the corresponding mask bit is not set). [round_note] -tmp[31:0] := a[63:32] -tmp[63:32] := a[63:32] -tmp[95:64] := a[127:96] -tmp[127:96] := a[127:96] -tmp[159:128] := a[191:160] -tmp[191:160] := a[191:160] -tmp[223:192] := a[255:224] -tmp[255:224] := a[255:224] FOR j := 0 to 7 - i := j*32 + i := j*64 IF k[j] - dst[i+31:i] := tmp[i+31:i] + dst[i+63:i] := -(a[i+63:i] * b[i+63:i]) - c[i+63:i] ELSE - dst[i+31:i] := src[i+31:i] + dst[i+63:i] := c[i+63:i] FI -ENDFOR -dst[MAX:256] := 0 +ENDFOR +dst[MAX:512] := 0 - + + +
immintrin.h
 - + Floating Point - AVX512VL - AVX512F - Move - - - Duplicate odd-indexed single-precision (32-bit) floating-point elements from "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + AVX512F/KNCNI + Arithmetic + + + + + + Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", subtract packed elements in "c" from the negated intermediate result, and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). -tmp[31:0] := a[63:32] -tmp[63:32] := a[63:32] -tmp[95:64] := a[127:96] -tmp[127:96] := a[127:96] -tmp[159:128] := a[191:160] -tmp[191:160] := a[191:160] -tmp[223:192] := a[255:224] -tmp[255:224] := a[255:224] FOR j := 0 to 7 - i := j*32 + i := j*64 IF k[j] - dst[i+31:i] := tmp[i+31:i] + dst[i+63:i] := -(a[i+63:i] * b[i+63:i]) - c[i+63:i] ELSE - dst[i+31:i] := 0 + dst[i+63:i] := a[i+63:i] FI -ENDFOR -dst[MAX:256] := 0 +ENDFOR +dst[MAX:512] := 0 - + + +
immintrin.h
 - + Floating Point - AVX512VL - AVX512F - Move - - - - Duplicate odd-indexed single-precision (32-bit) floating-point elements from "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + AVX512F/KNCNI + Arithmetic + + + + + + + Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", subtract packed elements in "c" from the negated intermediate result, and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). + [round_note] -tmp[31:0] := a[63:32] -tmp[63:32] := a[63:32] -tmp[95:64] := a[127:96] -tmp[127:96] := a[127:96] -FOR j := 0 to 3 - i := j*32 +FOR j := 0 to 7 + i := j*64 IF k[j] - dst[i+31:i] := tmp[i+31:i] + dst[i+63:i] := -(a[i+63:i] * b[i+63:i]) - c[i+63:i] ELSE - dst[i+31:i] := src[i+31:i] + dst[i+63:i] := a[i+63:i] FI -ENDFOR -dst[MAX:128] := 0 +ENDFOR +dst[MAX:512] := 0 - + + +
immintrin.h
 - + Floating Point - AVX512VL - AVX512F - Move - - - Duplicate odd-indexed single-precision (32-bit) floating-point elements from "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + AVX512F/KNCNI + Arithmetic + + + + + Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", subtract packed elements in "c" from the negated intermediate result, and store the results in "dst". -tmp[31:0] := a[63:32] -tmp[63:32] := a[63:32] -tmp[95:64] := a[127:96] -tmp[127:96] := a[127:96] -FOR j := 0 to 3 +FOR j := 0 to 15 i := j*32 - IF k[j] - dst[i+31:i] := tmp[i+31:i] - ELSE - dst[i+31:i] := 0 - FI -ENDFOR -dst[MAX:128] := 0 + dst[i+31:i] := -(a[i+31:i] * b[i+31:i]) - c[i+31:i] +ENDFOR +dst[MAX:512] := 0 - + + +
immintrin.h
 - + Floating Point - AVX512VL - AVX512F - Move - - - - Duplicate even-indexed single-precision (32-bit) floating-point elements from "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + AVX512F/KNCNI + Arithmetic + + + + + + Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", subtract packed elements in "c" from the negated intermediate result, and store the results in "dst". + [round_note] -tmp[31:0] := a[31:0] -tmp[63:32] := a[31:0] -tmp[95:64] := a[95:64] -tmp[127:96] := a[95:64] -tmp[159:128] := a[159:128] -tmp[191:160] := a[159:128] -tmp[223:192] := a[223:192] -tmp[255:224] := a[223:192] -FOR j := 0 to 7 +FOR j := 0 to 15 i := j*32 - IF k[j] - dst[i+31:i] := tmp[i+31:i] - ELSE - dst[i+31:i] := src[i+31:i] - FI + dst[i+31:i] := -(a[i+31:i] * b[i+31:i]) - c[i+31:i] ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - + + +
immintrin.h
 - + Floating Point - AVX512VL - AVX512F - Move - - - Duplicate even-indexed single-precision (32-bit) floating-point elements from "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + AVX512F/KNCNI + Arithmetic + + + + + + Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", subtract packed elements in "c" from the negated intermediate result, and store the results in "dst" using writemask "k" (elements are copied from "c" when the corresponding mask bit is not set). -tmp[31:0] := a[31:0] -tmp[63:32] := a[31:0] -tmp[95:64] := a[95:64] -tmp[127:96] := a[95:64] -tmp[159:128] := a[159:128] -tmp[191:160] := a[159:128] -tmp[223:192] := a[223:192] -tmp[255:224] := a[223:192] -FOR j := 0 to 7 +FOR j := 0 to 15 i := j*32 IF k[j] - dst[i+31:i] := tmp[i+31:i] + dst[i+31:i] := -(a[i+31:i] * b[i+31:i]) - c[i+31:i] ELSE - dst[i+31:i] := 0 + dst[i+31:i] := c[i+31:i] FI -ENDFOR -dst[MAX:256] := 0 +ENDFOR +dst[MAX:512] := 0 - + + +
immintrin.h
 - + Floating Point - AVX512VL - AVX512F - Move - - - - Duplicate even-indexed single-precision (32-bit) floating-point elements from "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + AVX512F/KNCNI + Arithmetic + + + + + + + Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", subtract packed elements in "c" from the negated intermediate result, and store the results in "dst" using writemask "k" (elements are copied from "c" when the corresponding mask bit is not set). [round_note] -tmp[31:0] := a[31:0] -tmp[63:32] := a[31:0] -tmp[95:64] := a[95:64] -tmp[127:96] := a[95:64] -FOR j := 0 to 3 +FOR j := 0 to 15 i := j*32 IF k[j] - dst[i+31:i] := tmp[i+31:i] + dst[i+31:i] := -(a[i+31:i] * b[i+31:i]) - c[i+31:i] ELSE - dst[i+31:i] := src[i+31:i] + dst[i+31:i] := c[i+31:i] FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - + + +
immintrin.h
 - + Floating Point - AVX512VL - AVX512F - Move - - - Duplicate even-indexed single-precision (32-bit) floating-point elements from "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + AVX512F/KNCNI + Arithmetic + + + + + + Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", subtract packed elements in "c" from the negated intermediate result, and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). -tmp[31:0] := a[31:0] -tmp[63:32] := a[31:0] -tmp[95:64] := a[95:64] -tmp[127:96] := a[95:64] -FOR j := 0 to 3 +FOR j := 0 to 15 i := j*32 IF k[j] - dst[i+31:i] := tmp[i+31:i] + dst[i+31:i] := -(a[i+31:i] * b[i+31:i]) - c[i+31:i] ELSE - dst[i+31:i] := 0 + dst[i+31:i] := a[i+31:i] FI -ENDFOR -dst[MAX:128] := 0 +ENDFOR +dst[MAX:512] := 0 - + + +
immintrin.h
 - + Floating Point - AVX512VL - AVX512F - Load - - - - Load packed double-precision (64-bit) floating-point elements from memoy into "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - "mem_addr" does not need to be aligned on any particular boundary. + AVX512F/KNCNI + Arithmetic + + + + + + + Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", subtract packed elements in "c" from the negated intermediate result, and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). + [round_note] -FOR j := 0 to 3 - i := j*64 +FOR j := 0 to 15 + i := j*32 IF k[j] - dst[i+63:i] := MEM[mem_addr+i+63:mem_addr+i] + dst[i+31:i] := -(a[i+31:i] * b[i+31:i]) - c[i+31:i] ELSE - dst[i+63:i] := src[i+63:i] + dst[i+31:i] := a[i+31:i] FI -ENDFOR -dst[MAX:256] := 0 +ENDFOR +dst[MAX:512] := 0 - + + +
immintrin.h
 - + Floating Point - AVX512VL - AVX512F - Store - - - - Store packed double-precision (64-bit) floating-point elements from "a" into memory using writemask "k". - "mem_addr" does not need to be aligned on any particular boundary. + AVX512F/KNCNI + Load + + + + + Gather single-precision (32-bit) floating-point elements from memory using 32-bit indices. 32-bit elements are loaded from addresses starting at "base_addr" and offset by each 32-bit element in "vindex" (each index is scaled by the factor in "scale"). Gathered elements are merged into "dst". "scale" should be 1, 2, 4 or 8. -FOR j := 0 to 3 - i := j*64 - IF k[j] - MEM[mem_addr+i+63:mem_addr+i] := a[i+63:i] - FI +FOR j := 0 to 15 + i := j*32 + m := j*32 + addr := base_addr + SignExtend64(vindex[m+31:m]) * ZeroExtend64(scale) * 8 + dst[i+31:i] := MEM[addr+31:addr] ENDFOR +dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - AVX512VL - AVX512F + AVX512F/KNCNI Load - - - Load packed double-precision (64-bit) floating-point elements from memoy into "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - "mem_addr" does not need to be aligned on any particular boundary. - + + + + + + + Gather single-precision (32-bit) floating-point elements from memory using 32-bit indices. 32-bit elements are loaded from addresses starting at "base_addr" and offset by each 32-bit element in "vindex" (each index is scaled by the factor in "scale"). Gathered elements are merged into "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). "scale" should be 1, 2, 4 or 8. -FOR j := 0 to 3 - i := j*64 +FOR j := 0 to 15 + i := j*32 + m := j*32 IF k[j] - dst[i+63:i] := MEM[mem_addr+i+63:mem_addr+i] + addr := base_addr + SignExtend64(vindex[m+31:m]) * ZeroExtend64(scale) * 8 + dst[i+31:i] := MEM[addr+31:addr] ELSE - dst[i+63:i] := 0 + dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - AVX512VL - AVX512F - Load - - - - Load packed double-precision (64-bit) floating-point elements from memoy into "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - "mem_addr" does not need to be aligned on any particular boundary. - -FOR j := 0 to 1 + AVX512F/KNCNI + Miscellaneous + + + Convert the exponent of each packed double-precision (64-bit) floating-point element in "a" to a double-precision (64-bit) floating-point number representing the integer exponent, and store the results in "dst". This intrinsic essentially calculates "floor(log2(x))" for each element. + FOR j := 0 to 7 i := j*64 - IF k[j] - dst[i+63:i] := MEM[mem_addr+i+63:mem_addr+i] - ELSE - dst[i+63:i] := src[i+63:i] - FI + dst[i+63:i] := ConvertExpFP64(a[i+63:i]) ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - AVX512VL - AVX512F - Store - - - - Store packed double-precision (64-bit) floating-point elements from "a" into memory using writemask "k". - "mem_addr" does not need to be aligned on any particular boundary. - -FOR j := 0 to 1 + AVX512F/KNCNI + Miscellaneous + + + + Convert the exponent of each packed double-precision (64-bit) floating-point element in "a" to a double-precision (64-bit) floating-point number representing the integer exponent, and store the results in "dst". This intrinsic essentially calculates "floor(log2(x))" for each element. + [sae_note] + FOR j := 0 to 7 i := j*64 - IF k[j] - MEM[mem_addr+i+63:mem_addr+i] := a[i+63:i] - FI + dst[i+63:i] := ConvertExpFP64(a[i+63:i]) ENDFOR +dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - AVX512VL - AVX512F - Load - - - Load packed double-precision (64-bit) floating-point elements from memoy into "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - "mem_addr" does not need to be aligned on any particular boundary. - - -FOR j := 0 to 1 + AVX512F/KNCNI + Miscellaneous + + + + + Convert the exponent of each packed double-precision (64-bit) floating-point element in "a" to a double-precision (64-bit) floating-point number representing the integer exponent, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). This intrinsic essentially calculates "floor(log2(x))" for each element. + FOR j := 0 to 7 i := j*64 IF k[j] - dst[i+63:i] := MEM[mem_addr+i+63:mem_addr+i] + dst[i+63:i] := ConvertExpFP64(a[i+63:i]) ELSE - dst[i+63:i] := 0 + dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - AVX512VL - AVX512F - Load - - - - Load packed single-precision (32-bit) floating-point elements from memory into "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - "mem_addr" does not need to be aligned on any particular boundary. - - -FOR j := 0 to 7 - i := j*32 + AVX512F/KNCNI + Miscellaneous + + + + + + Convert the exponent of each packed double-precision (64-bit) floating-point element in "a" to a double-precision (64-bit) floating-point number representing the integer exponent, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). This intrinsic essentially calculates "floor(log2(x))" for each element. + [sae_note] + FOR j := 0 to 7 + i := j*64 IF k[j] - dst[i+31:i] := MEM[mem_addr+i+31:mem_addr+i] + dst[i+63:i] := ConvertExpFP64(a[i+63:i]) ELSE - dst[i+31:i] := src[i+31:i] + dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - AVX512VL - AVX512F - Store - - - - Store packed single-precision (32-bit) floating-point elements from "a" into memory using writemask "k". - "mem_addr" does not need to be aligned on any particular boundary. - - -FOR j := 0 to 7 + AVX512F/KNCNI + Miscellaneous + + + Convert the exponent of each packed single-precision (32-bit) floating-point element in "a" to a single-precision (32-bit) floating-point number representing the integer exponent, and store the results in "dst". This intrinsic essentially calculates "floor(log2(x))" for each element. + FOR j := 0 to 15 i := j*32 - IF k[j] - MEM[mem_addr+i+31:mem_addr+i] := a[i+31:i] - FI + dst[i+31:i] := ConvertExpFP32(a[i+31:i]) ENDFOR +dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - AVX512VL - AVX512F - Load - - - Load packed single-precision (32-bit) floating-point elements from memory into "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - "mem_addr" does not need to be aligned on any particular boundary. - - -FOR j := 0 to 7 + AVX512F/KNCNI + Miscellaneous + + + + Convert the exponent of each packed single-precision (32-bit) floating-point element in "a" to a single-precision (32-bit) floating-point number representing the integer exponent, and store the results in "dst". This intrinsic essentially calculates "floor(log2(x))" for each element. + [sae_note] + FOR j := 0 to 15 i := j*32 - IF k[j] - dst[i+31:i] := MEM[mem_addr+i+31:mem_addr+i] - ELSE - dst[i+31:i] := 0 - FI + dst[i+31:i] := ConvertExpFP32(a[i+31:i]) ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - AVX512VL - AVX512F - Load - - - - Load packed single-precision (32-bit) floating-point elements from memory into "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - "mem_addr" does not need to be aligned on any particular boundary. - - -FOR j := 0 to 3 + AVX512F/KNCNI + Miscellaneous + + + + + Convert the exponent of each packed single-precision (32-bit) floating-point element in "a" to a single-precision (32-bit) floating-point number representing the integer exponent, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). This intrinsic essentially calculates "floor(log2(x))" for each element. + FOR j := 0 to 15 i := j*32 IF k[j] - dst[i+31:i] := MEM[mem_addr+i+31:mem_addr+i] + dst[i+31:i] := ConvertExpFP32(a[i+31:i]) ELSE dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:128] := 0 - - -
immintrin.h
- - - Floating Point - AVX512VL - AVX512F - Store - - - - Store packed single-precision (32-bit) floating-point elements from "a" into memory using writemask "k". - "mem_addr" does not need to be aligned on any particular boundary. - - -FOR j := 0 to 3 - i := j*32 - IF k[j] - MEM[mem_addr+i+31:mem_addr+i] := a[i+31:i] - FI -ENDFOR +dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - AVX512VL - AVX512F - Load - - - Load packed single-precision (32-bit) floating-point elements from memory into "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - "mem_addr" does not need to be aligned on any particular boundary. - - -FOR j := 0 to 3 + AVX512F/KNCNI + Miscellaneous + + + + + + Convert the exponent of each packed single-precision (32-bit) floating-point element in "a" to a single-precision (32-bit) floating-point number representing the integer exponent, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). This intrinsic essentially calculates "floor(log2(x))" for each element. + [sae_note] + FOR j := 0 to 15 i := j*32 IF k[j] - dst[i+31:i] := MEM[mem_addr+i+31:mem_addr+i] + dst[i+31:i] := ConvertExpFP32(a[i+31:i]) ELSE - dst[i+31:i] := 0 + dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - AVX512VL - AVX512F - Arithmetic - - - - - Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -FOR j := 0 to 3 + AVX512F/KNCNI + Miscellaneous + + + + + Normalize the mantissas of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst". This intrinsic essentially calculates "±(2^k)*|x.significand|", where "k" depends on the interval range defined by "interv" and the sign depends on "sc" and the source sign. + [getmant_note] + FOR j := 0 to 7 i := j*64 - IF k[j] - dst[i+63:i] := a[i+63:i] * b[i+63:i] - ELSE - dst[i+63:i] := src[i+63:i] - FI + dst[i+63:i] := GetNormalizedMantissa(a[i+63:i], sc, interv) ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - AVX512VL - AVX512F - Arithmetic - - - - Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - - -FOR j := 0 to 3 + AVX512F/KNCNI + Miscellaneous + + + + + + Normalize the mantissas of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst". This intrinsic essentially calculates "±(2^k)*|x.significand|", where "k" depends on the interval range defined by "interv" and the sign depends on "sc" and the source sign. + [getmant_note][sae_note] + FOR j := 0 to 7 i := j*64 - IF k[j] - dst[i+63:i] := a[i+63:i] * b[i+63:i] - ELSE - dst[i+63:i] := 0 - FI + dst[i+63:i] := GetNormalizedMantissa(a[i+63:i], sc, interv) ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - AVX512VL - AVX512F - Arithmetic - - - - - Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -FOR j := 0 to 1 + AVX512F/KNCNI + Miscellaneous + + + + + + + Normalize the mantissas of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). This intrinsic essentially calculates "±(2^k)*|x.significand|", where "k" depends on the interval range defined by "interv" and the sign depends on "sc" and the source sign. + [getmant_note] + FOR j := 0 to 7 i := j*64 IF k[j] - dst[i+63:i] := a[i+63:i] * b[i+63:i] + dst[i+63:i] := GetNormalizedMantissa(a[i+63:i], sc, interv) ELSE dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - AVX512VL - AVX512F - Arithmetic - - - - Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - - -FOR j := 0 to 1 + AVX512F/KNCNI + Miscellaneous + + + + + + + + Normalize the mantissas of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). This intrinsic essentially calculates "±(2^k)*|x.significand|", where "k" depends on the interval range defined by "interv" and the sign depends on "sc" and the source sign. + [getmant_note][sae_note] + FOR j := 0 to 7 i := j*64 IF k[j] - dst[i+63:i] := a[i+63:i] * b[i+63:i] + dst[i+63:i] := GetNormalizedMantissa(a[i+63:i], sc, interv) ELSE - dst[i+63:i] := 0 + dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - AVX512VL - AVX512F - Arithmetic - - - - - Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). RM. - -FOR j := 0 to 7 - i := j*32 - IF k[j] - dst[i+31:i] := a[i+31:i] * b[i+31:i] - ELSE - dst[i+31:i] := src[i+31:i] - FI + AVX512F/KNCNI + Miscellaneous + + + + + Normalize the mantissas of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst". This intrinsic essentially calculates "±(2^k)*|x.significand|", where "k" depends on the interval range defined by "interv" and the sign depends on "sc" and the source sign. + [getmant_note] + FOR j := 0 to 15 + i := j*32 + dst[i+31:i] := GetNormalizedMantissa(a[i+31:i], sc, interv) ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - AVX512VL - AVX512F - Arithmetic - - - - Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - - -FOR j := 0 to 7 + AVX512F/KNCNI + Miscellaneous + + + + + + Normalize the mantissas of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst". This intrinsic essentially calculates "±(2^k)*|x.significand|", where "k" depends on the interval range defined by "interv" and the sign depends on "sc" and the source sign. + [getmant_note][sae_note] + FOR j := 0 to 15 i := j*32 - IF k[j] - dst[i+31:i] := a[i+31:i] * b[i+31:i] - ELSE - dst[i+31:i] := 0 - FI + dst[i+31:i] := GetNormalizedMantissa(a[i+31:i], sc, interv) ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - AVX512VL - AVX512F - Arithmetic - - - - - Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). RM. - -FOR j := 0 to 3 + AVX512F/KNCNI + Miscellaneous + + + + + + + Normalize the mantissas of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). This intrinsic essentially calculates "±(2^k)*|x.significand|", where "k" depends on the interval range defined by "interv" and the sign depends on "sc" and the source sign. + [getmant_note] + FOR j := 0 to 15 i := j*32 IF k[j] - dst[i+31:i] := a[i+31:i] * b[i+31:i] + dst[i+31:i] := GetNormalizedMantissa(a[i+31:i], sc, interv) ELSE dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - AVX512VL - AVX512F - Arithmetic - - - - Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - - -FOR j := 0 to 3 + AVX512F/KNCNI + Miscellaneous + + + + + + + + Normalize the mantissas of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). This intrinsic essentially calculates "±(2^k)*|x.significand|", where "k" depends on the interval range defined by "interv" and the sign depends on "sc" and the source sign. + [getmant_note][sae_note] + FOR j := 0 to 15 i := j*32 IF k[j] - dst[i+31:i] := a[i+31:i] * b[i+31:i] + dst[i+31:i] := GetNormalizedMantissa(a[i+31:i], sc, interv) ELSE - dst[i+31:i] := 0 + dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - AVX512VL - AVX512DQ - Logical - - - - - Compute the bitwise OR of packed double-precision (64-bit) floating-point elements in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + AVX512F/KNCNI + Load + + + Load 512-bits (composed of 8 packed double-precision (64-bit) floating-point elements) from memory into "dst". + "mem_addr" must be aligned on a 64-byte boundary or a general-protection exception may be generated. -FOR j := 0 to 3 - i := j*64 - IF k[j] - dst[i+63:i] := a[i+63:i] BITWISE OR b[i+63:i] - ELSE - dst[i+63:i] := src[i+63:i] - FI -ENDFOR -dst[MAX:256] := 0 +dst[511:0] := MEM[mem_addr+511:mem_addr] +dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - AVX512VL - AVX512DQ - Logical - - - - Compute the bitwise OR of packed double-precision (64-bit) floating-point elements in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + AVX512F/KNCNI + Load + + + + + Load packed double-precision (64-bit) floating-point elements from memory into "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). "mem_addr" must be aligned on a 64-byte boundary or a general-protection exception may be generated. -FOR j := 0 to 3 +FOR j := 0 to 7 i := j*64 IF k[j] - dst[i+63:i] := a[i+63:i] BITWISE OR b[i+63:i] + dst[i+63:i] := MEM[mem_addr+i+63:mem_addr+i] ELSE - dst[i+63:i] := 0 + dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - AVX512DQ - Logical - - - - - Compute the bitwise OR of packed double-precision (64-bit) floating-point elements in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + AVX512F/KNCNI + Move + + + + + Move packed double-precision (64-bit) floating-point elements from "a" to "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). FOR j := 0 to 7 i := j*64 IF k[j] - dst[i+63:i] := a[i+63:i] BITWISE OR b[i+63:i] + dst[i+63:i] := a[i+63:i] ELSE dst[i+63:i] := src[i+63:i] FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - AVX512DQ - Logical - - - - Compute the bitwise OR of packed double-precision (64-bit) floating-point elements in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - + AVX512F/KNCNI + Store + + + + + Store packed double-precision (64-bit) floating-point elements from "a" into memory using writemask "k". + "mem_addr" must be aligned on a 64-byte boundary or a general-protection exception may be generated. FOR j := 0 to 7 i := j*64 IF k[j] - dst[i+63:i] := a[i+63:i] BITWISE OR b[i+63:i] - ELSE - dst[i+63:i] := 0 + MEM[mem_addr+i+63:mem_addr+i] := a[i+63:i] FI ENDFOR -dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - AVX512DQ - Logical - - - Compute the bitwise OR of packed double-precision (64-bit) floating-point elements in "a" and "b", and store the results in "dst". - + AVX512F/KNCNI + Store + + + + Store 512-bits (composed of 8 packed double-precision (64-bit) floating-point elements) from "a" into memory. + "mem_addr" must be aligned on a 64-byte boundary or a general-protection exception may be generated. -FOR j := 0 to 7 - i := j*64 - dst[i+63:i] := a[i+63:i] BITWISE OR b[i+63:i] -ENDFOR -dst[MAX:512] := 0 +MEM[mem_addr+511:mem_addr] := a[511:0] - +
immintrin.h
 - + Floating Point - AVX512VL - AVX512DQ - Logical - - - - - Compute the bitwise OR of packed double-precision (64-bit) floating-point elements in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + AVX512F/KNCNI + Load + + + Load 512-bits (composed of 16 packed single-precision (32-bit) floating-point elements) from memory into "dst". + "mem_addr" must be aligned on a 64-byte boundary or a general-protection exception may be generated. -FOR j := 0 to 1 - i := j*64 - IF k[j] - dst[i+63:i] := a[i+63:i] BITWISE OR b[i+63:i] - ELSE - dst[i+63:i] := src[i+63:i] - FI -ENDFOR -dst[MAX:128] := 0 +dst[511:0] := MEM[mem_addr+511:mem_addr] +dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - AVX512VL - AVX512DQ - Logical - - - - Compute the bitwise OR of packed double-precision (64-bit) floating-point elements in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + AVX512F/KNCNI + Load + + + + + Load packed single-precision (32-bit) floating-point elements from memory into "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). "mem_addr" must be aligned on a 64-byte boundary or a general-protection exception may be generated. -FOR j := 0 to 1 - i := j*64 +FOR j := 0 to 15 + i := j*32 IF k[j] - dst[i+63:i] := a[i+63:i] BITWISE OR b[i+63:i] + dst[i+31:i] := MEM[mem_addr+i+31:mem_addr+i] ELSE - dst[i+63:i] := 0 + dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - AVX512VL - AVX512DQ - Logical - - - - - Compute the bitwise OR of packed single-precision (32-bit) floating-point elements in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + AVX512F/KNCNI + Move + + + + + Move packed single-precision (32-bit) floating-point elements from "a" to "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 7 +FOR j := 0 to 15 i := j*32 IF k[j] - dst[i+31:i] := a[i+31:i] BITWISE OR b[i+31:i] + dst[i+31:i] := a[i+31:i] ELSE dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Floating Point - AVX512VL - AVX512DQ - Logical - - - - Compute the bitwise OR of packed single-precision (32-bit) floating-point elements in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + AVX512F/KNCNI + Store + + + + + Store packed single-precision (32-bit) floating-point elements from "a" into memory using writemask "k". + "mem_addr" must be aligned on a 64-byte boundary or a general-protection exception may be generated. -FOR j := 0 to 7 +FOR j := 0 to 15 i := j*32 IF k[j] - dst[i+31:i] := a[i+31:i] BITWISE OR b[i+31:i] - ELSE - dst[i+31:i] := 0 + MEM[mem_addr+i+31:mem_addr+i] := a[i+31:i] FI ENDFOR -dst[MAX:256] := 0 - +
immintrin.h
 - + Floating Point - AVX512DQ - Logical - - - - - Compute the bitwise OR of packed single-precision (32-bit) floating-point elements in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + AVX512F/KNCNI + Store + + + + Store 512-bits (composed of 16 packed single-precision (32-bit) floating-point elements) from "a" into memory. + "mem_addr" must be aligned on a 64-byte boundary or a general-protection exception may be generated. + +MEM[mem_addr+511:mem_addr] := a[511:0] + + +
immintrin.h
+ + + Integer + AVX512F/KNCNI + Load + + + Load 512-bits (composed of 16 packed 32-bit integers) from memory into "dst". + "mem_addr" must be aligned on a 64-byte boundary or a general-protection exception may be generated. + +dst[511:0] := MEM[mem_addr+511:mem_addr] +dst[MAX:512] := 0 + + +
immintrin.h
+ + + Integer + AVX512F/KNCNI + Load + + + Load 512-bits of integer data from memory into "dst". + "mem_addr" must be aligned on a 64-byte boundary or a general-protection exception may be generated. + +dst[511:0] := MEM[mem_addr+511:mem_addr] +dst[MAX:512] := 0 + + +
immintrin.h
+ + + Integer + AVX512F/KNCNI + Load + + + + + Load packed 32-bit integers from memory into "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + "mem_addr" must be aligned on a 64-byte boundary or a general-protection exception may be generated. FOR j := 0 to 15 i := j*32 IF k[j] - dst[i+31:i] := a[i+31:i] BITWISE OR b[i+31:i] + dst[i+31:i] := MEM[mem_addr+i+31:mem_addr+i] ELSE dst[i+31:i] := src[i+31:i] FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point - AVX512DQ - Logical - - - - Compute the bitwise OR of packed single-precision (32-bit) floating-point elements in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - + + Integer + AVX512F/KNCNI + Move + + + + + Move packed 32-bit integers from "a" to "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). FOR j := 0 to 15 i := j*32 IF k[j] - dst[i+31:i] := a[i+31:i] BITWISE OR b[i+31:i] + dst[i+31:i] := a[i+31:i] ELSE - dst[i+31:i] := 0 + dst[i+31:i] := src[i+31:i] FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point - AVX512DQ - Logical - - - Compute the bitwise OR of packed single-precision (32-bit) floating-point elements in "a" and "b", and store the results in "dst". - + + Integer + AVX512F/KNCNI + Store + + + + + Store packed 32-bit integers from "a" into memory using writemask "k". + "mem_addr" must be aligned on a 64-byte boundary or a general-protection exception may be generated. FOR j := 0 to 15 i := j*32 - dst[i+31:i] := a[i+31:i] BITWISE OR b[i+31:i] + IF k[j] + MEM[mem_addr+i+31:mem_addr+i] := a[i+31:i] + FI ENDFOR + + +
immintrin.h
+ + + Integer + AVX512F/KNCNI + Store + + + + Store 512-bits (composed of 16 packed 32-bit integers) from "a" into memory. + "mem_addr" must be aligned on a 64-byte boundary or a general-protection exception may be generated. + +MEM[mem_addr+511:mem_addr] := a[511:0] + + +
immintrin.h
+ + + Integer + AVX512F/KNCNI + Store + + + + Store 512-bits of integer data from "a" into memory. + "mem_addr" must be aligned on a 64-byte boundary or a general-protection exception may be generated. + +MEM[mem_addr+511:mem_addr] := a[511:0] + + +
immintrin.h
+ + + Integer + AVX512F/KNCNI + Load + + + Load 512-bits (composed of 8 packed 64-bit integers) from memory into "dst". + "mem_addr" must be aligned on a 64-byte boundary or a general-protection exception may be generated. + +dst[511:0] := MEM[mem_addr+511:mem_addr] dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point - AVX512VL - AVX512DQ - Logical - - - - - Compute the bitwise OR of packed single-precision (32-bit) floating-point elements in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Integer + AVX512F/KNCNI + Load + + + + + Load packed 64-bit integers from memory into "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + "mem_addr" must be aligned on a 64-byte boundary or a general-protection exception may be generated. -FOR j := 0 to 3 - i := j*32 +FOR j := 0 to 7 + i := j*64 IF k[j] - dst[i+31:i] := a[i+31:i] BITWISE OR b[i+31:i] + dst[i+63:i] := MEM[mem_addr+i+63:mem_addr+i] ELSE - dst[i+31:i] := src[i+31:i] + dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point - AVX512VL - AVX512DQ - Logical - - - - Compute the bitwise OR of packed single-precision (32-bit) floating-point elements in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Integer + AVX512F/KNCNI + Move + + + + + Move packed 64-bit integers from "a" to "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*32 +FOR j := 0 to 7 + i := j*64 IF k[j] - dst[i+31:i] := a[i+31:i] BITWISE OR b[i+31:i] + dst[i+63:i] := a[i+63:i] ELSE - dst[i+31:i] := 0 + dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX512VL - AVX512BW + AVX512F/KNCNI + Store + + + + + Store packed 64-bit integers from "a" into memory using writemask "k". + "mem_addr" must be aligned on a 64-byte boundary or a general-protection exception may be generated. + +FOR j := 0 to 7 + i := j*64 + IF k[j] + MEM[mem_addr+i+63:mem_addr+i] := a[i+63:i] + FI +ENDFOR + + +
immintrin.h
+ + + Integer + AVX512F/KNCNI + Store + + + + Store 512-bits (composed of 8 packed 64-bit integers) from "a" into memory. + "mem_addr" must be aligned on a 64-byte boundary or a general-protection exception may be generated. + +MEM[mem_addr+511:mem_addr] := a[511:0] + + +
immintrin.h
+ + + Floating Point + AVX512F/KNCNI Arithmetic - - - - Compute the absolute value of packed 8-bit integers in "a", and store the unsigned results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + + + + + Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). RM. -FOR j := 0 to 31 - i := j*8 +FOR j := 0 to 7 + i := j*64 IF k[j] - dst[i+7:i] := ABS(a[i+7:i]) + dst[i+63:i] := a[i+63:i] * b[i+63:i] ELSE - dst[i+7:i] := src[i+7:i] + dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512BW + + Floating Point + AVX512F/KNCNI Arithmetic - - - Compute the absolute value of packed 8-bit integers in "a", and store the unsigned results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + + + + + + Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + [round_note] -FOR j := 0 to 31 - i := j*8 +FOR j := 0 to 7 + i := j*64 IF k[j] - dst[i+7:i] := ABS(a[i+7:i]) + dst[i+63:i] := a[i+63:i] * b[i+63:i] ELSE - dst[i+7:i] := 0 + dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - AVX512BW + + Floating Point + AVX512F/KNCNI Arithmetic - - Compute the absolute value of packed 8-bit integers in "a", and store the unsigned results in "dst". + + + + Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", and store the results in "dst". -FOR j := 0 to 63 - i := j*8 - dst[i+7:i] := ABS(a[i+7:i]) +FOR j := 0 to 7 + i := j*64 + dst[i+63:i] := a[i+63:i] * b[i+63:i] ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - AVX512BW + + Floating Point + AVX512F/KNCNI Arithmetic - - - - Compute the absolute value of packed 8-bit integers in "a", and store the unsigned results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + + + + Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", and store the results in "dst". + [round_note] -FOR j := 0 to 63 - i := j*8 - IF k[j] - dst[i+7:i] := ABS(a[i+7:i]) - ELSE - dst[i+7:i] := src[i+7:i] - FI +FOR j := 0 to 7 + i := j*64 + dst[i+63:i] := a[i+63:i] * b[i+63:i] ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - AVX512BW + + Floating Point + AVX512F/KNCNI Arithmetic - - - Compute the absolute value of packed 8-bit integers in "a", and store the unsigned results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + + + + + Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). RM. -FOR j := 0 to 63 - i := j*8 +FOR j := 0 to 15 + i := j*32 IF k[j] - dst[i+7:i] := ABS(a[i+7:i]) + dst[i+31:i] := a[i+31:i] * b[i+31:i] ELSE - dst[i+7:i] := 0 + dst[i+31:i] := src[i+31:i] FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512BW + + Floating Point + AVX512F/KNCNI Arithmetic - - - - Compute the absolute value of packed 8-bit integers in "a", and store the unsigned results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + + + + + + Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + [round_note] FOR j := 0 to 15 - i := j*8 + i := j*32 IF k[j] - dst[i+7:i] := ABS(a[i+7:i]) + dst[i+31:i] := a[i+31:i] * b[i+31:i] ELSE - dst[i+7:i] := src[i+7:i] + dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512BW + + Floating Point + AVX512F/KNCNI Arithmetic - - - Compute the absolute value of packed 8-bit integers in "a", and store the unsigned results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + + + Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", and store the results in "dst". FOR j := 0 to 15 - i := j*8 - IF k[j] - dst[i+7:i] := ABS(a[i+7:i]) - ELSE - dst[i+7:i] := 0 - FI + i := j*32 + dst[i+31:i] := a[i+31:i] * b[i+31:i] ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512F + + Floating Point + AVX512F/KNCNI Arithmetic - - - - Compute the absolute value of packed 32-bit integers in "a", and store the unsigned results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + + + + Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", and store the results in "dst". + [round_note] -FOR j := 0 to 7 +FOR j := 0 to 15 i := j*32 - IF k[j] - dst[i+31:i] := ABS(a[i+31:i]) - ELSE - dst[i+31:i] := src[i+31:i] - FI + dst[i+31:i] := a[i+31:i] * b[i+31:i] ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX512VL - AVX512F + AVX512F/KNCNI Arithmetic - - - Compute the absolute value of packed 32-bit integers in "a", and store the unsigned results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + + + Add packed 32-bit integers in "a" and "b", and store the results in "dst". -FOR j := 0 to 7 +FOR j := 0 to 15 i := j*32 - IF k[j] - dst[i+31:i] := ABS(a[i+31:i]) - ELSE - dst[i+31:i] := 0 - FI + dst[i+31:i] := a[i+31:i] + b[i+31:i] ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX512VL - AVX512F + AVX512F/KNCNI Arithmetic - - - - Compute the absolute value of packed 32-bit integers in "a", and store the unsigned results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + + + + + Add packed 32-bit integers in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 3 +FOR j := 0 to 15 i := j*32 IF k[j] - dst[i+31:i] := ABS(a[i+31:i]) + dst[i+31:i] := a[i+31:i] + b[i+31:i] ELSE dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX512VL - AVX512F - Arithmetic - - - Compute the absolute value of packed 32-bit integers in "a", and store the unsigned results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + AVX512F/KNCNI + Logical + + + + Compute the bitwise AND of packed 32-bit integers in "a" and "b", and store the results in "dst". -FOR j := 0 to 3 +FOR j := 0 to 15 i := j*32 - IF k[j] - dst[i+31:i] := ABS(a[i+31:i]) - ELSE - dst[i+31:i] := 0 - FI + dst[i+31:i] := a[i+31:i] AND b[i+31:i] ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX512VL - AVX512F - Arithmetic - - Compute the absolute value of packed 64-bit integers in "a", and store the unsigned results in "dst". + AVX512F/KNCNI + Logical + + + + Compute the bitwise AND of 512 bits (representing integer data) in "a" and "b", and store the result in "dst". -FOR j := 0 to 3 - i := j*64 - dst[i+63:i] := ABS(a[i+63:i]) -ENDFOR -dst[MAX:256] := 0 +dst[511:0] := (a[511:0] AND b[511:0]) +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX512VL - AVX512F - Arithmetic - - - - Compute the absolute value of packed 64-bit integers in "a", and store the unsigned results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + AVX512F/KNCNI + Logical + + + + Compute the bitwise NOT of packed 32-bit integers in "a" and then AND with "b", and store the results in "dst". -FOR j := 0 to 3 - i := j*64 - IF k[j] - dst[i+63:i] := ABS(a[i+63:i]) - ELSE - dst[i+63:i] := src[i+63:i] - FI +FOR j := 0 to 15 + i := j*32 + dst[i+31:i] := (NOT a[i+31:i]) AND b[i+31:i] ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX512VL - AVX512F - Arithmetic - - - Compute the absolute value of packed 64-bit integers in "a", and store the unsigned results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + AVX512F/KNCNI + Logical + + + + Compute the bitwise NOT of 512 bits (representing integer data) in "a" and then AND with "b", and store the result in "dst". -FOR j := 0 to 3 - i := j*64 +dst[511:0] := ((NOT a[511:0]) AND b[511:0]) +dst[MAX:512] := 0 + + +
immintrin.h
+ + + Integer + AVX512F/KNCNI + Logical + + + + + + Compute the bitwise NOT of packed 32-bit integers in "a" and then AND with "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +FOR j := 0 to 15 + i := j*32 IF k[j] - dst[i+63:i] := ABS(a[i+63:i]) + dst[i+31:i] := ((NOT a[i+31:i]) AND b[i+31:i]) ELSE - dst[i+63:i] := 0 + dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX512VL - AVX512F - Arithmetic - - Compute the absolute value of packed 64-bit integers in "a", and store the unsigned results in "dst". + AVX512F/KNCNI + Logical + + + + Compute the bitwise NOT of 512 bits (composed of packed 64-bit integers) in "a" and then AND with "b", and store the results in "dst". -FOR j := 0 to 1 - i := j*64 - dst[i+63:i] := ABS(a[i+63:i]) -ENDFOR -dst[MAX:128] := 0 +dst[511:0] := ((NOT a[511:0]) AND b[511:0]) +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX512VL - AVX512F - Arithmetic - - - - Compute the absolute value of packed 64-bit integers in "a", and store the unsigned results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + AVX512F/KNCNI + Logical + + + + + + Compute the bitwise NOT of packed 64-bit integers in "a" and then AND with "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 1 +FOR j := 0 to 7 i := j*64 IF k[j] - dst[i+63:i] := ABS(a[i+63:i]) + dst[i+63:i] := ((NOT a[i+63:i]) AND b[i+63:i]) ELSE dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX512VL - AVX512F - Arithmetic - - - Compute the absolute value of packed 64-bit integers in "a", and store the unsigned results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + AVX512F/KNCNI + Logical + + + + Compute the bitwise AND of 512 bits (composed of packed 64-bit integers) in "a" and "b", and store the results in "dst". -FOR j := 0 to 1 +dst[511:0] := (a[511:0] AND b[511:0]) +dst[MAX:512] := 0 + + +
immintrin.h
+ + + Integer + AVX512F/KNCNI + Logical + + + + + + Compute the bitwise AND of packed 64-bit integers in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +FOR j := 0 to 7 i := j*64 IF k[j] - dst[i+63:i] := ABS(a[i+63:i]) + dst[i+63:i] := a[i+63:i] AND b[i+63:i] ELSE - dst[i+63:i] := 0 + dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX512VL - AVX512BW - Arithmetic - - - - Compute the absolute value of packed 16-bit integers in "a", and store the unsigned results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + AVX512F/KNCNI + Swizzle + + + + + Blend packed 32-bit integers from "a" and "b" using control mask "k", and store the results in "dst". FOR j := 0 to 15 - i := j*16 + i := j*32 IF k[j] - dst[i+15:i] := ABS(a[i+15:i]) + dst[i+31:i] := b[i+31:i] ELSE - dst[i+15:i] := src[i+15:i] + dst[i+31:i] := a[i+31:i] FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX512VL - AVX512BW - Arithmetic - - - Compute the absolute value of packed 16-bit integers in "a", and store the unsigned results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + AVX512F/KNCNI + Swizzle + + + + + Blend packed 64-bit integers from "a" and "b" using control mask "k", and store the results in "dst". -FOR j := 0 to 15 - i := j*16 +FOR j := 0 to 7 + i := j*64 IF k[j] - dst[i+15:i] := ABS(a[i+15:i]) + dst[i+63:i] := b[i+63:i] ELSE - dst[i+15:i] := 0 + dst[i+63:i] := a[i+63:i] FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX512BW - Arithmetic - - Compute the absolute value of packed 16-bit integers in "a", and store the unsigned results in "dst". - -FOR j := 0 to 31 - i := j*16 - dst[i+15:i] := ABS(a[i+15:i]) + Mask + AVX512F/KNCNI + Compare + + + + + Compare packed signed 32-bit integers in "a" and "b" based on the comparison operand specified by "imm8", and store the results in mask vector "k". + CASE (imm8[2:0]) OF +0: OP := _MM_CMPINT_EQ +1: OP := _MM_CMPINT_LT +2: OP := _MM_CMPINT_LE +3: OP := _MM_CMPINT_FALSE +4: OP := _MM_CMPINT_NE +5: OP := _MM_CMPINT_NLT +6: OP := _MM_CMPINT_NLE +7: OP := _MM_CMPINT_TRUE +ESAC +FOR j := 0 to 15 + i := j*32 + k[j] := ( a[i+31:i] OP b[i+31:i] ) ? 1 : 0 ENDFOR -dst[MAX:512] := 0 +k[MAX:16] := 0 - +
immintrin.h
 - + Integer - AVX512BW - Arithmetic - - - - Compute the absolute value of packed 16-bit integers in "a", and store the unsigned results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Mask + AVX512F/KNCNI + Compare + + + + Compare packed 32-bit integers in "a" and "b" for equality, and store the results in mask vector "k". -FOR j := 0 to 31 - i := j*16 - IF k[j] - dst[i+15:i] := ABS(a[i+15:i]) - ELSE - dst[i+15:i] := src[i+15:i] - FI +FOR j := 0 to 15 + i := j*32 + k[j] := ( a[i+31:i] == b[i+31:i] ) ? 1 : 0 ENDFOR -dst[MAX:512] := 0 +k[MAX:16] := 0 - +
immintrin.h
 - + Integer - AVX512BW - Arithmetic - - - Compute the absolute value of packed 16-bit integers in "a", and store the unsigned results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Mask + AVX512F/KNCNI + Compare + + + + Compare packed signed 32-bit integers in "a" and "b" for greater-than-or-equal, and store the results in mask vector "k". -FOR j := 0 to 31 - i := j*16 - IF k[j] - dst[i+15:i] := ABS(a[i+15:i]) - ELSE - dst[i+15:i] := 0 - FI +FOR j := 0 to 15 + i := j*32 + k[j] := ( a[i+31:i] >= b[i+31:i] ) ? 1 : 0 ENDFOR -dst[MAX:512] := 0 +k[MAX:16] := 0 - +
immintrin.h
 - + Integer - AVX512VL - AVX512BW - Arithmetic - - - - Compute the absolute value of packed 16-bit integers in "a", and store the unsigned results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Mask + AVX512F/KNCNI + Compare + + + + Compare packed signed 32-bit integers in "a" and "b" for greater-than, and store the results in mask vector "k". -FOR j := 0 to 7 - i := j*16 - IF k[j] - dst[i+15:i] := ABS(a[i+15:i]) - ELSE - dst[i+15:i] := src[i+15:i] - FI +FOR j := 0 to 15 + i := j*32 + k[j] := ( a[i+31:i] > b[i+31:i] ) ? 1 : 0 ENDFOR -dst[MAX:128] := 0 +k[MAX:16] := 0 - +
immintrin.h
 - + Integer - AVX512VL - AVX512BW - Arithmetic - - - Compute the absolute value of packed 16-bit integers in "a", and store the unsigned results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Mask + AVX512F/KNCNI + Compare + + + + Compare packed signed 32-bit integers in "a" and "b" for less-than-or-equal, and store the results in mask vector "k". -FOR j := 0 to 7 - i := j*16 - IF k[j] - dst[i+15:i] := ABS(a[i+15:i]) - ELSE - dst[i+15:i] := 0 - FI +FOR j := 0 to 15 + i := j*32 + k[j] := ( a[i+31:i] <= b[i+31:i] ) ? 1 : 0 ENDFOR -dst[MAX:128] := 0 +k[MAX:16] := 0 - +
immintrin.h
 - + Integer - AVX512VL - AVX512BW - Convert - Miscellaneous - - - - - Convert packed 32-bit integers from "a" and "b" to packed 16-bit integers using signed saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Mask + AVX512F/KNCNI + Compare + + + + Compare packed 32-bit integers in "a" and "b" for not-equal, and store the results in mask vector "k". -tmp_dst[15:0] := Saturate_Int32_To_Int16 (a[31:0]) -tmp_dst[31:16] := Saturate_Int32_To_Int16 (a[63:32]) -tmp_dst[47:32] := Saturate_Int32_To_Int16 (a[95:64]) -tmp_dst[63:48] := Saturate_Int32_To_Int16 (a[127:96]) -tmp_dst[79:64] := Saturate_Int32_To_Int16 (b[31:0]) -tmp_dst[95:80] := Saturate_Int32_To_Int16 (b[63:32]) -tmp_dst[111:96] := Saturate_Int32_To_Int16 (b[95:64]) -tmp_dst[127:112] := Saturate_Int32_To_Int16 (b[127:96]) -tmp_dst[143:128] := Saturate_Int32_To_Int16 (a[159:128]) -tmp_dst[159:144] := Saturate_Int32_To_Int16 (a[191:160]) -tmp_dst[175:160] := Saturate_Int32_To_Int16 (a[223:192]) -tmp_dst[191:176] := Saturate_Int32_To_Int16 (a[255:224]) -tmp_dst[207:192] := Saturate_Int32_To_Int16 (b[159:128]) -tmp_dst[223:208] := Saturate_Int32_To_Int16 (b[191:160]) -tmp_dst[239:224] := Saturate_Int32_To_Int16 (b[223:192]) -tmp_dst[255:240] := Saturate_Int32_To_Int16 (b[255:224]) - FOR j := 0 to 15 - i := j*16 - IF k[j] - dst[i+15:i] := tmp_dst[i+15:i] - ELSE - dst[i+15:i] := src[i+15:i] - FI + i := j*32 + k[j] := ( a[i+31:i] != b[i+31:i] ) ? 1 : 0 ENDFOR -dst[MAX:256] := 0 +k[MAX:16] := 0 - +
immintrin.h
 - + Integer - AVX512VL - AVX512BW - Convert - Miscellaneous - - - - Convert packed 32-bit integers from "a" and "b" to packed 16-bit integers using signed saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - - -tmp_dst[15:0] := Saturate_Int32_To_Int16 (a[31:0]) -tmp_dst[31:16] := Saturate_Int32_To_Int16 (a[63:32]) -tmp_dst[47:32] := Saturate_Int32_To_Int16 (a[95:64]) -tmp_dst[63:48] := Saturate_Int32_To_Int16 (a[127:96]) -tmp_dst[79:64] := Saturate_Int32_To_Int16 (b[31:0]) -tmp_dst[95:80] := Saturate_Int32_To_Int16 (b[63:32]) -tmp_dst[111:96] := Saturate_Int32_To_Int16 (b[95:64]) -tmp_dst[127:112] := Saturate_Int32_To_Int16 (b[127:96]) -tmp_dst[143:128] := Saturate_Int32_To_Int16 (a[159:128]) -tmp_dst[159:144] := Saturate_Int32_To_Int16 (a[191:160]) -tmp_dst[175:160] := Saturate_Int32_To_Int16 (a[223:192]) -tmp_dst[191:176] := Saturate_Int32_To_Int16 (a[255:224]) -tmp_dst[207:192] := Saturate_Int32_To_Int16 (b[159:128]) -tmp_dst[223:208] := Saturate_Int32_To_Int16 (b[191:160]) -tmp_dst[239:224] := Saturate_Int32_To_Int16 (b[223:192]) -tmp_dst[255:240] := Saturate_Int32_To_Int16 (b[255:224]) - + Mask + AVX512F/KNCNI + Compare + + + + + + Compare packed signed 32-bit integers in "a" and "b" based on the comparison operand specified by "imm8", and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). + CASE (imm8[2:0]) OF +0: OP := _MM_CMPINT_EQ +1: OP := _MM_CMPINT_LT +2: OP := _MM_CMPINT_LE +3: OP := _MM_CMPINT_FALSE +4: OP := _MM_CMPINT_NE +5: OP := _MM_CMPINT_NLT +6: OP := _MM_CMPINT_NLE +7: OP := _MM_CMPINT_TRUE +ESAC FOR j := 0 to 15 - i := j*16 - IF k[j] - dst[i+15:i] := tmp_dst[i+15:i] - ELSE - dst[i+15:i] := 0 - FI -ENDFOR -dst[MAX:256] := 0 - - -
immintrin.h
- - - Integer - AVX512BW - Convert - Miscellaneous - - - - - Convert packed 32-bit integers from "a" and "b" to packed 16-bit integers using signed saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -tmp_dst[15:0] := Saturate_Int32_To_Int16 (a[31:0]) -tmp_dst[31:16] := Saturate_Int32_To_Int16 (a[63:32]) -tmp_dst[47:32] := Saturate_Int32_To_Int16 (a[95:64]) -tmp_dst[63:48] := Saturate_Int32_To_Int16 (a[127:96]) -tmp_dst[79:64] := Saturate_Int32_To_Int16 (b[31:0]) -tmp_dst[95:80] := Saturate_Int32_To_Int16 (b[63:32]) -tmp_dst[111:96] := Saturate_Int32_To_Int16 (b[95:64]) -tmp_dst[127:112] := Saturate_Int32_To_Int16 (b[127:96]) -tmp_dst[143:128] := Saturate_Int32_To_Int16 (a[159:128]) -tmp_dst[159:144] := Saturate_Int32_To_Int16 (a[191:160]) -tmp_dst[175:160] := Saturate_Int32_To_Int16 (a[223:192]) -tmp_dst[191:176] := Saturate_Int32_To_Int16 (a[255:224]) -tmp_dst[207:192] := Saturate_Int32_To_Int16 (b[159:128]) -tmp_dst[223:208] := Saturate_Int32_To_Int16 (b[191:160]) -tmp_dst[239:224] := Saturate_Int32_To_Int16 (b[223:192]) -tmp_dst[255:240] := Saturate_Int32_To_Int16 (b[255:224]) -tmp_dst[271:256] := Saturate_Int32_To_Int16 (a[287:256]) -tmp_dst[287:272] := Saturate_Int32_To_Int16 (a[319:288]) -tmp_dst[303:288] := Saturate_Int32_To_Int16 (a[351:320]) -tmp_dst[319:304] := Saturate_Int32_To_Int16 (a[383:352]) -tmp_dst[335:320] := Saturate_Int32_To_Int16 (b[287:256]) -tmp_dst[351:336] := Saturate_Int32_To_Int16 (b[319:288]) -tmp_dst[367:352] := Saturate_Int32_To_Int16 (b[351:320]) -tmp_dst[383:368] := Saturate_Int32_To_Int16 (b[383:352]) -tmp_dst[399:384] := Saturate_Int32_To_Int16 (a[415:384]) -tmp_dst[415:400] := Saturate_Int32_To_Int16 (a[447:416]) -tmp_dst[431:416] := Saturate_Int32_To_Int16 (a[479:448]) -tmp_dst[447:432] := Saturate_Int32_To_Int16 (a[511:480]) -tmp_dst[463:448] := Saturate_Int32_To_Int16 (b[415:384]) -tmp_dst[479:464] := Saturate_Int32_To_Int16 (b[447:416]) -tmp_dst[495:480] := Saturate_Int32_To_Int16 (b[479:448]) -tmp_dst[511:496] := Saturate_Int32_To_Int16 (b[511:480]) - -FOR j := 0 to 31 - i := j*16 - IF k[j] - dst[i+15:i] := tmp_dst[i+15:i] - ELSE - dst[i+15:i] := src[i+15:i] - FI -ENDFOR -dst[MAX:512] := 0 - - -
immintrin.h
- - - Integer - AVX512BW - Convert - Miscellaneous - - - - Convert packed 32-bit integers from "a" and "b" to packed 16-bit integers using signed saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - - -tmp_dst[15:0] := Saturate_Int32_To_Int16 (a[31:0]) -tmp_dst[31:16] := Saturate_Int32_To_Int16 (a[63:32]) -tmp_dst[47:32] := Saturate_Int32_To_Int16 (a[95:64]) -tmp_dst[63:48] := Saturate_Int32_To_Int16 (a[127:96]) -tmp_dst[79:64] := Saturate_Int32_To_Int16 (b[31:0]) -tmp_dst[95:80] := Saturate_Int32_To_Int16 (b[63:32]) -tmp_dst[111:96] := Saturate_Int32_To_Int16 (b[95:64]) -tmp_dst[127:112] := Saturate_Int32_To_Int16 (b[127:96]) -tmp_dst[143:128] := Saturate_Int32_To_Int16 (a[159:128]) -tmp_dst[159:144] := Saturate_Int32_To_Int16 (a[191:160]) -tmp_dst[175:160] := Saturate_Int32_To_Int16 (a[223:192]) -tmp_dst[191:176] := Saturate_Int32_To_Int16 (a[255:224]) -tmp_dst[207:192] := Saturate_Int32_To_Int16 (b[159:128]) -tmp_dst[223:208] := Saturate_Int32_To_Int16 (b[191:160]) -tmp_dst[239:224] := Saturate_Int32_To_Int16 (b[223:192]) -tmp_dst[255:240] := Saturate_Int32_To_Int16 (b[255:224]) -tmp_dst[271:256] := Saturate_Int32_To_Int16 (a[287:256]) -tmp_dst[287:272] := Saturate_Int32_To_Int16 (a[319:288]) -tmp_dst[303:288] := Saturate_Int32_To_Int16 (a[351:320]) -tmp_dst[319:304] := Saturate_Int32_To_Int16 (a[383:352]) -tmp_dst[335:320] := Saturate_Int32_To_Int16 (b[287:256]) -tmp_dst[351:336] := Saturate_Int32_To_Int16 (b[319:288]) -tmp_dst[367:352] := Saturate_Int32_To_Int16 (b[351:320]) -tmp_dst[383:368] := Saturate_Int32_To_Int16 (b[383:352]) -tmp_dst[399:384] := Saturate_Int32_To_Int16 (a[415:384]) -tmp_dst[415:400] := Saturate_Int32_To_Int16 (a[447:416]) -tmp_dst[431:416] := Saturate_Int32_To_Int16 (a[479:448]) -tmp_dst[447:432] := Saturate_Int32_To_Int16 (a[511:480]) -tmp_dst[463:448] := Saturate_Int32_To_Int16 (b[415:384]) -tmp_dst[479:464] := Saturate_Int32_To_Int16 (b[447:416]) -tmp_dst[495:480] := Saturate_Int32_To_Int16 (b[479:448]) -tmp_dst[511:496] := Saturate_Int32_To_Int16 (b[511:480]) - -FOR j := 0 to 31 - i := j*16 - IF k[j] - dst[i+15:i] := tmp_dst[i+15:i] - ELSE - dst[i+15:i] := 0 + i := j*32 + IF k1[j] + k[j] := ( a[i+31:i] OP b[i+31:i] ) ? 1 : 0 + ELSE + k[j] := 0 FI ENDFOR -dst[MAX:512] := 0 +k[MAX:16] := 0 - +
immintrin.h
 - + Integer - AVX512BW - Convert - Miscellaneous - - - Convert packed 32-bit integers from "a" and "b" to packed 16-bit integers using signed saturation, and store the results in "dst". + Mask + AVX512F/KNCNI + Compare + + + + + Compare packed 32-bit integers in "a" and "b" for equality, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). -dst[15:0] := Saturate_Int32_To_Int16 (a[31:0]) -dst[31:16] := Saturate_Int32_To_Int16 (a[63:32]) -dst[47:32] := Saturate_Int32_To_Int16 (a[95:64]) -dst[63:48] := Saturate_Int32_To_Int16 (a[127:96]) -dst[79:64] := Saturate_Int32_To_Int16 (b[31:0]) -dst[95:80] := Saturate_Int32_To_Int16 (b[63:32]) -dst[111:96] := Saturate_Int32_To_Int16 (b[95:64]) -dst[127:112] := Saturate_Int32_To_Int16 (b[127:96]) -dst[143:128] := Saturate_Int32_To_Int16 (a[159:128]) -dst[159:144] := Saturate_Int32_To_Int16 (a[191:160]) -dst[175:160] := Saturate_Int32_To_Int16 (a[223:192]) -dst[191:176] := Saturate_Int32_To_Int16 (a[255:224]) -dst[207:192] := Saturate_Int32_To_Int16 (b[159:128]) -dst[223:208] := Saturate_Int32_To_Int16 (b[191:160]) -dst[239:224] := Saturate_Int32_To_Int16 (b[223:192]) -dst[255:240] := Saturate_Int32_To_Int16 (b[255:224]) -dst[271:256] := Saturate_Int32_To_Int16 (a[287:256]) -dst[287:272] := Saturate_Int32_To_Int16 (a[319:288]) -dst[303:288] := Saturate_Int32_To_Int16 (a[351:320]) -dst[319:304] := Saturate_Int32_To_Int16 (a[383:352]) -dst[335:320] := Saturate_Int32_To_Int16 (b[287:256]) -dst[351:336] := Saturate_Int32_To_Int16 (b[319:288]) -dst[367:352] := Saturate_Int32_To_Int16 (b[351:320]) -dst[383:368] := Saturate_Int32_To_Int16 (b[383:352]) -dst[399:384] := Saturate_Int32_To_Int16 (a[415:384]) -dst[415:400] := Saturate_Int32_To_Int16 (a[447:416]) -dst[431:416] := Saturate_Int32_To_Int16 (a[479:448]) -dst[447:432] := Saturate_Int32_To_Int16 (a[511:480]) -dst[463:448] := Saturate_Int32_To_Int16 (b[415:384]) -dst[479:464] := Saturate_Int32_To_Int16 (b[447:416]) -dst[495:480] := Saturate_Int32_To_Int16 (b[479:448]) -dst[511:496] := Saturate_Int32_To_Int16 (b[511:480]) -dst[MAX:512] := 0 +FOR j := 0 to 15 + i := j*32 + IF k1[j] + k[j] := ( a[i+31:i] == b[i+31:i] ) ? 1 : 0 + ELSE + k[j] := 0 + FI +ENDFOR +k[MAX:16] := 0 - +
immintrin.h
 - + Integer - AVX512VL - AVX512BW - Convert - Miscellaneous - - - - - Convert packed 32-bit integers from "a" and "b" to packed 16-bit integers using signed saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Mask + AVX512F/KNCNI + Compare + + + + + Compare packed signed 32-bit integers in "a" and "b" for greater-than-or-equal, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). -tmp_dst[15:0] := Saturate_Int32_To_Int16 (a[31:0]) -tmp_dst[31:16] := Saturate_Int32_To_Int16 (a[63:32]) -tmp_dst[47:32] := Saturate_Int32_To_Int16 (a[95:64]) -tmp_dst[63:48] := Saturate_Int32_To_Int16 (a[127:96]) -tmp_dst[79:64] := Saturate_Int32_To_Int16 (b[31:0]) -tmp_dst[95:80] := Saturate_Int32_To_Int16 (b[63:32]) -tmp_dst[111:96] := Saturate_Int32_To_Int16 (b[95:64]) -tmp_dst[127:112] := Saturate_Int32_To_Int16 (b[127:96]) - -FOR j := 0 to 7 - i := j*16 - IF k[j] - dst[i+15:i] := tmp_dst[i+15:i] - ELSE - dst[i+15:i] := src[i+15:i] +FOR j := 0 to 15 + i := j*32 + IF k1[j] + k[j] := ( a[i+31:i] >= b[i+31:i] ) ? 1 : 0 + ELSE + k[j] := 0 FI ENDFOR -dst[MAX:128] := 0 +k[MAX:16] := 0 - +
immintrin.h
 - + Integer - AVX512VL - AVX512BW - Convert - Miscellaneous - - - - Convert packed 32-bit integers from "a" and "b" to packed 16-bit integers using signed saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - + Mask + AVX512F/KNCNI + Compare + + + + + Compare packed signed 32-bit integers in "a" and "b" for greater-than, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). -tmp_dst[15:0] := Saturate_Int32_To_Int16 (a[31:0]) -tmp_dst[31:16] := Saturate_Int32_To_Int16 (a[63:32]) -tmp_dst[47:32] := Saturate_Int32_To_Int16 (a[95:64]) -tmp_dst[63:48] := Saturate_Int32_To_Int16 (a[127:96]) -tmp_dst[79:64] := Saturate_Int32_To_Int16 (b[31:0]) -tmp_dst[95:80] := Saturate_Int32_To_Int16 (b[63:32]) -tmp_dst[111:96] := Saturate_Int32_To_Int16 (b[95:64]) -tmp_dst[127:112] := Saturate_Int32_To_Int16 (b[127:96]) - -FOR j := 0 to 7 - i := j*16 - IF k[j] - dst[i+15:i] := tmp_dst[i+15:i] - ELSE - dst[i+15:i] := 0 +FOR j := 0 to 15 + i := j*32 + IF k1[j] + k[j] := ( a[i+31:i] > b[i+31:i] ) ? 1 : 0 + ELSE + k[j] := 0 FI ENDFOR -dst[MAX:128] := 0 +k[MAX:16] := 0 - -
immintrin.h
- - - Integer - AVX512VL - AVX512BW - Convert - Miscellaneous - - - - - Convert packed 16-bit integers from "a" and "b" to packed 8-bit integers using signed saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -tmp_dst[7:0] := Saturate_Int16_To_Int8 (a[15:0]) -tmp_dst[15:8] := Saturate_Int16_To_Int8 (a[31:16]) -tmp_dst[23:16] := Saturate_Int16_To_Int8 (a[47:32]) -tmp_dst[31:24] := Saturate_Int16_To_Int8 (a[63:48]) -tmp_dst[39:32] := Saturate_Int16_To_Int8 (a[79:64]) -tmp_dst[47:40] := Saturate_Int16_To_Int8 (a[95:80]) -tmp_dst[55:48] := Saturate_Int16_To_Int8 (a[111:96]) -tmp_dst[63:56] := Saturate_Int16_To_Int8 (a[127:112]) -tmp_dst[71:64] := Saturate_Int16_To_Int8 (b[15:0]) -tmp_dst[79:72] := Saturate_Int16_To_Int8 (b[31:16]) -tmp_dst[87:80] := Saturate_Int16_To_Int8 (b[47:32]) -tmp_dst[95:88] := Saturate_Int16_To_Int8 (b[63:48]) -tmp_dst[103:96] := Saturate_Int16_To_Int8 (b[79:64]) -tmp_dst[111:104] := Saturate_Int16_To_Int8 (b[95:80]) -tmp_dst[119:112] := Saturate_Int16_To_Int8 (b[111:96]) -tmp_dst[127:120] := Saturate_Int16_To_Int8 (b[127:112]) -tmp_dst[135:128] := Saturate_Int16_To_Int8 (a[143:128]) -tmp_dst[143:136] := Saturate_Int16_To_Int8 (a[159:144]) -tmp_dst[151:144] := Saturate_Int16_To_Int8 (a[175:160]) -tmp_dst[159:152] := Saturate_Int16_To_Int8 (a[191:176]) -tmp_dst[167:160] := Saturate_Int16_To_Int8 (a[207:192]) -tmp_dst[175:168] := Saturate_Int16_To_Int8 (a[223:208]) -tmp_dst[183:176] := Saturate_Int16_To_Int8 (a[239:224]) -tmp_dst[191:184] := Saturate_Int16_To_Int8 (a[255:240]) -tmp_dst[199:192] := Saturate_Int16_To_Int8 (b[143:128]) -tmp_dst[207:200] := Saturate_Int16_To_Int8 (b[159:144]) -tmp_dst[215:208] := Saturate_Int16_To_Int8 (b[175:160]) -tmp_dst[223:216] := Saturate_Int16_To_Int8 (b[191:176]) -tmp_dst[231:224] := Saturate_Int16_To_Int8 (b[207:192]) -tmp_dst[239:232] := Saturate_Int16_To_Int8 (b[223:208]) -tmp_dst[247:240] := Saturate_Int16_To_Int8 (b[239:224]) -tmp_dst[255:248] := Saturate_Int16_To_Int8 (b[255:240]) - -FOR j := 0 to 31 - i := j*8 - IF k[j] - dst[i+7:i] := tmp_dst[i+7:i] - ELSE - dst[i+7:i] := src[i+7:i] + +
immintrin.h
+ + + Integer + Mask + AVX512F/KNCNI + Compare + + + + + Compare packed signed 32-bit integers in "a" and "b" for less-than, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). + +FOR j := 0 to 15 + i := j*32 + IF k1[j] + k[j] := ( a[i+31:i] <= b[i+31:i] ) ? 1 : 0 + ELSE + k[j] := 0 FI ENDFOR -dst[MAX:256] := 0 +k[MAX:16] := 0 - -
immintrin.h
- - - Integer - AVX512VL - AVX512BW - Convert - Miscellaneous - - - - Convert packed 16-bit integers from "a" and "b" to packed 8-bit integers using signed saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - - -tmp_dst[7:0] := Saturate_Int16_To_Int8 (a[15:0]) -tmp_dst[15:8] := Saturate_Int16_To_Int8 (a[31:16]) -tmp_dst[23:16] := Saturate_Int16_To_Int8 (a[47:32]) -tmp_dst[31:24] := Saturate_Int16_To_Int8 (a[63:48]) -tmp_dst[39:32] := Saturate_Int16_To_Int8 (a[79:64]) -tmp_dst[47:40] := Saturate_Int16_To_Int8 (a[95:80]) -tmp_dst[55:48] := Saturate_Int16_To_Int8 (a[111:96]) -tmp_dst[63:56] := Saturate_Int16_To_Int8 (a[127:112]) -tmp_dst[71:64] := Saturate_Int16_To_Int8 (b[15:0]) -tmp_dst[79:72] := Saturate_Int16_To_Int8 (b[31:16]) -tmp_dst[87:80] := Saturate_Int16_To_Int8 (b[47:32]) -tmp_dst[95:88] := Saturate_Int16_To_Int8 (b[63:48]) -tmp_dst[103:96] := Saturate_Int16_To_Int8 (b[79:64]) -tmp_dst[111:104] := Saturate_Int16_To_Int8 (b[95:80]) -tmp_dst[119:112] := Saturate_Int16_To_Int8 (b[111:96]) -tmp_dst[127:120] := Saturate_Int16_To_Int8 (b[127:112]) -tmp_dst[135:128] := Saturate_Int16_To_Int8 (a[143:128]) -tmp_dst[143:136] := Saturate_Int16_To_Int8 (a[159:144]) -tmp_dst[151:144] := Saturate_Int16_To_Int8 (a[175:160]) -tmp_dst[159:152] := Saturate_Int16_To_Int8 (a[191:176]) -tmp_dst[167:160] := Saturate_Int16_To_Int8 (a[207:192]) -tmp_dst[175:168] := Saturate_Int16_To_Int8 (a[223:208]) -tmp_dst[183:176] := Saturate_Int16_To_Int8 (a[239:224]) -tmp_dst[191:184] := Saturate_Int16_To_Int8 (a[255:240]) -tmp_dst[199:192] := Saturate_Int16_To_Int8 (b[143:128]) -tmp_dst[207:200] := Saturate_Int16_To_Int8 (b[159:144]) -tmp_dst[215:208] := Saturate_Int16_To_Int8 (b[175:160]) -tmp_dst[223:216] := Saturate_Int16_To_Int8 (b[191:176]) -tmp_dst[231:224] := Saturate_Int16_To_Int8 (b[207:192]) -tmp_dst[239:232] := Saturate_Int16_To_Int8 (b[223:208]) -tmp_dst[247:240] := Saturate_Int16_To_Int8 (b[239:224]) -tmp_dst[255:248] := Saturate_Int16_To_Int8 (b[255:240]) - -FOR j := 0 to 31 - i := j*8 - IF k[j] - dst[i+7:i] := tmp_dst[i+7:i] - ELSE - dst[i+7:i] := 0 + +
immintrin.h
+ + + Integer + Mask + AVX512F/KNCNI + Compare + + + + + Compare packed 32-bit integers in "a" and "b" for not-equal, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). + +FOR j := 0 to 15 + i := j*32 + IF k1[j] + k[j] := ( a[i+31:i] != b[i+31:i] ) ? 1 : 0 + ELSE + k[j] := 0 FI ENDFOR -dst[MAX:256] := 0 +k[MAX:16] := 0 - -
immintrin.h
- - - Integer - AVX512BW - Convert - Miscellaneous - - - - - Convert packed 16-bit integers from "a" and "b" to packed 8-bit integers using signed saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -tmp_dst[7:0] := Saturate_Int16_To_Int8 (a[15:0]) -tmp_dst[15:8] := Saturate_Int16_To_Int8 (a[31:16]) -tmp_dst[23:16] := Saturate_Int16_To_Int8 (a[47:32]) -tmp_dst[31:24] := Saturate_Int16_To_Int8 (a[63:48]) -tmp_dst[39:32] := Saturate_Int16_To_Int8 (a[79:64]) -tmp_dst[47:40] := Saturate_Int16_To_Int8 (a[95:80]) -tmp_dst[55:48] := Saturate_Int16_To_Int8 (a[111:96]) -tmp_dst[63:56] := Saturate_Int16_To_Int8 (a[127:112]) -tmp_dst[71:64] := Saturate_Int16_To_Int8 (b[15:0]) -tmp_dst[79:72] := Saturate_Int16_To_Int8 (b[31:16]) -tmp_dst[87:80] := Saturate_Int16_To_Int8 (b[47:32]) -tmp_dst[95:88] := Saturate_Int16_To_Int8 (b[63:48]) -tmp_dst[103:96] := Saturate_Int16_To_Int8 (b[79:64]) -tmp_dst[111:104] := Saturate_Int16_To_Int8 (b[95:80]) -tmp_dst[119:112] := Saturate_Int16_To_Int8 (b[111:96]) -tmp_dst[127:120] := Saturate_Int16_To_Int8 (b[127:112]) -tmp_dst[135:128] := Saturate_Int16_To_Int8 (a[143:128]) -tmp_dst[143:136] := Saturate_Int16_To_Int8 (a[159:144]) -tmp_dst[151:144] := Saturate_Int16_To_Int8 (a[175:160]) -tmp_dst[159:152] := Saturate_Int16_To_Int8 (a[191:176]) -tmp_dst[167:160] := Saturate_Int16_To_Int8 (a[207:192]) -tmp_dst[175:168] := Saturate_Int16_To_Int8 (a[223:208]) -tmp_dst[183:176] := Saturate_Int16_To_Int8 (a[239:224]) -tmp_dst[191:184] := Saturate_Int16_To_Int8 (a[255:240]) -tmp_dst[199:192] := Saturate_Int16_To_Int8 (b[143:128]) -tmp_dst[207:200] := Saturate_Int16_To_Int8 (b[159:144]) -tmp_dst[215:208] := Saturate_Int16_To_Int8 (b[175:160]) -tmp_dst[223:216] := Saturate_Int16_To_Int8 (b[191:176]) -tmp_dst[231:224] := Saturate_Int16_To_Int8 (b[207:192]) -tmp_dst[239:232] := Saturate_Int16_To_Int8 (b[223:208]) -tmp_dst[247:240] := Saturate_Int16_To_Int8 (b[239:224]) -tmp_dst[255:248] := Saturate_Int16_To_Int8 (b[255:240]) -tmp_dst[263:256] := Saturate_Int16_To_Int8 (a[271:256]) -tmp_dst[271:264] := Saturate_Int16_To_Int8 (a[287:272]) -tmp_dst[279:272] := Saturate_Int16_To_Int8 (a[303:288]) -tmp_dst[287:280] := Saturate_Int16_To_Int8 (a[319:304]) -tmp_dst[295:288] := Saturate_Int16_To_Int8 (a[335:320]) -tmp_dst[303:296] := Saturate_Int16_To_Int8 (a[351:336]) -tmp_dst[311:304] := Saturate_Int16_To_Int8 (a[367:352]) -tmp_dst[319:312] := Saturate_Int16_To_Int8 (a[383:368]) -tmp_dst[327:320] := Saturate_Int16_To_Int8 (b[271:256]) -tmp_dst[335:328] := Saturate_Int16_To_Int8 (b[287:272]) -tmp_dst[343:336] := Saturate_Int16_To_Int8 (b[303:288]) -tmp_dst[351:344] := Saturate_Int16_To_Int8 (b[319:304]) -tmp_dst[359:352] := Saturate_Int16_To_Int8 (b[335:320]) -tmp_dst[367:360] := Saturate_Int16_To_Int8 (b[351:336]) -tmp_dst[375:368] := Saturate_Int16_To_Int8 (b[367:352]) -tmp_dst[383:376] := Saturate_Int16_To_Int8 (b[383:368]) -tmp_dst[391:384] := Saturate_Int16_To_Int8 (a[399:384]) -tmp_dst[399:392] := Saturate_Int16_To_Int8 (a[415:400]) -tmp_dst[407:400] := Saturate_Int16_To_Int8 (a[431:416]) -tmp_dst[415:408] := Saturate_Int16_To_Int8 (a[447:432]) -tmp_dst[423:416] := Saturate_Int16_To_Int8 (a[463:448]) -tmp_dst[431:424] := Saturate_Int16_To_Int8 (a[479:464]) -tmp_dst[439:432] := Saturate_Int16_To_Int8 (a[495:480]) -tmp_dst[447:440] := Saturate_Int16_To_Int8 (a[511:496]) -tmp_dst[455:448] := Saturate_Int16_To_Int8 (b[399:384]) -tmp_dst[463:456] := Saturate_Int16_To_Int8 (b[415:400]) -tmp_dst[471:464] := Saturate_Int16_To_Int8 (b[431:416]) -tmp_dst[479:472] := Saturate_Int16_To_Int8 (b[447:432]) -tmp_dst[487:480] := Saturate_Int16_To_Int8 (b[463:448]) -tmp_dst[495:488] := Saturate_Int16_To_Int8 (b[479:464]) -tmp_dst[503:496] := Saturate_Int16_To_Int8 (b[495:480]) -tmp_dst[511:504] := Saturate_Int16_To_Int8 (b[511:496]) - -FOR j := 0 to 63 - i := j*8 - IF k[j] - dst[i+7:i] := tmp_dst[i+7:i] - ELSE - dst[i+7:i] := src[i+7:i] - FI + +
immintrin.h
+ + + Integer + Mask + AVX512F/KNCNI + Compare + + + + + Compare packed unsigned 32-bit integers in "a" and "b" based on the comparison operand specified by "imm8", and store the results in mask vector "k". + CASE (imm8[2:0]) OF +0: OP := _MM_CMPINT_EQ +1: OP := _MM_CMPINT_LT +2: OP := _MM_CMPINT_LE +3: OP := _MM_CMPINT_FALSE +4: OP := _MM_CMPINT_NE +5: OP := _MM_CMPINT_NLT +6: OP := _MM_CMPINT_NLE +7: OP := _MM_CMPINT_TRUE +ESAC +FOR j := 0 to 15 + i := j*32 + k[j] := ( a[i+31:i] OP b[i+31:i] ) ? 1 : 0 ENDFOR -dst[MAX:512] := 0 +k[MAX:16] := 0 - -
immintrin.h
- - - Integer - AVX512BW - Convert - Miscellaneous - - - - Convert packed 16-bit integers from "a" and "b" to packed 8-bit integers using signed saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - - -tmp_dst[7:0] := Saturate_Int16_To_Int8 (a[15:0]) -tmp_dst[15:8] := Saturate_Int16_To_Int8 (a[31:16]) -tmp_dst[23:16] := Saturate_Int16_To_Int8 (a[47:32]) -tmp_dst[31:24] := Saturate_Int16_To_Int8 (a[63:48]) -tmp_dst[39:32] := Saturate_Int16_To_Int8 (a[79:64]) -tmp_dst[47:40] := Saturate_Int16_To_Int8 (a[95:80]) -tmp_dst[55:48] := Saturate_Int16_To_Int8 (a[111:96]) -tmp_dst[63:56] := Saturate_Int16_To_Int8 (a[127:112]) -tmp_dst[71:64] := Saturate_Int16_To_Int8 (b[15:0]) -tmp_dst[79:72] := Saturate_Int16_To_Int8 (b[31:16]) -tmp_dst[87:80] := Saturate_Int16_To_Int8 (b[47:32]) -tmp_dst[95:88] := Saturate_Int16_To_Int8 (b[63:48]) -tmp_dst[103:96] := Saturate_Int16_To_Int8 (b[79:64]) -tmp_dst[111:104] := Saturate_Int16_To_Int8 (b[95:80]) -tmp_dst[119:112] := Saturate_Int16_To_Int8 (b[111:96]) -tmp_dst[127:120] := Saturate_Int16_To_Int8 (b[127:112]) -tmp_dst[135:128] := Saturate_Int16_To_Int8 (a[143:128]) -tmp_dst[143:136] := Saturate_Int16_To_Int8 (a[159:144]) -tmp_dst[151:144] := Saturate_Int16_To_Int8 (a[175:160]) -tmp_dst[159:152] := Saturate_Int16_To_Int8 (a[191:176]) -tmp_dst[167:160] := Saturate_Int16_To_Int8 (a[207:192]) -tmp_dst[175:168] := Saturate_Int16_To_Int8 (a[223:208]) -tmp_dst[183:176] := Saturate_Int16_To_Int8 (a[239:224]) -tmp_dst[191:184] := Saturate_Int16_To_Int8 (a[255:240]) -tmp_dst[199:192] := Saturate_Int16_To_Int8 (b[143:128]) -tmp_dst[207:200] := Saturate_Int16_To_Int8 (b[159:144]) -tmp_dst[215:208] := Saturate_Int16_To_Int8 (b[175:160]) -tmp_dst[223:216] := Saturate_Int16_To_Int8 (b[191:176]) -tmp_dst[231:224] := Saturate_Int16_To_Int8 (b[207:192]) -tmp_dst[239:232] := Saturate_Int16_To_Int8 (b[223:208]) -tmp_dst[247:240] := Saturate_Int16_To_Int8 (b[239:224]) -tmp_dst[255:248] := Saturate_Int16_To_Int8 (b[255:240]) -tmp_dst[263:256] := Saturate_Int16_To_Int8 (a[271:256]) -tmp_dst[271:264] := Saturate_Int16_To_Int8 (a[287:272]) -tmp_dst[279:272] := Saturate_Int16_To_Int8 (a[303:288]) -tmp_dst[287:280] := Saturate_Int16_To_Int8 (a[319:304]) -tmp_dst[295:288] := Saturate_Int16_To_Int8 (a[335:320]) -tmp_dst[303:296] := Saturate_Int16_To_Int8 (a[351:336]) -tmp_dst[311:304] := Saturate_Int16_To_Int8 (a[367:352]) -tmp_dst[319:312] := Saturate_Int16_To_Int8 (a[383:368]) -tmp_dst[327:320] := Saturate_Int16_To_Int8 (b[271:256]) -tmp_dst[335:328] := Saturate_Int16_To_Int8 (b[287:272]) -tmp_dst[343:336] := Saturate_Int16_To_Int8 (b[303:288]) -tmp_dst[351:344] := Saturate_Int16_To_Int8 (b[319:304]) -tmp_dst[359:352] := Saturate_Int16_To_Int8 (b[335:320]) -tmp_dst[367:360] := Saturate_Int16_To_Int8 (b[351:336]) -tmp_dst[375:368] := Saturate_Int16_To_Int8 (b[367:352]) -tmp_dst[383:376] := Saturate_Int16_To_Int8 (b[383:368]) -tmp_dst[391:384] := Saturate_Int16_To_Int8 (a[399:384]) -tmp_dst[399:392] := Saturate_Int16_To_Int8 (a[415:400]) -tmp_dst[407:400] := Saturate_Int16_To_Int8 (a[431:416]) -tmp_dst[415:408] := Saturate_Int16_To_Int8 (a[447:432]) -tmp_dst[423:416] := Saturate_Int16_To_Int8 (a[463:448]) -tmp_dst[431:424] := Saturate_Int16_To_Int8 (a[479:464]) -tmp_dst[439:432] := Saturate_Int16_To_Int8 (a[495:480]) -tmp_dst[447:440] := Saturate_Int16_To_Int8 (a[511:496]) -tmp_dst[455:448] := Saturate_Int16_To_Int8 (b[399:384]) -tmp_dst[463:456] := Saturate_Int16_To_Int8 (b[415:400]) -tmp_dst[471:464] := Saturate_Int16_To_Int8 (b[431:416]) -tmp_dst[479:472] := Saturate_Int16_To_Int8 (b[447:432]) -tmp_dst[487:480] := Saturate_Int16_To_Int8 (b[463:448]) -tmp_dst[495:488] := Saturate_Int16_To_Int8 (b[479:464]) -tmp_dst[503:496] := Saturate_Int16_To_Int8 (b[495:480]) -tmp_dst[511:504] := Saturate_Int16_To_Int8 (b[511:496]) - -FOR j := 0 to 63 - i := j*8 - IF k[j] - dst[i+7:i] := tmp_dst[i+7:i] - ELSE - dst[i+7:i] := 0 - FI + +
immintrin.h
+ + + Integer + Mask + AVX512F/KNCNI + Compare + + + + Compare packed unsigned 32-bit integers in "a" and "b" for equality, and store the results in mask vector "k". + +FOR j := 0 to 15 + i := j*32 + k[j] := ( a[i+31:i] == b[i+31:i] ) ? 1 : 0 ENDFOR -dst[MAX:512] := 0 +k[MAX:16] := 0 - -
immintrin.h
- - - Integer - AVX512BW - Convert - Miscellaneous - - - Convert packed 16-bit integers from "a" and "b" to packed 8-bit integers using signed saturation, and store the results in "dst". - - -dst[7:0] := Saturate_Int16_To_Int8 (a[15:0]) -dst[15:8] := Saturate_Int16_To_Int8 (a[31:16]) -dst[23:16] := Saturate_Int16_To_Int8 (a[47:32]) -dst[31:24] := Saturate_Int16_To_Int8 (a[63:48]) -dst[39:32] := Saturate_Int16_To_Int8 (a[79:64]) -dst[47:40] := Saturate_Int16_To_Int8 (a[95:80]) -dst[55:48] := Saturate_Int16_To_Int8 (a[111:96]) -dst[63:56] := Saturate_Int16_To_Int8 (a[127:112]) -dst[71:64] := Saturate_Int16_To_Int8 (b[15:0]) -dst[79:72] := Saturate_Int16_To_Int8 (b[31:16]) -dst[87:80] := Saturate_Int16_To_Int8 (b[47:32]) -dst[95:88] := Saturate_Int16_To_Int8 (b[63:48]) -dst[103:96] := Saturate_Int16_To_Int8 (b[79:64]) -dst[111:104] := Saturate_Int16_To_Int8 (b[95:80]) -dst[119:112] := Saturate_Int16_To_Int8 (b[111:96]) -dst[127:120] := Saturate_Int16_To_Int8 (b[127:112]) -dst[135:128] := Saturate_Int16_To_Int8 (a[143:128]) -dst[143:136] := Saturate_Int16_To_Int8 (a[159:144]) -dst[151:144] := Saturate_Int16_To_Int8 (a[175:160]) -dst[159:152] := Saturate_Int16_To_Int8 (a[191:176]) -dst[167:160] := Saturate_Int16_To_Int8 (a[207:192]) -dst[175:168] := Saturate_Int16_To_Int8 (a[223:208]) -dst[183:176] := Saturate_Int16_To_Int8 (a[239:224]) -dst[191:184] := Saturate_Int16_To_Int8 (a[255:240]) -dst[199:192] := Saturate_Int16_To_Int8 (b[143:128]) -dst[207:200] := Saturate_Int16_To_Int8 (b[159:144]) -dst[215:208] := Saturate_Int16_To_Int8 (b[175:160]) -dst[223:216] := Saturate_Int16_To_Int8 (b[191:176]) -dst[231:224] := Saturate_Int16_To_Int8 (b[207:192]) -dst[239:232] := Saturate_Int16_To_Int8 (b[223:208]) -dst[247:240] := Saturate_Int16_To_Int8 (b[239:224]) -dst[255:248] := Saturate_Int16_To_Int8 (b[255:240]) -dst[263:256] := Saturate_Int16_To_Int8 (a[271:256]) -dst[271:264] := Saturate_Int16_To_Int8 (a[287:272]) -dst[279:272] := Saturate_Int16_To_Int8 (a[303:288]) -dst[287:280] := Saturate_Int16_To_Int8 (a[319:304]) -dst[295:288] := Saturate_Int16_To_Int8 (a[335:320]) -dst[303:296] := Saturate_Int16_To_Int8 (a[351:336]) -dst[311:304] := Saturate_Int16_To_Int8 (a[367:352]) -dst[319:312] := Saturate_Int16_To_Int8 (a[383:368]) -dst[327:320] := Saturate_Int16_To_Int8 (b[271:256]) -dst[335:328] := Saturate_Int16_To_Int8 (b[287:272]) -dst[343:336] := Saturate_Int16_To_Int8 (b[303:288]) -dst[351:344] := Saturate_Int16_To_Int8 (b[319:304]) -dst[359:352] := Saturate_Int16_To_Int8 (b[335:320]) -dst[367:360] := Saturate_Int16_To_Int8 (b[351:336]) -dst[375:368] := Saturate_Int16_To_Int8 (b[367:352]) -dst[383:376] := Saturate_Int16_To_Int8 (b[383:368]) -dst[391:384] := Saturate_Int16_To_Int8 (a[399:384]) -dst[399:392] := Saturate_Int16_To_Int8 (a[415:400]) -dst[407:400] := Saturate_Int16_To_Int8 (a[431:416]) -dst[415:408] := Saturate_Int16_To_Int8 (a[447:432]) -dst[423:416] := Saturate_Int16_To_Int8 (a[463:448]) -dst[431:424] := Saturate_Int16_To_Int8 (a[479:464]) -dst[439:432] := Saturate_Int16_To_Int8 (a[495:480]) -dst[447:440] := Saturate_Int16_To_Int8 (a[511:496]) -dst[455:448] := Saturate_Int16_To_Int8 (b[399:384]) -dst[463:456] := Saturate_Int16_To_Int8 (b[415:400]) -dst[471:464] := Saturate_Int16_To_Int8 (b[431:416]) -dst[479:472] := Saturate_Int16_To_Int8 (b[447:432]) -dst[487:480] := Saturate_Int16_To_Int8 (b[463:448]) -dst[495:488] := Saturate_Int16_To_Int8 (b[479:464]) -dst[503:496] := Saturate_Int16_To_Int8 (b[495:480]) -dst[511:504] := Saturate_Int16_To_Int8 (b[511:496]) -dst[MAX:512] := 0 - - -
immintrin.h
- - - Integer - AVX512VL - AVX512BW - Convert - Miscellaneous - - - - - Convert packed 16-bit integers from "a" and "b" to packed 8-bit integers using signed saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -tmp_dst[7:0] := Saturate_Int16_To_Int8 (a[15:0]) -tmp_dst[15:8] := Saturate_Int16_To_Int8 (a[31:16]) -tmp_dst[23:16] := Saturate_Int16_To_Int8 (a[47:32]) -tmp_dst[31:24] := Saturate_Int16_To_Int8 (a[63:48]) -tmp_dst[39:32] := Saturate_Int16_To_Int8 (a[79:64]) -tmp_dst[47:40] := Saturate_Int16_To_Int8 (a[95:80]) -tmp_dst[55:48] := Saturate_Int16_To_Int8 (a[111:96]) -tmp_dst[63:56] := Saturate_Int16_To_Int8 (a[127:112]) -tmp_dst[71:64] := Saturate_Int16_To_Int8 (b[15:0]) -tmp_dst[79:72] := Saturate_Int16_To_Int8 (b[31:16]) -tmp_dst[87:80] := Saturate_Int16_To_Int8 (b[47:32]) -tmp_dst[95:88] := Saturate_Int16_To_Int8 (b[63:48]) -tmp_dst[103:96] := Saturate_Int16_To_Int8 (b[79:64]) -tmp_dst[111:104] := Saturate_Int16_To_Int8 (b[95:80]) -tmp_dst[119:112] := Saturate_Int16_To_Int8 (b[111:96]) -tmp_dst[127:120] := Saturate_Int16_To_Int8 (b[127:112]) - + +
immintrin.h
+ + + Integer + Mask + AVX512F/KNCNI + Compare + + + + Compare packed unsigned 32-bit integers in "a" and "b" for greater-than-or-equal, and store the results in mask vector "k". + FOR j := 0 to 15 - i := j*8 - IF k[j] - dst[i+7:i] := tmp_dst[i+7:i] - ELSE - dst[i+7:i] := src[i+7:i] - FI + i := j*32 + k[j] := ( a[i+31:i] >= b[i+31:i] ) ? 1 : 0 ENDFOR -dst[MAX:128] := 0 +k[MAX:16] := 0 - +
immintrin.h
 - + Integer - AVX512VL - AVX512BW - Convert - Miscellaneous - - - - Convert packed 16-bit integers from "a" and "b" to packed 8-bit integers using signed saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - + Mask + AVX512F/KNCNI + Compare + + + + Compare packed unsigned 32-bit integers in "a" and "b" for greater-than, and store the results in mask vector "k". -tmp_dst[7:0] := Saturate_Int16_To_Int8 (a[15:0]) -tmp_dst[15:8] := Saturate_Int16_To_Int8 (a[31:16]) -tmp_dst[23:16] := Saturate_Int16_To_Int8 (a[47:32]) -tmp_dst[31:24] := Saturate_Int16_To_Int8 (a[63:48]) -tmp_dst[39:32] := Saturate_Int16_To_Int8 (a[79:64]) -tmp_dst[47:40] := Saturate_Int16_To_Int8 (a[95:80]) -tmp_dst[55:48] := Saturate_Int16_To_Int8 (a[111:96]) -tmp_dst[63:56] := Saturate_Int16_To_Int8 (a[127:112]) -tmp_dst[71:64] := Saturate_Int16_To_Int8 (b[15:0]) -tmp_dst[79:72] := Saturate_Int16_To_Int8 (b[31:16]) -tmp_dst[87:80] := Saturate_Int16_To_Int8 (b[47:32]) -tmp_dst[95:88] := Saturate_Int16_To_Int8 (b[63:48]) -tmp_dst[103:96] := Saturate_Int16_To_Int8 (b[79:64]) -tmp_dst[111:104] := Saturate_Int16_To_Int8 (b[95:80]) -tmp_dst[119:112] := Saturate_Int16_To_Int8 (b[111:96]) -tmp_dst[127:120] := Saturate_Int16_To_Int8 (b[127:112]) - FOR j := 0 to 15 - i := j*8 - IF k[j] - dst[i+7:i] := tmp_dst[i+7:i] - ELSE - dst[i+7:i] := 0 - FI + i := j*32 + k[j] := ( a[i+31:i] > b[i+31:i] ) ? 1 : 0 ENDFOR -dst[MAX:128] := 0 +k[MAX:16] := 0 - +
immintrin.h
 - + Integer - AVX512VL - AVX512BW - Convert - Miscellaneous - - - - - Convert packed 32-bit integers from "a" and "b" to packed 16-bit integers using unsigned saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Mask + AVX512F/KNCNI + Compare + + + + Compare packed unsigned 32-bit integers in "a" and "b" for less-than-or-equal, and store the results in mask vector "k". -tmp_dst[15:0] := Saturate_Int32_To_UnsignedInt16 (a[31:0]) -tmp_dst[31:16] := Saturate_Int32_To_UnsignedInt16 (a[63:32]) -tmp_dst[47:32] := Saturate_Int32_To_UnsignedInt16 (a[95:64]) -tmp_dst[63:48] := Saturate_Int32_To_UnsignedInt16 (a[127:96]) -tmp_dst[79:64] := Saturate_Int32_To_UnsignedInt16 (b[31:0]) -tmp_dst[95:80] := Saturate_Int32_To_UnsignedInt16 (b[63:32]) -tmp_dst[111:96] := Saturate_Int32_To_UnsignedInt16 (b[95:64]) -tmp_dst[127:112] := Saturate_Int32_To_UnsignedInt16 (b[127:96]) -tmp_dst[143:128] := Saturate_Int32_To_UnsignedInt16 (a[159:128]) -tmp_dst[159:144] := Saturate_Int32_To_UnsignedInt16 (a[191:160]) -tmp_dst[175:160] := Saturate_Int32_To_UnsignedInt16 (a[223:192]) -tmp_dst[191:176] := Saturate_Int32_To_UnsignedInt16 (a[255:224]) -tmp_dst[207:192] := Saturate_Int32_To_UnsignedInt16 (b[159:128]) -tmp_dst[223:208] := Saturate_Int32_To_UnsignedInt16 (b[191:160]) -tmp_dst[239:224] := Saturate_Int32_To_UnsignedInt16 (b[223:192]) -tmp_dst[255:240] := Saturate_Int32_To_UnsignedInt16 (b[255:224]) - FOR j := 0 to 15 - i := j*16 - IF k[j] - dst[i+15:i] := tmp_dst[i+15:i] - ELSE - dst[i+15:i] := src[i+15:i] - FI + i := j*32 + k[j] := ( a[i+31:i] <= b[i+31:i] ) ? 1 : 0 ENDFOR -dst[MAX:256] := 0 +k[MAX:16] := 0 - +
immintrin.h
 - + Integer - AVX512VL - AVX512BW - Convert - Miscellaneous - - - - Convert packed 32-bit integers from "a" and "b" to packed 16-bit integers using unsigned saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Mask + AVX512F/KNCNI + Compare + + + + Compare packed unsigned 32-bit integers in "a" and "b" for less-than, and store the results in mask vector "k". -tmp_dst[15:0] := Saturate_Int32_To_UnsignedInt16 (a[31:0]) -tmp_dst[31:16] := Saturate_Int32_To_UnsignedInt16 (a[63:32]) -tmp_dst[47:32] := Saturate_Int32_To_UnsignedInt16 (a[95:64]) -tmp_dst[63:48] := Saturate_Int32_To_UnsignedInt16 (a[127:96]) -tmp_dst[79:64] := Saturate_Int32_To_UnsignedInt16 (b[31:0]) -tmp_dst[95:80] := Saturate_Int32_To_UnsignedInt16 (b[63:32]) -tmp_dst[111:96] := Saturate_Int32_To_UnsignedInt16 (b[95:64]) -tmp_dst[127:112] := Saturate_Int32_To_UnsignedInt16 (b[127:96]) -tmp_dst[143:128] := Saturate_Int32_To_UnsignedInt16 (a[159:128]) -tmp_dst[159:144] := Saturate_Int32_To_UnsignedInt16 (a[191:160]) -tmp_dst[175:160] := Saturate_Int32_To_UnsignedInt16 (a[223:192]) -tmp_dst[191:176] := Saturate_Int32_To_UnsignedInt16 (a[255:224]) -tmp_dst[207:192] := Saturate_Int32_To_UnsignedInt16 (b[159:128]) -tmp_dst[223:208] := Saturate_Int32_To_UnsignedInt16 (b[191:160]) -tmp_dst[239:224] := Saturate_Int32_To_UnsignedInt16 (b[223:192]) -tmp_dst[255:240] := Saturate_Int32_To_UnsignedInt16 (b[255:224]) - FOR j := 0 to 15 - i := j*16 - IF k[j] - dst[i+15:i] := tmp_dst[i+15:i] - ELSE - dst[i+15:i] := 0 - FI + i := j*32 + k[j] := ( a[i+31:i] < b[i+31:i] ) ? 1 : 0 ENDFOR -dst[MAX:256] := 0 +k[MAX:16] := 0 - -
immintrin.h
- - - Integer - AVX512BW - Convert - Miscellaneous - - - - - Convert packed 32-bit integers from "a" and "b" to packed 16-bit integers using unsigned saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -tmp_dst[15:0] := Saturate_Int32_To_UnsignedInt16 (a[31:0]) -tmp_dst[31:16] := Saturate_Int32_To_UnsignedInt16 (a[63:32]) -tmp_dst[47:32] := Saturate_Int32_To_UnsignedInt16 (a[95:64]) -tmp_dst[63:48] := Saturate_Int32_To_UnsignedInt16 (a[127:96]) -tmp_dst[79:64] := Saturate_Int32_To_UnsignedInt16 (b[31:0]) -tmp_dst[95:80] := Saturate_Int32_To_UnsignedInt16 (b[63:32]) -tmp_dst[111:96] := Saturate_Int32_To_UnsignedInt16 (b[95:64]) -tmp_dst[127:112] := Saturate_Int32_To_UnsignedInt16 (b[127:96]) -tmp_dst[143:128] := Saturate_Int32_To_UnsignedInt16 (a[159:128]) -tmp_dst[159:144] := Saturate_Int32_To_UnsignedInt16 (a[191:160]) -tmp_dst[175:160] := Saturate_Int32_To_UnsignedInt16 (a[223:192]) -tmp_dst[191:176] := Saturate_Int32_To_UnsignedInt16 (a[255:224]) -tmp_dst[207:192] := Saturate_Int32_To_UnsignedInt16 (b[159:128]) -tmp_dst[223:208] := Saturate_Int32_To_UnsignedInt16 (b[191:160]) -tmp_dst[239:224] := Saturate_Int32_To_UnsignedInt16 (b[223:192]) -tmp_dst[255:240] := Saturate_Int32_To_UnsignedInt16 (b[255:224]) -tmp_dst[271:256] := Saturate_Int32_To_UnsignedInt16 (a[287:256]) -tmp_dst[287:272] := Saturate_Int32_To_UnsignedInt16 (a[319:288]) -tmp_dst[303:288] := Saturate_Int32_To_UnsignedInt16 (a[351:320]) -tmp_dst[319:304] := Saturate_Int32_To_UnsignedInt16 (a[383:352]) -tmp_dst[335:320] := Saturate_Int32_To_UnsignedInt16 (b[287:256]) -tmp_dst[351:336] := Saturate_Int32_To_UnsignedInt16 (b[319:288]) -tmp_dst[367:352] := Saturate_Int32_To_UnsignedInt16 (b[351:320]) -tmp_dst[383:368] := Saturate_Int32_To_UnsignedInt16 (b[383:352]) -tmp_dst[399:384] := Saturate_Int32_To_UnsignedInt16 (a[415:384]) -tmp_dst[415:400] := Saturate_Int32_To_UnsignedInt16 (a[447:416]) -tmp_dst[431:416] := Saturate_Int32_To_UnsignedInt16 (a[479:448]) -tmp_dst[447:432] := Saturate_Int32_To_UnsignedInt16 (a[511:480]) -tmp_dst[463:448] := Saturate_Int32_To_UnsignedInt16 (b[415:384]) -tmp_dst[479:464] := Saturate_Int32_To_UnsignedInt16 (b[447:416]) -tmp_dst[495:480] := Saturate_Int32_To_UnsignedInt16 (b[479:448]) -tmp_dst[511:496] := Saturate_Int32_To_UnsignedInt16 (b[511:480]) - -FOR j := 0 to 31 - i := j*16 - IF k[j] - dst[i+15:i] := tmp_dst[i+15:i] - ELSE - dst[i+15:i] := src[i+15:i] - FI + +
immintrin.h
+ + + Integer + Mask + AVX512F/KNCNI + Compare + + + + Compare packed unsigned 32-bit integers in "a" and "b" for not-equal, and store the results in mask vector "k". + +FOR j := 0 to 15 + i := j*32 + k[j] := ( a[i+31:i] != b[i+31:i] ) ? 1 : 0 ENDFOR -dst[MAX:512] := 0 +k[MAX:16] := 0 - -
immintrin.h
- - - Integer - AVX512BW - Convert - Miscellaneous - - - - Convert packed 32-bit integers from "a" and "b" to packed 16-bit integers using unsigned saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - -tmp_dst[15:0] := Saturate_Int32_To_UnsignedInt16 (a[31:0]) -tmp_dst[31:16] := Saturate_Int32_To_UnsignedInt16 (a[63:32]) -tmp_dst[47:32] := Saturate_Int32_To_UnsignedInt16 (a[95:64]) -tmp_dst[63:48] := Saturate_Int32_To_UnsignedInt16 (a[127:96]) -tmp_dst[79:64] := Saturate_Int32_To_UnsignedInt16 (b[31:0]) -tmp_dst[95:80] := Saturate_Int32_To_UnsignedInt16 (b[63:32]) -tmp_dst[111:96] := Saturate_Int32_To_UnsignedInt16 (b[95:64]) -tmp_dst[127:112] := Saturate_Int32_To_UnsignedInt16 (b[127:96]) -tmp_dst[143:128] := Saturate_Int32_To_UnsignedInt16 (a[159:128]) -tmp_dst[159:144] := Saturate_Int32_To_UnsignedInt16 (a[191:160]) -tmp_dst[175:160] := Saturate_Int32_To_UnsignedInt16 (a[223:192]) -tmp_dst[191:176] := Saturate_Int32_To_UnsignedInt16 (a[255:224]) -tmp_dst[207:192] := Saturate_Int32_To_UnsignedInt16 (b[159:128]) -tmp_dst[223:208] := Saturate_Int32_To_UnsignedInt16 (b[191:160]) -tmp_dst[239:224] := Saturate_Int32_To_UnsignedInt16 (b[223:192]) -tmp_dst[255:240] := Saturate_Int32_To_UnsignedInt16 (b[255:224]) -tmp_dst[271:256] := Saturate_Int32_To_UnsignedInt16 (a[287:256]) -tmp_dst[287:272] := Saturate_Int32_To_UnsignedInt16 (a[319:288]) -tmp_dst[303:288] := Saturate_Int32_To_UnsignedInt16 (a[351:320]) -tmp_dst[319:304] := Saturate_Int32_To_UnsignedInt16 (a[383:352]) -tmp_dst[335:320] := Saturate_Int32_To_UnsignedInt16 (b[287:256]) -tmp_dst[351:336] := Saturate_Int32_To_UnsignedInt16 (b[319:288]) -tmp_dst[367:352] := Saturate_Int32_To_UnsignedInt16 (b[351:320]) -tmp_dst[383:368] := Saturate_Int32_To_UnsignedInt16 (b[383:352]) -tmp_dst[399:384] := Saturate_Int32_To_UnsignedInt16 (a[415:384]) -tmp_dst[415:400] := Saturate_Int32_To_UnsignedInt16 (a[447:416]) -tmp_dst[431:416] := Saturate_Int32_To_UnsignedInt16 (a[479:448]) -tmp_dst[447:432] := Saturate_Int32_To_UnsignedInt16 (a[511:480]) -tmp_dst[463:448] := Saturate_Int32_To_UnsignedInt16 (b[415:384]) -tmp_dst[479:464] := Saturate_Int32_To_UnsignedInt16 (b[447:416]) -tmp_dst[495:480] := Saturate_Int32_To_UnsignedInt16 (b[479:448]) -tmp_dst[511:496] := Saturate_Int32_To_UnsignedInt16 (b[511:480]) - -FOR j := 0 to 31 - i := j*16 - IF k[j] - dst[i+15:i] := tmp_dst[i+15:i] - ELSE - dst[i+15:i] := 0 + +
immintrin.h
+ + + Integer + Mask + AVX512F/KNCNI + Compare + + + + + + Compare packed unsigned 32-bit integers in "a" and "b" based on the comparison operand specified by "imm8", and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). + CASE (imm8[2:0]) OF +0: OP := _MM_CMPINT_EQ +1: OP := _MM_CMPINT_LT +2: OP := _MM_CMPINT_LE +3: OP := _MM_CMPINT_FALSE +4: OP := _MM_CMPINT_NE +5: OP := _MM_CMPINT_NLT +6: OP := _MM_CMPINT_NLE +7: OP := _MM_CMPINT_TRUE +ESAC +FOR j := 0 to 15 + i := j*32 + IF k1[j] + k[j] := ( a[i+31:i] OP b[i+31:i] ) ? 1 : 0 + ELSE + k[j] := 0 FI ENDFOR -dst[MAX:512] := 0 +k[MAX:16] := 0 - +
immintrin.h
 - + Integer - AVX512BW - Convert - Miscellaneous - - - Convert packed 32-bit integers from "a" and "b" to packed 16-bit integers using unsigned saturation, and store the results in "dst". + Mask + AVX512F/KNCNI + Compare + + + + + Compare packed unsigned 32-bit integers in "a" and "b" for equality, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). -dst[15:0] := Saturate_Int32_To_UnsignedInt16 (a[31:0]) -dst[31:16] := Saturate_Int32_To_UnsignedInt16 (a[63:32]) -dst[47:32] := Saturate_Int32_To_UnsignedInt16 (a[95:64]) -dst[63:48] := Saturate_Int32_To_UnsignedInt16 (a[127:96]) -dst[79:64] := Saturate_Int32_To_UnsignedInt16 (b[31:0]) -dst[95:80] := Saturate_Int32_To_UnsignedInt16 (b[63:32]) -dst[111:96] := Saturate_Int32_To_UnsignedInt16 (b[95:64]) -dst[127:112] := Saturate_Int32_To_UnsignedInt16 (b[127:96]) -dst[143:128] := Saturate_Int32_To_UnsignedInt16 (a[159:128]) -dst[159:144] := Saturate_Int32_To_UnsignedInt16 (a[191:160]) -dst[175:160] := Saturate_Int32_To_UnsignedInt16 (a[223:192]) -dst[191:176] := Saturate_Int32_To_UnsignedInt16 (a[255:224]) -dst[207:192] := Saturate_Int32_To_UnsignedInt16 (b[159:128]) -dst[223:208] := Saturate_Int32_To_UnsignedInt16 (b[191:160]) -dst[239:224] := Saturate_Int32_To_UnsignedInt16 (b[223:192]) -dst[255:240] := Saturate_Int32_To_UnsignedInt16 (b[255:224]) -dst[271:256] := Saturate_Int32_To_UnsignedInt16 (a[287:256]) -dst[287:272] := Saturate_Int32_To_UnsignedInt16 (a[319:288]) -dst[303:288] := Saturate_Int32_To_UnsignedInt16 (a[351:320]) -dst[319:304] := Saturate_Int32_To_UnsignedInt16 (a[383:352]) -dst[335:320] := Saturate_Int32_To_UnsignedInt16 (b[287:256]) -dst[351:336] := Saturate_Int32_To_UnsignedInt16 (b[319:288]) -dst[367:352] := Saturate_Int32_To_UnsignedInt16 (b[351:320]) -dst[383:368] := Saturate_Int32_To_UnsignedInt16 (b[383:352]) -dst[399:384] := Saturate_Int32_To_UnsignedInt16 (a[415:384]) -dst[415:400] := Saturate_Int32_To_UnsignedInt16 (a[447:416]) -dst[431:416] := Saturate_Int32_To_UnsignedInt16 (a[479:448]) -dst[447:432] := Saturate_Int32_To_UnsignedInt16 (a[511:480]) -dst[463:448] := Saturate_Int32_To_UnsignedInt16 (b[415:384]) -dst[479:464] := Saturate_Int32_To_UnsignedInt16 (b[447:416]) -dst[495:480] := Saturate_Int32_To_UnsignedInt16 (b[479:448]) -dst[511:496] := Saturate_Int32_To_UnsignedInt16 (b[511:480]) -dst[MAX:512] := 0 +FOR j := 0 to 15 + i := j*32 + IF k1[j] + k[j] := ( a[i+31:i] == b[i+31:i] ) ? 1 : 0 + ELSE + k[j] := 0 + FI +ENDFOR +k[MAX:16] := 0 - +
immintrin.h
 - + Integer - AVX512VL - AVX512BW - Convert - Miscellaneous - - - - - Convert packed 32-bit integers from "a" and "b" to packed 16-bit integers using unsigned saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Mask + AVX512F/KNCNI + Compare + + + + + Compare packed unsigned 32-bit integers in "a" and "b" for greater-than-or-equal, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). -tmp_dst[15:0] := Saturate_Int32_To_UnsignedInt16 (a[31:0]) -tmp_dst[31:16] := Saturate_Int32_To_UnsignedInt16 (a[63:32]) -tmp_dst[47:32] := Saturate_Int32_To_UnsignedInt16 (a[95:64]) -tmp_dst[63:48] := Saturate_Int32_To_UnsignedInt16 (a[127:96]) -tmp_dst[79:64] := Saturate_Int32_To_UnsignedInt16 (b[31:0]) -tmp_dst[95:80] := Saturate_Int32_To_UnsignedInt16 (b[63:32]) -tmp_dst[111:96] := Saturate_Int32_To_UnsignedInt16 (b[95:64]) -tmp_dst[127:112] := Saturate_Int32_To_UnsignedInt16 (b[127:96]) - -FOR j := 0 to 7 - i := j*16 - IF k[j] - dst[i+15:i] := tmp_dst[i+15:i] - ELSE - dst[i+15:i] := src[i+15:i] +FOR j := 0 to 15 + i := j*32 + IF k1[j] + k[j] := ( a[i+31:i] >= b[i+31:i] ) ? 1 : 0 + ELSE + k[j] := 0 FI ENDFOR -dst[MAX:128] := 0 +k[MAX:16] := 0 - +
immintrin.h
 - + Integer - AVX512VL - AVX512BW - Convert - Miscellaneous - - - - Convert packed 32-bit integers from "a" and "b" to packed 16-bit integers using unsigned saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Mask + AVX512F/KNCNI + Compare + + + + + Compare packed unsigned 32-bit integers in "a" and "b" for greater-than, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). -tmp_dst[15:0] := Saturate_Int32_To_UnsignedInt16 (a[31:0]) -tmp_dst[31:16] := Saturate_Int32_To_UnsignedInt16 (a[63:32]) -tmp_dst[47:32] := Saturate_Int32_To_UnsignedInt16 (a[95:64]) -tmp_dst[63:48] := Saturate_Int32_To_UnsignedInt16 (a[127:96]) -tmp_dst[79:64] := Saturate_Int32_To_UnsignedInt16 (b[31:0]) -tmp_dst[95:80] := Saturate_Int32_To_UnsignedInt16 (b[63:32]) -tmp_dst[111:96] := Saturate_Int32_To_UnsignedInt16 (b[95:64]) -tmp_dst[127:112] := Saturate_Int32_To_UnsignedInt16 (b[127:96]) - -FOR j := 0 to 7 - i := j*16 - IF k[j] - dst[i+15:i] := tmp_dst[i+15:i] - ELSE - dst[i+15:i] := 0 - FI -ENDFOR -dst[MAX:128] := 0 - - -
immintrin.h
- - - Integer - AVX512VL - AVX512BW - Convert - Miscellaneous - - - - - Convert packed 16-bit integers from "a" and "b" to packed 8-bit integers using unsigned saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -tmp_dst[7:0] := Saturate_Int16_To_UnsignedInt8 (a[15:0]) -tmp_dst[15:8] := Saturate_Int16_To_UnsignedInt8 (a[31:16]) -tmp_dst[23:16] := Saturate_Int16_To_UnsignedInt8 (a[47:32]) -tmp_dst[31:24] := Saturate_Int16_To_UnsignedInt8 (a[63:48]) -tmp_dst[39:32] := Saturate_Int16_To_UnsignedInt8 (a[79:64]) -tmp_dst[47:40] := Saturate_Int16_To_UnsignedInt8 (a[95:80]) -tmp_dst[55:48] := Saturate_Int16_To_UnsignedInt8 (a[111:96]) -tmp_dst[63:56] := Saturate_Int16_To_UnsignedInt8 (a[127:112]) -tmp_dst[71:64] := Saturate_Int16_To_UnsignedInt8 (b[15:0]) -tmp_dst[79:72] := Saturate_Int16_To_UnsignedInt8 (b[31:16]) -tmp_dst[87:80] := Saturate_Int16_To_UnsignedInt8 (b[47:32]) -tmp_dst[95:88] := Saturate_Int16_To_UnsignedInt8 (b[63:48]) -tmp_dst[103:96] := Saturate_Int16_To_UnsignedInt8 (b[79:64]) -tmp_dst[111:104] := Saturate_Int16_To_UnsignedInt8 (b[95:80]) -tmp_dst[119:112] := Saturate_Int16_To_UnsignedInt8 (b[111:96]) -tmp_dst[127:120] := Saturate_Int16_To_UnsignedInt8 (b[127:112]) -tmp_dst[135:128] := Saturate_Int16_To_UnsignedInt8 (a[143:128]) -tmp_dst[143:136] := Saturate_Int16_To_UnsignedInt8 (a[159:144]) -tmp_dst[151:144] := Saturate_Int16_To_UnsignedInt8 (a[175:160]) -tmp_dst[159:152] := Saturate_Int16_To_UnsignedInt8 (a[191:176]) -tmp_dst[167:160] := Saturate_Int16_To_UnsignedInt8 (a[207:192]) -tmp_dst[175:168] := Saturate_Int16_To_UnsignedInt8 (a[223:208]) -tmp_dst[183:176] := Saturate_Int16_To_UnsignedInt8 (a[239:224]) -tmp_dst[191:184] := Saturate_Int16_To_UnsignedInt8 (a[255:240]) -tmp_dst[199:192] := Saturate_Int16_To_UnsignedInt8 (b[143:128]) -tmp_dst[207:200] := Saturate_Int16_To_UnsignedInt8 (b[159:144]) -tmp_dst[215:208] := Saturate_Int16_To_UnsignedInt8 (b[175:160]) -tmp_dst[223:216] := Saturate_Int16_To_UnsignedInt8 (b[191:176]) -tmp_dst[231:224] := Saturate_Int16_To_UnsignedInt8 (b[207:192]) -tmp_dst[239:232] := Saturate_Int16_To_UnsignedInt8 (b[223:208]) -tmp_dst[247:240] := Saturate_Int16_To_UnsignedInt8 (b[239:224]) -tmp_dst[255:248] := Saturate_Int16_To_UnsignedInt8 (b[255:240]) - -FOR j := 0 to 31 - i := j*8 - IF k[j] - dst[i+7:i] := tmp_dst[i+7:i] - ELSE - dst[i+7:i] := src[i+7:i] +FOR j := 0 to 15 + i := j*32 + IF k1[j] + k[j] := ( a[i+31:i] > b[i+31:i] ) ? 1 : 0 + ELSE + k[j] := 0 FI ENDFOR -dst[MAX:256] := 0 +k[MAX:16] := 0 - +
immintrin.h
 - + Integer - AVX512VL - AVX512BW - Convert - Miscellaneous - - - - Convert packed 16-bit integers from "a" and "b" to packed 8-bit integers using unsigned saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Mask + AVX512F/KNCNI + Compare + + + + + Compare packed unsigned 32-bit integers in "a" and "b" for less-than, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). -tmp_dst[7:0] := Saturate_Int16_To_UnsignedInt8 (a[15:0]) -tmp_dst[15:8] := Saturate_Int16_To_UnsignedInt8 (a[31:16]) -tmp_dst[23:16] := Saturate_Int16_To_UnsignedInt8 (a[47:32]) -tmp_dst[31:24] := Saturate_Int16_To_UnsignedInt8 (a[63:48]) -tmp_dst[39:32] := Saturate_Int16_To_UnsignedInt8 (a[79:64]) -tmp_dst[47:40] := Saturate_Int16_To_UnsignedInt8 (a[95:80]) -tmp_dst[55:48] := Saturate_Int16_To_UnsignedInt8 (a[111:96]) -tmp_dst[63:56] := Saturate_Int16_To_UnsignedInt8 (a[127:112]) -tmp_dst[71:64] := Saturate_Int16_To_UnsignedInt8 (b[15:0]) -tmp_dst[79:72] := Saturate_Int16_To_UnsignedInt8 (b[31:16]) -tmp_dst[87:80] := Saturate_Int16_To_UnsignedInt8 (b[47:32]) -tmp_dst[95:88] := Saturate_Int16_To_UnsignedInt8 (b[63:48]) -tmp_dst[103:96] := Saturate_Int16_To_UnsignedInt8 (b[79:64]) -tmp_dst[111:104] := Saturate_Int16_To_UnsignedInt8 (b[95:80]) -tmp_dst[119:112] := Saturate_Int16_To_UnsignedInt8 (b[111:96]) -tmp_dst[127:120] := Saturate_Int16_To_UnsignedInt8 (b[127:112]) -tmp_dst[135:128] := Saturate_Int16_To_UnsignedInt8 (a[143:128]) -tmp_dst[143:136] := Saturate_Int16_To_UnsignedInt8 (a[159:144]) -tmp_dst[151:144] := Saturate_Int16_To_UnsignedInt8 (a[175:160]) -tmp_dst[159:152] := Saturate_Int16_To_UnsignedInt8 (a[191:176]) -tmp_dst[167:160] := Saturate_Int16_To_UnsignedInt8 (a[207:192]) -tmp_dst[175:168] := Saturate_Int16_To_UnsignedInt8 (a[223:208]) -tmp_dst[183:176] := Saturate_Int16_To_UnsignedInt8 (a[239:224]) -tmp_dst[191:184] := Saturate_Int16_To_UnsignedInt8 (a[255:240]) -tmp_dst[199:192] := Saturate_Int16_To_UnsignedInt8 (b[143:128]) -tmp_dst[207:200] := Saturate_Int16_To_UnsignedInt8 (b[159:144]) -tmp_dst[215:208] := Saturate_Int16_To_UnsignedInt8 (b[175:160]) -tmp_dst[223:216] := Saturate_Int16_To_UnsignedInt8 (b[191:176]) -tmp_dst[231:224] := Saturate_Int16_To_UnsignedInt8 (b[207:192]) -tmp_dst[239:232] := Saturate_Int16_To_UnsignedInt8 (b[223:208]) -tmp_dst[247:240] := Saturate_Int16_To_UnsignedInt8 (b[239:224]) -tmp_dst[255:248] := Saturate_Int16_To_UnsignedInt8 (b[255:240]) - -FOR j := 0 to 31 - i := j*8 - IF k[j] - dst[i+7:i] := tmp_dst[i+7:i] - ELSE - dst[i+7:i] := 0 - FI -ENDFOR -dst[MAX:256] := 0 - - -
immintrin.h
- - - Integer - AVX512BW - Convert - Miscellaneous - - - - - Convert packed 16-bit integers from "a" and "b" to packed 8-bit integers using unsigned saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -tmp_dst[7:0] := Saturate_Int16_To_UnsignedInt8 (a[15:0]) -tmp_dst[15:8] := Saturate_Int16_To_UnsignedInt8 (a[31:16]) -tmp_dst[23:16] := Saturate_Int16_To_UnsignedInt8 (a[47:32]) -tmp_dst[31:24] := Saturate_Int16_To_UnsignedInt8 (a[63:48]) -tmp_dst[39:32] := Saturate_Int16_To_UnsignedInt8 (a[79:64]) -tmp_dst[47:40] := Saturate_Int16_To_UnsignedInt8 (a[95:80]) -tmp_dst[55:48] := Saturate_Int16_To_UnsignedInt8 (a[111:96]) -tmp_dst[63:56] := Saturate_Int16_To_UnsignedInt8 (a[127:112]) -tmp_dst[71:64] := Saturate_Int16_To_UnsignedInt8 (b[15:0]) -tmp_dst[79:72] := Saturate_Int16_To_UnsignedInt8 (b[31:16]) -tmp_dst[87:80] := Saturate_Int16_To_UnsignedInt8 (b[47:32]) -tmp_dst[95:88] := Saturate_Int16_To_UnsignedInt8 (b[63:48]) -tmp_dst[103:96] := Saturate_Int16_To_UnsignedInt8 (b[79:64]) -tmp_dst[111:104] := Saturate_Int16_To_UnsignedInt8 (b[95:80]) -tmp_dst[119:112] := Saturate_Int16_To_UnsignedInt8 (b[111:96]) -tmp_dst[127:120] := Saturate_Int16_To_UnsignedInt8 (b[127:112]) -tmp_dst[135:128] := Saturate_Int16_To_UnsignedInt8 (a[143:128]) -tmp_dst[143:136] := Saturate_Int16_To_UnsignedInt8 (a[159:144]) -tmp_dst[151:144] := Saturate_Int16_To_UnsignedInt8 (a[175:160]) -tmp_dst[159:152] := Saturate_Int16_To_UnsignedInt8 (a[191:176]) -tmp_dst[167:160] := Saturate_Int16_To_UnsignedInt8 (a[207:192]) -tmp_dst[175:168] := Saturate_Int16_To_UnsignedInt8 (a[223:208]) -tmp_dst[183:176] := Saturate_Int16_To_UnsignedInt8 (a[239:224]) -tmp_dst[191:184] := Saturate_Int16_To_UnsignedInt8 (a[255:240]) -tmp_dst[199:192] := Saturate_Int16_To_UnsignedInt8 (b[143:128]) -tmp_dst[207:200] := Saturate_Int16_To_UnsignedInt8 (b[159:144]) -tmp_dst[215:208] := Saturate_Int16_To_UnsignedInt8 (b[175:160]) -tmp_dst[223:216] := Saturate_Int16_To_UnsignedInt8 (b[191:176]) -tmp_dst[231:224] := Saturate_Int16_To_UnsignedInt8 (b[207:192]) -tmp_dst[239:232] := Saturate_Int16_To_UnsignedInt8 (b[223:208]) -tmp_dst[247:240] := Saturate_Int16_To_UnsignedInt8 (b[239:224]) -tmp_dst[255:248] := Saturate_Int16_To_UnsignedInt8 (b[255:240]) -tmp_dst[263:256] := Saturate_Int16_To_UnsignedInt8 (a[271:256]) -tmp_dst[271:264] := Saturate_Int16_To_UnsignedInt8 (a[287:272]) -tmp_dst[279:272] := Saturate_Int16_To_UnsignedInt8 (a[303:288]) -tmp_dst[287:280] := Saturate_Int16_To_UnsignedInt8 (a[319:304]) -tmp_dst[295:288] := Saturate_Int16_To_UnsignedInt8 (a[335:320]) -tmp_dst[303:296] := Saturate_Int16_To_UnsignedInt8 (a[351:336]) -tmp_dst[311:304] := Saturate_Int16_To_UnsignedInt8 (a[367:352]) -tmp_dst[319:312] := Saturate_Int16_To_UnsignedInt8 (a[383:368]) -tmp_dst[327:320] := Saturate_Int16_To_UnsignedInt8 (b[271:256]) -tmp_dst[335:328] := Saturate_Int16_To_UnsignedInt8 (b[287:272]) -tmp_dst[343:336] := Saturate_Int16_To_UnsignedInt8 (b[303:288]) -tmp_dst[351:344] := Saturate_Int16_To_UnsignedInt8 (b[319:304]) -tmp_dst[359:352] := Saturate_Int16_To_UnsignedInt8 (b[335:320]) -tmp_dst[367:360] := Saturate_Int16_To_UnsignedInt8 (b[351:336]) -tmp_dst[375:368] := Saturate_Int16_To_UnsignedInt8 (b[367:352]) -tmp_dst[383:376] := Saturate_Int16_To_UnsignedInt8 (b[383:368]) -tmp_dst[391:384] := Saturate_Int16_To_UnsignedInt8 (a[399:384]) -tmp_dst[399:392] := Saturate_Int16_To_UnsignedInt8 (a[415:400]) -tmp_dst[407:400] := Saturate_Int16_To_UnsignedInt8 (a[431:416]) -tmp_dst[415:408] := Saturate_Int16_To_UnsignedInt8 (a[447:432]) -tmp_dst[423:416] := Saturate_Int16_To_UnsignedInt8 (a[463:448]) -tmp_dst[431:424] := Saturate_Int16_To_UnsignedInt8 (a[479:464]) -tmp_dst[439:432] := Saturate_Int16_To_UnsignedInt8 (a[495:480]) -tmp_dst[447:440] := Saturate_Int16_To_UnsignedInt8 (a[511:496]) -tmp_dst[455:448] := Saturate_Int16_To_UnsignedInt8 (b[399:384]) -tmp_dst[463:456] := Saturate_Int16_To_UnsignedInt8 (b[415:400]) -tmp_dst[471:464] := Saturate_Int16_To_UnsignedInt8 (b[431:416]) -tmp_dst[479:472] := Saturate_Int16_To_UnsignedInt8 (b[447:432]) -tmp_dst[487:480] := Saturate_Int16_To_UnsignedInt8 (b[463:448]) -tmp_dst[495:488] := Saturate_Int16_To_UnsignedInt8 (b[479:464]) -tmp_dst[503:496] := Saturate_Int16_To_UnsignedInt8 (b[495:480]) -tmp_dst[511:504] := Saturate_Int16_To_UnsignedInt8 (b[511:496]) - -FOR j := 0 to 63 - i := j*8 - IF k[j] - dst[i+7:i] := tmp_dst[i+7:i] - ELSE - dst[i+7:i] := src[i+7:i] - FI -ENDFOR -dst[MAX:512] := 0 - - -
immintrin.h
- - - Integer - AVX512BW - Convert - Miscellaneous - - - - Convert packed 16-bit integers from "a" and "b" to packed 8-bit integers using unsigned saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - -tmp_dst[7:0] := Saturate_Int16_To_UnsignedInt8 (a[15:0]) -tmp_dst[15:8] := Saturate_Int16_To_UnsignedInt8 (a[31:16]) -tmp_dst[23:16] := Saturate_Int16_To_UnsignedInt8 (a[47:32]) -tmp_dst[31:24] := Saturate_Int16_To_UnsignedInt8 (a[63:48]) -tmp_dst[39:32] := Saturate_Int16_To_UnsignedInt8 (a[79:64]) -tmp_dst[47:40] := Saturate_Int16_To_UnsignedInt8 (a[95:80]) -tmp_dst[55:48] := Saturate_Int16_To_UnsignedInt8 (a[111:96]) -tmp_dst[63:56] := Saturate_Int16_To_UnsignedInt8 (a[127:112]) -tmp_dst[71:64] := Saturate_Int16_To_UnsignedInt8 (b[15:0]) -tmp_dst[79:72] := Saturate_Int16_To_UnsignedInt8 (b[31:16]) -tmp_dst[87:80] := Saturate_Int16_To_UnsignedInt8 (b[47:32]) -tmp_dst[95:88] := Saturate_Int16_To_UnsignedInt8 (b[63:48]) -tmp_dst[103:96] := Saturate_Int16_To_UnsignedInt8 (b[79:64]) -tmp_dst[111:104] := Saturate_Int16_To_UnsignedInt8 (b[95:80]) -tmp_dst[119:112] := Saturate_Int16_To_UnsignedInt8 (b[111:96]) -tmp_dst[127:120] := Saturate_Int16_To_UnsignedInt8 (b[127:112]) -tmp_dst[135:128] := Saturate_Int16_To_UnsignedInt8 (a[143:128]) -tmp_dst[143:136] := Saturate_Int16_To_UnsignedInt8 (a[159:144]) -tmp_dst[151:144] := Saturate_Int16_To_UnsignedInt8 (a[175:160]) -tmp_dst[159:152] := Saturate_Int16_To_UnsignedInt8 (a[191:176]) -tmp_dst[167:160] := Saturate_Int16_To_UnsignedInt8 (a[207:192]) -tmp_dst[175:168] := Saturate_Int16_To_UnsignedInt8 (a[223:208]) -tmp_dst[183:176] := Saturate_Int16_To_UnsignedInt8 (a[239:224]) -tmp_dst[191:184] := Saturate_Int16_To_UnsignedInt8 (a[255:240]) -tmp_dst[199:192] := Saturate_Int16_To_UnsignedInt8 (b[143:128]) -tmp_dst[207:200] := Saturate_Int16_To_UnsignedInt8 (b[159:144]) -tmp_dst[215:208] := Saturate_Int16_To_UnsignedInt8 (b[175:160]) -tmp_dst[223:216] := Saturate_Int16_To_UnsignedInt8 (b[191:176]) -tmp_dst[231:224] := Saturate_Int16_To_UnsignedInt8 (b[207:192]) -tmp_dst[239:232] := Saturate_Int16_To_UnsignedInt8 (b[223:208]) -tmp_dst[247:240] := Saturate_Int16_To_UnsignedInt8 (b[239:224]) -tmp_dst[255:248] := Saturate_Int16_To_UnsignedInt8 (b[255:240]) -tmp_dst[263:256] := Saturate_Int16_To_UnsignedInt8 (a[271:256]) -tmp_dst[271:264] := Saturate_Int16_To_UnsignedInt8 (a[287:272]) -tmp_dst[279:272] := Saturate_Int16_To_UnsignedInt8 (a[303:288]) -tmp_dst[287:280] := Saturate_Int16_To_UnsignedInt8 (a[319:304]) -tmp_dst[295:288] := Saturate_Int16_To_UnsignedInt8 (a[335:320]) -tmp_dst[303:296] := Saturate_Int16_To_UnsignedInt8 (a[351:336]) -tmp_dst[311:304] := Saturate_Int16_To_UnsignedInt8 (a[367:352]) -tmp_dst[319:312] := Saturate_Int16_To_UnsignedInt8 (a[383:368]) -tmp_dst[327:320] := Saturate_Int16_To_UnsignedInt8 (b[271:256]) -tmp_dst[335:328] := Saturate_Int16_To_UnsignedInt8 (b[287:272]) -tmp_dst[343:336] := Saturate_Int16_To_UnsignedInt8 (b[303:288]) -tmp_dst[351:344] := Saturate_Int16_To_UnsignedInt8 (b[319:304]) -tmp_dst[359:352] := Saturate_Int16_To_UnsignedInt8 (b[335:320]) -tmp_dst[367:360] := Saturate_Int16_To_UnsignedInt8 (b[351:336]) -tmp_dst[375:368] := Saturate_Int16_To_UnsignedInt8 (b[367:352]) -tmp_dst[383:376] := Saturate_Int16_To_UnsignedInt8 (b[383:368]) -tmp_dst[391:384] := Saturate_Int16_To_UnsignedInt8 (a[399:384]) -tmp_dst[399:392] := Saturate_Int16_To_UnsignedInt8 (a[415:400]) -tmp_dst[407:400] := Saturate_Int16_To_UnsignedInt8 (a[431:416]) -tmp_dst[415:408] := Saturate_Int16_To_UnsignedInt8 (a[447:432]) -tmp_dst[423:416] := Saturate_Int16_To_UnsignedInt8 (a[463:448]) -tmp_dst[431:424] := Saturate_Int16_To_UnsignedInt8 (a[479:464]) -tmp_dst[439:432] := Saturate_Int16_To_UnsignedInt8 (a[495:480]) -tmp_dst[447:440] := Saturate_Int16_To_UnsignedInt8 (a[511:496]) -tmp_dst[455:448] := Saturate_Int16_To_UnsignedInt8 (b[399:384]) -tmp_dst[463:456] := Saturate_Int16_To_UnsignedInt8 (b[415:400]) -tmp_dst[471:464] := Saturate_Int16_To_UnsignedInt8 (b[431:416]) -tmp_dst[479:472] := Saturate_Int16_To_UnsignedInt8 (b[447:432]) -tmp_dst[487:480] := Saturate_Int16_To_UnsignedInt8 (b[463:448]) -tmp_dst[495:488] := Saturate_Int16_To_UnsignedInt8 (b[479:464]) -tmp_dst[503:496] := Saturate_Int16_To_UnsignedInt8 (b[495:480]) -tmp_dst[511:504] := Saturate_Int16_To_UnsignedInt8 (b[511:496]) - -FOR j := 0 to 63 - i := j*8 - IF k[j] - dst[i+7:i] := tmp_dst[i+7:i] - ELSE - dst[i+7:i] := 0 +FOR j := 0 to 15 + i := j*32 + IF k1[j] + k[j] := ( a[i+31:i] <= b[i+31:i] ) ? 1 : 0 + ELSE + k[j] := 0 FI ENDFOR -dst[MAX:512] := 0 +k[MAX:16] := 0 - -
immintrin.h
- - - Integer - AVX512BW - Convert - Miscellaneous - - - Convert packed 16-bit integers from "a" and "b" to packed 8-bit integers using unsigned saturation, and store the results in "dst". - -dst[7:0] := Saturate_Int16_To_UnsignedInt8 (a[15:0]) -dst[15:8] := Saturate_Int16_To_UnsignedInt8 (a[31:16]) -dst[23:16] := Saturate_Int16_To_UnsignedInt8 (a[47:32]) -dst[31:24] := Saturate_Int16_To_UnsignedInt8 (a[63:48]) -dst[39:32] := Saturate_Int16_To_UnsignedInt8 (a[79:64]) -dst[47:40] := Saturate_Int16_To_UnsignedInt8 (a[95:80]) -dst[55:48] := Saturate_Int16_To_UnsignedInt8 (a[111:96]) -dst[63:56] := Saturate_Int16_To_UnsignedInt8 (a[127:112]) -dst[71:64] := Saturate_Int16_To_UnsignedInt8 (b[15:0]) -dst[79:72] := Saturate_Int16_To_UnsignedInt8 (b[31:16]) -dst[87:80] := Saturate_Int16_To_UnsignedInt8 (b[47:32]) -dst[95:88] := Saturate_Int16_To_UnsignedInt8 (b[63:48]) -dst[103:96] := Saturate_Int16_To_UnsignedInt8 (b[79:64]) -dst[111:104] := Saturate_Int16_To_UnsignedInt8 (b[95:80]) -dst[119:112] := Saturate_Int16_To_UnsignedInt8 (b[111:96]) -dst[127:120] := Saturate_Int16_To_UnsignedInt8 (b[127:112]) -dst[135:128] := Saturate_Int16_To_UnsignedInt8 (a[143:128]) -dst[143:136] := Saturate_Int16_To_UnsignedInt8 (a[159:144]) -dst[151:144] := Saturate_Int16_To_UnsignedInt8 (a[175:160]) -dst[159:152] := Saturate_Int16_To_UnsignedInt8 (a[191:176]) -dst[167:160] := Saturate_Int16_To_UnsignedInt8 (a[207:192]) -dst[175:168] := Saturate_Int16_To_UnsignedInt8 (a[223:208]) -dst[183:176] := Saturate_Int16_To_UnsignedInt8 (a[239:224]) -dst[191:184] := Saturate_Int16_To_UnsignedInt8 (a[255:240]) -dst[199:192] := Saturate_Int16_To_UnsignedInt8 (b[143:128]) -dst[207:200] := Saturate_Int16_To_UnsignedInt8 (b[159:144]) -dst[215:208] := Saturate_Int16_To_UnsignedInt8 (b[175:160]) -dst[223:216] := Saturate_Int16_To_UnsignedInt8 (b[191:176]) -dst[231:224] := Saturate_Int16_To_UnsignedInt8 (b[207:192]) -dst[239:232] := Saturate_Int16_To_UnsignedInt8 (b[223:208]) -dst[247:240] := Saturate_Int16_To_UnsignedInt8 (b[239:224]) -dst[255:248] := Saturate_Int16_To_UnsignedInt8 (b[255:240]) -dst[263:256] := Saturate_Int16_To_UnsignedInt8 (a[271:256]) -dst[271:264] := Saturate_Int16_To_UnsignedInt8 (a[287:272]) -dst[279:272] := Saturate_Int16_To_UnsignedInt8 (a[303:288]) -dst[287:280] := Saturate_Int16_To_UnsignedInt8 (a[319:304]) -dst[295:288] := Saturate_Int16_To_UnsignedInt8 (a[335:320]) -dst[303:296] := Saturate_Int16_To_UnsignedInt8 (a[351:336]) -dst[311:304] := Saturate_Int16_To_UnsignedInt8 (a[367:352]) -dst[319:312] := Saturate_Int16_To_UnsignedInt8 (a[383:368]) -dst[327:320] := Saturate_Int16_To_UnsignedInt8 (b[271:256]) -dst[335:328] := Saturate_Int16_To_UnsignedInt8 (b[287:272]) -dst[343:336] := Saturate_Int16_To_UnsignedInt8 (b[303:288]) -dst[351:344] := Saturate_Int16_To_UnsignedInt8 (b[319:304]) -dst[359:352] := Saturate_Int16_To_UnsignedInt8 (b[335:320]) -dst[367:360] := Saturate_Int16_To_UnsignedInt8 (b[351:336]) -dst[375:368] := Saturate_Int16_To_UnsignedInt8 (b[367:352]) -dst[383:376] := Saturate_Int16_To_UnsignedInt8 (b[383:368]) -dst[391:384] := Saturate_Int16_To_UnsignedInt8 (a[399:384]) -dst[399:392] := Saturate_Int16_To_UnsignedInt8 (a[415:400]) -dst[407:400] := Saturate_Int16_To_UnsignedInt8 (a[431:416]) -dst[415:408] := Saturate_Int16_To_UnsignedInt8 (a[447:432]) -dst[423:416] := Saturate_Int16_To_UnsignedInt8 (a[463:448]) -dst[431:424] := Saturate_Int16_To_UnsignedInt8 (a[479:464]) -dst[439:432] := Saturate_Int16_To_UnsignedInt8 (a[495:480]) -dst[447:440] := Saturate_Int16_To_UnsignedInt8 (a[511:496]) -dst[455:448] := Saturate_Int16_To_UnsignedInt8 (b[399:384]) -dst[463:456] := Saturate_Int16_To_UnsignedInt8 (b[415:400]) -dst[471:464] := Saturate_Int16_To_UnsignedInt8 (b[431:416]) -dst[479:472] := Saturate_Int16_To_UnsignedInt8 (b[447:432]) -dst[487:480] := Saturate_Int16_To_UnsignedInt8 (b[463:448]) -dst[495:488] := Saturate_Int16_To_UnsignedInt8 (b[479:464]) -dst[503:496] := Saturate_Int16_To_UnsignedInt8 (b[495:480]) -dst[511:504] := Saturate_Int16_To_UnsignedInt8 (b[511:496]) -dst[MAX:512] := 0 - - -
immintrin.h
- - - Integer - AVX512VL - AVX512BW - Convert - Miscellaneous - - - - - Convert packed 16-bit integers from "a" and "b" to packed 8-bit integers using unsigned saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -tmp_dst[7:0] := Saturate_Int16_To_UnsignedInt8 (a[15:0]) -tmp_dst[15:8] := Saturate_Int16_To_UnsignedInt8 (a[31:16]) -tmp_dst[23:16] := Saturate_Int16_To_UnsignedInt8 (a[47:32]) -tmp_dst[31:24] := Saturate_Int16_To_UnsignedInt8 (a[63:48]) -tmp_dst[39:32] := Saturate_Int16_To_UnsignedInt8 (a[79:64]) -tmp_dst[47:40] := Saturate_Int16_To_UnsignedInt8 (a[95:80]) -tmp_dst[55:48] := Saturate_Int16_To_UnsignedInt8 (a[111:96]) -tmp_dst[63:56] := Saturate_Int16_To_UnsignedInt8 (a[127:112]) -tmp_dst[71:64] := Saturate_Int16_To_UnsignedInt8 (b[15:0]) -tmp_dst[79:72] := Saturate_Int16_To_UnsignedInt8 (b[31:16]) -tmp_dst[87:80] := Saturate_Int16_To_UnsignedInt8 (b[47:32]) -tmp_dst[95:88] := Saturate_Int16_To_UnsignedInt8 (b[63:48]) -tmp_dst[103:96] := Saturate_Int16_To_UnsignedInt8 (b[79:64]) -tmp_dst[111:104] := Saturate_Int16_To_UnsignedInt8 (b[95:80]) -tmp_dst[119:112] := Saturate_Int16_To_UnsignedInt8 (b[111:96]) -tmp_dst[127:120] := Saturate_Int16_To_UnsignedInt8 (b[127:112]) - + +
immintrin.h
+ + + Integer + Mask + AVX512F/KNCNI + Compare + + + + + Compare packed unsigned 32-bit integers in "a" and "b" for less-than-or-equal, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). + FOR j := 0 to 15 - i := j*8 - IF k[j] - dst[i+7:i] := tmp_dst[i+7:i] - ELSE - dst[i+7:i] := src[i+7:i] + i := j*32 + IF k1[j] + k[j] := ( a[i+31:i] < b[i+31:i] ) ? 1 : 0 + ELSE + k[j] := 0 FI ENDFOR -dst[MAX:128] := 0 +k[MAX:16] := 0 - +
immintrin.h
 - + Integer - AVX512VL - AVX512BW - Convert - Miscellaneous - - - - Convert packed 16-bit integers from "a" and "b" to packed 8-bit integers using unsigned saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Mask + AVX512F/KNCNI + Compare + + + + + Compare packed unsigned 32-bit integers in "a" and "b" for not-equal, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). -tmp_dst[7:0] := Saturate_Int16_To_UnsignedInt8 (a[15:0]) -tmp_dst[15:8] := Saturate_Int16_To_UnsignedInt8 (a[31:16]) -tmp_dst[23:16] := Saturate_Int16_To_UnsignedInt8 (a[47:32]) -tmp_dst[31:24] := Saturate_Int16_To_UnsignedInt8 (a[63:48]) -tmp_dst[39:32] := Saturate_Int16_To_UnsignedInt8 (a[79:64]) -tmp_dst[47:40] := Saturate_Int16_To_UnsignedInt8 (a[95:80]) -tmp_dst[55:48] := Saturate_Int16_To_UnsignedInt8 (a[111:96]) -tmp_dst[63:56] := Saturate_Int16_To_UnsignedInt8 (a[127:112]) -tmp_dst[71:64] := Saturate_Int16_To_UnsignedInt8 (b[15:0]) -tmp_dst[79:72] := Saturate_Int16_To_UnsignedInt8 (b[31:16]) -tmp_dst[87:80] := Saturate_Int16_To_UnsignedInt8 (b[47:32]) -tmp_dst[95:88] := Saturate_Int16_To_UnsignedInt8 (b[63:48]) -tmp_dst[103:96] := Saturate_Int16_To_UnsignedInt8 (b[79:64]) -tmp_dst[111:104] := Saturate_Int16_To_UnsignedInt8 (b[95:80]) -tmp_dst[119:112] := Saturate_Int16_To_UnsignedInt8 (b[111:96]) -tmp_dst[127:120] := Saturate_Int16_To_UnsignedInt8 (b[127:112]) - FOR j := 0 to 15 - i := j*8 - IF k[j] - dst[i+7:i] := tmp_dst[i+7:i] - ELSE - dst[i+7:i] := 0 + i := j*32 + IF k1[j] + k[j] := ( a[i+31:i] != b[i+31:i] ) ? 1 : 0 + ELSE + k[j] := 0 FI ENDFOR -dst[MAX:128] := 0 +k[MAX:16] := 0 - +
immintrin.h
 - + Integer - AVX512VL - AVX512BW - Arithmetic - - - - - Add packed 8-bit integers in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + AVX512F/KNCNI + Swizzle + + + + + + Shuffle 32-bit integers in "a" across lanes using the corresponding index in "idx", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). Note that this intrinsic shuffles across 128-bit lanes, unlike past intrinsics that use the "permutevar" name. This intrinsic is identical to "_mm512_mask_permutexvar_epi32", and it is recommended that you use that intrinsic name. -FOR j := 0 to 31 - i := j*8 +FOR j := 0 to 15 + i := j*32 + id := idx[i+3:i]*32 IF k[j] - dst[i+7:i] := a[i+7:i] + b[i+7:i] + dst[i+31:i] := a[id+31:id] ELSE - dst[i+7:i] := src[i+7:i] + dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX512VL - AVX512BW - Arithmetic - - - - Add packed 8-bit integers in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + AVX512F/KNCNI + Swizzle + + + + Shuffle 32-bit integers in "a" across lanes using the corresponding index in "idx", and store the results in "dst". Note that this intrinsic shuffles across 128-bit lanes, unlike past intrinsics that use the "permutevar" name. This intrinsic is identical to "_mm512_permutexvar_epi32", and it is recommended that you use that intrinsic name. -FOR j := 0 to 31 - i := j*8 - IF k[j] - dst[i+7:i] := a[i+7:i] + b[i+7:i] - ELSE - dst[i+7:i] := 0 - FI +FOR j := 0 to 15 + i := j*32 + id := idx[i+3:i]*32 + dst[i+31:i] := a[id+31:id] ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX512BW - Arithmetic - - - Add packed 8-bit integers in "a" and "b", and store the results in "dst". + AVX512F/KNCNI + Load + + + + + Gather 32-bit integers from memory using 32-bit indices. 32-bit elements are loaded from addresses starting at "base_addr" and offset by each 32-bit element in "vindex" (each index is scaled by the factor in "scale"). Gathered elements are merged into "dst". "scale" should be 1, 2, 4 or 8. -FOR j := 0 to 63 - i := j*8 - dst[i+7:i] := a[i+7:i] + b[i+7:i] +FOR j := 0 to 15 + i := j*32 + m := j*32 + addr := base_addr + SignExtend64(vindex[m+31:m]) * ZeroExtend64(scale) * 8 + dst[i+31:i] := MEM[addr+31:addr] ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX512BW - Arithmetic - - - - - Add packed 8-bit integers in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + AVX512F/KNCNI + Load + + + + + + + Gather 32-bit integers from memory using 32-bit indices. 32-bit elements are loaded from addresses starting at "base_addr" and offset by each 32-bit element in "vindex" (each index is scaled by the factor in "scale"). Gathered elements are merged into "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). "scale" should be 1, 2, 4 or 8. -FOR j := 0 to 63 - i := j*8 +FOR j := 0 to 15 + i := j*32 + m := j*32 IF k[j] - dst[i+7:i] := a[i+7:i] + b[i+7:i] + addr := base_addr + SignExtend64(vindex[m+31:m]) * ZeroExtend64(scale) * 8 + dst[i+31:i] := MEM[addr+31:addr] ELSE - dst[i+7:i] := src[i+7:i] + dst[i+31:i] := src[i+31:i] FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX512BW - Arithmetic - - - - Add packed 8-bit integers in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + AVX512F/KNCNI + Special Math Functions + + + + + + Compare packed signed 32-bit integers in "a" and "b", and store packed maximum values in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 63 - i := j*8 +FOR j := 0 to 15 + i := j*32 IF k[j] - dst[i+7:i] := a[i+7:i] + b[i+7:i] + dst[i+31:i] := MAX(a[i+31:i], b[i+31:i]) ELSE - dst[i+7:i] := 0 + dst[i+31:i] := src[i+31:i] FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX512VL - AVX512BW - Arithmetic - - - - - Add packed 8-bit integers in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + AVX512F/KNCNI + Special Math Functions + + + + Compare packed signed 32-bit integers in "a" and "b", and store packed maximum values in "dst". FOR j := 0 to 15 - i := j*8 - IF k[j] - dst[i+7:i] := a[i+7:i] + b[i+7:i] - ELSE - dst[i+7:i] := src[i+7:i] - FI + i := j*32 + dst[i+31:i] := MAX(a[i+31:i], b[i+31:i]) ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX512VL - AVX512BW - Arithmetic - - - - Add packed 8-bit integers in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + AVX512F/KNCNI + Special Math Functions + + + + + + Compare packed unsigned 32-bit integers in "a" and "b", and store packed maximum values in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). FOR j := 0 to 15 - i := j*8 + i := j*32 IF k[j] - dst[i+7:i] := a[i+7:i] + b[i+7:i] + dst[i+31:i] := MAX(a[i+31:i], b[i+31:i]) ELSE - dst[i+7:i] := 0 + dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX512VL - AVX512F - Arithmetic - - - - - Add packed 32-bit integers in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + AVX512F/KNCNI + Special Math Functions + + + + Compare packed unsigned 32-bit integers in "a" and "b", and store packed maximum values in "dst". -FOR j := 0 to 7 +FOR j := 0 to 15 + i := j*32 + dst[i+31:i] := MAX(a[i+31:i], b[i+31:i]) +ENDFOR +dst[MAX:512] := 0 + + +
immintrin.h
+ + + Integer + AVX512F/KNCNI + Special Math Functions + + + + + + Compare packed signed 32-bit integers in "a" and "b", and store packed minimum values in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +FOR j := 0 to 15 i := j*32 IF k[j] - dst[i+31:i] := a[i+31:i] + b[i+31:i] + dst[i+31:i] := MIN(a[i+31:i], b[i+31:i]) ELSE dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX512VL - AVX512F - Arithmetic - - - - Add packed 32-bit integers in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + AVX512F/KNCNI + Special Math Functions + + + + Compare packed signed 32-bit integers in "a" and "b", and store packed minimum values in "dst". -FOR j := 0 to 7 +FOR j := 0 to 15 i := j*32 - IF k[j] - dst[i+31:i] := a[i+31:i] + b[i+31:i] - ELSE - dst[i+31:i] := 0 - FI + dst[i+31:i] := MIN(a[i+31:i], b[i+31:i]) ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX512VL - AVX512F - Arithmetic - - - - - Add packed 32-bit integers in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + AVX512F/KNCNI + Special Math Functions + + + + + + Compare packed unsigned 32-bit integers in "a" and "b", and store packed minimum values in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 3 +FOR j := 0 to 15 i := j*32 IF k[j] - dst[i+31:i] := a[i+31:i] + b[i+31:i] + dst[i+31:i] := MIN(a[i+31:i], b[i+31:i]) ELSE dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX512VL - AVX512F - Arithmetic - - - - Add packed 32-bit integers in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + AVX512F/KNCNI + Special Math Functions + + + + Compare packed unsigned 32-bit integers in "a" and "b", and store packed minimum values in "dst". -FOR j := 0 to 3 +FOR j := 0 to 15 i := j*32 - IF k[j] - dst[i+31:i] := a[i+31:i] + b[i+31:i] - ELSE - dst[i+31:i] := 0 - FI + dst[i+31:i] := MIN(a[i+31:i], b[i+31:i]) ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX512VL - AVX512F + AVX512F/KNCNI Arithmetic - - - - - Add packed 64-bit integers in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + + + + + Multiply the packed 32-bit integers in "a" and "b", producing intermediate 64-bit integers, and store the low 32 bits of the intermediate integers in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*64 +FOR j := 0 to 15 + i := j*32 IF k[j] - dst[i+63:i] := a[i+63:i] + b[i+63:i] + tmp[63:0] := a[i+31:i] * b[i+31:i] + dst[i+31:i] := tmp[31:0] ELSE - dst[i+63:i] := src[i+63:i] + dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX512VL - AVX512F + AVX512F/KNCNI Arithmetic - - - - Add packed 64-bit integers in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + + + Multiply the packed 32-bit integers in "a" and "b", producing intermediate 64-bit integers, and store the low 32 bits of the intermediate integers in "dst". -FOR j := 0 to 3 - i := j*64 - IF k[j] - dst[i+63:i] := a[i+63:i] + b[i+63:i] - ELSE - dst[i+63:i] :=0 - FI +FOR j := 0 to 15 + i := j*32 + tmp[63:0] := a[i+31:i] * b[i+31:i] + dst[i+31:i] := tmp[31:0] ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX512VL - AVX512F - Arithmetic - - - - - Add packed 64-bit integers in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + AVX512F/KNCNI + Logical + + + + + + Compute the bitwise OR of packed 32-bit integers in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 1 - i := j*64 +FOR j := 0 to 15 + i := j*32 IF k[j] - dst[i+63:i] := a[i+63:i] + b[i+63:i] + dst[i+31:i] := a[i+31:i] OR b[i+31:i] ELSE - dst[i+63:i] := src[i+63:i] + dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX512VL - AVX512F - Arithmetic - - - - Add packed 64-bit integers in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + AVX512F/KNCNI + Logical + + + + Compute the bitwise OR of packed 32-bit integers in "a" and "b", and store the results in "dst". -FOR j := 0 to 1 - i := j*64 - IF k[j] - dst[i+63:i] := a[i+63:i] + b[i+63:i] - ELSE - dst[i+63:i] := 0 - FI +FOR j := 0 to 15 + i := j*32 + dst[i+31:i] := a[i+31:i] OR b[i+31:i] ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX512VL - AVX512BW - Arithmetic - - - - - Add packed 8-bit integers in "a" and "b" using saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + AVX512F/KNCNI + Logical + + + + Compute the bitwise OR of 512 bits (representing integer data) in "a" and "b", and store the result in "dst". -FOR j := 0 to 31 - i := j*8 - IF k[j] - dst[i+7:i] := Saturate_To_Int8( a[i+7:i] + b[i+7:i] ) - ELSE - dst[i+7:i] := src[i+7:i] - FI -ENDFOR -dst[MAX:256] := 0 +dst[511:0] := (a[511:0] OR b[511:0]) +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX512VL - AVX512BW - Arithmetic - - - - Add packed 8-bit integers in "a" and "b" using saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - + AVX512F/KNCNI + Logical + + + + + + Compute the bitwise OR of packed 64-bit integers in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 31 - i := j*8 +FOR j := 0 to 7 + i := j*64 IF k[j] - dst[i+7:i] := Saturate_To_Int8( a[i+7:i] + b[i+7:i] ) + dst[i+63:i] := a[i+63:i] OR b[i+63:i] ELSE - dst[i+7:i] := 0 + dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX512BW - Arithmetic - - - Add packed 8-bit integers in "a" and "b" using saturation, and store the results in "dst". + AVX512F/KNCNI + Logical + + + + Compute the bitwise OR of packed 64-bit integers in "a" and "b", and store the resut in "dst". -FOR j := 0 to 63 - i := j*8 - dst[i+7:i] := Saturate_To_Int8( a[i+7:i] + b[i+7:i] ) +FOR j := 0 to 7 + i := j*64 + dst[i+63:i] := a[i+63:i] OR b[i+63:i] ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX512BW - Arithmetic - - - - - Add packed 8-bit integers in "a" and "b" using saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + AVX512F/KNCNI + Store + + + + + + Scatter 32-bit integers from "a" into memory using 32-bit indices. 32-bit elements are stored at addresses starting at "base_addr" and offset by each 32-bit element in "vindex" (each index is scaled by the factor in "scale"). "scale" should be 1, 2, 4 or 8. -FOR j := 0 to 63 - i := j*8 - IF k[j] - dst[i+7:i] := Saturate_To_Int8( a[i+7:i] + b[i+7:i] ) - ELSE - dst[i+7:i] := src[i+7:i] - FI +FOR j := 0 to 15 + i := j*32 + m := j*32 + addr := base_addr + SignExtend64(vindex[m+31:m]) * ZeroExtend64(scale) * 8 + MEM[addr+31:addr] := a[i+31:i] ENDFOR -dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX512BW - Arithmetic - - - - Add packed 8-bit integers in "a" and "b" using saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - + AVX512F/KNCNI + Store + + + + + + + Scatter 32-bit integers from "a" into memory using 32-bit indices. 32-bit elements are stored at addresses starting at "base_addr" and offset by each 32-bit element in "vindex" (each index is scaled by the factor in "scale") subject to mask "k" (elements are not stored when the corresponding mask bit is not set). "scale" should be 1, 2, 4 or 8. -FOR j := 0 to 63 - i := j*8 +FOR j := 0 to 15 + i := j*32 + m := j*32 IF k[j] - dst[i+7:i] := Saturate_To_Int8( a[i+7:i] + b[i+7:i] ) - ELSE - dst[i+7:i] := 0 + addr := base_addr + SignExtend64(vindex[m+31:m]) * ZeroExtend64(scale) * 8 + MEM[addr+31:addr] := a[i+31:i] FI ENDFOR -dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX512VL - AVX512BW - Arithmetic - - - - - Add packed 8-bit integers in "a" and "b" using saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + AVX512F/KNCNI + Swizzle + + + + + + Shuffle 32-bit integers in "a" within 128-bit lanes using the control in "imm8", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). +DEFINE SELECT4(src, control) { + CASE(control[1:0]) OF + 0: tmp[31:0] := src[31:0] + 1: tmp[31:0] := src[63:32] + 2: tmp[31:0] := src[95:64] + 3: tmp[31:0] := src[127:96] + ESAC + RETURN tmp[31:0] +} +tmp_dst[31:0] := SELECT4(a[127:0], imm8[1:0]) +tmp_dst[63:32] := SELECT4(a[127:0], imm8[3:2]) +tmp_dst[95:64] := SELECT4(a[127:0], imm8[5:4]) +tmp_dst[127:96] := SELECT4(a[127:0], imm8[7:6]) +tmp_dst[159:128] := SELECT4(a[255:128], imm8[1:0]) +tmp_dst[191:160] := SELECT4(a[255:128], imm8[3:2]) +tmp_dst[223:192] := SELECT4(a[255:128], imm8[5:4]) +tmp_dst[255:224] := SELECT4(a[255:128], imm8[7:6]) +tmp_dst[287:256] := SELECT4(a[383:256], imm8[1:0]) +tmp_dst[319:288] := SELECT4(a[383:256], imm8[3:2]) +tmp_dst[351:320] := SELECT4(a[383:256], imm8[5:4]) +tmp_dst[383:352] := SELECT4(a[383:256], imm8[7:6]) +tmp_dst[415:384] := SELECT4(a[511:384], imm8[1:0]) +tmp_dst[447:416] := SELECT4(a[511:384], imm8[3:2]) +tmp_dst[479:448] := SELECT4(a[511:384], imm8[5:4]) +tmp_dst[511:480] := SELECT4(a[511:384], imm8[7:6]) FOR j := 0 to 15 - i := j*8 + i := j*32 IF k[j] - dst[i+7:i] := Saturate_To_Int8( a[i+7:i] + b[i+7:i] ) + dst[i+31:i] := tmp_dst[i+31:i] ELSE - dst[i+7:i] := src[i+7:i] + dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 + + +
immintrin.h
+ + + Integer + AVX512F/KNCNI + Swizzle + + + + Shuffle 32-bit integers in "a" within 128-bit lanes using the control in "imm8", and store the results in "dst". + +DEFINE SELECT4(src, control) { + CASE(control[1:0]) OF + 0: tmp[31:0] := src[31:0] + 1: tmp[31:0] := src[63:32] + 2: tmp[31:0] := src[95:64] + 3: tmp[31:0] := src[127:96] + ESAC + RETURN tmp[31:0] +} +dst[31:0] := SELECT4(a[127:0], imm8[1:0]) +dst[63:32] := SELECT4(a[127:0], imm8[3:2]) +dst[95:64] := SELECT4(a[127:0], imm8[5:4]) +dst[127:96] := SELECT4(a[127:0], imm8[7:6]) +dst[159:128] := SELECT4(a[255:128], imm8[1:0]) +dst[191:160] := SELECT4(a[255:128], imm8[3:2]) +dst[223:192] := SELECT4(a[255:128], imm8[5:4]) +dst[255:224] := SELECT4(a[255:128], imm8[7:6]) +dst[287:256] := SELECT4(a[383:256], imm8[1:0]) +dst[319:288] := SELECT4(a[383:256], imm8[3:2]) +dst[351:320] := SELECT4(a[383:256], imm8[5:4]) +dst[383:352] := SELECT4(a[383:256], imm8[7:6]) +dst[415:384] := SELECT4(a[511:384], imm8[1:0]) +dst[447:416] := SELECT4(a[511:384], imm8[3:2]) +dst[479:448] := SELECT4(a[511:384], imm8[5:4]) +dst[511:480] := SELECT4(a[511:384], imm8[7:6]) +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX512VL - AVX512BW - Arithmetic - - - - Add packed 8-bit integers in "a" and "b" using saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - + AVX512F/KNCNI + Shift + + + + + + Shift packed 32-bit integers in "a" left by "imm8" while shifting in zeros, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). FOR j := 0 to 15 - i := j*8 + i := j*32 IF k[j] - dst[i+7:i] := Saturate_To_Int8( a[i+7:i] + b[i+7:i] ) + IF imm8[7:0] > 31 + dst[i+31:i] := 0 + ELSE + dst[i+31:i] := ZeroExtend32(a[i+31:i] << imm8[7:0]) + FI ELSE - dst[i+7:i] := 0 + dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX512VL - AVX512BW - Arithmetic - - - - - Add packed 16-bit integers in "a" and "b" using saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + AVX512F/KNCNI + Shift + + + + Shift packed 32-bit integers in "a" left by "imm8" while shifting in zeros, and store the results in "dst". FOR j := 0 to 15 - i := j*16 - IF k[j] - dst[i+15:i] := Saturate_To_Int16( a[i+15:i] + b[i+15:i] ) + i := j*32 + IF imm8[7:0] > 31 + dst[i+31:i] := 0 ELSE - dst[i+15:i] := src[i+15:i] + dst[i+31:i] := ZeroExtend32(a[i+31:i] << imm8[7:0]) FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX512VL - AVX512BW - Arithmetic - - - - Add packed 16-bit integers in "a" and "b" using saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - + AVX512F/KNCNI + Shift + + + + + + Shift packed 32-bit integers in "a" left by the amount specified by the corresponding element in "count" while shifting in zeros, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). FOR j := 0 to 15 - i := j*16 + i := j*32 IF k[j] - dst[i+15:i] := Saturate_To_Int16( a[i+15:i] + b[i+15:i] ) + IF count[i+31:i] < 32 + dst[i+31:i] := ZeroExtend32(a[i+31:i] << count[i+31:i]) + ELSE + dst[i+31:i] := 0 + FI ELSE - dst[i+15:i] := 0 - FI + dst[i+31:i] := src[i+31:i] + FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX512BW - Arithmetic - - - Add packed 16-bit integers in "a" and "b" using saturation, and store the results in "dst". + AVX512F/KNCNI + Shift + + + + Shift packed 32-bit integers in "a" left by the amount specified by the corresponding element in "count" while shifting in zeros, and store the results in "dst". -FOR j := 0 to 31 - i := j*16 - dst[i+15:i] := Saturate_To_Int16( a[i+15:i] + b[i+15:i] ) +FOR j := 0 to 15 + i := j*32 + IF count[i+31:i] < 32 + dst[i+31:i] := ZeroExtend32(a[i+31:i] << count[i+31:i]) + ELSE + dst[i+31:i] := 0 + FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX512BW - Arithmetic - - - - - Add packed 16-bit integers in "a" and "b" using saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + AVX512F/KNCNI + Shift + + + + + + Shift packed 32-bit integers in "a" right by "imm8" while shifting in sign bits, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 31 - i := j*16 +FOR j := 0 to 15 + i := j*32 IF k[j] - dst[i+15:i] := Saturate_To_Int16( a[i+15:i] + b[i+15:i] ) + IF imm8[7:0] > 31 + dst[i+31:i] := (a[i+31] ? 0xFFFFFFFF : 0x0) + ELSE + dst[i+31:i] := SignExtend32(a[i+31:i] >> imm8[7:0]) + FI ELSE - dst[i+15:i] := src[i+15:i] + dst[i+31:i] := src[i+31:i] FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX512BW - Arithmetic - - - - Add packed 16-bit integers in "a" and "b" using saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - + AVX512F/KNCNI + Shift + + + + Shift packed 32-bit integers in "a" right by "imm8" while shifting in sign bits, and store the results in "dst". -FOR j := 0 to 31 - i := j*16 - IF k[j] - dst[i+15:i] := Saturate_To_Int16( a[i+15:i] + b[i+15:i] ) +FOR j := 0 to 15 + i := j*32 + IF imm8[7:0] > 31 + dst[i+31:i] := (a[i+31] ? 0xFFFFFFFF : 0x0) ELSE - dst[i+15:i] := 0 + dst[i+31:i] := SignExtend32(a[i+31:i] >> imm8[7:0]) FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX512VL - AVX512BW - Arithmetic - - - - - Add packed 16-bit integers in "a" and "b" using saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + AVX512F/KNCNI + Shift + + + + + + Shift packed 32-bit integers in "a" right by the amount specified by the corresponding element in "count" while shifting in sign bits, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := j*16 +FOR j := 0 to 15 + i := j*32 IF k[j] - dst[i+15:i] := Saturate_To_Int16( a[i+15:i] + b[i+15:i] ) + IF count[i+31:i] < 32 + dst[i+31:i] := SignExtend32(a[i+31:i] >> count[i+31:i]) + ELSE + dst[i+31:i] := (a[i+31] ? 0xFFFFFFFF : 0) + FI ELSE - dst[i+15:i] := src[i+15:i] - FI + dst[i+31:i] := src[i+31:i] + FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX512VL - AVX512BW - Arithmetic - - - - Add packed 16-bit integers in "a" and "b" using saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - + AVX512F/KNCNI + Shift + + + + Shift packed 32-bit integers in "a" right by the amount specified by the corresponding element in "count" while shifting in sign bits, and store the results in "dst". -FOR j := 0 to 7 - i := j*16 - IF k[j] - dst[i+15:i] := Saturate_To_Int16( a[i+15:i] + b[i+15:i] ) +FOR j := 0 to 15 + i := j*32 + IF count[i+31:i] < 32 + dst[i+31:i] := SignExtend32(a[i+31:i] >> count[i+31:i]) ELSE - dst[i+15:i] := 0 + dst[i+31:i] := (a[i+31] ? 0xFFFFFFFF : 0) FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX512VL - AVX512BW - Arithmetic - - - - - Add packed unsigned 8-bit integers in "a" and "b" using saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + AVX512F/KNCNI + Shift + + + + + + Shift packed 32-bit integers in "a" right by "imm8" while shifting in zeros, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 31 - i := j*8 +FOR j := 0 to 15 + i := j*32 IF k[j] - dst[i+7:i] := Saturate_To_UnsignedInt8( a[i+7:i] + b[i+7:i] ) + IF imm8[7:0] > 31 + dst[i+31:i] := 0 + ELSE + dst[i+31:i] := ZeroExtend32(a[i+31:i] >> imm8[7:0]) + FI ELSE - dst[i+7:i] := src[i+7:i] + dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX512VL - AVX512BW - Arithmetic - - - - Add packed unsigned 8-bit integers in "a" and "b" using saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - + AVX512F/KNCNI + Shift + + + + Shift packed 32-bit integers in "a" right by "imm8" while shifting in zeros, and store the results in "dst". -FOR j := 0 to 31 - i := j*8 - IF k[j] - dst[i+7:i] := Saturate_To_UnsignedInt8( a[i+7:i] + b[i+7:i] ) +FOR j := 0 to 15 + i := j*32 + IF imm8[7:0] > 31 + dst[i+31:i] := 0 ELSE - dst[i+7:i] := 0 + dst[i+31:i] := ZeroExtend32(a[i+31:i] >> imm8[7:0]) FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX512BW - Arithmetic - - - Add packed unsigned 8-bit integers in "a" and "b" using saturation, and store the results in "dst". + AVX512F/KNCNI + Shift + + + + + + Shift packed 32-bit integers in "a" right by the amount specified by the corresponding element in "count" while shifting in zeros, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 63 - i := j*8 - dst[i+7:i] := Saturate_To_UnsignedInt8( a[i+7:i] + b[i+7:i] ) +FOR j := 0 to 15 + i := j*32 + IF k[j] + IF count[i+31:i] < 32 + dst[i+31:i] := ZeroExtend32(a[i+31:i] >> count[i+31:i]) + ELSE + dst[i+31:i] := 0 + FI + ELSE + dst[i+31:i] := src[i+31:i] + FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX512BW - Arithmetic - - - - - Add packed unsigned 8-bit integers in "a" and "b" using saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + AVX512F/KNCNI + Shift + + + + Shift packed 32-bit integers in "a" right by the amount specified by the corresponding element in "count" while shifting in zeros, and store the results in "dst". -FOR j := 0 to 63 - i := j*8 - IF k[j] - dst[i+7:i] := Saturate_To_UnsignedInt8( a[i+7:i] + b[i+7:i] ) +FOR j := 0 to 15 + i := j*32 + IF count[i+31:i] < 32 + dst[i+31:i] := ZeroExtend32(a[i+31:i] >> count[i+31:i]) ELSE - dst[i+7:i] := src[i+7:i] + dst[i+31:i] := 0 FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX512BW + AVX512F/KNCNI Arithmetic - - - - Add packed unsigned 8-bit integers in "a" and "b" using saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - + + + + + + Subtract packed 32-bit integers in "b" from packed 32-bit integers in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 63 - i := j*8 +FOR j := 0 to 15 + i := j*32 IF k[j] - dst[i+7:i] := Saturate_To_UnsignedInt8( a[i+7:i] + b[i+7:i] ) + dst[i+31:i] := a[i+31:i] - b[i+31:i] ELSE - dst[i+7:i] := 0 - FI + dst[i+31:i] := src[i+31:i] + FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX512VL - AVX512BW + AVX512F/KNCNI Arithmetic - - - - - Add packed unsigned 8-bit integers in "a" and "b" using saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + + + Subtract packed 32-bit integers in "b" from packed 32-bit integers in "a", and store the results in "dst". FOR j := 0 to 15 - i := j*8 - IF k[j] - dst[i+7:i] := Saturate_To_UnsignedInt8( a[i+7:i] + b[i+7:i] ) - ELSE - dst[i+7:i] := src[i+7:i] - FI + i := j*32 + dst[i+31:i] := a[i+31:i] - b[i+31:i] ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX512VL - AVX512BW - Arithmetic - - - - Add packed unsigned 8-bit integers in "a" and "b" using saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - + Mask + AVX512F/KNCNI + Logical + + + + + Compute the bitwise AND of packed 32-bit integers in "a" and "b", producing intermediate 32-bit values, and set the corresponding bit in result mask "k" (subject to writemask "k") if the intermediate value is non-zero. FOR j := 0 to 15 - i := j*8 - IF k[j] - dst[i+7:i] := Saturate_To_UnsignedInt8( a[i+7:i] + b[i+7:i] ) + i := j*32 + IF k1[j] + k[j] := ((a[i+31:i] AND b[i+31:i]) != 0) ? 1 : 0 ELSE - dst[i+7:i] := 0 + k[j] := 0 FI ENDFOR -dst[MAX:128] := 0 +k[MAX:16] := 0 - +
immintrin.h
 - + Integer - AVX512VL - AVX512BW - Arithmetic - - - - - Add packed unsigned 16-bit integers in "a" and "b" using saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Mask + AVX512F/KNCNI + Logical + + + + Compute the bitwise AND of packed 32-bit integers in "a" and "b", producing intermediate 32-bit values, and set the corresponding bit in result mask "k" if the intermediate value is non-zero. FOR j := 0 to 15 - i := j*16 - IF k[j] - dst[i+15:i] := Saturate_To_UnsignedInt16( a[i+15:i] + b[i+15:i] ) - ELSE - dst[i+15:i] := src[i+15:i] - FI + i := j*32 + k[j] := ((a[i+31:i] AND b[i+31:i]) != 0) ? 1 : 0 ENDFOR -dst[MAX:256] := 0 +k[MAX:16] := 0 - +
immintrin.h
 - + Integer - AVX512VL - AVX512BW - Arithmetic - - - - Add packed unsigned 16-bit integers in "a" and "b" using saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - + AVX512F/KNCNI + Logical + + + + + + Compute the bitwise XOR of packed 32-bit integers in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). FOR j := 0 to 15 - i := j*16 + i := j*32 IF k[j] - dst[i+15:i] := Saturate_To_UnsignedInt16( a[i+15:i] + b[i+15:i] ) + dst[i+31:i] := a[i+31:i] XOR b[i+31:i] ELSE - dst[i+15:i] := 0 + dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:256] := 0 - - -
immintrin.h
- - - Integer - AVX512BW - Arithmetic - - - Add packed unsigned 16-bit integers in "a" and "b" using saturation, and store the results in "dst". - -FOR j := 0 to 31 - i := j*16 - dst[i+15:i] := Saturate_To_UnsignedInt16( a[i+15:i] + b[i+15:i] ) -ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX512BW - Arithmetic - - - - - Add packed unsigned 16-bit integers in "a" and "b" using saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + AVX512F/KNCNI + Logical + + + + Compute the bitwise XOR of packed 32-bit integers in "a" and "b", and store the results in "dst". -FOR j := 0 to 31 - i := j*16 - IF k[j] - dst[i+15:i] := Saturate_To_UnsignedInt16( a[i+15:i] + b[i+15:i] ) - ELSE - dst[i+15:i] := src[i+15:i] - FI +FOR j := 0 to 15 + i := j*32 + dst[i+31:i] := a[i+31:i] XOR b[i+31:i] ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX512BW - Arithmetic - - - - Add packed unsigned 16-bit integers in "a" and "b" using saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - + AVX512F/KNCNI + Logical + + + + Compute the bitwise XOR of 512 bits (representing integer data) in "a" and "b", and store the result in "dst". -FOR j := 0 to 31 - i := j*16 - IF k[j] - dst[i+15:i] := Saturate_To_UnsignedInt16( a[i+15:i] + b[i+15:i] ) - ELSE - dst[i+15:i] := 0 - FI -ENDFOR +dst[511:0] := (a[511:0] XOR b[511:0]) dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX512VL - AVX512BW - Arithmetic - - - - - Add packed unsigned 16-bit integers in "a" and "b" using saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + AVX512F/KNCNI + Logical + + + + + + Compute the bitwise XOR of packed 64-bit integers in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). FOR j := 0 to 7 - i := j*16 + i := j*64 IF k[j] - dst[i+15:i] := Saturate_To_UnsignedInt16( a[i+15:i] + b[i+15:i] ) + dst[i+63:i] := a[i+63:i] XOR b[i+63:i] ELSE - dst[i+15:i] := src[i+15:i] + dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX512VL - AVX512BW - Arithmetic - - - - Add packed unsigned 16-bit integers in "a" and "b" using saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - + AVX512F/KNCNI + Logical + + + + Compute the bitwise XOR of packed 64-bit integers in "a" and "b", and store the results in "dst". FOR j := 0 to 7 - i := j*16 - IF k[j] - dst[i+15:i] := Saturate_To_UnsignedInt16( a[i+15:i] + b[i+15:i] ) - ELSE - dst[i+15:i] := 0 - FI + i := j*64 + dst[i+63:i] := a[i+63:i] XOR b[i+63:i] ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512BW - Arithmetic - - - - - Add packed 16-bit integers in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Floating Point + AVX512F/KNCNI + Store + + + + + + Scatter single-precision (32-bit) floating-point elements from "a" into memory using 32-bit indices. 32-bit elements are stored at addresses starting at "base_addr" and offset by each 32-bit element in "vindex" (each index is scaled by the factor in "scale"). "scale" should be 1, 2, 4 or 8. FOR j := 0 to 15 - i := j*16 - IF k[j] - dst[i+15:i] := a[i+15:i] + b[i+15:i] - ELSE - dst[i+15:i] := src[i+15:i] - FI + i := j*32 + m := j*32 + addr := base_addr + SignExtend64(vindex[m+31:m]) * ZeroExtend64(scale) * 8 + MEM[addr+31:addr] := a[i+31:i] ENDFOR -dst[MAX:256] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512BW - Arithmetic - - - - Add packed 16-bit integers in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Floating Point + AVX512F/KNCNI + Store + + + + + + + Scatter single-precision (32-bit) floating-point elements from "a" into memory using 32-bit indices. 32-bit elements are stored at addresses starting at "base_addr" and offset by each 32-bit element in "vindex" (each index is scaled by the factor in "scale") subject to mask "k" (elements are not stored when the corresponding mask bit is not set). "scale" should be 1, 2, 4 or 8. FOR j := 0 to 15 - i := j*16 + i := j*32 + m := j*32 IF k[j] - dst[i+15:i] := a[i+15:i] + b[i+15:i] - ELSE - dst[i+15:i] := 0 + addr := base_addr + SignExtend64(vindex[m+31:m]) * ZeroExtend64(scale) * 8 + MEM[addr+31:addr] := a[i+31:i] FI ENDFOR -dst[MAX:256] := 0 - - -
immintrin.h
- - - Integer - AVX512BW - Arithmetic - - - Add packed 16-bit integers in "a" and "b", and store the results in "dst". - -FOR j := 0 to 31 - i := j*16 - dst[i+15:i] := a[i+15:i] + b[i+15:i] -ENDFOR -dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - AVX512BW + + Floating Point + AVX512F/KNCNI Arithmetic - - - - - Add packed 16-bit integers in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + + + + + Subtract packed double-precision (64-bit) floating-point elements in "b" from packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 31 - i := j*16 +FOR j := 0 to 7 + i := j*64 IF k[j] - dst[i+15:i] := a[i+15:i] + b[i+15:i] + dst[i+63:i] := a[i+63:i] - b[i+63:i] ELSE - dst[i+15:i] := src[i+15:i] + dst[i+63:i] := src[i+63:i] FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - AVX512BW + + Floating Point + AVX512F/KNCNI Arithmetic - - - - Add packed 16-bit integers in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + + + + + + Subtract packed double-precision (64-bit) floating-point elements in "b" from packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + [round_note] -FOR j := 0 to 31 - i := j*16 +FOR j := 0 to 7 + i := j*64 IF k[j] - dst[i+15:i] := a[i+15:i] + b[i+15:i] + dst[i+63:i] := a[i+63:i] - b[i+63:i] ELSE - dst[i+15:i] := 0 + dst[i+63:i] := src[i+63:i] FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512BW + + Floating Point + AVX512F/KNCNI Arithmetic - - - - - Add packed 16-bit integers in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + + + Subtract packed double-precision (64-bit) floating-point elements in "b" from packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst". FOR j := 0 to 7 - i := j*16 - IF k[j] - dst[i+15:i] := a[i+15:i] + b[i+15:i] - ELSE - dst[i+15:i] := src[i+15:i] - FI + i := j*64 + dst[i+63:i] := a[i+63:i] - b[i+63:i] ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512BW + + Floating Point + AVX512F/KNCNI Arithmetic - - - - Add packed 16-bit integers in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + + + + Subtract packed double-precision (64-bit) floating-point elements in "b" from packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst". + [round_note] FOR j := 0 to 7 - i := j*16 - IF k[j] - dst[i+15:i] := a[i+15:i] + b[i+15:i] - ELSE - dst[i+15:i] := 0 - FI -ENDFOR -dst[MAX:128] := 0 - - -
immintrin.h
- - - Integer - AVX512VL - AVX512BW - Miscellaneous - - - - - - Concatenate pairs of 16-byte blocks in "a" and "b" into a 32-byte temporary result, shift the result right by "count" bytes, and store the low 16 bytes in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -FOR j := 0 to 1 - i := j*128 - tmp[255:0] := ((a[i+127:i] << 128) OR b[i+127:i]) >> (count[7:0]*8) - tmp_dst[i+127:i] := tmp[127:0] -ENDFOR - -FOR j := 0 to 31 - i := j*8 - IF k[j] - dst[i+7:i] := tmp_dst[i+7:i] - ELSE - dst[i+7:i] := src[i+7:i] - FI + i := j*64 + dst[i+63:i] := a[i+63:i] - b[i+63:i] ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512BW - Miscellaneous - - - - - Concatenate pairs of 16-byte blocks in "a" and "b" into a 32-byte temporary result, shift the result right by "count" bytes, and store the low 16 bytes in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Floating Point + AVX512F/KNCNI + Arithmetic + + + + + + Subtract packed single-precision (32-bit) floating-point elements in "b" from packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 1 - i := j*128 - tmp[255:0] := ((a[i+127:i] << 128) OR b[i+127:i]) >> (count[7:0]*8) - tmp_dst[i+127:i] := tmp[127:0] -ENDFOR - -FOR j := 0 to 31 - i := j*8 +FOR j := 0 to 15 + i := j*32 IF k[j] - dst[i+7:i] := tmp_dst[i+7:i] + dst[i+31:i] := a[i+31:i] - b[i+31:i] ELSE - dst[i+7:i] := 0 - FI -ENDFOR -dst[MAX:256] := 0 - - -
immintrin.h
- - - Integer - AVX512BW - Miscellaneous - - - - Concatenate pairs of 16-byte blocks in "a" and "b" into a 32-byte temporary result, shift the result right by "count" bytes, and store the low 16 bytes in "dst". - -FOR j := 0 to 3 - i := j*128 - tmp[255:0] := ((a[i+127:i] << 128) OR b[i+127:i]) >> (count[7:0]*8) - dst[i+127:i] := tmp[127:0] + dst[i+31:i] := src[i+31:i] + FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - AVX512BW - Miscellaneous - - - - - - Concatenate pairs of 16-byte blocks in "a" and "b" into a 32-byte temporary result, shift the result right by "count" bytes, and store the low 16 bytes in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Floating Point + AVX512F/KNCNI + Arithmetic + + + + + + + Subtract packed single-precision (32-bit) floating-point elements in "b" from packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + [round_note] -FOR j := 0 to 3 - i := j*128 - tmp[255:0] := ((a[i+127:i] << 128) OR b[i+127:i]) >> (count[7:0]*8) - tmp_dst[i+127:i] := tmp[127:0] -ENDFOR - -FOR j := 0 to 63 - i := j*8 +FOR j := 0 to 15 + i := j*32 IF k[j] - dst[i+7:i] := tmp_dst[i+7:i] + dst[i+31:i] := a[i+31:i] - b[i+31:i] ELSE - dst[i+7:i] := src[i+7:i] - FI + dst[i+31:i] := src[i+31:i] + FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
- - - Integer - AVX512BW - Miscellaneous - - - - - Concatenate pairs of 16-byte blocks in "a" and "b" into a 32-byte temporary result, shift the result right by "count" bytes, and store the low 16 bytes in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - -FOR j := 0 to 3 - i := j*128 - tmp[255:0] := ((a[i+127:i] << 128) OR b[i+127:i]) >> (count[7:0]*8) - tmp_dst[i+127:i] := tmp[127:0] -ENDFOR - -FOR j := 0 to 63 - i := j*8 - IF k[j] - dst[i+7:i] := tmp_dst[i+7:i] - ELSE - dst[i+7:i] := 0 - FI + + + Floating Point + AVX512F/KNCNI + Arithmetic + + + + Subtract packed single-precision (32-bit) floating-point elements in "b" from packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst". + +FOR j := 0 to 15 + i := j*32 + dst[i+31:i] := a[i+31:i] - b[i+31:i] ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512BW - Miscellaneous - - - - - - Concatenate pairs of 16-byte blocks in "a" and "b" into a 32-byte temporary result, shift the result right by "count" bytes, and store the low 16 bytes in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Floating Point + AVX512F/KNCNI + Arithmetic + + + + + Subtract packed single-precision (32-bit) floating-point elements in "b" from packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst". + [round_note] -tmp_dst[255:0] := ((a[127:0] << 128) OR b[127:0]) >> (count[7:0]*8) - FOR j := 0 to 15 - i := j*8 - IF k[j] - dst[i+7:i] := tmp_dst[i+7:i] - ELSE - dst[i+7:i] := src[i+7:i] - FI + i := j*32 + dst[i+31:i] := a[i+31:i] - b[i+31:i] ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - + +
immintrin.h
+ + + Floating Point + AVX512F/KNCNI + Cast + + + Cast vector of type __m512d to type __m512. + This intrinsic is only used for compilation and does not generate any instructions, thus it has zero latency.
immintrin.h
 - + + Floating Point Integer - AVX512VL - AVX512BW - Miscellaneous - - - - - Concatenate pairs of 16-byte blocks in "a" and "b" into a 32-byte temporary result, shift the result right by "count" bytes, and store the low 16 bytes in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + AVX512F/KNCNI + Cast + + + Cast vector of type __m512d to type __m512i. + This intrinsic is only used for compilation and does not generate any instructions, thus it has zero latency. +
immintrin.h
+ + + Floating Point + AVX512F/KNCNI + Cast + + + Cast vector of type __m512 to type __m512d. + This intrinsic is only used for compilation and does not generate any instructions, thus it has zero latency. +
immintrin.h
+ + + Floating Point + Integer + AVX512F/KNCNI + Cast + + + Cast vector of type __m512 to type __m512i. + This intrinsic is only used for compilation and does not generate any instructions, thus it has zero latency. +
immintrin.h
+ + + Floating Point + AVX512F/KNCNI + Cast + + + Cast vector of type __m512i to type __m512d. + This intrinsic is only used for compilation and does not generate any instructions, thus it has zero latency. +
immintrin.h
+ + + Floating Point + AVX512F/KNCNI + Cast + + + Cast vector of type __m512i to type __m512. + This intrinsic is only used for compilation and does not generate any instructions, thus it has zero latency. +
immintrin.h
+ + + Integer + AVX512F/KNCNI + Arithmetic + + + + Reduce the packed 32-bit integers in "a" by addition using mask "k". Returns the sum of all active elements in "a". -tmp_dst[255:0] := ((a[127:0] << 128) OR b[127:0]) >> (count[7:0]*8) - +dst[31:0] := 0 FOR j := 0 to 15 - i := j*8 + i := j*32 IF k[j] - dst[i+7:i] := tmp_dst[i+7:i] - ELSE - dst[i+7:i] := 0 + dst[31:0] := dst[31:0] + a[i+31:i] FI ENDFOR -dst[MAX:128] := 0 -
immintrin.h
 - + Integer - AVX512VL - AVX512F - Logical - - - - - Compute the bitwise AND of packed 32-bit integers in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + AVX512F/KNCNI + Arithmetic + + + + Reduce the packed 64-bit integers in "a" by addition using mask "k". Returns the sum of all active elements in "a". +dst[63:0] := 0 FOR j := 0 to 7 - i := j*32 + i := j*64 IF k[j] - dst[i+31:i] := a[i+31:i] BITWISE AND b[i+31:i] - ELSE - dst[i+31:i] := src[i+31:i] + dst[63:0] := dst[63:0] + a[i+63:i] FI ENDFOR -dst[MAX:256] := 0 -
immintrin.h
 - - Integer - AVX512VL - AVX512F - Logical - - - - Compute the bitwise AND of packed 32-bit integers in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Floating Point + AVX512F/KNCNI + Arithmetic + + + + Reduce the packed double-precision (64-bit) floating-point elements in "a" by addition using mask "k". Returns the sum of all active elements in "a". +dst[63:0] := 0.0 FOR j := 0 to 7 - i := j*32 + i := j*64 IF k[j] - dst[i+31:i] := a[i+31:i] BITWISE AND b[i+31:i] - ELSE - dst[i+31:i] := 0 + dst[63:0] := dst[63:0] + a[i+63:i] FI ENDFOR -dst[MAX:256] := 0 -
immintrin.h
 - - Integer - AVX512VL - AVX512F - Logical - - - - - Compute the bitwise AND of packed 32-bit integers in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Floating Point + AVX512F/KNCNI + Arithmetic + + + + Reduce the packed single-precision (32-bit) floating-point elements in "a" by addition using mask "k". Returns the sum of all active elements in "a". -FOR j := 0 to 3 +dst[31:0] := 0.0 +FOR j := 0 to 15 i := j*32 IF k[j] - dst[i+31:i] := a[i+31:i] AND b[i+31:i] - ELSE - dst[i+31:i] := src[i+31:i] + dst[31:0] := dst[31:0] + a[i+31:i] FI ENDFOR -dst[MAX:128] := 0 -
immintrin.h
 - + Integer - AVX512VL - AVX512F + AVX512F/KNCNI Logical - - - - Compute the bitwise AND of packed 32-bit integers in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + + + Reduce the packed 32-bit integers in "a" by bitwise AND using mask "k". Returns the bitwise AND of all active elements in "a". -FOR j := 0 to 3 +dst[31:0] := 0xFFFFFFFF +FOR j := 0 to 15 i := j*32 IF k[j] - dst[i+31:i] := a[i+31:i] AND b[i+31:i] - ELSE - dst[i+31:i] := 0 + dst[31:0] := dst[31:0] AND a[i+31:i] FI ENDFOR -dst[MAX:128] := 0 -
immintrin.h
 - + Integer - AVX512VL - AVX512F + AVX512F/KNCNI Logical - - - - - Compute the bitwise NOT of packed 32-bit integers in "a" and then AND with "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + + + Reduce the packed 64-bit integers in "a" by bitwise AND using mask "k". Returns the bitwise AND of all active elements in "a". +dst[63:0] := 0xFFFFFFFFFFFFFFFF FOR j := 0 to 7 - i := j*32 + i := j*64 IF k[j] - dst[i+31:i] := ((NOT a[i+31:i]) AND b[i+31:i]) - ELSE - dst[i+31:i] := src[i+31:i] + dst[63:0] := dst[63:0] AND a[i+63:i] FI ENDFOR -dst[MAX:256] := 0 -
immintrin.h
 - + Integer - AVX512VL - AVX512F - Logical - - - - Compute the bitwise NOT of packed 32-bit integers in "a" and then AND with "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + AVX512F/KNCNI + Special Math Functions + + + + Reduce the packed signed 32-bit integers in "a" by maximum using mask "k". Returns the maximum of all active elements in "a". -FOR j := 0 to 7 +dst[31:0] := Int32(-0x80000000) +FOR j := 0 to 15 i := j*32 IF k[j] - dst[i+31:i] := (NOT a[i+31:i]) AND b[i+31:i] - ELSE - dst[i+31:i] := 0 + dst[31:0] := (dst[31:0] > a[i+31:i] ? dst[31:0] : a[i+31:i]) FI ENDFOR -dst[MAX:256] := 0 -
immintrin.h
 - + Integer - AVX512VL - AVX512F - Logical - - - - - Compute the bitwise NOT of packed 32-bit integers in "a" and then AND with "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + AVX512F/KNCNI + Special Math Functions + + + + Reduce the packed signed 64-bit integers in "a" by maximum using mask "k". Returns the maximum of all active elements in "a". -FOR j := 0 to 3 - i := j*32 +dst[63:0] := Int64(-0x8000000000000000) +FOR j := 0 to 7 + i := j*64 IF k[j] - dst[i+31:i] := ((NOT a[i+31:i]) AND b[i+31:i]) - ELSE - dst[i+31:i] := src[i+31:i] + dst[63:0] := (dst[63:0] > a[i+63:i] ? dst[63:0] : a[i+63:i]) FI ENDFOR -dst[MAX:128] := 0 -
immintrin.h
 - + Integer - AVX512VL - AVX512F - Logical - - - - Compute the bitwise NOT of packed 32-bit integers in "a" and then AND with "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + AVX512F/KNCNI + Special Math Functions + + + + Reduce the packed unsigned 32-bit integers in "a" by maximum using mask "k". Returns the maximum of all active elements in "a". -FOR j := 0 to 3 +dst[31:0] := 0 +FOR j := 0 to 15 i := j*32 IF k[j] - dst[i+31:i] := (NOT a[i+31:i]) AND b[i+31:i] - ELSE - dst[i+31:i] := 0 + dst[31:0] := (dst[31:0] > a[i+31:i] ? dst[31:0] : a[i+31:i]) FI ENDFOR -dst[MAX:128] := 0 -
immintrin.h
 - + Integer - AVX512VL - AVX512F - Logical - - - - - Compute the bitwise NOT of packed 64-bit integers in "a" and then AND with "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + AVX512F/KNCNI + Special Math Functions + + + + Reduce the packed unsigned 64-bit integers in "a" by maximum using mask "k". Returns the maximum of all active elements in "a". -FOR j := 0 to 3 +dst[63:0] := 0 +FOR j := 0 to 7 i := j*64 IF k[j] - dst[i+63:i] := ((NOT a[i+63:i]) AND b[i+63:i]) - ELSE - dst[i+63:i] := src[i+63:i] + dst[63:0] := (dst[63:0] > a[i+63:i] ? dst[63:0] : a[i+63:i]) FI ENDFOR -dst[MAX:256] := 0 -
immintrin.h
 - - Integer - AVX512VL - AVX512F - Logical - - - - Compute the bitwise NOT of packed 64-bit integers in "a" and then AND with "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Floating Point + AVX512F/KNCNI + Special Math Functions + + + + Reduce the packed double-precision (64-bit) floating-point elements in "a" by maximum using mask "k". Returns the maximum of all active elements in "a". -FOR j := 0 to 3 +dst[63:0] := Cast_FP64(0xFFEFFFFFFFFFFFFF) +FOR j := 0 to 7 i := j*64 IF k[j] - dst[i+63:i] := (NOT a[i+63:i]) AND b[i+63:i] - ELSE - dst[i+63:i] := 0 + dst[63:0] := (dst[63:0] > a[i+63:i] ? dst[63:0] : a[i+63:i]) FI ENDFOR -dst[MAX:256] := 0 -
immintrin.h
 - - Integer - AVX512VL - AVX512F - Logical - - - - - Compute the bitwise NOT of packed 64-bit integers in "a" and then AND with "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Floating Point + AVX512F/KNCNI + Special Math Functions + + + + Reduce the packed single-precision (32-bit) floating-point elements in "a" by maximum using mask "k". Returns the maximum of all active elements in "a". -FOR j := 0 to 1 - i := j*64 +dst[31:0] := Cast_FP32(0xFF7FFFFF) +FOR j := 0 to 15 + i := j*32 IF k[j] - dst[i+63:i] := ((NOT a[i+63:i]) AND b[i+63:i]) - ELSE - dst[i+63:i] := src[i+63:i] + dst[31:0] := (dst[31:0] > a[i+31:i] ? dst[31:0] : a[i+31:i]) FI ENDFOR -dst[MAX:128] := 0 -
immintrin.h
 - + Integer - AVX512VL - AVX512F - Logical - - - - Compute the bitwise NOT of packed 64-bit integers in "a" and then AND with "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + AVX512F/KNCNI + Special Math Functions + + + + Reduce the packed signed 32-bit integers in "a" by maximum using mask "k". Returns the minimum of all active elements in "a". -FOR j := 0 to 1 - i := j*64 +dst[31:0] := Int32(0x7FFFFFFF) +FOR j := 0 to 15 + i := j*32 IF k[j] - dst[i+63:i] := (NOT a[i+63:i]) AND b[i+63:i] - ELSE - dst[i+63:i] := 0 + dst[31:0] := (dst[31:0] < a[i+31:i] ? dst[31:0] : a[i+31:i]) FI ENDFOR -dst[MAX:128] := 0 -
immintrin.h
 - + Integer - AVX512VL - AVX512F - Logical - - - - - Compute the bitwise AND of packed 64-bit integers in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + AVX512F/KNCNI + Special Math Functions + + + + Reduce the packed signed 64-bit integers in "a" by maximum using mask "k". Returns the minimum of all active elements in "a". -FOR j := 0 to 3 +dst[63:0] := Int64(0x7FFFFFFFFFFFFFFF) +FOR j := 0 to 7 i := j*64 IF k[j] - dst[i+63:i] := a[i+63:i] AND b[i+63:i] - ELSE - dst[i+63:i] := src[i+63:i] + dst[63:0] := (dst[63:0] < a[i+63:i] ? dst[63:0] : a[i+63:i]) FI ENDFOR -dst[MAX:256] := 0 -
immintrin.h
 - + Integer - AVX512VL - AVX512F - Logical - - - - Compute the bitwise AND of packed 64-bit integers in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + AVX512F/KNCNI + Special Math Functions + + + + Reduce the packed unsigned 32-bit integers in "a" by maximum using mask "k". Returns the minimum of all active elements in "a". -FOR j := 0 to 3 - i := j*64 +dst[31:0] := 0xFFFFFFFF +FOR j := 0 to 15 + i := j*32 IF k[j] - dst[i+63:i] := a[i+63:i] AND b[i+63:i] - ELSE - dst[i+63:i] := 0 + dst[31:0] := (dst[31:0] < a[i+31:i] ? dst[31:0] : a[i+31:i]) FI ENDFOR -dst[MAX:256] := 0 -
immintrin.h
 - + Integer - AVX512VL - AVX512F - Logical - - - - - Compute the bitwise AND of packed 64-bit integers in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + AVX512F/KNCNI + Special Math Functions + + + + Reduce the packed unsigned 64-bit integers in "a" by minimum using mask "k". Returns the minimum of all active elements in "a". -FOR j := 0 to 1 +dst[63:0] := 0xFFFFFFFFFFFFFFFF +FOR j := 0 to 7 i := j*64 IF k[j] - dst[i+63:i] := a[i+63:i] AND b[i+63:i] - ELSE - dst[i+63:i] := src[i+63:i] + dst[63:0] := (dst[63:0] < a[i+63:i] ? dst[63:0] : a[i+63:i]) FI ENDFOR -dst[MAX:128] := 0 -
immintrin.h
 - - Integer - AVX512VL - AVX512F - Logical - - - - Compute the bitwise AND of packed 64-bit integers in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Floating Point + AVX512F/KNCNI + Special Math Functions + + + + Reduce the packed double-precision (64-bit) floating-point elements in "a" by maximum using mask "k". Returns the minimum of all active elements in "a". -FOR j := 0 to 1 +dst[63:0] := Cast_FP64(0x7FEFFFFFFFFFFFFF) +FOR j := 0 to 7 i := j*64 IF k[j] - dst[i+63:i] := a[i+63:i] AND b[i+63:i] - ELSE - dst[i+63:i] := 0 + dst[63:0] := (dst[63:0] < a[i+63:i] ? dst[63:0] : a[i+63:i]) FI ENDFOR -dst[MAX:128] := 0 -
immintrin.h
 - - Integer - AVX512VL - AVX512BW - Arithmetic - - - - - Average packed unsigned 8-bit integers in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Floating Point + AVX512F/KNCNI + Special Math Functions + + + + Reduce the packed single-precision (32-bit) floating-point elements in "a" by maximum using mask "k". Returns the minimum of all active elements in "a". -FOR j := 0 to 31 - i := j*8 +dst[31:0] := Cast_FP32(0x7F7FFFFF) +FOR j := 0 to 15 + i := j*32 IF k[j] - dst[i+7:i] := (a[i+7:i] + b[i+7:i] + 1) >> 1 - ELSE - dst[i+7:i] := src[i+7:i] + dst[31:0] := (dst[31:0] < a[i+31:i] ? dst[31:0] : a[i+31:i]) FI ENDFOR -dst[MAX:256] := 0 -
immintrin.h
 - + Integer - AVX512VL - AVX512BW + AVX512F/KNCNI Arithmetic - - - - Average packed unsigned 8-bit integers in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + + + Reduce the packed 32-bit integers in "a" by multiplication using mask "k". Returns the product of all active elements in "a". -FOR j := 0 to 31 - i := j*8 +dst[31:0] := 1 +FOR j := 0 to 15 + i := j*32 IF k[j] - dst[i+7:i] := (a[i+7:i] + b[i+7:i] + 1) >> 1 - ELSE - dst[i+7:i] := 0 + dst[31:0] := dst[31:0] * a[i+31:i] FI ENDFOR -dst[MAX:256] := 0 -
immintrin.h
 - + Integer - AVX512BW + AVX512F/KNCNI Arithmetic - - - Average packed unsigned 8-bit integers in "a" and "b", and store the results in "dst". + + + + Reduce the packed 64-bit integers in "a" by multiplication using mask "k". Returns the product of all active elements in "a". -FOR j := 0 to 63 - i := j*8 - dst[i+7:i] := (a[i+7:i] + b[i+7:i] + 1) >> 1 +dst[63:0] := 1 +FOR j := 0 to 7 + i := j*64 + IF k[j] + dst[63:0] := dst[63:0] * a[i+63:i] + FI ENDFOR -dst[MAX:512] := 0 -
immintrin.h
 - - Integer - AVX512BW + + Floating Point + AVX512F/KNCNI Arithmetic - - - - - Average packed unsigned 8-bit integers in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + + + Reduce the packed double-precision (64-bit) floating-point elements in "a" by multiplication using mask "k". Returns the product of all active elements in "a". -FOR j := 0 to 63 - i := j*8 +dst[63:0] := 1.0 +FOR j := 0 to 7 + i := j*64 IF k[j] - dst[i+7:i] := (a[i+7:i] + b[i+7:i] + 1) >> 1 - ELSE - dst[i+7:i] := src[i+7:i] + dst[63:0] := dst[63:0] * a[i+63:i] FI ENDFOR -dst[MAX:512] := 0 -
immintrin.h
 - - Integer - AVX512BW + + Floating Point + AVX512F/KNCNI Arithmetic - - - - Average packed unsigned 8-bit integers in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + + + Reduce the packed single-precision (32-bit) floating-point elements in "a" by multiplication using mask "k". Returns the product of all active elements in "a". -FOR j := 0 to 63 - i := j*8 +dst[31:0] := FP32(1.0) +FOR j := 0 to 15 + i := j*32 IF k[j] - dst[i+7:i] := (a[i+7:i] + b[i+7:i] + 1) >> 1 - ELSE - dst[i+7:i] := 0 + dst[31:0] := dst[31:0] * a[i+31:i] FI ENDFOR -dst[MAX:512] := 0 -
immintrin.h
 - + Integer - AVX512VL - AVX512BW - Arithmetic - - - - - Average packed unsigned 8-bit integers in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + AVX512F/KNCNI + Logical + + + + Reduce the packed 32-bit integers in "a" by bitwise OR using mask "k". Returns the bitwise OR of all active elements in "a". +dst[31:0] := 0 FOR j := 0 to 15 - i := j*8 + i := j*32 IF k[j] - dst[i+7:i] := (a[i+7:i] + b[i+7:i] + 1) >> 1 - ELSE - dst[i+7:i] := src[i+7:i] + dst[31:0] := dst[31:0] OR a[i+31:i] FI ENDFOR -dst[MAX:128] := 0 -
immintrin.h
 - + Integer - AVX512VL - AVX512BW - Arithmetic - - - - Average packed unsigned 8-bit integers in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + AVX512F/KNCNI + Logical + + + + Reduce the packed 64-bit integers in "a" by bitwise OR using mask "k". Returns the bitwise OR of all active elements in "a". -FOR j := 0 to 15 - i := j*8 +dst[63:0] := 0 +FOR j := 0 to 7 + i := j*64 IF k[j] - dst[i+7:i] := (a[i+7:i] + b[i+7:i] + 1) >> 1 - ELSE - dst[i+7:i] := 0 + dst[63:0] := dst[63:0] OR a[i+63:i] FI ENDFOR -dst[MAX:128] := 0 -
immintrin.h
 - + Integer - AVX512VL - AVX512BW + AVX512F/KNCNI Arithmetic - - - - - Average packed unsigned 16-bit integers in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + + Reduce the packed 32-bit integers in "a" by addition. Returns the sum of all elements in "a". +dst[31:0] := 0 FOR j := 0 to 15 - i := j*16 - IF k[j] - dst[i+15:i] := (a[i+15:i] + b[i+15:i] + 1) >> 1 - ELSE - dst[i+15:i] := src[i+15:i] - FI + i := j*32 + dst[31:0] := dst[31:0] + a[i+31:i] ENDFOR -dst[MAX:256] := 0 -
immintrin.h
 - + Integer - AVX512VL - AVX512BW + AVX512F/KNCNI Arithmetic - - - - Average packed unsigned 16-bit integers in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - + + + Reduce the packed 64-bit integers in "a" by addition. Returns the sum of all elements in "a". -FOR j := 0 to 15 - i := j*16 - IF k[j] - dst[i+15:i] := (a[i+15:i] + b[i+15:i] + 1) >> 1 - ELSE - dst[i+15:i] := 0 - FI +dst[63:0] := 0 +FOR j := 0 to 7 + i := j*64 + dst[63:0] := dst[63:0] + a[i+63:i] ENDFOR -dst[MAX:256] := 0 -
immintrin.h
 - - Integer - AVX512BW + + Floating Point + AVX512F/KNCNI Arithmetic - - - Average packed unsigned 16-bit integers in "a" and "b", and store the results in "dst". + + + Reduce the packed double-precision (64-bit) floating-point elements in "a" by addition. Returns the sum of all elements in "a". -FOR j := 0 to 31 - i := j*16 - dst[i+15:i] := (a[i+15:i] + b[i+15:i] + 1) >> 1 +dst[63:0] := 0.0 +FOR j := 0 to 7 + i := j*64 + dst[63:0] := dst[63:0] + a[i+63:i] ENDFOR -dst[MAX:512] := 0 -
immintrin.h
 - - Integer - AVX512BW + + Floating Point + AVX512F/KNCNI Arithmetic - - - - - Average packed unsigned 16-bit integers in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + + Reduce the packed single-precision (32-bit) floating-point elements in "a" by addition. Returns the sum of all elements in "a". -FOR j := 0 to 31 - i := j*16 - IF k[j] - dst[i+15:i] := (a[i+15:i] + b[i+15:i] + 1) >> 1 - ELSE - dst[i+15:i] := src[i+15:i] - FI +dst[31:0] := 0.0 +FOR j := 0 to 15 + i := j*32 + dst[31:0] := dst[31:0] + a[i+31:i] ENDFOR -dst[MAX:512] := 0 -
immintrin.h
 - + Integer - AVX512BW - Arithmetic - - - - Average packed unsigned 16-bit integers in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - + AVX512F/KNCNI + Logical + + + Reduce the packed 32-bit integers in "a" by bitwise AND. Returns the bitwise AND of all elements in "a". -FOR j := 0 to 31 - i := j*16 - IF k[j] - dst[i+15:i] := (a[i+15:i] + b[i+15:i] + 1) >> 1 - ELSE - dst[i+15:i] := 0 - FI +dst[31:0] := 0xFFFFFFFF +FOR j := 0 to 15 + i := j*32 + dst[31:0] := dst[31:0] AND a[i+31:i] ENDFOR -dst[MAX:512] := 0 -
immintrin.h
 - + Integer - AVX512VL - AVX512BW - Arithmetic - - - - - Average packed unsigned 16-bit integers in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + AVX512F/KNCNI + Logical + + + Reduce the packed 64-bit integers in "a" by bitwise AND. Returns the bitwise AND of all elements in "a". +dst[63:0] := 0xFFFFFFFFFFFFFFFF FOR j := 0 to 7 - i := j*16 - IF k[j] - dst[i+15:i] := (a[i+15:i] + b[i+15:i] + 1) >> 1 - ELSE - dst[i+15:i] := src[i+15:i] - FI + i := j*64 + dst[63:0] := dst[63:0] AND a[i+63:i] ENDFOR -dst[MAX:128] := 0 -
immintrin.h
 - + Integer - AVX512VL - AVX512BW - Arithmetic - - - - Average packed unsigned 16-bit integers in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - + AVX512F/KNCNI + Special Math Functions + + + Reduce the packed signed 32-bit integers in "a" by maximum. Returns the maximum of all elements in "a". + +dst[31:0] := Int32(-0x80000000) +FOR j := 0 to 15 + i := j*32 + dst[31:0] := (dst[31:0] > a[i+31:i] ? dst[31:0] : a[i+31:i]) +ENDFOR + +
immintrin.h
+ + + Integer + AVX512F/KNCNI + Special Math Functions + + + Reduce the packed signed 64-bit integers in "a" by maximum. Returns the maximum of all elements in "a". +dst[63:0] := Int64(-0x8000000000000000) FOR j := 0 to 7 - i := j*16 - IF k[j] - dst[i+15:i] := (a[i+15:i] + b[i+15:i] + 1) >> 1 - ELSE - dst[i+15:i] := 0 - FI + i := j*64 + dst[63:0] := (dst[63:0] > a[i+63:i] ? dst[63:0] : a[i+63:i]) ENDFOR -dst[MAX:128] := 0 -
immintrin.h
 - + Integer - AVX512VL - AVX512BW - Miscellaneous - - - - Blend packed 8-bit integers from "a" and "b" using control mask "k", and store the results in "dst". + AVX512F/KNCNI + Special Math Functions + + + Reduce the packed unsigned 32-bit integers in "a" by maximum. Returns the maximum of all elements in "a". -FOR j := 0 to 31 - i := j*8 - IF k[j] - dst[i+7:i] := b[i+7:i] - ELSE - dst[i+7:i] := a[i+7:i] - FI +dst[31:0] := 0 +FOR j := 0 to 15 + i := j*32 + dst[31:0] := (dst[31:0] > a[i+31:i] ? dst[31:0] : a[i+31:i]) ENDFOR -dst[MAX:256] := 0 -
immintrin.h
 - + Integer - AVX512BW - Miscellaneous - - - - Blend packed 8-bit integers from "a" and "b" using control mask "k", and store the results in "dst". + AVX512F/KNCNI + Special Math Functions + + + Reduce the packed unsigned 64-bit integers in "a" by maximum. Returns the maximum of all elements in "a". -FOR j := 0 to 63 - i := j*8 - IF k[j] - dst[i+7:i] := b[i+7:i] - ELSE - dst[i+7:i] := a[i+7:i] - FI +dst[63:0] := 0 +FOR j := 0 to 7 + i := j*64 + dst[63:0] := (dst[63:0] > a[i+63:i] ? dst[63:0] : a[i+63:i]) ENDFOR -dst[MAX:512] := 0 -
immintrin.h
 - - Integer - AVX512VL - AVX512BW - Miscellaneous - - - - Blend packed 8-bit integers from "a" and "b" using control mask "k", and store the results in "dst". + + Floating Point + AVX512F/KNCNI + Special Math Functions + + + Reduce the packed double-precision (64-bit) floating-point elements in "a" by maximum. Returns the maximum of all elements in "a". -FOR j := 0 to 15 - i := j*8 - IF k[j] - dst[i+7:i] := b[i+7:i] - ELSE - dst[i+7:i] := a[i+7:i] - FI +dst[63:0] := Cast_FP64(0xFFEFFFFFFFFFFFFF) +FOR j := 0 to 7 + i := j*64 + dst[63:0] := (dst[63:0] > a[i+63:i] ? dst[63:0] : a[i+63:i]) ENDFOR -dst[MAX:128] := 0 -
immintrin.h
 - - Integer - AVX512VL - AVX512F - Miscellaneous - - - - Blend packed 32-bit integers from "a" and "b" using control mask "k", and store the results in "dst". + + Floating Point + AVX512F/KNCNI + Special Math Functions + + + Reduce the packed single-precision (32-bit) floating-point elements in "a" by maximum. Returns the maximum of all elements in "a". -FOR j := 0 to 7 +dst[31:0] := Cast_FP32(0xFF7FFFFF) +FOR j := 0 to 15 i := j*32 - IF k[j] - dst[i+31:i] := b[i+31:i] - ELSE - dst[i+31:i] := a[i+31:i] - FI + dst[31:0] := (dst[31:0] > a[i+31:i] ? dst[31:0] : a[i+31:i]) ENDFOR -dst[MAX:256] := 0 -
immintrin.h
 - + Integer - AVX512VL - AVX512F - Miscellaneous - - - - Blend packed 32-bit integers from "a" and "b" using control mask "k", and store the results in "dst". + AVX512F/KNCNI + Special Math Functions + + + Reduce the packed signed 32-bit integers in "a" by minimum. Returns the minimum of all elements in "a". -FOR j := 0 to 3 +dst[31:0] := Int32(0x7FFFFFFF) +FOR j := 0 to 15 i := j*32 - IF k[j] - dst[i+31:i] := b[i+31:i] - ELSE - dst[i+31:i] := a[i+31:i] - FI + dst[31:0] := (dst[31:0] < a[i+31:i] ? dst[31:0] : a[i+31:i]) ENDFOR -dst[MAX:128] := 0 -
immintrin.h
 - + Integer - AVX512VL - AVX512F - Miscellaneous - - - - Blend packed 64-bit integers from "a" and "b" using control mask "k", and store the results in "dst". + AVX512F/KNCNI + Special Math Functions + + + Reduce the packed signed 64-bit integers in "a" by minimum. Returns the minimum of all elements in "a". -FOR j := 0 to 3 +dst[63:0] := Int64(0x7FFFFFFFFFFFFFFF) +FOR j := 0 to 7 i := j*64 - IF k[j] - dst[i+63:i] := b[i+63:i] - ELSE - dst[i+63:i] := a[i+63:i] - FI + dst[63:0] := (dst[63:0] < a[i+63:i] ? dst[63:0] : a[i+63:i]) ENDFOR -dst[MAX:256] := 0 -
immintrin.h
 - + Integer - AVX512VL - AVX512F - Miscellaneous - - - - Blend packed 64-bit integers from "a" and "b" using control mask "k", and store the results in "dst". + AVX512F/KNCNI + Special Math Functions + + + Reduce the packed unsigned 32-bit integers in "a" by minimum. Returns the minimum of all elements in "a". -FOR j := 0 to 1 - i := j*64 - IF k[j] - dst[i+63:i] := b[i+63:i] - ELSE - dst[i+63:i] := a[i+63:i] - FI +dst[31:0] := 0xFFFFFFFF +FOR j := 0 to 15 + i := j*32 + dst[31:0] := (dst[31:0] < a[i+31:i] ? dst[31:0] : a[i+31:i]) ENDFOR -dst[MAX:128] := 0 -
immintrin.h
 - + Integer - AVX512VL - AVX512BW - Miscellaneous - - - - Blend packed 16-bit integers from "a" and "b" using control mask "k", and store the results in "dst". + AVX512F/KNCNI + Special Math Functions + + + Reduce the packed unsigned 64-bit integers in "a" by minimum. Returns the minimum of all elements in "a". -FOR j := 0 to 15 - i := j*16 - IF k[j] - dst[i+15:i] := b[i+15:i] - ELSE - dst[i+15:i] := a[i+15:i] - FI +dst[63:0] := 0xFFFFFFFFFFFFFFFF +FOR j := 0 to 7 + i := j*64 + dst[63:0] := (dst[63:0] < a[i+63:i] ? dst[63:0] : a[i+63:i]) ENDFOR -dst[MAX:256] := 0 -
immintrin.h
 - - Integer - AVX512BW - Miscellaneous - - - - Blend packed 16-bit integers from "a" and "b" using control mask "k", and store the results in "dst". + + Floating Point + AVX512F/KNCNI + Special Math Functions + + + Reduce the packed double-precision (64-bit) floating-point elements in "a" by minimum. Returns the minimum of all elements in "a". -FOR j := 0 to 31 - i := j*16 - IF k[j] - dst[i+15:i] := b[i+15:i] - ELSE - dst[i+15:i] := a[i+15:i] - FI +dst[63:0] := Cast_FP64(0x7FEFFFFFFFFFFFFF) +FOR j := 0 to 7 + i := j*64 + dst[63:0] := (dst[63:0] < a[i+63:i] ? dst[63:0] : a[i+63:i]) ENDFOR -dst[MAX:512] := 0 -
immintrin.h
 - - Integer - AVX512VL - AVX512BW - Miscellaneous - - - - Blend packed 16-bit integers from "a" and "b" using control mask "k", and store the results in "dst". + + Floating Point + AVX512F/KNCNI + Special Math Functions + + + Reduce the packed single-precision (32-bit) floating-point elements in "a" by minimum. Returns the minimum of all elements in "a". -FOR j := 0 to 7 - i := j*16 - IF k[j] - dst[i+15:i] := b[i+15:i] - ELSE - dst[i+15:i] := a[i+15:i] - FI +dst[31:0] := Cast_FP32(0x7F7FFFFF) +FOR j := 0 to 15 + i := j*32 + dst[31:0] := (dst[31:0] < a[i+31:i] ? dst[31:0] : a[i+31:i]) ENDFOR -dst[MAX:128] := 0 -
immintrin.h
 - + Integer - AVX512VL - AVX512BW - Miscellaneous - - - - Broadcast the low packed 8-bit integer from "a" to all elements of "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + AVX512F/KNCNI + Arithmetic + + + Reduce the packed 32-bit integers in "a" by multiplication. Returns the product of all elements in "a". -FOR j := 0 to 31 - i := j*8 - IF k[j] - dst[i+7:i] := a[7:0] - ELSE - dst[i+7:i] := src[i+7:i] - FI +dst[31:0] := 1 +FOR j := 0 to 15 + i := j*32 + dst[31:0] := dst[31:0] * a[i+31:i] ENDFOR -dst[MAX:256] := 0 -
immintrin.h
 - + Integer - AVX512VL - AVX512BW - Set - - - - Broadcast 8-bit integer "a" to all elements of "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + AVX512F/KNCNI + Arithmetic + + + Reduce the packed 64-bit integers in "a" by multiplication. Returns the product of all elements in "a". -FOR j := 0 to 31 - i := j*8 - IF k[j] - dst[i+7:i] := a[7:0] - ELSE - dst[i+7:i] := src[i+7:i] - FI +dst[63:0] := 1 +FOR j := 0 to 7 + i := j*64 + dst[63:0] := dst[63:0] * a[i+63:i] ENDFOR -dst[MAX:256] := 0 -
immintrin.h
 - - Integer - AVX512VL - AVX512BW - Miscellaneous - - - Broadcast the low packed 8-bit integer from "a" to all elements of "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Floating Point + AVX512F/KNCNI + Arithmetic + + + Reduce the packed double-precision (64-bit) floating-point elements in "a" by multiplication. Returns the product of all elements in "a". -FOR j := 0 to 31 - i := j*8 - IF k[j] - dst[i+7:i] := a[7:0] - ELSE - dst[i+7:i] := 0 - FI +dst[63:0] := 1.0 +FOR j := 0 to 7 + i := j*64 + dst[63:0] := dst[63:0] * a[i+63:i] ENDFOR -dst[MAX:256] := 0 -
immintrin.h
 - - Integer - AVX512VL - AVX512BW - Set - - - Broadcast 8-bit integer "a" to all elements of "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Floating Point + AVX512F/KNCNI + Arithmetic + + + Reduce the packed single-precision (32-bit) floating-point elements in "a" by multiplication. Returns the product of all elements in "a". -FOR j := 0 to 31 - i := j*8 - IF k[j] - dst[i+7:i] := a[7:0] - ELSE - dst[i+7:i] := 0 - FI +dst[31:0] := FP32(1.0) +FOR j := 0 to 15 + i := j*32 + dst[31:0] := dst[31:0] * a[i+31:i] ENDFOR -dst[MAX:256] := 0 -
immintrin.h
 - + Integer - AVX512BW - Miscellaneous - - Broadcast the low packed 8-bit integer from "a" to all elements of "dst". + AVX512F/KNCNI + Logical + + + Reduce the packed 32-bit integers in "a" by bitwise OR. Returns the bitwise OR of all elements in "a". -FOR j := 0 to 63 - i := j*8 - dst[i+7:i] := a[7:0] +dst[31:0] := 0 +FOR j := 0 to 15 + i := j*32 + dst[31:0] := dst[31:0] OR a[i+31:i] ENDFOR -dst[MAX:512] := 0 -
immintrin.h
 - + Integer - AVX512BW - Miscellaneous - - - - Broadcast the low packed 8-bit integer from "a" to all elements of "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + AVX512F/KNCNI + Logical + + + Reduce the packed 64-bit integers in "a" by bitwise OR. Returns the bitwise OR of all elements in "a". -FOR j := 0 to 63 - i := j*8 - IF k[j] - dst[i+7:i] := a[7:0] - ELSE - dst[i+7:i] := src[i+7:i] - FI +dst[63:0] := 0 +FOR j := 0 to 7 + i := j*64 + dst[63:0] := dst[63:0] OR a[i+63:i] ENDFOR -dst[MAX:512] := 0 -
immintrin.h
 - + Integer - AVX512BW - Set - - - - Broadcast 8-bit integer "a" to all elements of "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + AVX512F/KNCNI + Logical + + + + + + Performs element-by-element bitwise AND between packed 32-bit integer elements of "v2" and "v3", storing the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 63 - i := j*8 +FOR j := 0 to 15 + i := j*32 IF k[j] - dst[i+7:i] := a[7:0] + dst[i+31:i] := v2[i+31:i] & v3[i+31:i] ELSE - dst[i+7:i] := src[i+7:i] + dst[i+31:i] := src[i+31:i] FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - AVX512BW - Miscellaneous - - - Broadcast the low packed 8-bit integer from "a" to all elements of "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Floating Point + AVX512F/KNCNI + Convert + + + Performs element-by-element conversion of the lower half of packed single-precision (32-bit) floating-point elements in "v2" to packed double-precision (64-bit) floating-point elements, storing the results in "dst". -FOR j := 0 to 63 - i := j*8 - IF k[j] - dst[i+7:i] := a[7:0] - ELSE - dst[i+7:i] := 0 - FI +FOR j := 0 to 7 + i := j*32 + n := j*64 + dst[n+63:n] := Convert_FP32_To_FP64(v2[i+31:i]) ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - AVX512BW - Set - - - Broadcast 8-bit integer "a" to all elements of "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Floating Point + AVX512F/KNCNI + Convert + + + + + Performs element-by-element conversion of the lower half of packed single-precision (32-bit) floating-point elements in "v2" to packed double-precision (64-bit) floating-point elements, storing the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 63 - i := j*8 +FOR j := 0 to 7 + i := j*32 + l := j*64 IF k[j] - dst[i+7:i] := a[7:0] + dst[l+63:l] := Convert_FP32_To_FP64(v2[i+31:i]) ELSE - dst[i+7:i] := 0 + dst[l+63:l] := src[l+63:l] FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + + Floating Point Integer - AVX512VL - AVX512BW - Miscellaneous - - - - Broadcast the low packed 8-bit integer from "a" to all elements of "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + AVX512F/KNCNI + Convert + + + Performs element-by-element conversion of the lower half of packed 32-bit integer elements in "v2" to packed double-precision (64-bit) floating-point elements, storing the results in "dst". -FOR j := 0 to 15 - i := j*8 - IF k[j] - dst[i+7:i] := a[7:0] - ELSE - dst[i+7:i] := src[i+7:i] - FI +FOR j := 0 to 7 + i := j*32 + l := j*64 + dst[l+63:l] := Convert_Int32_To_FP64(v2[i+31:i]) ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + + Floating Point Integer - AVX512VL - AVX512BW - Set - - - - Broadcast 8-bit integer "a" to all elements of "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + AVX512F/KNCNI + Convert + + + + + Performs element-by-element conversion of the lower half of packed 32-bit integer elements in "v2" to packed double-precision (64-bit) floating-point elements, storing the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*8 +FOR j := 0 to 7 + i := j*32 + n := j*64 IF k[j] - dst[i+7:i] := a[7:0] + dst[n+63:n] := Convert_Int32_To_FP64(v2[i+31:i]) ELSE - dst[i+7:i] := src[i+7:i] + dst[n+63:n] := src[n+63:n] FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + + Floating Point Integer - AVX512VL - AVX512BW - Miscellaneous - - - Broadcast the low packed 8-bit integer from "a" to all elements of "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + AVX512F/KNCNI + Convert + + + Performs element-by-element conversion of the lower half of packed 32-bit unsigned integer elements in "v2" to packed double-precision (64-bit) floating-point elements, storing the results in "dst". -FOR j := 0 to 15 - i := j*8 - IF k[j] - dst[i+7:i] := a[7:0] - ELSE - dst[i+7:i] := 0 - FI +FOR j := 0 to 7 + i := j*32 + n := j*64 + dst[n+63:n] := Convert_Int32_To_FP64(v2[i+31:i]) ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + + Floating Point Integer - AVX512VL - AVX512BW - Set - - - Broadcast 8-bit integer "a" to all elements of "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + AVX512F/KNCNI + Convert + + + + + Performs element-by-element conversion of the lower half of 32-bit unsigned integer elements in "v2" to packed double-precision (64-bit) floating-point elements, storing the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*8 +FOR j := 0 to 7 + i := j*32 + l := j*64 IF k[j] - dst[i+7:i] := a[7:0] + dst[l+63:l] := Convert_Int32_To_FP64(v2[i+31:i]) ELSE - dst[i+7:i] := 0 + dst[l+63:l] := src[l+63:l] FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX512VL - AVX512F - Miscellaneous - - - - Broadcast the low packed 32-bit integer from "a" to all elements of "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + AVX512F/KNCNI + Load + + + + + + + Up-converts 16 memory locations starting at location "base_addr" at packed 32-bit integer indices stored in "vindex" scaled by "scale" using "conv" to 32-bit integer elements and stores them in "dst". AVX512 only supports _MM_UPCONV_EPI32_NONE. -FOR j := 0 to 7 +FOR j := 0 to 15 i := j*32 - IF k[j] - dst[i+31:i] := a[31:0] - ELSE - dst[i+31:i] := src[i+31:i] - FI + m := j*32 + addr := base_addr + SignExtend64(vindex[m+31:m]) * ZeroExtend64(scale) * 8 + + CASE conv OF + _MM_UPCONV_EPI32_NONE: dst[i+31:i] := MEM[addr+31:addr] + _MM_UPCONV_EPI32_UINT8: dst[i+31:i] := ZeroExtend32(MEM[addr+7:addr]) + _MM_UPCONV_EPI32_SINT8: dst[i+31:i] := SignExtend32(MEM[addr+7:addr]) + _MM_UPCONV_EPI32_UINT16: dst[i+31:i] := ZeroExtend32(MEM[addr+15:addr]) + _MM_UPCONV_EPI32_SINT16: dst[i+31:i] := SignExtend32(MEM[addr+15:addr]) + ESAC ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX512VL - AVX512F - Set - - - - Broadcast 32-bit integer "a" to all elements of "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + AVX512F/KNCNI + Load + + + + + + + + + Up-converts 16 single-precision (32-bit) memory locations starting at location "base_addr" at packed 32-bit integer indices stored in "vindex" scaled by "scale" using "conv" to 32-bit integer elements and stores them in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). AVX512 only supports _MM_UPCONV_EPI32_NONE. -FOR j := 0 to 7 +FOR j := 0 to 15 i := j*32 + m := j*32 + addr := base_addr + SignExtend64(vindex[m+31:m]) * ZeroExtend64(scale) * 8 + IF k[j] - dst[i+31:i] := a[31:0] + CASE conv OF + _MM_UPCONV_EPI32_NONE: dst[i+31:i] := MEM[addr+31:addr] + _MM_UPCONV_EPI32_UINT8: dst[i+31:i] := ZeroExtend32(MEM[addr+7:addr]) + _MM_UPCONV_EPI32_SINT8: dst[i+31:i] := SignExtend32(MEM[addr+7:addr]) + _MM_UPCONV_EPI32_UINT16: dst[i+31:i] := ZeroExtend32(MEM[addr+15:addr]) + _MM_UPCONV_EPI32_SINT16: dst[i+31:i] := SignExtend32(MEM[addr+15:addr]) + ESAC ELSE dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX512VL - AVX512F - Miscellaneous - - - Broadcast the low packed 32-bit integer from "a" to all elements of "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + AVX512F/KNCNI + Load + + + + + + + Up-converts 8 double-precision (64-bit) memory locations starting at location "base_addr" at packed 32-bit integer indices stored in the lower half of "vindex" scaled by "scale" using "conv" to 64-bit integer elements and stores them in "dst". FOR j := 0 to 7 - i := j*32 - IF k[j] - dst[i+31:i] := a[31:0] - ELSE - dst[i+31:i] := 0 - FI + i := j*64 + m := j*32 + addr := base_addr + SignExtend64(vindex[m+31:m]) * ZeroExtend64(scale) * 8 + + CASE conv OF + _MM_UPCONV_EPI64_NONE: dst[i+63:i] := MEM[addr+63:addr] + ESAC ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX512VL - AVX512F - Set - - - Broadcast 32-bit integer "a" to all elements of "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + AVX512F/KNCNI + Load + + + + + + + + + Up-converts 8 double-precision (64-bit) memory locations starting at location "base_addr" at packed 32-bit integer indices stored in the lower half of "vindex" scaled by "scale" using "conv" to 64-bit integer elements and stores them in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). FOR j := 0 to 7 - i := j*32 + i := j*64 + m := j*32 + addr := base_addr + SignExtend64(vindex[m+31:m]) * ZeroExtend64(scale) * 8 + IF k[j] - dst[i+31:i] := a[31:0] + CASE conv OF + _MM_UPCONV_EPI64_NONE: dst[i+63:i] := MEM[addr+63:addr] + ESAC ELSE - dst[i+31:i] := 0 + dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Miscellaneous - - - - Broadcast the low packed 32-bit integer from "a" to all elements of "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Floating Point + AVX512F/KNCNI + Load + + + + + + + Up-converts 16 memory locations starting at location "base_addr" at packed 32-bit integer indices stored in "vindex" scaled by "scale" using "conv" to single-precision (32-bit) floating-point elements and stores them in "dst". AVX512 only supports _MM_UPCONV_PS_NONE. -FOR j := 0 to 3 +FOR j := 0 to 15 i := j*32 - IF k[j] - dst[i+31:i] := a[31:0] - ELSE - dst[i+31:i] := src[i+31:i] - FI + m := j*32 + addr := base_addr + SignExtend64(vindex[m+31:m]) * ZeroExtend64(scale) * 8 + + CASE conv OF + _MM_UPCONV_PS_NONE: dst[i+31:i] := MEM[addr+31:addr] + _MM_UPCONV_PS_FLOAT16: dst[i+31:i] := Convert_FP16_To_FP32(MEM[addr+15:addr]) + _MM_UPCONV_PS_UINT8: dst[i+31:i] := Convert_UInt8_To_FP32(MEM[addr+7:addr]) + _MM_UPCONV_PS_SINT8: dst[i+31:i] := Convert_Int8_To_FP32(MEM[addr+7:addr]) + _MM_UPCONV_PS_UINT16: dst[i+31:i] := Convert_UInt16_To_FP32(MEM[addr+15:addr]) + _MM_UPCONV_PS_SINT16: dst[i+31:i] := Convert_Int16_To_FP32(MEM[addr+15:addr]) + ESAC ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Set - - - - Broadcast 32-bit integer "a" to all elements of "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Floating Point + AVX512F/KNCNI + Load + + + + + + + + + Up-converts 16 single-precision (32-bit) memory locations starting at location "base_addr" at packed 32-bit integer indices stored in "vindex" scaled by "scale" using "conv" to single-precision (32-bit) floating-point elements and stores them in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). AVX512 only supports _MM_UPCONV_PS_NONE. -FOR j := 0 to 3 +FOR j := 0 to 15 i := j*32 + m := j*32 + addr := base_addr + SignExtend64(vindex[m+31:m]) * ZeroExtend64(scale) * 8 + IF k[j] - dst[i+31:i] := a[31:0] + CASE conv OF + _MM_UPCONV_PS_NONE: dst[i+31:i] := MEM[addr+31:addr] + _MM_UPCONV_PS_FLOAT16: dst[i+31:i] := Convert_FP16_To_FP32(MEM[addr+15:addr]) + _MM_UPCONV_PS_UINT8: dst[i+31:i] := Convert_UInt8_To_FP32(MEM[addr+7:addr]) + _MM_UPCONV_PS_SINT8: dst[i+31:i] := Convert_Int8_To_FP32(MEM[addr+7:addr]) + _MM_UPCONV_PS_UINT16: dst[i+31:i] := Convert_UInt16_To_FP32(MEM[addr+15:addr]) + _MM_UPCONV_PS_SINT16: dst[i+31:i] := Convert_Int16_To_FP32(MEM[addr+15:addr]) + ESAC ELSE dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Miscellaneous - - - Broadcast the low packed 32-bit integer from "a" to all elements of "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Floating Point + AVX512F/KNCNI + Load + + + + + + + Up-converts 8 double-precision (64-bit) floating-point elements in memory locations starting at location "base_addr" at packed 32-bit integer indices stored in the lower half of "vindex" scaled by "scale" using "conv" to 64-bit floating-point elements and stores them in "dst". -FOR j := 0 to 3 - i := j*32 - IF k[j] - dst[i+31:i] := a[31:0] - ELSE - dst[i+31:i] := 0 - FI +FOR j := 0 to 7 + i := j*64 + m := j*32 + addr := base_addr + SignExtend64(vindex[m+31:m]) * ZeroExtend64(scale) * 8 + + CASE conv OF + _MM_UPCONV_PD_NONE: dst[i+63:i] := MEM[addr+63:addr] + ESAC ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Set - - - Broadcast 32-bit integer "a" to all elements of "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Floating Point + AVX512F/KNCNI + Load + + + + + + + + + Up-converts 8 double-precision (64-bit) floating-point elements in memory locations starting at location "base_addr" at packed 32-bit integer indices stored in the lower half of "vindex" scaled by "scale" using "conv" to 64-bit floating-point elements and stores them in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*32 +FOR j := 0 to 7 + i := j*64 + m := j*32 + addr := base_addr + SignExtend64(vindex[m+31:m]) * ZeroExtend64(scale) * 8 + IF k[j] - dst[i+31:i] := a[31:0] + CASE conv OF + _MM_UPCONV_PD_NONE: + dst[i+63:i] := MEM[addr+63:addr] + ESAC ELSE - dst[i+31:i] := 0 + dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512CD - Miscellaneous - - Broadcast the low 8-bits from input mask "k" to all 64-bit elements of "dst". + + Floating Point + AVX512F/KNCNI + Store + + + + + + + + Down-converts 16 packed single-precision (32-bit) floating-point elements in "a" and stores them in memory locations starting at location "base_addr" at packed 32-bit integer indices stored in "vindex" scaled by "scale" using "conv". AVX512 only supports _MM_DOWNCONV_PS_NONE. -FOR j := 0 to 3 - i := j*64 - dst[i+63:i] := ZeroExtend(k[7:0]) +FOR j := 0 to 15 + i := j*32 + m := j*32 + addr := base_addr + SignExtend64(vindex[m+31:m]) * ZeroExtend64(scale) * 8 + + CASE conv OF + _MM_DOWNCONV_PS_NONE: MEM[addr+31:addr] := a[i+31:i] + _MM_DOWNCONV_PS_FLOAT16: MEM[addr+15:addr] := Convert_FP32_To_FP16(a[i+31:i]) + _MM_DOWNCONV_PS_UINT8: MEM[addr+ 7:addr] := Convert_FP32_To_UInt8(a[i+31:i]) + _MM_DOWNCONV_PS_SINT8: MEM[addr+ 7:addr] := Convert_FP32_To_Int8(a[i+31:i]) + _MM_DOWNCONV_PS_UINT16: MEM[addr+15:addr] := Convert_FP32_To_UInt16(a[i+31:i]) + _MM_DOWNCONV_PS_SINT16: MEM[addr+15:addr] := Convert_FP32_To_Int16(a[i+31:i]) + ESAC ENDFOR -dst[MAX:256] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512CD - Miscellaneous - - Broadcast the low 8-bits from input mask "k" to all 64-bit elements of "dst". + + Floating Point + AVX512F/KNCNI + Store + + + + + + + + + Down-converts 16 packed single-precision (32-bit) floating-point elements in "a" according to "conv" and stores them in memory locations starting at location "base_addr" at packed 32-bit integer indices stored in "vindex" scaled by "scale" using writemask "k" (elements are written only when the corresponding mask bit is not set). AVX512 only supports _MM_DOWNCONV_PS_NONE. -FOR j := 0 to 1 - i := j*64 - dst[i+63:i] := ZeroExtend(k[7:0]) +FOR j := 0 to 15 + i := j*32 + m := j*32 + addr := base_addr + SignExtend64(vindex[m+31:m]) * ZeroExtend64(scale) * 8 + + IF k[j] + CASE conv OF + _MM_DOWNCONV_PS_NONE: MEM[addr+31:addr] := a[i+31:i] + _MM_DOWNCONV_PS_FLOAT16: MEM[addr+15:addr] := Convert_FP32_To_FP16(a[i+31:i]) + _MM_DOWNCONV_PS_UINT8: MEM[addr+ 7:addr] := Convert_FP32_To_UInt8(a[i+31:i]) + _MM_DOWNCONV_PS_SINT8: MEM[addr+ 7:addr] := Convert_FP32_To_Int8(a[i+31:i]) + _MM_DOWNCONV_PS_UINT16: MEM[addr+15:addr] := Convert_FP32_To_UInt16(a[i+31:i]) + _MM_DOWNCONV_PS_SINT16: MEM[addr+15:addr] := Convert_FP32_To_Int16(a[i+31:i]) + ESAC + FI ENDFOR -dst[MAX:128] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512CD - Miscellaneous - - Broadcast the low 16-bits from input mask "k" to all 32-bit elements of "dst". + + Floating Point + AVX512F/KNCNI + Store + + + + + + + + Down-converts 8 packed double-precision (64-bit) floating-point elements in "a" and stores them in memory locations starting at location "base_addr" at packed 32-bit integer indices stored in "vindex" scaled by "scale" using "conv". FOR j := 0 to 7 - i := j*32 - dst[i+31:i] := ZeroExtend(k[15:0]) + i := j*64 + m := j*32 + addr := base_addr + SignExtend64(vindex[m+31:m]) * ZeroExtend64(scale) * 8 + + CASE conv OF + _MM_DOWNCONV_PD_NONE: MEM[addr+63:addr] := a[i+63:i] + ESAC ENDFOR -dst[MAX:256] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512CD - Miscellaneous - - Broadcast the low 16-bits from input mask "k" to all 32-bit elements of "dst". + + Floating Point + AVX512F/KNCNI + Store + + + + + + + + + Down-converts 8 packed double-precision (64-bit) floating-point elements in "a" and stores them in memory locations starting at location "base_addr" at packed 32-bit integer indices stored in "vindex" scaled by "scale" using "conv". Only those elements whose corresponding mask bit is set in writemask "k" are written to memory. -FOR j := 0 to 3 - i := j*32 - dst[i+31:i] := ZeroExtend(k[15:0]) +FOR j := 0 to 7 + i := j*64 + m := j*32 + addr := base_addr + SignExtend64(vindex[m+31:m]) * ZeroExtend64(scale) * 8 + + IF k[j] + CASE conv OF + _MM_DOWNCONV_PD_NONE: MEM[addr+63:addr] := a[i+63:i] + ESAC + FI ENDFOR -dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - AVX512VL - AVX512F - Miscellaneous - - - - Broadcast the low packed 64-bit integer from "a" to all elements of "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + AVX512F/KNCNI + Store + + + + + + + + Down-converts 8 packed 64-bit integer elements in "a" and stores them in memory locations starting at location "base_addr" at packed 32-bit integer indices stored in "vindex" scaled by "scale" using "conv". -FOR j := 0 to 3 +FOR j := 0 to 7 i := j*64 - IF k[j] - dst[i+63:i] := a[63:0] - ELSE - dst[i+63:i] := src[i+63:i] - FI + m := j*32 + addr := base_addr + SignExtend64(vindex[m+31:m]) * ZeroExtend64(scale) * 8 + + CASE conv OF + _MM_DOWNCONV_EPI64_NONE: MEM[addr+63:addr] := a[i+63:i] + ESAC ENDFOR -dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - AVX512VL - AVX512F - Set - - - - Broadcast 64-bit integer "a" to all elements of "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + AVX512F/KNCNI + Store + + + + + + + + + Down-converts 8 packed 64-bit integer elements in "a" and stores them in memory locations starting at location "base_addr" at packed 32-bit integer indices stored in "vindex" scaled by "scale" using "conv". Only those elements whose corresponding mask bit is set in writemask "k" are written to memory. -FOR j := 0 to 3 +FOR j := 0 to 7 i := j*64 + m := j*32 + addr := base_addr + SignExtend64(vindex[m+31:m]) * ZeroExtend64(scale) * 8 + IF k[j] - dst[i+63:i] := a[63:0] - ELSE - dst[i+63:i] := src[i+63:i] + CASE conv OF + _MM_DOWNCONV_EPI64_NONE: MEM[addr+63:addr] := a[i+63:i] + ESAC FI ENDFOR -dst[MAX:256] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Miscellaneous - - - Broadcast the low packed 64-bit integer from "a" to all elements of "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Floating Point + AVX512F/KNCNI + Convert + + + Performs an element-by-element conversion of packed double-precision (64-bit) floating-point elements in "v2" to single-precision (32-bit) floating-point elements and stores them in "dst". The elements are stored in the lower half of the results vector, while the remaining upper half locations are set to 0. -FOR j := 0 to 3 +FOR j := 0 to 7 i := j*64 - IF k[j] - dst[i+63:i] := a[63:0] - ELSE - dst[i+63:i] := 0 - FI + k := j*32 + dst[k+31:k] := Convert_FP64_To_FP32(v2[i+63:i]) ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Set - - - Broadcast 64-bit integer "a" to all elements of "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Floating Point + AVX512F/KNCNI + Convert + + + + + Performs an element-by-element conversion of packed double-precision (64-bit) floating-point elements in "v2" to single-precision (32-bit) floating-point elements and stores them in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). The elements are stored in the lower half of the results vector, while the remaining upper half locations are set to 0. -FOR j := 0 to 3 +FOR j := 0 to 7 i := j*64 + l := j*32 IF k[j] - dst[i+63:i] := a[63:0] + dst[l+31:l] := Convert_FP64_To_FP32(v2[i+63:i]) ELSE - dst[i+63:i] := 0 + dst[l+31:l] := src[l+31:l] FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX512VL - AVX512F - Miscellaneous - - - - Broadcast the low packed 64-bit integer from "a" to all elements of "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + AVX512F/KNCNI + Load + + + + + Loads 8 64-bit integer elements from memory starting at location "base_addr" at packed 32-bit integer indices stored in the lower half of "vindex" scaled by "scale" and stores them in "dst". -FOR j := 0 to 1 +FOR j := 0 to 7 i := j*64 - IF k[j] - dst[i+63:i] := a[63:0] - ELSE - dst[i+63:i] := src[i+63:i] - FI + m := j*32 + addr := base_addr + SignExtend64(vindex[m+31:m]) * ZeroExtend64(scale) * 8 + dst[i+63:i] := MEM[addr+63:addr] ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX512VL - AVX512F - Set - - - - Broadcast 64-bit integer "a" to all elements of "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + AVX512F/KNCNI + Load + + + + + + + Loads 8 64-bit integer elements from memory starting at location "base_addr" at packed 32-bit integer indices stored in the lower half of "vindex" scaled by "scale" and stores them in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 1 +FOR j := 0 to 7 i := j*64 + m := j*32 IF k[j] - dst[i+63:i] := a[63:0] + addr := base_addr + SignExtend64(vindex[m+31:m]) * ZeroExtend64(scale) * 8 + dst[i+63:i] := MEM[addr+63:addr] ELSE dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Miscellaneous - - - Broadcast the low packed 64-bit integer from "a" to all elements of "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Floating Point + AVX512F/KNCNI + Load + + + + + Loads 8 double-precision (64-bit) floating-point elements stored at memory locations starting at location "base_addr" at packed 32-bit integer indices stored in the lower half of "vindex" scaled by "scale" them in "dst". -FOR j := 0 to 1 +FOR j := 0 to 7 i := j*64 - IF k[j] - dst[i+63:i] := a[63:0] - ELSE - dst[i+63:i] := 0 - FI + m := j*32 + addr := base_addr + SignExtend64(vindex[m+31:m]) * ZeroExtend64(scale) * 8 + dst[i+63:i] := MEM[addr+63:addr] ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Set - - - Broadcast 64-bit integer "a" to all elements of "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Floating Point + AVX512F/KNCNI + Load + + + + + + + Loads 8 double-precision (64-bit) floating-point elements from memory starting at location "base_addr" at packed 32-bit integer indices stored in the lower half of "vindex" scaled by "scale" into "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 1 +FOR j := 0 to 7 i := j*64 + m := j*32 IF k[j] - dst[i+63:i] := a[63:0] + addr := base_addr + SignExtend64(vindex[m+31:m]) * ZeroExtend64(scale) * 8 + dst[i+63:i] := MEM[addr+63:addr] ELSE - dst[i+63:i] := 0 + dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512BW - Miscellaneous - - - - Broadcast the low packed 16-bit integer from "a" to all elements of "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Floating Point + AVX512F/KNCNI + Store + + + + + + Stores 8 packed double-precision (64-bit) floating-point elements in "a" and to memory locations starting at location "base_addr" at packed 32-bit integer indices stored in "vindex" scaled by "scale". -FOR j := 0 to 15 - i := j*16 - IF k[j] - dst[i+15:i] := a[15:0] - ELSE - dst[i+15:i] := src[i+15:i] - FI +FOR j := 0 to 7 + i := j*64 + m := j*32 + addr := base_addr + SignExtend64(vindex[m+31:m]) * ZeroExtend64(scale) * 8 + MEM[addr+63:addr] := a[i+63:i] ENDFOR -dst[MAX:256] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512BW - Set - - - - Broadcast the low packed 16-bit integer from "a" to all elements of "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Floating Point + AVX512F/KNCNI + Store + + + + + + + Stores 8 packed double-precision (64-bit) floating-point elements in "a" to memory locations starting at location "base_addr" at packed 32-bit integer indices stored in "vindex" scaled by "scale". Only those elements whose corresponding mask bit is set in writemask "k" are written to memory. -FOR j := 0 to 15 - i := j*16 +FOR j := 0 to 7 + i := j*64 + m := j*32 IF k[j] - dst[i+15:i] := a[15:0] - ELSE - dst[i+15:i] := src[i+15:i] + addr := base_addr + SignExtend64(vindex[m+31:m]) * ZeroExtend64(scale) * 8 + MEM[addr+63:addr] := a[i+63:i] FI ENDFOR -dst[MAX:256] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512BW - Miscellaneous - - - Broadcast the low packed 16-bit integer from "a" to all elements of "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Floating Point + AVX512F/KNCNI + Arithmetic + + + Finds the absolute value of each packed single-precision (32-bit) floating-point element in "v2", storing the results in "dst". FOR j := 0 to 15 - i := j*16 - IF k[j] - dst[i+15:i] := a[15:0] - ELSE - dst[i+15:i] := 0 - FI + i := j*32 + dst[i+31:i] := ABS(v2[i+31:i]) ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512BW - Set - - - Broadcast 16-bit integer "a" to all elements of "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - + + Floating Point + AVX512F/KNCNI + Arithmetic + + + + + Finds the absolute value of each packed single-precision (32-bit) floating-point element in "v2", storing the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). FOR j := 0 to 15 - i := j*16 + i := j*32 IF k[j] - dst[i+15:i] := a[15:0] + dst[i+31:i] := ABS(v2[i+31:i]) ELSE - dst[i+15:i] := 0 + dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - AVX512BW - Miscellaneous - - Broadcast the low packed 16-bit integer from "a" to all elements of "dst". + + Floating Point + AVX512F/KNCNI + Arithmetic + + + Finds the absolute value of each packed double-precision (64-bit) floating-point element in "v2", storing the results in "dst". -FOR j := 0 to 31 - i := j*16 - dst[i+15:i] := a[15:0] +FOR j := 0 to 7 + i := j*64 + dst[i+63:i] := ABS(v2[i+63:i]) ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - AVX512BW - Miscellaneous - - - - Broadcast the low packed 16-bit integer from "a" to all elements of "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Floating Point + AVX512F/KNCNI + Arithmetic + + + + + Finds the absolute value of each packed double-precision (64-bit) floating-point element in "v2", storing the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 31 - i := j*16 +FOR j := 0 to 7 + i := j*64 IF k[j] - dst[i+15:i] := a[15:0] + dst[i+63:i] := ABS(v2[i+63:i]) ELSE - dst[i+15:i] := src[i+15:i] + dst[i+63:i] := src[i+63:i] FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - AVX512BW - Set - - - - Broadcast 16-bit integer "a" to all elements of "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Floating Point + AVX512F/KNCNI + Elementary Math Functions + + + Compute the base-2 logarithm of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst". -FOR j := 0 to 31 - i := j*16 - IF k[j] - dst[i+15:i] := a[15:0] - ELSE - dst[i+15:i] := src[i+15:i] - FI +FOR j := 0 to 15 + i := j*32 + dst[i+31:i] := LOG(a[i+31:i]) / LOG(2.0) ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - AVX512BW - Miscellaneous - - - Broadcast the low packed 16-bit integer from "a" to all elements of "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Floating Point + AVX512F/KNCNI + Elementary Math Functions + + + + + Compute the base-2 logarithm of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 31 - i := j*16 +FOR j := 0 to 15 + i := j*32 IF k[j] - dst[i+15:i] := a[15:0] + dst[i+31:i] := LOG(a[i+31:i]) / LOG(2.0) ELSE - dst[i+15:i] := 0 + dst[i+31:i] := src[i+31:i] FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX512BW - Set - - - Broadcast the low packed 16-bit integer from "a" to all elements of "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - + AVX512F/KNCNI + Store + + + + + + + + Down-converts 16 packed 32-bit integer elements in "a" using "conv" and stores them in memory locations starting at location "base_addr" at packed 32-bit integer indices stored in "vindex" scaled by "scale". "hint" indicates to the processor whether the data is non-temporal. AVX512 only supports _MM_DOWNCONV_EPI32_NONE. -FOR j := 0 to 31 - i := j*16 - IF k[j] - dst[i+15:i] := a[15:0] - ELSE - dst[i+15:i] := 0 - FI +FOR j := 0 to 15 + i := j*32 + m := j*32 + addr := base_addr + SignExtend64(vindex[m+31:m]) * ZeroExtend64(scale) * 8 + + CASE conv OF + _MM_DOWNCONV_EPI32_NONE: MEM[addr+31:addr] := a[i+31:i] + _MM_DOWNCONV_EPI32_UINT8: MEM[addr+ 7:addr] := Truncate8(a[i+31:i]) + _MM_DOWNCONV_EPI32_SINT8: MEM[addr+ 7:addr] := Saturate8(a[i+31:i]) + _MM_DOWNCONV_EPI32_UINT16: MEM[addr+15:addr] := Truncate16(a[i+31:i]) + _MM_DOWNCONV_EPI32_SINT16: MEM[addr+15:addr] := Saturate16(a[i+15:i]) + ESAC ENDFOR -dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX512VL - AVX512BW - Miscellaneous - - - - Broadcast the low packed 16-bit integer from "a" to all elements of "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + AVX512F/KNCNI + Store + + + + + + + + + Down-converts 16 packed 32-bit integer elements in "a" using "conv" and stores them in memory locations starting at location "base_addr" at packed 32-bit integer indices stored in "vindex" scaled by "scale". Elements are written using writemask "k" (elements are only written when the corresponding mask bit is set; otherwise, elements are left unchanged in memory). "hint" indicates to the processor whether the data is non-temporal. AVX512 only supports _MM_DOWNCONV_EPI32_NONE. -FOR j := 0 to 7 - i := j*16 +FOR j := 0 to 15 + i := j*32 + m := j*32 + addr := base_addr + SignExtend64(vindex[m+31:m]) * ZeroExtend64(scale) * 8 + IF k[j] - dst[i+15:i] := a[15:0] - ELSE - dst[i+15:i] := src[i+15:i] - FI + CASE conv OF + _MM_DOWNCONV_EPI32_NONE: MEM[addr+31:addr] := a[i+31:i] + _MM_DOWNCONV_EPI32_UINT8: MEM[addr+ 7:addr] := Truncate8(a[i+31:i]) + _MM_DOWNCONV_EPI32_SINT8: MEM[addr+ 7:addr] := Saturate8(a[i+31:i]) + _MM_DOWNCONV_EPI32_UINT16: MEM[addr+15:addr] := Truncate16(a[i+31:i]) + _MM_DOWNCONV_EPI32_SINT16: MEM[addr+15:addr] := Saturate16(a[i+15:i]) + ESAC + FI ENDFOR -dst[MAX:128] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512BW - Set - - - - Broadcast the low packed 16-bit integer from "a" to all elements of "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + AVX512IFMA52 + Arithmetic + + + + + Multiply packed unsigned 52-bit integers in each 64-bit element of "b" and "c" to form a 104-bit intermediate result. Add the low 52-bit unsigned integer from the intermediate result with the corresponding unsigned 64-bit integer in "a", and store the results in "dst". FOR j := 0 to 7 - i := j*16 - IF k[j] - dst[i+15:i] := a[15:0] - ELSE - dst[i+15:i] := src[i+15:i] - FI + i := j*64 + tmp[127:0] := ZeroExtend64(b[i+51:i]) * ZeroExtend64(c[i+51:i]) + dst[i+63:i] := a[i+63:i] + ZeroExtend64(tmp[51:0]) ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512BW - Miscellaneous - - - Broadcast the low packed 16-bit integer from "a" to all elements of "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + AVX512IFMA52 + Arithmetic + + + + + + Multiply packed unsigned 52-bit integers in each 64-bit element of "b" and "c" to form a 104-bit intermediate result. Add the low 52-bit unsigned integer from the intermediate result with the corresponding unsigned 64-bit integer in "a", and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). FOR j := 0 to 7 - i := j*16 + i := j*64 IF k[j] - dst[i+15:i] := a[15:0] + tmp[127:0] := ZeroExtend64(b[i+51:i]) * ZeroExtend64(c[i+51:i]) + dst[i+63:i] := a[i+63:i] + ZeroExtend64(tmp[51:0]) ELSE - dst[i+15:i] := 0 + dst[i+63:i] := a[i+63:i] FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512BW - Set - - - Broadcast the low packed 16-bit integer from "a" to all elements of "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - + + AVX512IFMA52 + Arithmetic + + + + + + Multiply packed unsigned 52-bit integers in each 64-bit element of "b" and "c" to form a 104-bit intermediate result. Add the low 52-bit unsigned integer from the intermediate result with the corresponding unsigned 64-bit integer in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). FOR j := 0 to 7 - i := j*16 + i := j*64 IF k[j] - dst[i+15:i] := a[15:0] + tmp[127:0] := ZeroExtend64(b[i+51:i]) * ZeroExtend64(c[i+51:i]) + dst[i+63:i] := a[i+63:i] + ZeroExtend64(tmp[51:0]) ELSE - dst[i+15:i] := 0 + dst[i+63:i] := 0 FI ENDFOR -dst[MAX:128] := 0 - - -
immintrin.h
- - - Integer - Mask - AVX512VL - AVX512BW - Compare - - - - Compare packed 8-bit integers in "a" and "b" based on the comparison operand specified by "imm8", and store the results in mask vector "k". - -CASE (imm8[7:0]) OF -0: OP := _MM_CMPINT_EQ -1: OP := _MM_CMPINT_LT -2: OP := _MM_CMPINT_LE -3: OP := _MM_CMPINT_FALSE -4: OP := _MM_CMPINT_NE -5: OP := _MM_CMPINT_NLT -6: OP := _MM_CMPINT_NLE -7: OP := _MM_CMPINT_TRUE -ESAC -FOR j := 0 to 31 - i := j*8 - k[j] := ( a[i+7:i] OP b[i+7:i] ) ? 1 : 0 -ENDFOR -k[MAX:32] := 0 - - -
immintrin.h
- - - Integer - Mask - AVX512VL - AVX512BW - Compare - - - Compare packed 8-bit integers in "a" and "b" for equality, and store the results in mask vector "k". - -FOR j := 0 to 31 - i := j*8 - k[j] := ( a[i+7:i] == b[i+7:i] ) ? 1 : 0 -ENDFOR -k[MAX:32] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - Mask + + AVX512IFMA52 AVX512VL - AVX512BW - Compare - - - Compare packed 8-bit integers in "a" and "b" for greater-than-or-equal, and store the results in mask vector "k". + Arithmetic + + + + + Multiply packed unsigned 52-bit integers in each 64-bit element of "b" and "c" to form a 104-bit intermediate result. Add the low 52-bit unsigned integer from the intermediate result with the corresponding unsigned 64-bit integer in "a", and store the results in "dst". -FOR j := 0 to 31 - i := j*8 - k[j] := ( a[i+7:i] >= b[i+7:i] ) ? 1 : 0 +FOR j := 0 to 3 + i := j*64 + tmp[127:0] := ZeroExtend64(b[i+51:i]) * ZeroExtend64(c[i+51:i]) + dst[i+63:i] := a[i+63:i] + ZeroExtend64(tmp[51:0]) ENDFOR -k[MAX:32] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - - Integer - Mask + + AVX512IFMA52 AVX512VL - AVX512BW - Compare - - - Compare packed 8-bit integers in "a" and "b" for greater-than, and store the results in mask vector "k". + Arithmetic + + + + + + Multiply packed unsigned 52-bit integers in each 64-bit element of "b" and "c" to form a 104-bit intermediate result. Add the low 52-bit unsigned integer from the intermediate result with the corresponding unsigned 64-bit integer in "a", and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). -FOR j := 0 to 31 - i := j*8 - k[j] := ( a[i+7:i] > b[i+7:i] ) ? 1 : 0 +FOR j := 0 to 3 + i := j*64 + IF k[j] + tmp[127:0] := ZeroExtend64(b[i+51:i]) * ZeroExtend64(c[i+51:i]) + dst[i+63:i] := a[i+63:i] + ZeroExtend64(tmp[51:0]) + ELSE + dst[i+63:i] := a[i+63:i] + FI ENDFOR -k[MAX:32] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - - Integer - Mask + + AVX512IFMA52 AVX512VL - AVX512BW - Compare - - - Compare packed 8-bit integers in "a" and "b" for less-than-or-equal, and store the results in mask vector "k". + Arithmetic + + + + + + Multiply packed unsigned 52-bit integers in each 64-bit element of "b" and "c" to form a 104-bit intermediate result. Add the low 52-bit unsigned integer from the intermediate result with the corresponding unsigned 64-bit integer in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 31 - i := j*8 - k[j] := ( a[i+7:i] <= b[i+7:i] ) ? 1 : 0 +FOR j := 0 to 3 + i := j*64 + IF k[j] + tmp[127:0] := ZeroExtend64(b[i+51:i]) * ZeroExtend64(c[i+51:i]) + dst[i+63:i] := a[i+63:i] + ZeroExtend64(tmp[51:0]) + ELSE + dst[i+63:i] := 0 + FI ENDFOR -k[MAX:32] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - - Integer - Mask + + AVX512IFMA52 AVX512VL - AVX512BW - Compare - - - Compare packed 8-bit integers in "a" and "b" for less-than, and store the results in mask vector "k". + Arithmetic + + + + + Multiply packed unsigned 52-bit integers in each 64-bit element of "b" and "c" to form a 104-bit intermediate result. Add the low 52-bit unsigned integer from the intermediate result with the corresponding unsigned 64-bit integer in "a", and store the results in "dst". -FOR j := 0 to 31 - i := j*8 - k[j] := ( a[i+7:i] < b[i+7:i] ) ? 1 : 0 +FOR j := 0 to 1 + i := j*64 + tmp[127:0] := ZeroExtend64(b[i+51:i]) * ZeroExtend64(c[i+51:i]) + dst[i+63:i] := a[i+63:i] + ZeroExtend64(tmp[51:0]) ENDFOR -k[MAX:32] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - - Integer - Mask + + AVX512IFMA52 AVX512VL - AVX512BW - Compare - - - Compare packed 8-bit integers in "a" and "b" for not-equal, and store the results in mask vector "k". + Arithmetic + + + + + + Multiply packed unsigned 52-bit integers in each 64-bit element of "b" and "c" to form a 104-bit intermediate result. Add the low 52-bit unsigned integer from the intermediate result with the corresponding unsigned 64-bit integer in "a", and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). -FOR j := 0 to 31 - i := j*8 - k[j] := ( a[i+7:i] != b[i+7:i] ) ? 1 : 0 +FOR j := 0 to 1 + i := j*64 + IF k[j] + tmp[127:0] := ZeroExtend64(b[i+51:i]) * ZeroExtend64(c[i+51:i]) + dst[i+63:i] := a[i+63:i] + ZeroExtend64(tmp[51:0]) + ELSE + dst[i+63:i] := a[i+63:i] + FI ENDFOR -k[MAX:32] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - - Integer - Mask + + AVX512IFMA52 AVX512VL - AVX512BW - Compare - - - - - Compare packed 8-bit integers in "a" and "b" based on the comparison operand specified by "imm8", and store the results in mask vector "k1" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Arithmetic + + + + + + Multiply packed unsigned 52-bit integers in each 64-bit element of "b" and "c" to form a 104-bit intermediate result. Add the low 52-bit unsigned integer from the intermediate result with the corresponding unsigned 64-bit integer in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -CASE (imm8[7:0]) OF -0: OP := _MM_CMPINT_EQ -1: OP := _MM_CMPINT_LT -2: OP := _MM_CMPINT_LE -3: OP := _MM_CMPINT_FALSE -4: OP := _MM_CMPINT_NE -5: OP := _MM_CMPINT_NLT -6: OP := _MM_CMPINT_NLE -7: OP := _MM_CMPINT_TRUE -ESAC -FOR j := 0 to 31 - i := j*8 - IF k1[j] - k[j] := ( a[i+7:i] OP b[i+7:i] ) ? 1 : 0 +FOR j := 0 to 1 + i := j*64 + IF k[j] + tmp[127:0] := ZeroExtend64(b[i+51:i]) * ZeroExtend64(c[i+51:i]) + dst[i+63:i] := a[i+63:i] + ZeroExtend64(tmp[51:0]) ELSE - k[j] := 0 + dst[i+63:i] := 0 FI ENDFOR -k[MAX:32] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - - Integer - Mask - AVX512VL - AVX512BW - Compare - - - - Compare packed 8-bit integers in "a" and "b" for equality, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). + + AVX512IFMA52 + Arithmetic + + + + + Multiply packed unsigned 52-bit integers in each 64-bit element of "b" and "c" to form a 104-bit intermediate result. Add the high 52-bit unsigned integer from the intermediate result with the corresponding unsigned 64-bit integer in "a", and store the results in "dst". -FOR j := 0 to 31 - i := j*8 - IF k1[j] - k[j] := ( a[i+7:i] == b[i+7:i] ) ? 1 : 0 +FOR j := 0 to 7 + i := j*64 + tmp[127:0] := ZeroExtend64(b[i+51:i]) * ZeroExtend64(c[i+51:i]) + dst[i+63:i] := a[i+63:i] + ZeroExtend64(tmp[103:52]) +ENDFOR +dst[MAX:512] := 0 + + +
immintrin.h
+ + + AVX512IFMA52 + Arithmetic + + + + + + Multiply packed unsigned 52-bit integers in each 64-bit element of "b" and "c" to form a 104-bit intermediate result. Add the high 52-bit unsigned integer from the intermediate result with the corresponding unsigned 64-bit integer in "a", and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). + +FOR j := 0 to 7 + i := j*64 + IF k[j] + tmp[127:0] := ZeroExtend64(b[i+51:i]) * ZeroExtend64(c[i+51:i]) + dst[i+63:i] := a[i+63:i] + ZeroExtend64(tmp[103:52]) ELSE - k[j] := 0 + dst[i+63:i] := a[i+63:i] FI ENDFOR -k[MAX:32] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - Mask - AVX512VL - AVX512BW - Compare - - - - Compare packed 8-bit integers in "a" and "b" for greater-than-or-equal, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). + + AVX512IFMA52 + Arithmetic + + + + + + Multiply packed unsigned 52-bit integers in each 64-bit element of "b" and "c" to form a 104-bit intermediate result. Add the high 52-bit unsigned integer from the intermediate result with the corresponding unsigned 64-bit integer in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 31 - i := j*8 - IF k1[j] - k[j] := ( a[i+7:i] >= b[i+7:i] ) ? 1 : 0 +FOR j := 0 to 7 + i := j*64 + IF k[j] + tmp[127:0] := ZeroExtend64(b[i+51:i]) * ZeroExtend64(c[i+51:i]) + dst[i+63:i] := a[i+63:i] + ZeroExtend64(tmp[103:52]) ELSE - k[j] := 0 + dst[i+63:i] := 0 FI ENDFOR -k[MAX:32] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - Mask + + AVX512IFMA52 AVX512VL - AVX512BW - Compare - - - - Compare packed 8-bit integers in "a" and "b" for greater-than, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). + Arithmetic + + + + + Multiply packed unsigned 52-bit integers in each 64-bit element of "b" and "c" to form a 104-bit intermediate result. Add the high 52-bit unsigned integer from the intermediate result with the corresponding unsigned 64-bit integer in "a", and store the results in "dst". -FOR j := 0 to 31 - i := j*8 - IF k1[j] - k[j] := ( a[i+7:i] > b[i+7:i] ) ? 1 : 0 - ELSE - k[j] := 0 - FI +FOR j := 0 to 3 + i := j*64 + tmp[127:0] := ZeroExtend64(b[i+51:i]) * ZeroExtend64(c[i+51:i]) + dst[i+63:i] := a[i+63:i] + ZeroExtend64(tmp[103:52]) ENDFOR -k[MAX:32] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - - Integer - Mask + + AVX512IFMA52 AVX512VL - AVX512BW - Compare - - - - Compare packed 8-bit integers in "a" and "b" for less-than-or-equal, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). + Arithmetic + + + + + + Multiply packed unsigned 52-bit integers in each 64-bit element of "b" and "c" to form a 104-bit intermediate result. Add the high 52-bit unsigned integer from the intermediate result with the corresponding unsigned 64-bit integer in "a", and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). -FOR j := 0 to 31 - i := j*8 - IF k1[j] - k[j] := ( a[i+7:i] <= b[i+7:i] ) ? 1 : 0 +FOR j := 0 to 3 + i := j*64 + IF k[j] + tmp[127:0] := ZeroExtend64(b[i+51:i]) * ZeroExtend64(c[i+51:i]) + dst[i+63:i] := a[i+63:i] + ZeroExtend64(tmp[103:52]) ELSE - k[j] := 0 + dst[i+63:i] := a[i+63:i] FI ENDFOR -k[MAX:32] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - - Integer - Mask + + AVX512IFMA52 AVX512VL - AVX512BW - Compare - - - - Compare packed 8-bit integers in "a" and "b" for less-than, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). + Arithmetic + + + + + + Multiply packed unsigned 52-bit integers in each 64-bit element of "b" and "c" to form a 104-bit intermediate result. Add the high 52-bit unsigned integer from the intermediate result with the corresponding unsigned 64-bit integer in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 31 - i := j*8 - IF k1[j] - k[j] := ( a[i+7:i] < b[i+7:i] ) ? 1 : 0 +FOR j := 0 to 3 + i := j*64 + IF k[j] + tmp[127:0] := ZeroExtend64(b[i+51:i]) * ZeroExtend64(c[i+51:i]) + dst[i+63:i] := a[i+63:i] + ZeroExtend64(tmp[103:52]) ELSE - k[j] := 0 + dst[i+63:i] := 0 FI ENDFOR -k[MAX:32] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - - Integer - Mask + + AVX512IFMA52 AVX512VL - AVX512BW - Compare - - - - Compare packed 8-bit integers in "a" and "b" for not-equal, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). + Arithmetic + + + + + Multiply packed unsigned 52-bit integers in each 64-bit element of "b" and "c" to form a 104-bit intermediate result. Add the high 52-bit unsigned integer from the intermediate result with the corresponding unsigned 64-bit integer in "a", and store the results in "dst". -FOR j := 0 to 31 - i := j*8 - IF k1[j] - k[j] := ( a[i+7:i] != b[i+7:i] ) ? 1 : 0 - ELSE - k[j] := 0 - FI +FOR j := 0 to 1 + i := j*64 + tmp[127:0] := ZeroExtend64(b[i+51:i]) * ZeroExtend64(c[i+51:i]) + dst[i+63:i] := a[i+63:i] + ZeroExtend64(tmp[103:52]) ENDFOR -k[MAX:32] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - - Integer - Mask - AVX512BW - Compare - - - - Compare packed 8-bit integers in "a" and "b" based on the comparison operand specified by "imm8", and store the results in mask vector "k". + + AVX512IFMA52 + AVX512VL + Arithmetic + + + + + + Multiply packed unsigned 52-bit integers in each 64-bit element of "b" and "c" to form a 104-bit intermediate result. Add the high 52-bit unsigned integer from the intermediate result with the corresponding unsigned 64-bit integer in "a", and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). -CASE (imm8[7:0]) OF -0: OP := _MM_CMPINT_EQ -1: OP := _MM_CMPINT_LT -2: OP := _MM_CMPINT_LE -3: OP := _MM_CMPINT_FALSE -4: OP := _MM_CMPINT_NE -5: OP := _MM_CMPINT_NLT -6: OP := _MM_CMPINT_NLE -7: OP := _MM_CMPINT_TRUE -ESAC -FOR j := 0 to 63 - i := j*8 - k[j] := ( a[i+7:i] OP b[i+7:i] ) ? 1 : 0 +FOR j := 0 to 1 + i := j*64 + IF k[j] + tmp[127:0] := ZeroExtend64(b[i+51:i]) * ZeroExtend64(c[i+51:i]) + dst[i+63:i] := a[i+63:i] + ZeroExtend64(tmp[103:52]) + ELSE + dst[i+63:i] := a[i+63:i] + FI ENDFOR -k[MAX:64] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - - Integer - Mask - AVX512BW - Compare - - - Compare packed 8-bit integers in "a" and "b" for equality, and store the results in mask vector "k". + + AVX512IFMA52 + AVX512VL + Arithmetic + + + + + + Multiply packed unsigned 52-bit integers in each 64-bit element of "b" and "c" to form a 104-bit intermediate result. Add the high 52-bit unsigned integer from the intermediate result with the corresponding unsigned 64-bit integer in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 63 - i := j*8 - k[j] := ( a[i+7:i] == b[i+7:i] ) ? 1 : 0 +FOR j := 0 to 1 + i := j*64 + IF k[j] + tmp[127:0] := ZeroExtend64(b[i+51:i]) * ZeroExtend64(c[i+51:i]) + dst[i+63:i] := a[i+63:i] + ZeroExtend64(tmp[103:52]) + ELSE + dst[i+63:i] := 0 + FI ENDFOR -k[MAX:64] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - - Integer - Mask - AVX512BW - Compare - - - Compare packed 8-bit integers in "a" and "b" for greater-than-or-equal, and store the results in mask vector "k". + + Floating Point + AVX512PF + Load + + + + + + Prefetch single-precision (32-bit) floating-point elements from memory using 64-bit indices. 32-bit elements are loaded from addresses starting at "base_addr" and offset by each 64-bit element in "vindex" (each index is scaled by the factor in "scale"). Gathered elements are merged in cache. "scale" should be 1, 2, 4 or 8. The "hint" parameter may be 1 (_MM_HINT_T0) for prefetching to L1 cache, or 2 (_MM_HINT_T1) for prefetching to L2 cache. -FOR j := 0 to 63 - i := j*8 - k[j] := ( a[i+7:i] >= b[i+7:i] ) ? 1 : 0 +FOR j:= 0 to 7 + i := j*64 + m := j*64 + addr := base_addr + vindex[m+63:m] * ZeroExtend64(scale) * 8 + Prefetch(MEM[addr+31:addr], hint) ENDFOR -k[MAX:64] := 0 - + +
immintrin.h
 - - Integer - Mask - AVX512BW - Compare - - - Compare packed 8-bit integers in "a" and "b" for greater-than, and store the results in mask vector "k". + + Floating Point + AVX512PF + Load + + + + + + + Prefetch single-precision (32-bit) floating-point elements from memory using 64-bit indices. 32-bit elements are loaded from addresses starting at "base_addr" and offset by each 64-bit element in "vindex" (each index is scaled by the factor in "scale"). Gathered elements are merged in cache using writemask "k" (elements are only brought into cache when their corresponding mask bit is set). "scale" should be 1, 2, 4 or 8.. The "hint" parameter may be 1 (_MM_HINT_T0) for prefetching to L1 cache, or 2 (_MM_HINT_T1) for prefetching to L2 cache. -FOR j := 0 to 63 - i := j*8 - k[j] := ( a[i+7:i] > b[i+7:i] ) ? 1 : 0 +FOR j:= 0 to 7 + i := j*64 + m := j*64 + IF k[j] + addr := base_addr + vindex[m+63:m] * ZeroExtend64(scale) * 8 + Prefetch(MEM[addr+31:addr], hint) + FI ENDFOR -k[MAX:64] := 0 - + +
immintrin.h
 - - Integer - Mask - AVX512BW - Compare - - - Compare packed 8-bit integers in "a" and "b" for less-than-or-equal, and store the results in mask vector "k". + + Floating Point + AVX512PF + Store + + + + + + Prefetch single-precision (32-bit) floating-point elements with intent to write into memory using 64-bit indices. Elements are prefetched into cache level "hint", where "hint" is 0 or 1. 32-bit elements are stored at addresses starting at "base_addr" and offset by each 64-bit element in "vindex" (each index is scaled by the factor in "scale"). "scale" should be 1, 2, 4 or 8. -FOR j := 0 to 63 - i := j*8 - k[j] := ( a[i+7:i] <= b[i+7:i] ) ? 1 : 0 +FOR j := 0 to 7 + i := j*64 + m := j*64 + addr := base_addr + vindex[m+63:m] * ZeroExtend64(scale) * 8 + Prefetch(MEM[addr+31:addr], hint) ENDFOR -k[MAX:64] := 0 - + +
immintrin.h
 - - Integer - Mask - AVX512BW - Compare - - - Compare packed 8-bit integers in "a" and "b" for less-than, and store the results in mask vector "k". + + Floating Point + AVX512PF + Store + + + + + + + Prefetch single-precision (32-bit) floating-point elements with intent to write into memory using 64-bit indices. The "hint" parameter may be 1 (_MM_HINT_T0) for prefetching to L1 cache, or 2 (_MM_HINT_T1) for prefetching to L2 cache. 32-bit elements are stored at addresses starting at "base_addr" and offset by each 64-bit element in "vindex" (each index is scaled by the factor in "scale") subject to mask "k" (elements are not brought into cache when the corresponding mask bit is not set). "scale" should be 1, 2, 4 or 8. -FOR j := 0 to 63 - i := j*8 - k[j] := ( a[i+7:i] < b[i+7:i] ) ? 1 : 0 +FOR j := 0 to 7 + i := j*64 + m := j*64 + IF k[j] + addr := base_addr + vindex[m+63:m] * ZeroExtend64(scale) * 8 + Prefetch(MEM[addr+31:addr], hint) + FI ENDFOR -k[MAX:64] := 0 - + +
immintrin.h
 - - Integer - Mask - AVX512BW - Compare - - - Compare packed 8-bit integers in "a" and "b" for not-equal, and store the results in mask vector "k". + + Floating Point + AVX512PF + Load + + + + + + Prefetch double-precision (64-bit) floating-point elements from memory using 32-bit indices. 64-bit elements are loaded from addresses starting at "base_addr" and offset by each 32-bit element in "vindex" (each index is scaled by the factor in "scale"). Gathered elements are merged in cache. "scale" should be 1, 2, 4 or 8. The "hint" parameter may be 1 (_MM_HINT_T0) for prefetching to L1 cache, or 2 (_MM_HINT_T1) for prefetching to L2 cache. -FOR j := 0 to 63 - i := j*8 - k[j] := ( a[i+7:i] != b[i+7:i] ) ? 1 : 0 +FOR j := 0 to 7 + i := j*64 + m := j*32 + addr := base_addr + SignExtend64(vindex[m+31:m]) * ZeroExtend64(scale) * 8 + Prefetch(MEM[addr+63:addr], hint) ENDFOR -k[MAX:64] := 0 - + +
immintrin.h
 - - Integer - Mask - AVX512BW - Compare - - - - - Compare packed 8-bit integers in "a" and "b" based on the comparison operand specified by "imm8", and store the results in mask vector "k1" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Floating Point + AVX512PF + Load + + + + + + + Prefetch double-precision (64-bit) floating-point elements from memory using 32-bit indices. 64-bit elements are loaded from addresses starting at "base_addr" and offset by each 32-bit element in "vindex" (each index is scaled by the factor in "scale"). Gathered elements are merged in cache using writemask "k" (elements are brought into cache only when their corresponding mask bits are set). "scale" should be 1, 2, 4 or 8. The "hint" parameter may be 1 (_MM_HINT_T0) for prefetching to L1 cache, or 2 (_MM_HINT_T1) for prefetching to L2 cache. -CASE (imm8[7:0]) OF -0: OP := _MM_CMPINT_EQ -1: OP := _MM_CMPINT_LT -2: OP := _MM_CMPINT_LE -3: OP := _MM_CMPINT_FALSE -4: OP := _MM_CMPINT_NE -5: OP := _MM_CMPINT_NLT -6: OP := _MM_CMPINT_NLE -7: OP := _MM_CMPINT_TRUE -ESAC -FOR j := 0 to 63 - i := j*8 - IF k1[j] - k[j] := ( a[i+7:i] OP b[i+7:i] ) ? 1 : 0 - ELSE - k[j] := 0 +FOR j := 0 to 7 + i := j*64 + m := j*32 + IF k[j] + addr := base_addr + SignExtend64(vindex[m+31:m]) * ZeroExtend64(scale) * 8 + Prefetch(MEM[addr+63:addr], hint) FI ENDFOR -k[MAX:64] := 0 - + +
immintrin.h
 - - Integer - Mask - AVX512BW - Compare - - - - Compare packed 8-bit integers in "a" and "b" for equality, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). + + Floating Point + AVX512PF + Store + + + + + + Prefetch double-precision (64-bit) floating-point elements with intent to write using 32-bit indices. The "hint" parameter may be 1 (_MM_HINT_T0) for prefetching to L1 cache, or 2 (_MM_HINT_T1) for prefetching to L2 cache. 64-bit elements are brought into cache from addresses starting at "base_addr" and offset by each 32-bit element in "vindex" (each index is scaled by the factor in "scale"). "scale" should be 1, 2, 4 or 8. -FOR j := 0 to 63 - i := j*8 - IF k1[j] - k[j] := ( a[i+7:i] == b[i+7:i] ) ? 1 : 0 - ELSE - k[j] := 0 - FI +FOR j := 0 to 7 + i := j*64 + m := j*32 + addr := base_addr + SignExtend64(vindex[m+31:m]) * ZeroExtend64(scale) * 8 + Prefetch(MEM[addr+63:addr], hint) ENDFOR -k[MAX:64] := 0 - + +
immintrin.h
 - - Integer - Mask - AVX512BW - Compare - - - - Compare packed 8-bit integers in "a" and "b" for greater-than-or-equal, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). + + Floating Point + AVX512PF + Store + + + + + + + Prefetch double-precision (64-bit) floating-point elements with intent to write using 32-bit indices. The "hint" parameter may be 1 (_MM_HINT_T0) for prefetching to L1 cache, or 2 (_MM_HINT_T1) for prefetching to L2 cache. 64-bit elements are brought into cache from addresses starting at "base_addr" and offset by each 32-bit element in "vindex" (each index is scaled by the factor in "scale") subject to mask "k" (elements are not brought into cache when the corresponding mask bit is not set). "scale" should be 1, 2, 4 or 8. -FOR j := 0 to 63 - i := j*8 - IF k1[j] - k[j] := ( a[i+7:i] >= b[i+7:i] ) ? 1 : 0 - ELSE - k[j] := 0 +FOR j := 0 to 7 + i := j*64 + m := j*32 + IF k[j] + addr := base_addr + SignExtend64(vindex[m+31:m]) * ZeroExtend64(scale) * 8 + Prefetch(MEM[addr+63:addr], hint) FI ENDFOR -k[MAX:64] := 0 - + +
immintrin.h
 - - Integer - Mask - AVX512BW - Compare - - - - Compare packed 8-bit integers in "a" and "b" for greater-than, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). + + Floating Point + AVX512PF + Load + + + + + + Prefetch double-precision (64-bit) floating-point elements from memory into cache level specified by "hint" using 64-bit indices. 64-bit elements are loaded from addresses starting at "base_addr" and offset by each 64-bit element in "vindex" (each index is scaled by the factor in "scale"). "scale" should be 1, 2, 4 or 8. The "hint" parameter may be 1 (_MM_HINT_T0) for prefetching to L1 cache, or 2 (_MM_HINT_T1) for prefetching to L2 cache. -FOR j := 0 to 63 - i := j*8 - IF k1[j] - k[j] := ( a[i+7:i] > b[i+7:i] ) ? 1 : 0 - ELSE - k[j] := 0 - FI +FOR j := 0 to 7 + i := j*64 + m := j*64 + addr := base_addr + vindex[m+63:m] * ZeroExtend64(scale) * 8 + Prefetch(MEM[addr+63:addr], hint) ENDFOR -k[MAX:64] := 0 - + +
immintrin.h
 - - Integer - Mask - AVX512BW - Compare - - - - Compare packed 8-bit integers in "a" and "b" for less-than-or-equal, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). + + Floating Point + AVX512PF + Load + + + + + + + Prefetch double-precision (64-bit) floating-point elements from memory into cache level specified by "hint" using 64-bit indices. 64-bit elements are loaded from addresses starting at "base_addr" and offset by each 64-bit element in "vindex" (each index is scaled by the factor in "scale"). Prefetched elements are merged in cache using writemask "k" (elements are copied from memory when the corresponding mask bit is set). "scale" should be 1, 2, 4 or 8. The "hint" parameter may be 1 (_MM_HINT_T0) for prefetching to L1 cache, or 2 (_MM_HINT_T1) for prefetching to L2 cache. -FOR j := 0 to 63 - i := j*8 - IF k1[j] - k[j] := ( a[i+7:i] <= b[i+7:i] ) ? 1 : 0 - ELSE - k[j] := 0 +FOR j := 0 to 7 + i := j*64 + m := j*64 + IF k[j] + addr := base_addr + vindex[m+63:m] * ZeroExtend64(scale) * 8 + Prefetch(MEM[addr+63:addr], hint) FI ENDFOR -k[MAX:64] := 0 - + +
immintrin.h
 - - Integer - Mask - AVX512BW - Compare - - - - Compare packed 8-bit integers in "a" and "b" for less-than, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). + + Floating Point + AVX512PF + Store + + + + + + Prefetch double-precision (64-bit) floating-point elements with intent to write into memory using 64-bit indices. The "hint" parameter may be 1 (_MM_HINT_T0) for prefetching to L1 cache, or 2 (_MM_HINT_T1) for prefetching to L2 cache. 64-bit elements are brought into cache from addresses starting at "base_addr" and offset by each 64-bit element in "vindex" (each index is scaled by the factor in "scale"). "scale" should be 1, 2, 4 or 8. -FOR j := 0 to 63 - i := j*8 - IF k1[j] - k[j] := ( a[i+7:i] < b[i+7:i] ) ? 1 : 0 - ELSE - k[j] := 0 - FI +FOR j := 0 to 7 + i := j*64 + m := j*64 + addr := base_addr + vindex[m+63:m] * ZeroExtend64(scale) * 8 + Prefetch(MEM[addr+63:addr], hint) ENDFOR -k[MAX:64] := 0 - + +
immintrin.h
 - - Integer - Mask - AVX512BW - Compare - - - - Compare packed 8-bit integers in "a" and "b" for not-equal, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). + + Floating Point + AVX512PF + Store + + + + + + + Prefetch double-precision (64-bit) floating-point elements with intent to write into memory using 64-bit indices. The "hint" parameter may be 1 (_MM_HINT_T0) for prefetching to L1 cache, or 2 (_MM_HINT_T1) for prefetching to L2 cache. 64-bit elements are brought into cache from addresses starting at "base_addr" and offset by each 64-bit element in "vindex" (each index is scaled by the factor in "scale") subject to mask "k" (elements are not brought into cache when the corresponding mask bit is not set). "scale" should be 1, 2, 4 or 8. -FOR j := 0 to 63 - i := j*8 - IF k1[j] - k[j] := ( a[i+7:i] != b[i+7:i] ) ? 1 : 0 - ELSE - k[j] := 0 +FOR j := 0 to 7 + i := j*64 + m := j*64 + IF k[j] + addr := base_addr + vindex[m+63:m] * ZeroExtend64(scale) * 8 + Prefetch(MEM[addr+63:addr], hint) FI ENDFOR -k[MAX:64] := 0 - + +
immintrin.h
 - - Integer - Mask - AVX512VL - AVX512BW - Compare - - - - Compare packed 8-bit integers in "a" and "b" based on the comparison operand specified by "imm8", and store the results in mask vector "k". + + Floating Point + AVX512PF/KNCNI + Load + + + + + + + Prefetch single-precision (32-bit) floating-point elements from memory using 32-bit indices. 32-bit elements are loaded from addresses starting at "base_addr" and offset by each 32-bit element in "vindex" (each index is scaled by the factor in "scale"). Gathered elements are merged in cache using writemask "k" (elements are brought into cache only when their corresponding mask bits are set). "scale" should be 1, 2, 4 or 8. The "hint" parameter may be 1 (_MM_HINT_T0) for prefetching to L1 cache, or 2 (_MM_HINT_T1) for prefetching to L2 cache. -CASE (imm8[7:0]) OF -0: OP := _MM_CMPINT_EQ -1: OP := _MM_CMPINT_LT -2: OP := _MM_CMPINT_LE -3: OP := _MM_CMPINT_FALSE -4: OP := _MM_CMPINT_NE -5: OP := _MM_CMPINT_NLT -6: OP := _MM_CMPINT_NLE -7: OP := _MM_CMPINT_TRUE -ESAC FOR j := 0 to 15 - i := j*8 - k[j] := ( a[i+7:i] OP b[i+7:i] ) ? 1 : 0 + i := j*32 + m := j*32 + IF k[j] + addr := base_addr + SignExtend64(vindex[m+31:m]) * ZeroExtend64(scale) * 8 + Prefetch(MEM[addr+31:addr], hint) + FI ENDFOR -k[MAX:16] := 0 - + +
immintrin.h
 - - Integer - Mask - AVX512VL - AVX512BW - Compare - - - Compare packed 8-bit integers in "a" and "b" for equality, and store the results in mask vector "k". + + Floating Point + AVX512PF/KNCNI + Load + + + + + + + Prefetches a set of 16 single-precision (32-bit) memory locations pointed by base address "base_addr" and 32-bit integer index vector "vindex" with scale "scale" to L1 or L2 level of cache depending on the value of "hint". The "hint" parameter may be 1 (_MM_HINT_T0) for prefetching to L1 cache, or 2 (_MM_HINT_T1) for prefetching to L2 cache. +The "conv" parameter specifies the granularity used by compilers to better encode the instruction. It should be the same as the "conv" parameter specified for the subsequent gather intrinsic. FOR j := 0 to 15 - i := j*8 - k[j] := ( a[i+7:i] == b[i+7:i] ) ? 1 : 0 + i := j*32 + m := j*32 + addr := base_addr + SignExtend64(vindex[m+31:m]) * ZeroExtend64(scale) * 8 + Prefetch(MEM[addr+31:addr], hint) ENDFOR -k[MAX:16] := 0 - + +
immintrin.h
 - - Integer - Mask - AVX512VL - AVX512BW - Compare - - - Compare packed 8-bit integers in "a" and "b" for greater-than-or-equal, and store the results in mask vector "k". + + Floating Point + AVX512PF/KNCNI + Load + + + + + + + + Prefetches a set of 16 single-precision (32-bit) memory locations pointed by base address "base_addr" and 32-bit integer index vector "vindex" with scale "scale" to L1 or L2 level of cache depending on the value of "hint". Gathered elements are merged in cache using writemask "k" (elements are brought into cache only when their corresponding mask bits are set). The "hint" parameter may be 1 (_MM_HINT_T0) for prefetching to L1 cache, or 2 (_MM_HINT_T1) for prefetching to L2 cache. +The "conv" parameter specifies the granularity used by compilers to better encode the instruction. It should be the same as the "conv" parameter specified for the subsequent gather intrinsic. FOR j := 0 to 15 - i := j*8 - k[j] := ( a[i+7:i] >= b[i+7:i] ) ? 1 : 0 + i := j*32 + m := j*32 + IF k[j] + addr := base_addr + SignExtend64(vindex[m+31:m]) * ZeroExtend64(scale) * 8 + Prefetch(MEM[addr+31:addr], hint) + FI ENDFOR -k[MAX:16] := 0 - + +
immintrin.h
 - - Integer - Mask - AVX512VL - AVX512BW - Compare - - - Compare packed 8-bit integers in "a" and "b" for greater-than, and store the results in mask vector "k". + + Floating Point + AVX512PF/KNCNI + Store + + + + + + + Prefetches a set of 16 single-precision (32-bit) memory locations pointed by base address "base_addr" and 32-bit integer index vector "vindex" with scale "scale" to L1 or L2 level of cache depending on the value of "hint", with a request for exclusive ownership. The "hint" parameter may be one of the following: _MM_HINT_T0 = 1 for prefetching to L1 cache, _MM_HINT_T1 = 2 for prefetching to L2 cache, _MM_HINT_T2 = 3 for prefetching to L2 cache non-temporal, _MM_HINT_NTA = 0 for prefetching to L1 cache non-temporal. The "conv" parameter specifies the granularity used by compilers to better encode the instruction. It should be the same as the "conv" parameter specified for the subsequent scatter intrinsic. FOR j := 0 to 15 - i := j*8 - k[j] := ( a[i+7:i] > b[i+7:i] ) ? 1 : 0 + i := j*32 + m := j*32 + addr := base_addr + SignExtend64(vindex[m+31:m]) * ZeroExtend64(scale) * 8 + Prefetch(MEM[addr+31:addr], hint) ENDFOR -k[MAX:16] := 0 - + +
immintrin.h
 - - Integer - Mask - AVX512VL - AVX512BW - Compare - - - Compare packed 8-bit integers in "a" and "b" for less-than-or-equal, and store the results in mask vector "k". + + Floating Point + AVX512PF/KNCNI + Store + + + + + + + + Prefetches a set of 16 single-precision (32-bit) memory locations pointed by base address "base_addr" and 32-bit integer index vector "vindex" with scale "scale" to L1 or L2 level of cache depending on the value of "hint". The "hint" parameter may be 1 (_MM_HINT_T0) for prefetching to L1 cache, or 2 (_MM_HINT_T1) for prefetching to L2 cache. +The "conv" parameter specifies the granularity used by compilers to better encode the instruction. It should be the same as the "conv" parameter specified for the subsequent gather intrinsic. Only those elements whose corresponding mask bit in "k" is set are loaded into cache. +cachev := 0 FOR j := 0 to 15 - i := j*8 - k[j] := ( a[i+7:i] <= b[i+7:i] ) ? 1 : 0 + i := j*32 + m := j*32 + IF k[j] + addr := base_addr + SignExtend64(vindex[m+31:m]) * ZeroExtend64(scale) * 8 + Prefetch(MEM[addr+31:addr], hint) + FI ENDFOR -k[MAX:16] := 0 - + +
immintrin.h
 - - Integer - Mask - AVX512VL - AVX512BW - Compare - - - Compare packed 8-bit integers in "a" and "b" for less-than, and store the results in mask vector "k". + + Floating Point + AVX512PF/KNCNI + Load + + + + + + Prefetches 16 single-precision (32-bit) floating-point elements in memory starting at location "base_addr" at packed 32-bit integer indices stored in "vindex" scaled by "scale". The "hint" parameter may be 1 (_MM_HINT_T0) for prefetching to L1 cache, or 2 (_MM_HINT_T1) for prefetching to L2 cache. FOR j := 0 to 15 - i := j*8 - k[j] := ( a[i+7:i] < b[i+7:i] ) ? 1 : 0 + i := j*32 + m := j*32 + addr := base_addr + SignExtend64(vindex[m+31:m]) * ZeroExtend64(scale) * 8 + Prefetch(MEM[addr+31:addr], hint) ENDFOR -k[MAX:16] := 0 - + +
immintrin.h
 - - Integer - Mask - AVX512VL - AVX512BW - Compare - - - Compare packed 8-bit integers in "a" and "b" for not-equal, and store the results in mask vector "k". + + Floating Point + AVX512PF/KNCNI + Store + + + + + + Prefetches 16 single-precision (32-bit) floating-point elements in memory starting at location "base_addr" at packed 32-bit integer indices stored in "vindex" scaled by "scale". The "hint" parameter may be 1 (_MM_HINT_T0) for prefetching to L1 cache, or 2 (_MM_HINT_T1) for prefetching to L2 cache. FOR j := 0 to 15 - i := j*8 - k[j] := ( a[i+7:i] != b[i+7:i] ) ? 1 : 0 + i := j*32 + m := j*32 + addr := base_addr + SignExtend64(vindex[m+31:m]) * ZeroExtend64(scale) * 8 + Prefetch(MEM[addr+31:addr], hint) ENDFOR -k[MAX:16] := 0 - + +
immintrin.h
 - - Integer - Mask - AVX512VL - AVX512BW - Compare - - - - - Compare packed 8-bit integers in "a" and "b" based on the comparison operand specified by "imm8", and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). + + Floating Point + AVX512PF/KNCNI + Store + + + + + + + Prefetches 16 single-precision (32-bit) floating-point elements in memory starting at location "base_addr" at packed 32-bit integer indices stored in "vindex" scaled by "scale". The "hint" parameter may be 1 (_MM_HINT_T0) for prefetching to L1 cache, or 2 (_MM_HINT_T1) for prefetching to L2 cache. Only those elements whose corresponding mask bit in "k" is set are loaded into the desired cache. -CASE (imm8[7:0]) OF -0: OP := _MM_CMPINT_EQ -1: OP := _MM_CMPINT_LT -2: OP := _MM_CMPINT_LE -3: OP := _MM_CMPINT_FALSE -4: OP := _MM_CMPINT_NE -5: OP := _MM_CMPINT_NLT -6: OP := _MM_CMPINT_NLE -7: OP := _MM_CMPINT_TRUE -ESAC FOR j := 0 to 15 - i := j*8 - IF k1[j] - k[j] := ( a[i+7:i] OP b[i+7:i] ) ? 1 : 0 - ELSE - k[j] := 0 + i := j*32 + m := j*32 + IF k[j] + addr := base_addr + SignExtend64(vindex[m+31:m]) * ZeroExtend64(scale) * 8 + Prefetch(MEM[addr+31:addr], hint) FI ENDFOR -k[MAX:16] := 0 - + +
immintrin.h
 - + Integer - Mask + AVX512VPOPCNTDQ AVX512VL - AVX512BW - Compare - - - - Compare packed 8-bit integers in "a" and "b" for equality, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). + Bit Manipulation + + + + Count the number of logical 1 bits in packed 64-bit integers in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*8 - IF k1[j] - k[j] := ( a[i+7:i] == b[i+7:i] ) ? 1 : 0 +DEFINE POPCNT(a) { + count := 0 + DO WHILE a > 0 + count += a[0] + a >>= 1 + OD + RETURN count +} +FOR j := 0 to 3 + i := j*64 + IF k[j] + dst[i+63:i] := POPCNT(a[i+63:i]) ELSE - k[j] := 0 + dst[i+63:i] := 0 FI ENDFOR -k[MAX:16] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - Mask + AVX512VPOPCNTDQ AVX512VL - AVX512BW - Compare - - - - Compare packed 8-bit integers in "a" and "b" for greater-than-or-equal, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). + Bit Manipulation + + + + + Count the number of logical 1 bits in packed 64-bit integers in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*8 - IF k1[j] - k[j] := ( a[i+7:i] >= b[i+7:i] ) ? 1 : 0 +DEFINE POPCNT(a) { + count := 0 + DO WHILE a > 0 + count += a[0] + a >>= 1 + OD + RETURN count +} +FOR j := 0 to 3 + i := j*64 + IF k[j] + dst[i+63:i] := POPCNT(a[i+63:i]) ELSE - k[j] := 0 + dst[i+63:i] := src[i+63:i] FI ENDFOR -k[MAX:16] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - Mask + AVX512VPOPCNTDQ AVX512VL - AVX512BW - Compare - - - - Compare packed 8-bit integers in "a" and "b" for greater-than, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). + Bit Manipulation + + + Count the number of logical 1 bits in packed 64-bit integers in "a", and store the results in "dst". -FOR j := 0 to 15 - i := j*8 - IF k1[j] - k[j] := ( a[i+7:i] > b[i+7:i] ) ? 1 : 0 - ELSE - k[j] := 0 - FI +DEFINE POPCNT(a) { + count := 0 + DO WHILE a > 0 + count += a[0] + a >>= 1 + OD + RETURN count +} +FOR j := 0 to 3 + i := j*64 + dst[i+63:i] := POPCNT(a[i+63:i]) ENDFOR -k[MAX:16] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - Mask + AVX512VPOPCNTDQ AVX512VL - AVX512BW - Compare - - - - Compare packed 8-bit integers in "a" and "b" for less-than-or-equal, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). + Bit Manipulation + + + + Count the number of logical 1 bits in packed 64-bit integers in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*8 - IF k1[j] - k[j] := ( a[i+7:i] <= b[i+7:i] ) ? 1 : 0 +DEFINE POPCNT(a) { + count := 0 + DO WHILE a > 0 + count += a[0] + a >>= 1 + OD + RETURN count +} +FOR j := 0 to 1 + i := j*64 + IF k[j] + dst[i+63:i] := POPCNT(a[i+63:i]) ELSE - k[j] := 0 + dst[i+63:i] := 0 FI ENDFOR -k[MAX:16] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - Mask + AVX512VPOPCNTDQ AVX512VL - AVX512BW - Compare - - - - Compare packed 8-bit integers in "a" and "b" for less-than, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). + Bit Manipulation + + + + + Count the number of logical 1 bits in packed 64-bit integers in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*8 - IF k1[j] - k[j] := ( a[i+7:i] < b[i+7:i] ) ? 1 : 0 +DEFINE POPCNT(a) { + count := 0 + DO WHILE a > 0 + count += a[0] + a >>= 1 + OD + RETURN count +} +FOR j := 0 to 1 + i := j*64 + IF k[j] + dst[i+63:i] := POPCNT(a[i+63:i]) ELSE - k[j] := 0 + dst[i+63:i] := src[i+63:i] FI ENDFOR -k[MAX:16] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - Mask + AVX512VPOPCNTDQ AVX512VL - AVX512BW - Compare - - - - Compare packed 8-bit integers in "a" and "b" for not-equal, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). + Bit Manipulation + + + Count the number of logical 1 bits in packed 64-bit integers in "a", and store the results in "dst". -FOR j := 0 to 15 - i := j*8 - IF k1[j] - k[j] := ( a[i+7:i] != b[i+7:i] ) ? 1 : 0 - ELSE - k[j] := 0 - FI +DEFINE POPCNT(a) { + count := 0 + DO WHILE a > 0 + count += a[0] + a >>= 1 + OD + RETURN count +} +FOR j := 0 to 1 + i := j*64 + dst[i+63:i] := POPCNT(a[i+63:i]) ENDFOR -k[MAX:16] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - Mask + AVX512VPOPCNTDQ AVX512VL - AVX512F - Compare - - - - Compare packed 32-bit integers in "a" and "b" based on the comparison operand specified by "imm8", and store the results in mask vector "k". + Bit Manipulation + + + Count the number of logical 1 bits in packed 32-bit integers in "a", and store the results in "dst". -CASE (imm8[7:0]) OF -0: OP := _MM_CMPINT_EQ -1: OP := _MM_CMPINT_LT -2: OP := _MM_CMPINT_LE -3: OP := _MM_CMPINT_FALSE -4: OP := _MM_CMPINT_NE -5: OP := _MM_CMPINT_NLT -6: OP := _MM_CMPINT_NLE -7: OP := _MM_CMPINT_TRUE -ESAC +DEFINE POPCNT(a) { + count := 0 + DO WHILE a > 0 + count += a[0] + a >>= 1 + OD + RETURN count +} FOR j := 0 to 7 i := j*32 - k[j] := ( a[i+31:i] OP b[i+31:i] ) ? 1 : 0 + dst[i+31:i] := POPCNT(a[i+31:i]) ENDFOR -k[MAX:8] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - Mask + AVX512VPOPCNTDQ AVX512VL - AVX512F - Compare - - - Compare packed 32-bit integers in "a" and "b" for equality, and store the results in mask vector "k". + Bit Manipulation + + + + + Count the number of logical 1 bits in packed 32-bit integers in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). +DEFINE POPCNT(a) { + count := 0 + DO WHILE a > 0 + count += a[0] + a >>= 1 + OD + RETURN count +} FOR j := 0 to 7 i := j*32 - k[j] := ( a[i+31:i] == b[i+31:i] ) ? 1 : 0 + IF k[j] + dst[i+31:i] := POPCNT(a[i+31:i]) + ELSE + dst[i+31:i] := src[i+31:i] + FI ENDFOR -k[MAX:8] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - Mask + AVX512VPOPCNTDQ AVX512VL - AVX512F - Compare - - - Compare packed 32-bit integers in "a" and "b" for greater-than-or-equal, and store the results in mask vector "k". + Bit Manipulation + + + + Count the number of logical 1 bits in packed 32-bit integers in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). +DEFINE POPCNT(a) { + count := 0 + DO WHILE a > 0 + count += a[0] + a >>= 1 + OD + RETURN count +} FOR j := 0 to 7 i := j*32 - k[j] := ( a[i+31:i] >= b[i+31:i] ) ? 1 : 0 + IF k[j] + dst[i+31:i] := POPCNT(a[i+31:i]) + ELSE + dst[i+31:i] := 0 + FI ENDFOR -k[MAX:8] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - Mask + AVX512VPOPCNTDQ AVX512VL - AVX512F - Compare - - - Compare packed 32-bit integers in "a" and "b" for greater-than, and store the results in mask vector "k". + Bit Manipulation + + + Count the number of logical 1 bits in packed 32-bit integers in "a", and store the results in "dst". -FOR j := 0 to 7 +DEFINE POPCNT(a) { + count := 0 + DO WHILE a > 0 + count += a[0] + a >>= 1 + OD + RETURN count +} +FOR j := 0 to 3 i := j*32 - k[j] := ( a[i+31:i] > b[i+31:i] ) ? 1 : 0 + dst[i+31:i] := POPCNT(a[i+31:i]) ENDFOR -k[MAX:8] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - Mask + AVX512VPOPCNTDQ AVX512VL - AVX512F - Compare - - - Compare packed 32-bit integers in "a" and "b" for less-than-or-equal, and store the results in mask vector "k". + Bit Manipulation + + + + + Count the number of logical 1 bits in packed 32-bit integers in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 7 +DEFINE POPCNT(a) { + count := 0 + DO WHILE a > 0 + count += a[0] + a >>= 1 + OD + RETURN count +} +FOR j := 0 to 3 i := j*32 - k[j] := ( a[i+31:i] <= b[i+31:i] ) ? 1 : 0 + IF k[j] + dst[i+31:i] := POPCNT(a[i+31:i]) + ELSE + dst[i+31:i] := src[i+31:i] + FI ENDFOR -k[MAX:8] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - Mask + AVX512VPOPCNTDQ AVX512VL - AVX512F - Compare - - - Compare packed 32-bit integers in "a" and "b" for less-than, and store the results in mask vector "k". + Bit Manipulation + + + + Count the number of logical 1 bits in packed 32-bit integers in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 +DEFINE POPCNT(a) { + count := 0 + DO WHILE a > 0 + count += a[0] + a >>= 1 + OD + RETURN count +} +FOR j := 0 to 3 i := j*32 - k[j] := ( a[i+31:i] < b[i+31:i] ) ? 1 : 0 + IF k[j] + dst[i+31:i] := POPCNT(a[i+31:i]) + ELSE + dst[i+31:i] := 0 + FI ENDFOR -k[MAX:8] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - Mask - AVX512VL - AVX512F - Compare - - - Compare packed 32-bit integers in "a" and "b" for not-equal, and store the results in mask vector "k". + AVX512VPOPCNTDQ + Bit Manipulation + + + Count the number of logical 1 bits in packed 32-bit integers in "a", and store the results in "dst". -FOR j := 0 to 7 +DEFINE POPCNT(a) { + count := 0 + DO WHILE a > 0 + count += a[0] + a >>= 1 + OD + RETURN count +} +FOR j := 0 to 15 i := j*32 - k[j] := ( a[i+31:i] != b[i+31:i] ) ? 1 : 0 + dst[i+31:i] := POPCNT(a[i+31:i]) ENDFOR -k[MAX:8] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - Mask - AVX512VL - AVX512F - Compare - - - - - Compare packed 32-bit integers in "a" and "b" based on the comparison operand specified by "imm8", and store the results in mask vector "k1" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + AVX512VPOPCNTDQ + Bit Manipulation + + + + + Count the number of logical 1 bits in packed 32-bit integers in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -CASE (imm8[7:0]) OF -0: OP := _MM_CMPINT_EQ -1: OP := _MM_CMPINT_LT -2: OP := _MM_CMPINT_LE -3: OP := _MM_CMPINT_FALSE -4: OP := _MM_CMPINT_NE -5: OP := _MM_CMPINT_NLT -6: OP := _MM_CMPINT_NLE -7: OP := _MM_CMPINT_TRUE -ESAC -FOR j := 0 to 7 +DEFINE POPCNT(a) { + count := 0 + DO WHILE a > 0 + count += a[0] + a >>= 1 + OD + RETURN count +} +FOR j := 0 to 15 i := j*32 - IF k1[j] - k[j] := ( a[i+31:i] OP b[i+31:i] ) ? 1 : 0 - ELSE - k[j] := 0 + IF k[j] + dst[i+31:i] := POPCNT(a[i+31:i]) + ELSE + dst[i+31:i] := src[i+31:i] FI ENDFOR -k[MAX:8] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - Mask - AVX512VL - AVX512F - Compare - - - - Compare packed 32-bit integers in "a" and "b" for equality, and store the results in mask vector "k1" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + AVX512VPOPCNTDQ + Bit Manipulation + + + + Count the number of logical 1 bits in packed 32-bit integers in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 +DEFINE POPCNT(a) { + count := 0 + DO WHILE a > 0 + count += a[0] + a >>= 1 + OD + RETURN count +} +FOR j := 0 to 15 i := j*32 - IF k1[j] - k[j] := ( a[i+31:i] == b[i+31:i] ) ? 1 : 0 - ELSE - k[j] := 0 + IF k[j] + dst[i+31:i] := POPCNT(a[i+31:i]) + ELSE + dst[i+31:i] := 0 FI ENDFOR -k[MAX:8] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - Mask - AVX512VL - AVX512F - Compare - - - - Compare packed 32-bit integers in "a" and "b" for greater-than-or-equal, and store the results in mask vector "k1" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + AVX512VPOPCNTDQ + Bit Manipulation + + + Count the number of logical 1 bits in packed 64-bit integers in "a", and store the results in "dst". +DEFINE POPCNT(a) { + count := 0 + DO WHILE a > 0 + count += a[0] + a >>= 1 + OD + RETURN count +} FOR j := 0 to 7 - i := j*32 - IF k1[j] - k[j] := ( a[i+31:i] >= b[i+31:i] ) ? 1 : 0 - ELSE - k[j] := 0 - FI + i := j*64 + dst[i+63:i] := POPCNT(a[i+63:i]) ENDFOR -k[MAX:8] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - Mask - AVX512VL - AVX512F - Compare - - - - Compare packed 32-bit integers in "a" and "b" for greater-than, and store the results in mask vector "k1" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + AVX512VPOPCNTDQ + Bit Manipulation + + + + + Count the number of logical 1 bits in packed 64-bit integers in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). +DEFINE POPCNT(a) { + count := 0 + DO WHILE a > 0 + count += a[0] + a >>= 1 + OD + RETURN count +} FOR j := 0 to 7 - i := j*32 - IF k1[j] - k[j] := ( a[i+31:i] > b[i+31:i] ) ? 1 : 0 - ELSE - k[j] := 0 + i := j*64 + IF k[j] + dst[i+63:i] := POPCNT(a[i+63:i]) + ELSE + dst[i+63:i] := src[i+63:i] FI ENDFOR -k[MAX:8] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - Mask - AVX512VL - AVX512F - Compare - - - - Compare packed 32-bit integers in "a" and "b" for less-than-or-equal, and store the results in mask vector "k1" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + AVX512VPOPCNTDQ + Bit Manipulation + + + + Count the number of logical 1 bits in packed 64-bit integers in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). +DEFINE POPCNT(a) { + count := 0 + DO WHILE a > 0 + count += a[0] + a >>= 1 + OD + RETURN count +} FOR j := 0 to 7 - i := j*32 - IF k1[j] - k[j] := ( a[i+31:i] <= b[i+31:i] ) ? 1 : 0 - ELSE - k[j] := 0 + i := j*64 + IF k[j] + dst[i+63:i] := POPCNT(a[i+63:i]) + ELSE + dst[i+63:i] := 0 FI ENDFOR -k[MAX:8] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - Mask - AVX512VL - AVX512F - Compare - - - - Compare packed 32-bit integers in "a" and "b" for less-than, and store the results in mask vector "k1" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Floating Point + AVX512_4FMAPS + Arithmetic + + + + + + + + Multiply packed single-precision (32-bit) floating-point elements specified in 4 consecutive operands "a0" through "a3" by the 4 corresponding packed elements in "b", accumulate with the corresponding elements in "src", and store the results in "dst". -FOR j := 0 to 7 - i := j*32 - IF k1[j] - k[j] := ( a[i+31:i] < b[i+31:i] ) ? 1 : 0 - ELSE - k[j] := 0 - FI +dst[511:0] := src[511:0] +FOR i := 0 to 15 + FOR m := 0 to 3 + addr := b + m * 32 + dst.fp32[i] := dst.fp32[i] + a{m}.fp32[i] * Cast_FP32(MEM[addr+31:addr]) + ENDFOR ENDFOR -k[MAX:8] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - Mask - AVX512VL - AVX512F - Compare - - - - Compare packed 32-bit integers in "a" and "b" for not-equal, and store the results in mask vector "k1" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Floating Point + AVX512_4FMAPS + Arithmetic + + + + + + + + + Multiply packed single-precision (32-bit) floating-point elements specified in 4 consecutive operands "a0" through "a3" by the 4 corresponding packed elements in "b", accumulate with the corresponding elements in "src", and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := j*32 - IF k1[j] - k[j] := ( a[i+31:i] != b[i+31:i] ) ? 1 : 0 - ELSE - k[j] := 0 - FI +dst[511:0] := src[511:0] +FOR i := 0 to 15 + FOR m := 0 to 3 + addr := b + m * 32 + IF k[i] + dst.fp32[i] := dst.fp32[i] + a{m}.fp32[i] * Cast_FP32(MEM[addr+31:addr]) + FI + ENDFOR ENDFOR -k[MAX:8] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - Mask - AVX512VL - AVX512F - Compare - - - - Compare packed 32-bit integers in "a" and "b" based on the comparison operand specified by "imm8", and store the results in mask vector "k". + + Floating Point + AVX512_4FMAPS + Arithmetic + + + + + + + + + Multiply packed single-precision (32-bit) floating-point elements specified in 4 consecutive operands "a0" through "a3" by the 4 corresponding packed elements in "b", accumulate with the corresponding elements in "src", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -CASE (imm8[7:0]) OF -0: OP := _MM_CMPINT_EQ -1: OP := _MM_CMPINT_LT -2: OP := _MM_CMPINT_LE -3: OP := _MM_CMPINT_FALSE -4: OP := _MM_CMPINT_NE -5: OP := _MM_CMPINT_NLT -6: OP := _MM_CMPINT_NLE -7: OP := _MM_CMPINT_TRUE -ESAC -FOR j := 0 to 3 - i := j*32 - k[j] := ( a[i+31:i] OP b[i+31:i] ) ? 1 : 0 +dst[511:0] := src[511:0] +FOR i := 0 to 15 + FOR m := 0 to 3 + addr := b + m * 32 + IF k[i] + dst.fp32[i] := dst.fp32[i] + a{m}.fp32[i] * Cast_FP32(MEM[addr+31:addr]) + ELSE + dst.fp32[i] := 0 + FI + ENDFOR ENDFOR -k[MAX:4] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - Mask - AVX512VL - AVX512F - Compare - - - Compare packed 32-bit integers in "a" and "b" for equality, and store the results in mask vector "k". + + Floating Point + AVX512_4FMAPS + Arithmetic + + + + + + + + Multiply packed single-precision (32-bit) floating-point elements specified in 4 consecutive operands "a0" through "a3" by the 4 corresponding packed elements in "b", accumulate the negated intermediate result with the corresponding elements in "src", and store the results in "dst". -FOR j := 0 to 3 - i := j*32 - k[j] := ( a[i+31:i] == b[i+31:i] ) ? 1 : 0 +dst[511:0] := src[511:0] +FOR i := 0 to 15 + FOR m := 0 to 3 + addr := b + m * 32 + dst.fp32[i] := dst.fp32[i] - a{m}.fp32[i] * Cast_FP32(MEM[addr+31:addr]) + ENDFOR ENDFOR -k[MAX:4] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - Mask - AVX512VL - AVX512F - Compare - - - Compare packed 32-bit integers in "a" and "b" for greater-than-or-equal, and store the results in mask vector "k". + + Floating Point + AVX512_4FMAPS + Arithmetic + + + + + + + + + Multiply packed single-precision (32-bit) floating-point elements specified in 4 consecutive operands "a0" through "a3" by the 4 corresponding packed elements in "b", accumulate the negated intermediate result with the corresponding elements in "src", and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*32 - k[j] := ( a[i+31:i] >= b[i+31:i] ) ? 1 : 0 +dst[511:0] := src[511:0] +FOR i := 0 to 15 + FOR m := 0 to 3 + addr := b + m * 32 + IF k[i] + dst.fp32[i] := dst.fp32[i] - a{m}.fp32[i] * Cast_FP32(MEM[addr+31:addr]) + FI + ENDFOR ENDFOR -k[MAX:4] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - Mask - AVX512VL - AVX512F - Compare - - - Compare packed 32-bit integers in "a" and "b" for greater-than, and store the results in mask vector "k". + + Floating Point + AVX512_4FMAPS + Arithmetic + + + + + + + + + Multiply packed single-precision (32-bit) floating-point elements specified in 4 consecutive operands "a0" through "a3" by the 4 corresponding packed elements in "b", accumulate the negated intermediate result with the corresponding elements in "src", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*32 - k[j] := ( a[i+31:i] > b[i+31:i] ) ? 1 : 0 +dst[511:0] := src[511:0] +FOR i := 0 to 15 + FOR m := 0 to 3 + addr := b + m * 32 + IF k[i] + dst.fp32[i] := dst.fp32[i] - a{m}.fp32[i] * Cast_FP32(MEM[addr+31:addr]) + ELSE + dst.fp32[i] := 0 + FI + ENDFOR ENDFOR -k[MAX:4] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - Mask - AVX512VL - AVX512F - Compare - - - Compare packed 32-bit integers in "a" and "b" for less-than-or-equal, and store the results in mask vector "k". + + Floating Point + AVX512_4FMAPS + Arithmetic + + + + + + + + Multiply the lower single-precision (32-bit) floating-point elements specified in 4 consecutive operands "a0" through "a3" by corresponding element in "b", accumulate with the lower element in "a", and store the result in the lower element of "dst". -FOR j := 0 to 3 - i := j*32 - k[j] := ( a[i+31:i] <= b[i+31:i] ) ? 1 : 0 +dst[127:0] := src[127:0] +FOR m := 0 to 3 + addr := b + m * 32 + dst.fp32[0] := dst.fp32[0] + a{m}.fp32[0] * Cast_FP32(MEM[addr+31:addr]) ENDFOR -k[MAX:4] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - - Integer - Mask - AVX512VL - AVX512F - Compare - - - Compare packed 32-bit integers in "a" and "b" for less-than, and store the results in mask vector "k". - -FOR j := 0 to 3 - i := j*32 - k[j] := ( a[i+31:i] < b[i+31:i] ) ? 1 : 0 -ENDFOR -k[MAX:4] := 0 + + Floating Point + AVX512_4FMAPS + Arithmetic + + + + + + + + + Multiply the lower single-precision (32-bit) floating-point elements specified in 4 consecutive operands "a0" through "a3" by corresponding element in "b", accumulate with the lower element in "a", and store the result in the lower element of "dst" using writemask "k" (the element is copied from "a" when mask bit 0 is not set). + +dst[127:0] := src[127:0] +IF k[0] + FOR m := 0 to 3 + addr := b + m * 32 + dst.fp32[0] := dst.fp32[0] + a{m}.fp32[0] * Cast_FP32(MEM[addr+31:addr]) + ENDFOR +FI +dst[MAX:128] := 0 - +
immintrin.h
 - - Integer - Mask - AVX512VL - AVX512F - Compare - - - Compare packed 32-bit integers in "a" and "b" for not-equal, and store the results in mask vector "k". - -FOR j := 0 to 3 - i := j*32 - k[j] := ( a[i+31:i] != b[i+31:i] ) ? 1 : 0 -ENDFOR -k[MAX:4] := 0 + + Floating Point + AVX512_4FMAPS + Arithmetic + + + + + + + + + Multiply the lower single-precision (32-bit) floating-point elements specified in 4 consecutive operands "a0" through "a3" by corresponding element in "b", accumulate with the lower element in "a", and store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set). + +dst[127:0] := src[127:0] +IF k[0] + FOR m := 0 to 3 + addr := b + m * 32 + dst.fp32[0] := dst.fp32[0] + a{m}.fp32[0] * Cast_FP32(MEM[addr+31:addr]) + ENDFOR +ELSE + dst.fp32[0] := 0 +FI +dst[MAX:128] := 0 - +
immintrin.h
 - - Integer - Mask - AVX512VL - AVX512F - Compare - - - - - Compare packed 32-bit integers in "a" and "b" based on the comparison operand specified by "imm8", and store the results in mask vector "k1" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Floating Point + AVX512_4FMAPS + Arithmetic + + + + + + + + Multiply the lower single-precision (32-bit) floating-point elements specified in 4 consecutive operands "a0" through "a3" by corresponding element in "b", accumulate the negated intermediate result with the lower element in "src", and store the result in the lower element of "dst". -CASE (imm8[7:0]) OF -0: OP := _MM_CMPINT_EQ -1: OP := _MM_CMPINT_LT -2: OP := _MM_CMPINT_LE -3: OP := _MM_CMPINT_FALSE -4: OP := _MM_CMPINT_NE -5: OP := _MM_CMPINT_NLT -6: OP := _MM_CMPINT_NLE -7: OP := _MM_CMPINT_TRUE -ESAC -FOR j := 0 to 3 - i := j*32 - IF k1[j] - k[j] := ( a[i+31:i] OP b[i+31:i] ) ? 1 : 0 - ELSE - k[j] := 0 - FI +dst[127:0] := src[127:0] +FOR m := 0 to 3 + addr := b + m * 32 + dst.fp32[0] := dst.fp32[0] - a{m}.fp32[0] * Cast_FP32(MEM[addr+31:addr]) ENDFOR -k[MAX:4] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - - Integer - Mask - AVX512VL - AVX512F - Compare - - - - Compare packed 32-bit integers in "a" and "b" for equality, and store the results in mask vector "k1" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - -FOR j := 0 to 3 - i := j*32 - IF k1[j] - k[j] := ( a[i+31:i] == b[i+31:i] ) ? 1 : 0 - ELSE - k[j] := 0 - FI -ENDFOR -k[MAX:4] := 0 + + Floating Point + AVX512_4FMAPS + Arithmetic + + + + + + + + + Multiply the lower single-precision (32-bit) floating-point elements specified in 4 consecutive operands "a0" through "a3" by corresponding element in "b", accumulate the negated intermediate result with the lower element in "src", and store the result in the lower element of "dst" using writemask "k" (the element is copied from "a" when mask bit 0 is not set). + +dst[127:0] := src[127:0] +IF k[0] + FOR m := 0 to 3 + addr := b + m * 32 + dst.fp32[0] := dst.fp32[0] - a{m}.fp32[0] * Cast_FP32(MEM[addr+31:addr]) + ENDFOR +FI +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Floating Point + AVX512_4FMAPS + Arithmetic + + + + + + + + + Multiply the lower single-precision (32-bit) floating-point elements specified in 4 consecutive operands "a0" through "a3" by corresponding element in "b", accumulate the negated intermediate result with the lower element in "src", and store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set). + +dst[127:0] := src[127:0] +IF k[0] + FOR m := 0 to 3 + addr := b + m * 32 + dst.fp32[0] := dst.fp32[0] - a{m}.fp32[0] * Cast_FP32(MEM[addr+31:addr]) + ENDFOR +ELSE + dst.fp32[0] := 0 +FI +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - Mask - AVX512VL - AVX512F - Compare - - - - Compare packed 32-bit integers in "a" and "b" for greater-than-or-equal, and store the results in mask vector "k1" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + AVX512_4VNNIW + Arithmetic + + + + + + + + Compute 4 sequential operand source-block dot-products of two signed 16-bit element operands with 32-bit element accumulation, and store the results in "dst". -FOR j := 0 to 3 - i := j*32 - IF k1[j] - k[j] := ( a[i+31:i] >= b[i+31:i] ) ? 1 : 0 - ELSE - k[j] := 0 - FI +dst[511:0] := src[511:0] +FOR i := 0 to 15 + FOR m := 0 to 3 + lim_base := b + m*32 + t.dword := MEM[lim_base+31:lim_base] + p1.dword := SignExtend32(a{m}.word[2*i+0]) * SignExtend32(Cast_Int16(t.word[0])) + p2.dword := SignExtend32(a{m}.word[2*i+1]) * SignExtend32(Cast_Int16(t.word[1])) + dst.dword[i] := dst.dword[i] + p1.dword + p2.dword + ENDFOR ENDFOR -k[MAX:4] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - Mask - AVX512VL - AVX512F - Compare - - - - parameter varname='b' type='__m128i'/> - Compare packed 32-bit integers in "a" and "b" for greater-than, and store the results in mask vector "k1" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + AVX512_4VNNIW + Arithmetic + + + + + + + + + Compute 4 sequential operand source-block dot-products of two signed 16-bit element operands with 32-bit element accumulation with mask, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*32 - IF k1[j] - k[j] := ( a[i+31:i] > b[i+31:i] ) ? 1 : 0 - ELSE - k[j] := 0 +dst[511:0] := src[511:0] +FOR i := 0 to 15 + IF k[i] + FOR m := 0 to 3 + lim_base := b + m*32 + t.dword := MEM[lim_base+31:lim_base] + p1.dword := SignExtend32(a{m}.word[2*i+0]) * SignExtend32(Cast_Int16(t.word[0])) + p2.dword := SignExtend32(a{m}.word[2*i+1]) * SignExtend32(Cast_Int16(t.word[1])) + dst.dword[i] := dst.dword[i] + p1.dword + p2.dword + ENDFOR + ELSE + dst.dword[i] := src.dword[i] FI ENDFOR -k[MAX:4] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - Mask - AVX512VL - AVX512F - Compare - - - - parameter varname='b' type='__m128i'/> - Compare packed 32-bit integers in "a" and "b" for less-than-or-equal, and store the results in mask vector "k1" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + AVX512_4VNNIW + Arithmetic + + + + + + + + + Compute 4 sequential operand source-block dot-products of two signed 16-bit element operands with 32-bit element accumulation with mask, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*32 - IF k1[j] - k[j] := ( a[i+31:i] <= b[i+31:i] ) ? 1 : 0 - ELSE - k[j] := 0 +dst[511:0] := src[511:0] +FOR i := 0 to 15 + IF k[i] + FOR m := 0 to 3 + lim_base := b + m*32 + t.dword := MEM[lim_base+31:lim_base] + p1.dword := SignExtend32(a{m}.word[2*i+0]) * SignExtend32(Cast_Int16(t.word[0])) + p2.dword := SignExtend32(a{m}.word[2*i+1]) * SignExtend32(Cast_Int16(t.word[1])) + dst.dword[i] := dst.dword[i] + p1.dword + p2.dword + ENDFOR + ELSE + dst.dword[i] := 0 FI ENDFOR -k[MAX:4] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - Mask - AVX512VL - AVX512F - Compare - - - - parameter varname='b' type='__m128i'/> - Compare packed 32-bit integers in "a" and "b" for less-than, and store the results in mask vector "k1" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + AVX512_4VNNIW + Arithmetic + + + + + + + + Compute 4 sequential operand source-block dot-products of two signed 16-bit element operands with 32-bit element accumulation and signed saturation, and store the results in "dst". -FOR j := 0 to 3 - i := j*32 - IF k1[j] - k[j] := ( a[i+31:i] < b[i+31:i] ) ? 1 : 0 - ELSE - k[j] := 0 - FI +dst[511:0] := src[511:0] +FOR i := 0 to 15 + FOR m := 0 to 3 + lim_base := b + m*32 + t.dword := MEM[lim_base+31:lim_base] + p1.dword := SignExtend32(a{m}.word[2*i+0]) * SignExtend32(Cast_Int16(t.word[0])) + p2.dword := SignExtend32(a{m}.word[2*i+1]) * SignExtend32(Cast_Int16(t.word[1])) + dst.dword[i] := Saturate32(dst.dword[i] + p1.dword + p2.dword) + ENDFOR ENDFOR -k[MAX:4] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - Mask - AVX512VL - AVX512F - Compare - - - - parameter varname='b' type='__m128i'/> - Compare packed 32-bit integers in "a" and "b" for not-equal, and store the results in mask vector "k1" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + AVX512_4VNNIW + Arithmetic + + + + + + + + + Compute 4 sequential operand source-block dot-products of two signed 16-bit element operands with 32-bit element accumulation with mask and signed saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set).. -FOR j := 0 to 3 - i := j*32 - IF k1[j] - k[j] := ( a[i+31:i] != b[i+31:i] ) ? 1 : 0 - ELSE - k[j] := 0 +dst[511:0] := src[511:0] +FOR i := 0 to 15 + IF k[i] + FOR m := 0 to 3 + lim_base := b + m*32 + t.dword := MEM[lim_base+31:lim_base] + p1.dword := SignExtend32(a{m}.word[2*i+0]) * SignExtend32(Cast_Int16(t.word[0])) + p2.dword := SignExtend32(a{m}.word[2*i+1]) * SignExtend32(Cast_Int16(t.word[1])) + dst.dword[i] := Saturate32(dst.dword[i] + p1.dword + p2.dword) + ENDFOR + ELSE + dst.dword[i] := src.dword[i] FI ENDFOR -k[MAX:4] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - Mask - AVX512VL - AVX512F - Compare - - - - Compare packed 64-bit integers in "a" and "b" based on the comparison operand specified by "imm8", and store the results in mask vector "k". + AVX512_4VNNIW + Arithmetic + + + + + + + + + Compute 4 sequential operand source-block dot-products of two signed 16-bit element operands with 32-bit element accumulation with mask and signed saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set).. -CASE (imm8[7:0]) OF -0: OP := _MM_CMPINT_EQ -1: OP := _MM_CMPINT_LT -2: OP := _MM_CMPINT_LE -3: OP := _MM_CMPINT_FALSE -4: OP := _MM_CMPINT_NE -5: OP := _MM_CMPINT_NLT -6: OP := _MM_CMPINT_NLE -7: OP := _MM_CMPINT_TRUE -ESAC -FOR j := 0 to 3 - i := j*64 - k[j] := ( a[i+63:i] OP b[i+63:i] ) ? 1 : 0 +dst[511:0] := src[511:0] +FOR i := 0 to 15 + IF k[i] + FOR m := 0 to 3 + lim_base := b + m*32 + t.dword := MEM[lim_base+31:lim_base] + p1.dword := SignExtend32(a{m}.word[2*i+0]) * SignExtend32(Cast_Int16(t.word[0])) + p2.dword := SignExtend32(a{m}.word[2*i+1]) * SignExtend32(Cast_Int16(t.word[1])) + dst.dword[i] := Saturate32(dst.dword[i] + p1.dword + p2.dword) + ENDFOR + ELSE + dst.dword[i] := 0 + FI ENDFOR -k[MAX:4] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - Mask + + Floating Point + AVX512_BF16 AVX512VL - AVX512F - Compare - - - Compare packed 64-bit integers in "a" and "b" for equality, and store the results in mask vector "k". + Convert + + + + Convert packed single-precision (32-bit) floating-point elements in two vectors "a" and "b" to packed BF16 (16-bit) floating-point elements, and store the results in single vector "dst". -FOR j := 0 to 3 - i := j*64 - k[j] := ( a[i+63:i] == b[i+63:i] ) ? 1 : 0 +FOR j := 0 to 7 + IF j < 4 + t := b.fp32[j] + ELSE + t := a.fp32[j-4] + FI + dst.word[j] := Convert_FP32_To_BF16(t) ENDFOR -k[MAX:4] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - - Integer - Mask + + Floating Point + AVX512_BF16 AVX512VL - AVX512F - Compare - - - Compare packed 64-bit integers in "a" and "b" for greater-than-or-equal, and store the results in mask vector "k". + Convert + + + + + + Convert packed single-precision (32-bit) floating-point elements in two vectors "a" and "b" to packed BF16 (16-bit) floating-point elements, and store the results in single vector "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*64 - k[j] := ( a[i+63:i] >= b[i+63:i] ) ? 1 : 0 +FOR j := 0 to 7 + IF k[j] + IF j < 4 + t := b.fp32[j] + ELSE + t := a.fp32[j-4] + FI + dst.word[j] := Convert_FP32_To_BF16(t) + ELSE + dst.word[j] := src.word[j] + FI ENDFOR -k[MAX:4] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - - Integer - Mask + + Floating Point + AVX512_BF16 AVX512VL - AVX512F - Compare - - - Compare packed 64-bit integers in "a" and "b" for greater-than, and store the results in mask vector "k". + Convert + + + + + Convert packed single-precision (32-bit) floating-point elements in two vectors "a" and "b" to packed BF16 (16-bit) floating-point elements, and store the results in single vector "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*64 - k[j] := ( a[i+63:i] > b[i+63:i] ) ? 1 : 0 +FOR j := 0 to 7 + IF k[j] + IF j < 4 + t := b.fp32[j] + ELSE + t := a.fp32[j-4] + FI + dst.word[j] := Convert_FP32_To_BF16(t) + ELSE + dst.word[j] := 0 + FI ENDFOR -k[MAX:4] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - - Integer - Mask + + Floating Point + AVX512_BF16 AVX512VL - AVX512F - Compare - - - Compare packed 64-bit integers in "a" and "b" for less-than-or-equal, and store the results in mask vector "k". + Convert + + + + Convert packed single-precision (32-bit) floating-point elements in two vectors "a" and "b" to packed BF16 (16-bit) floating-point elements, and store the results in single vector "dst". -FOR j := 0 to 3 - i := j*64 - k[j] := ( a[i+63:i] <= b[i+63:i] ) ? 1 : 0 +FOR j := 0 to 15 + IF j < 8 + t := b.fp32[j] + ELSE + t := a.fp32[j-8] + FI + dst.word[j] := Convert_FP32_To_BF16(t) ENDFOR -k[MAX:4] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - - Integer - Mask + + Floating Point + AVX512_BF16 AVX512VL - AVX512F - Compare - - - Compare packed 64-bit integers in "a" and "b" for less-than, and store the results in mask vector "k". + Convert + + + + + + Convert packed single-precision (32-bit) floating-point elements in two vectors "a" and "b" to packed BF16 (16-bit) floating-point elements, and store the results in single vector "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*64 - k[j] := ( a[i+63:i] < b[i+63:i] ) ? 1 : 0 +FOR j := 0 to 15 + IF k[j] + IF j < 8 + t := b.fp32[j] + ELSE + t := a.fp32[j-8] + FI + dst.word[j] := Convert_FP32_To_BF16(t) + ELSE + dst.word[j] := src.word[j] + FI ENDFOR -k[MAX:4] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - - Integer - Mask + + Floating Point + AVX512_BF16 AVX512VL - AVX512F - Compare - - - Compare packed 64-bit integers in "a" and "b" for not-equal, and store the results in mask vector "k". + Convert + + + + + Convert packed single-precision (32-bit) floating-point elements in two vectors "a" and "b" to packed BF16 (16-bit) floating-point elements, and store the results in single vector "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*64 - k[j] := ( a[i+63:i] != b[i+63:i] ) ? 1 : 0 +FOR j := 0 to 15 + IF k[j] + IF j < 8 + t := b.fp32[j] + ELSE + t := a.fp32[j-8] + FI + dst.word[j] := Convert_FP32_To_BF16(t) + ELSE + dst.word[j] := 0 + FI ENDFOR -k[MAX:4] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - - Integer - Mask - AVX512VL + + Floating Point + AVX512_BF16 AVX512F - Compare - - - - - Compare packed 64-bit integers in "a" and "b" based on the comparison operand specified by "imm8", and store the results in mask vector "k1" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Convert + + + + Convert packed single-precision (32-bit) floating-point elements in two vectors "a" and "b" to packed BF16 (16-bit) floating-point elements, and store the results in single vector "dst". -CASE (imm8[7:0]) OF -0: OP := _MM_CMPINT_EQ -1: OP := _MM_CMPINT_LT -2: OP := _MM_CMPINT_LE -3: OP := _MM_CMPINT_FALSE -4: OP := _MM_CMPINT_NE -5: OP := _MM_CMPINT_NLT -6: OP := _MM_CMPINT_NLE -7: OP := _MM_CMPINT_TRUE -ESAC -FOR j := 0 to 3 - i := j*64 - IF k1[j] - k[j] := ( a[i+63:i] OP b[i+63:i] ) ? 1 : 0 - ELSE - k[j] := 0 +FOR j := 0 to 31 + IF j < 16 + t := b.fp32[j] + ELSE + t := a.fp32[j-16] FI + dst.word[j] := Convert_FP32_To_BF16(t) ENDFOR -k[MAX:4] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - Mask - AVX512VL + + Floating Point + AVX512_BF16 AVX512F - Compare - - - - Compare packed 64-bit integers in "a" and "b" for equality, and store the results in mask vector "k1" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Convert + + + + + + Convert packed single-precision (32-bit) floating-point elements in two vectors "a" and "b" to packed BF16 (16-bit) floating-point elements, and store the results in single vector "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*64 - IF k1[j] - k[j] := ( a[i+63:i] == b[i+63:i] ) ? 1 : 0 - ELSE - k[j] := 0 +FOR j := 0 to 31 + IF k[j] + IF j < 16 + t := b.fp32[j] + ELSE + t := a.fp32[j-16] + FI + dst.word[j] := Convert_FP32_To_BF16(t) + ELSE + dst.word[j] := src.word[j] FI ENDFOR -k[MAX:4] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - Mask - AVX512VL + + Floating Point + AVX512_BF16 AVX512F - Compare - - - - Compare packed 64-bit integers in "a" and "b" for greater-than-or-equal, and store the results in mask vector "k1" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Convert + + + + + Convert packed single-precision (32-bit) floating-point elements in two vectors "a" and "b" to packed BF16 (16-bit) floating-point elements, and store the results in single vector "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*64 - IF k1[j] - k[j] := ( a[i+63:i] >= b[i+63:i] ) ? 1 : 0 - ELSE - k[j] := 0 +FOR j := 0 to 31 + IF k[j] + IF j < 16 + t := b.fp32[j] + ELSE + t := a.fp32[j-16] + FI + dst.word[j] := Convert_FP32_To_BF16(t) + ELSE + dst.word[j] := 0 FI ENDFOR -k[MAX:4] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - Mask + + Floating Point + AVX512_BF16 AVX512VL - AVX512F - Compare - - - - Compare packed 64-bit integers in "a" and "b" for greater-than, and store the results in mask vector "k1" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Convert + + + Convert packed single-precision (32-bit) floating-point elements in "a" to packed BF16 (16-bit) floating-point elements, and store the results in "dst". FOR j := 0 to 3 - i := j*64 - IF k1[j] - k[j] := ( a[i+63:i] > b[i+63:i] ) ? 1 : 0 - ELSE - k[j] := 0 - FI + dst.word[j] := Convert_FP32_To_BF16(a.fp32[j]) ENDFOR -k[MAX:4] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - - Integer - Mask + + Floating Point + AVX512_BF16 AVX512VL - AVX512F - Compare - - - - Compare packed 64-bit integers in "a" and "b" for less-than-or-equal, and store the results in mask vector "k1" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Convert + + + + + Convert packed single-precision (32-bit) floating-point elements in "a" to packed BF16 (16-bit) floating-point elements, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). FOR j := 0 to 3 - i := j*64 - IF k1[j] - k[j] := ( a[i+63:i] <= b[i+63:i] ) ? 1 : 0 - ELSE - k[j] := 0 + IF k[j] + dst.word[j] := Convert_FP32_To_BF16(a.fp32[j]) + ELSE + dst.word[j] := src.word[j] FI ENDFOR -k[MAX:4] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - - Integer - Mask + + Floating Point + AVX512_BF16 AVX512VL - AVX512F - Compare - - - - Compare packed 64-bit integers in "a" and "b" for less-than, and store the results in mask vector "k1" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Convert + + + + Convert packed single-precision (32-bit) floating-point elements in "a" to packed BF16 (16-bit) floating-point elements, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). FOR j := 0 to 3 - i := j*64 - IF k1[j] - k[j] := ( a[i+63:i] < b[i+63:i] ) ? 1 : 0 - ELSE - k[j] := 0 + IF k[j] + dst.word[j] := Convert_FP32_To_BF16(a.fp32[j]) + ELSE + dst.word[j] := 0 FI ENDFOR -k[MAX:4] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - - Integer - Mask + + Floating Point + AVX512_BF16 AVX512VL - AVX512F - Compare - - - - Compare packed 64-bit integers in "a" and "b" for not-equal, and store the results in mask vector "k1" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Convert + + + Convert packed single-precision (32-bit) floating-point elements in "a" to packed BF16 (16-bit) floating-point elements, and store the results in "dst". -FOR j := 0 to 3 - i := j*64 - IF k1[j] - k[j] := ( a[i+63:i] != b[i+63:i] ) ? 1 : 0 - ELSE - k[j] := 0 - FI +FOR j := 0 to 7 + dst.word[j] := Convert_FP32_To_BF16(a.fp32[j]) ENDFOR -k[MAX:4] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - - Integer - Mask + + Floating Point + AVX512_BF16 AVX512VL - AVX512F - Compare - - - - Compare packed 64-bit integers in "a" and "b" based on the comparison operand specified by "imm8", and store the results in mask vector "k". + Convert + + + + + Convert packed single-precision (32-bit) floating-point elements in "a" to packed BF16 (16-bit) floating-point elements, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -CASE (imm8[7:0]) OF -0: OP := _MM_CMPINT_EQ -1: OP := _MM_CMPINT_LT -2: OP := _MM_CMPINT_LE -3: OP := _MM_CMPINT_FALSE -4: OP := _MM_CMPINT_NE -5: OP := _MM_CMPINT_NLT -6: OP := _MM_CMPINT_NLE -7: OP := _MM_CMPINT_TRUE -ESAC -FOR j := 0 to 1 - i := j*64 - k[j] := ( a[i+63:i] OP b[i+63:i] ) ? 1 : 0 +FOR j := 0 to 7 + IF k[j] + dst.word[j] := Convert_FP32_To_BF16(a.fp32[j]) + ELSE + dst.word[j] := src.word[j] + FI ENDFOR -k[MAX:2] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - - Integer - Mask + + Floating Point + AVX512_BF16 AVX512VL - AVX512F - Compare - - - Compare packed 64-bit integers in "a" and "b" for equality, and store the results in mask vector "k". + Convert + + + + Convert packed single-precision (32-bit) floating-point elements in "a" to packed BF16 (16-bit) floating-point elements, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 1 - i := j*64 - k[j] := ( a[i+63:i] == b[i+63:i] ) ? 1 : 0 +FOR j := 0 to 7 + IF k[j] + dst.word[j] := Convert_FP32_To_BF16(a.fp32[j]) + ELSE + dst.word[j] := 0 + FI ENDFOR -k[MAX:2] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - - Integer - Mask - AVX512VL + + Floating Point + AVX512_BF16 AVX512F - Compare - - - Compare packed 64-bit integers in "a" and "b" for greater-than-or-equal, and store the results in mask vector "k". + Convert + + + Convert packed single-precision (32-bit) floating-point elements in "a" to packed BF16 (16-bit) floating-point elements, and store the results in "dst". -FOR j := 0 to 1 - i := j*64 - k[j] := ( a[i+63:i] >= b[i+63:i] ) ? 1 : 0 +FOR j := 0 to 15 + dst.word[j] := Convert_FP32_To_BF16(a.fp32[j]) ENDFOR -k[MAX:2] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - - Integer - Mask - AVX512VL + + Floating Point + AVX512_BF16 AVX512F - Compare - - - Compare packed 64-bit integers in "a" and "b" for greater-than, and store the results in mask vector "k". + Convert + + + + + Convert packed single-precision (32-bit) floating-point elements in "a" to packed BF16 (16-bit) floating-point elements, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 1 - i := j*64 - k[j] := ( a[i+63:i] > b[i+63:i] ) ? 1 : 0 +FOR j := 0 to 15 + IF k[j] + dst.word[j] := Convert_FP32_To_BF16(a.fp32[j]) + ELSE + dst.word[j] := src.word[j] + FI ENDFOR -k[MAX:2] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - - Integer - Mask - AVX512VL + + Floating Point + AVX512_BF16 AVX512F - Compare - - - Compare packed 64-bit integers in "a" and "b" for less-than-or-equal, and store the results in mask vector "k". + Convert + + + + Convert packed single-precision (32-bit) floating-point elements in "a" to packed BF16 (16-bit) floating-point elements, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 1 - i := j*64 - k[j] := ( a[i+63:i] <= b[i+63:i] ) ? 1 : 0 +FOR j := 0 to 15 + IF k[j] + dst.word[j] := Convert_FP32_To_BF16(a.fp32[j]) + ELSE + dst.word[j] := 0 + FI ENDFOR -k[MAX:2] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - - Integer - Mask + + Floating Point + AVX512_BF16 AVX512VL - AVX512F - Compare - - - Compare packed 64-bit integers in "a" and "b" for less-than, and store the results in mask vector "k". + Arithmetic + + + + + Compute dot-product of BF16 (16-bit) floating-point pairs in "a" and "b", accumulating the intermediate single-precision (32-bit) floating-point elements with elements in "src", and store the results in "dst". -FOR j := 0 to 1 - i := j*64 - k[j] := ( a[i+63:i] < b[i+63:i] ) ? 1 : 0 +DEFINE make_fp32(x[15:0]) { + y.fp32 := 0.0 + y[31:16] := x[15:0] + RETURN y +} +dst := src +FOR j := 0 to 3 + dst.fp32[j] += make_fp32(a.bf16[2*j+1]) * make_fp32(b.bf16[2*j+1]) + dst.fp32[j] += make_fp32(a.bf16[2*j+0]) * make_fp32(b.bf16[2*j+0]) ENDFOR -k[MAX:2] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - - Integer - Mask + + Floating Point + AVX512_BF16 AVX512VL - AVX512F - Compare - - - Compare packed 64-bit integers in "a" and "b" for not-equal, and store the results in mask vector "k". + Arithmetic + + + + + + Compute dot-product of BF16 (16-bit) floating-point pairs in "a" and "b", accumulating the intermediate single-precision (32-bit) floating-point elements with elements in "src", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 1 - i := j*64 - k[j] := ( a[i+63:i] != b[i+63:i] ) ? 1 : 0 +DEFINE make_fp32(x[15:0]) { + y.fp32 := 0.0 + y[31:16] := x[15:0] + RETURN y +} +dst := src +FOR j := 0 to 3 + IF k[j] + dst.fp32[j] += make_fp32(a.bf16[2*j+1]) * make_fp32(b.bf16[2*j+1]) + dst.fp32[j] += make_fp32(a.bf16[2*j+0]) * make_fp32(b.bf16[2*j+0]) + ELSE + dst.dword[j] := src.dword[j] + FI ENDFOR -k[MAX:2] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - - Integer - Mask + + Floating Point + AVX512_BF16 AVX512VL - AVX512F - Compare - - - - - Compare packed 64-bit integers in "a" and "b" based on the comparison operand specified by "imm8", and store the results in mask vector "k1" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Arithmetic + + + + + + Compute dot-product of BF16 (16-bit) floating-point pairs in "a" and "b", accumulating the intermediate single-precision (32-bit) floating-point elements with elements in "src", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -CASE (imm8[7:0]) OF -0: OP := _MM_CMPINT_EQ -1: OP := _MM_CMPINT_LT -2: OP := _MM_CMPINT_LE -3: OP := _MM_CMPINT_FALSE -4: OP := _MM_CMPINT_NE -5: OP := _MM_CMPINT_NLT -6: OP := _MM_CMPINT_NLE -7: OP := _MM_CMPINT_TRUE -ESAC -FOR j := 0 to 1 - i := j*64 - IF k1[j] - k[j] := ( a[i+63:i] OP b[i+63:i] ) ? 1 : 0 - ELSE - k[j] := 0 +DEFINE make_fp32(x[15:0]) { + y.fp32 := 0.0 + y[31:16] := x[15:0] + RETURN y +} +dst := src +FOR j := 0 to 3 + IF k[j] + dst.fp32[j] += make_fp32(a.bf16[2*j+1]) * make_fp32(b.bf16[2*j+1]) + dst.fp32[j] += make_fp32(a.bf16[2*j+0]) * make_fp32(b.bf16[2*j+0]) + ELSE + dst.dword[j] := 0 FI ENDFOR -k[MAX:2] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - - Integer - Mask + + Floating Point + AVX512_BF16 AVX512VL - AVX512F - Compare - - - - Compare packed 64-bit integers in "a" and "b" for equality, and store the results in mask vector "k1" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Arithmetic + + + + + Compute dot-product of BF16 (16-bit) floating-point pairs in "a" and "b", accumulating the intermediate single-precision (32-bit) floating-point elements with elements in "src", and store the results in "dst". -FOR j := 0 to 1 - i := j*64 - IF k1[j] - k[j] := ( a[i+63:i] == b[i+63:i] ) ? 1 : 0 - ELSE - k[j] := 0 - FI +DEFINE make_fp32(x[15:0]) { + y.fp32 := 0.0 + y[31:16] := x[15:0] + RETURN y +} +dst := src +FOR j := 0 to 7 + dst.fp32[j] += make_fp32(a.bf16[2*j+1]) * make_fp32(b.bf16[2*j+1]) + dst.fp32[j] += make_fp32(a.bf16[2*j+0]) * make_fp32(b.bf16[2*j+0]) ENDFOR -k[MAX:2] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - - Integer - Mask + + Floating Point + AVX512_BF16 AVX512VL - AVX512F - Compare - - - - Compare packed 64-bit integers in "a" and "b" for greater-than-or-equal, and store the results in mask vector "k1" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Arithmetic + + + + + + Compute dot-product of BF16 (16-bit) floating-point pairs in "a" and "b", accumulating the intermediate single-precision (32-bit) floating-point elements with elements in "src", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 1 - i := j*64 - IF k1[j] - k[j] := ( a[i+63:i] >= b[i+63:i] ) ? 1 : 0 - ELSE - k[j] := 0 +DEFINE make_fp32(x[15:0]) { + y.fp32 := 0.0 + y[31:16] := x[15:0] + RETURN y +} +dst := src +FOR j := 0 to 7 + IF k[j] + dst.fp32[j] += make_fp32(a.bf16[2*j+1]) * make_fp32(b.bf16[2*j+1]) + dst.fp32[j] += make_fp32(a.bf16[2*j+0]) * make_fp32(b.bf16[2*j+0]) + ELSE + dst.dword[j] := src.dword[j] FI ENDFOR -k[MAX:2] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - - Integer - Mask + + Floating Point + AVX512_BF16 AVX512VL - AVX512F - Compare - - - - Compare packed 64-bit integers in "a" and "b" for greater-than, and store the results in mask vector "k1" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Arithmetic + + + + + + Compute dot-product of BF16 (16-bit) floating-point pairs in "a" and "b", accumulating the intermediate single-precision (32-bit) floating-point elements with elements in "src", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 1 - i := j*64 - IF k1[j] - k[j] := ( a[i+63:i] > b[i+63:i] ) ? 1 : 0 - ELSE - k[j] := 0 +DEFINE make_fp32(x[15:0]) { + y.fp32 := 0.0 + y[31:16] := x[15:0] + RETURN y +} +dst := src +FOR j := 0 to 7 + IF k[j] + dst.fp32[j] += make_fp32(a.bf16[2*j+1]) * make_fp32(b.bf16[2*j+1]) + dst.fp32[j] += make_fp32(a.bf16[2*j+0]) * make_fp32(b.bf16[2*j+0]) + ELSE + dst.dword[j] := 0 FI ENDFOR -k[MAX:2] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - - Integer - Mask - AVX512VL + + Floating Point + AVX512_BF16 AVX512F - Compare - - - - Compare packed 64-bit integers in "a" and "b" for less-than-or-equal, and store the results in mask vector "k1" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Arithmetic + + + + + Compute dot-product of BF16 (16-bit) floating-point pairs in "a" and "b", accumulating the intermediate single-precision (32-bit) floating-point elements with elements in "src", and store the results in "dst". -FOR j := 0 to 1 - i := j*64 - IF k1[j] - k[j] := ( a[i+63:i] <= b[i+63:i] ) ? 1 : 0 - ELSE - k[j] := 0 - FI +DEFINE make_fp32(x[15:0]) { + y.fp32 := 0.0 + y[31:16] := x[15:0] + RETURN y +} +dst := src +FOR j := 0 to 15 + dst.fp32[j] += make_fp32(a.bf16[2*j+1]) * make_fp32(b.bf16[2*j+1]) + dst.fp32[j] += make_fp32(a.bf16[2*j+0]) * make_fp32(b.bf16[2*j+0]) ENDFOR -k[MAX:2] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - Mask - AVX512VL + + Floating Point + AVX512_BF16 AVX512F - Compare - - - - Compare packed 64-bit integers in "a" and "b" for less-than, and store the results in mask vector "k1" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Arithmetic + + + + + + Compute dot-product of BF16 (16-bit) floating-point pairs in "a" and "b", accumulating the intermediate single-precision (32-bit) floating-point elements with elements in "src", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 1 - i := j*64 - IF k1[j] - k[j] := ( a[i+63:i] < b[i+63:i] ) ? 1 : 0 - ELSE - k[j] := 0 +DEFINE make_fp32(x[15:0]) { + y.fp32 := 0.0 + y[31:16] := x[15:0] + RETURN y +} +dst := src +FOR j := 0 to 15 + IF k[j] + dst.fp32[j] += make_fp32(a.bf16[2*j+1]) * make_fp32(b.bf16[2*j+1]) + dst.fp32[j] += make_fp32(a.bf16[2*j+0]) * make_fp32(b.bf16[2*j+0]) + ELSE + dst.dword[j] := src.dword[j] FI ENDFOR -k[MAX:2] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - Mask - AVX512VL + + Floating Point + AVX512_BF16 AVX512F - Compare - - - - Compare packed 64-bit integers in "a" and "b" for not-equal, and store the results in mask vector "k1" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Arithmetic + + + + + + Compute dot-product of BF16 (16-bit) floating-point pairs in "a" and "b", accumulating the intermediate single-precision (32-bit) floating-point elements with elements in "src", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 1 - i := j*64 - IF k1[j] - k[j] := ( a[i+63:i] != b[i+63:i] ) ? 1 : 0 - ELSE - k[j] := 0 +DEFINE make_fp32(x[15:0]) { + y.fp32 := 0.0 + y[31:16] := x[15:0] + RETURN y +} +dst := src +FOR j := 0 to 15 + IF k[j] + dst.fp32[j] += make_fp32(a.bf16[2*j+1]) * make_fp32(b.bf16[2*j+1]) + dst.fp32[j] += make_fp32(a.bf16[2*j+0]) * make_fp32(b.bf16[2*j+0]) + ELSE + dst.dword[j] := 0 FI ENDFOR -k[MAX:2] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer Mask - AVX512VL - AVX512BW - Compare - - - - Compare packed unsigned 8-bit integers in "a" and "b" based on the comparison operand specified by "imm8", and store the results in mask vector "k". - -CASE (imm8[7:0]) OF -0: OP := _MM_CMPINT_EQ -1: OP := _MM_CMPINT_LT -2: OP := _MM_CMPINT_LE -3: OP := _MM_CMPINT_FALSE -4: OP := _MM_CMPINT_NE -5: OP := _MM_CMPINT_NLT -6: OP := _MM_CMPINT_NLE -7: OP := _MM_CMPINT_TRUE -ESAC -FOR j := 0 to 31 - i := j*8 - k[j] := ( a[i+7:i] OP b[i+7:i] ) ? 1 : 0 + AVX512_BITALG + Bit Manipulation + + + + + Gather 64 bits from "b" using selection bits in "c". For each 64-bit element in "b", gather 8 bits from the 64-bit element in "b" at 8 bit position controlled by the 8 corresponding 8-bit elements of "c", and store the result in the corresponding 8-bit element of "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + +FOR i := 0 to 7 //Qword + FOR j := 0 to 7 // Byte + IF k[i*8+j] + m := c.qword[i].byte[j] & 0x3F + dst[i*8+j] := b.qword[i].bit[m] + ELSE + dst[i*8+j] := 0 + FI + ENDFOR ENDFOR -k[MAX:32] := 0 +dst[MAX:64] := 0 - +
immintrin.h
 - + Integer Mask - AVX512VL - AVX512BW - Compare - - - Compare packed unsigned 8-bit integers in "a" and "b" for equality, and store the results in mask vector "k". - -FOR j := 0 to 31 - i := j*8 - k[j] := ( a[i+7:i] == b[i+7:i] ) ? 1 : 0 + AVX512_BITALG + Bit Manipulation + + + + Gather 64 bits from "b" using selection bits in "c". For each 64-bit element in "b", gather 8 bits from the 64-bit element in "b" at 8 bit position controlled by the 8 corresponding 8-bit elements of "c", and store the result in the corresponding 8-bit element of "dst". + +FOR i := 0 to 7 //Qword + FOR j := 0 to 7 // Byte + m := c.qword[i].byte[j] & 0x3F + dst[i*8+j] := b.qword[i].bit[m] + ENDFOR ENDFOR -k[MAX:32] := 0 +dst[MAX:64] := 0 - +
immintrin.h
 - + Integer Mask + AVX512_BITALG AVX512VL - AVX512BW - Compare - - - Compare packed unsigned 8-bit integers in "a" and "b" for greater-than-or-equal, and store the results in mask vector "k". - -FOR j := 0 to 31 - i := j*8 - k[j] := ( a[i+7:i] >= b[i+7:i] ) ? 1 : 0 + Bit Manipulation + + + + + Gather 64 bits from "b" using selection bits in "c". For each 64-bit element in "b", gather 8 bits from the 64-bit element in "b" at 8 bit position controlled by the 8 corresponding 8-bit elements of "c", and store the result in the corresponding 8-bit element of "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + +FOR i := 0 to 3 //Qword + FOR j := 0 to 7 // Byte + IF k[i*8+j] + m := c.qword[i].byte[j] & 0x3F + dst[i*8+j] := b.qword[i].bit[m] + ELSE + dst[i*8+j] := 0 + FI + ENDFOR ENDFOR -k[MAX:32] := 0 +dst[MAX:32] := 0 - +
immintrin.h
 - + Integer Mask + AVX512_BITALG AVX512VL - AVX512BW - Compare - - - Compare packed unsigned 8-bit integers in "a" and "b" for greater-than, and store the results in mask vector "k". - -FOR j := 0 to 31 - i := j*8 - k[j] := ( a[i+7:i] > b[i+7:i] ) ? 1 : 0 + Bit Manipulation + + + + Gather 64 bits from "b" using selection bits in "c". For each 64-bit element in "b", gather 8 bits from the 64-bit element in "b" at 8 bit position controlled by the 8 corresponding 8-bit elements of "c", and store the result in the corresponding 8-bit element of "dst". + +FOR i := 0 to 3 //Qword + FOR j := 0 to 7 // Byte + m := c.qword[i].byte[j] & 0x3F + dst[i*8+j] := b.qword[i].bit[m] + ENDFOR ENDFOR -k[MAX:32] := 0 +dst[MAX:32] := 0 - +
immintrin.h
 - + Integer Mask + AVX512_BITALG AVX512VL - AVX512BW - Compare - - - Compare packed unsigned 8-bit integers in "a" and "b" for less-than-or-equal, and store the results in mask vector "k". - -FOR j := 0 to 31 - i := j*8 - k[j] := ( a[i+7:i] <= b[i+7:i] ) ? 1 : 0 + Bit Manipulation + + + + + Gather 64 bits from "b" using selection bits in "c". For each 64-bit element in "b", gather 8 bits from the 64-bit element in "b" at 8 bit position controlled by the 8 corresponding 8-bit elements of "c", and store the result in the corresponding 8-bit element of "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + +FOR i := 0 to 1 //Qword + FOR j := 0 to 7 // Byte + IF k[i*8+j] + m := c.qword[i].byte[j] & 0x3F + dst[i*8+j] := b.qword[i].bit[m] + ELSE + dst[i*8+j] := 0 + FI + ENDFOR ENDFOR -k[MAX:32] := 0 +dst[MAX:16] := 0 - +
immintrin.h
 - + Integer Mask + AVX512_BITALG AVX512VL - AVX512BW - Compare - - - Compare packed unsigned 8-bit integers in "a" and "b" for less-than, and store the results in mask vector "k". - -FOR j := 0 to 31 - i := j*8 - k[j] := ( a[i+7:i] < b[i+7:i] ) ? 1 : 0 + Bit Manipulation + + + + Gather 64 bits from "b" using selection bits in "c". For each 64-bit element in "b", gather 8 bits from the 64-bit element in "b" at 8 bit position controlled by the 8 corresponding 8-bit elements of "c", and store the result in the corresponding 8-bit element of "dst". + +FOR i := 0 to 1 //Qword + FOR j := 0 to 7 // Byte + m := c.qword[i].byte[j] & 0x3F + dst[i*8+j] := b.qword[i].bit[m] + ENDFOR ENDFOR -k[MAX:32] := 0 +dst[MAX:16] := 0 - +
immintrin.h
 - + Integer - Mask - AVX512VL - AVX512BW - Compare - - - Compare packed unsigned 8-bit integers in "a" and "b" for not-equal, and store the results in mask vector "k". - + AVX512_BITALG + Bit Manipulation + + + Count the number of logical 1 bits in packed 16-bit integers in "a", and store the results in "dst". + +DEFINE POPCNT(a) { + count := 0 + DO WHILE a > 0 + count += a[0] + a >>= 1 + OD + RETURN count +} FOR j := 0 to 31 - i := j*8 - k[j] := ( a[i+7:i] != b[i+7:i] ) ? 1 : 0 + i := j*16 + dst[i+15:i] := POPCNT(a[i+15:i]) ENDFOR -k[MAX:32] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - Mask - AVX512VL - AVX512BW - Compare - - - - - Compare packed unsigned 8-bit integers in "a" and "b" based on the comparison operand specified by "imm8", and store the results in mask vector "k1" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - -CASE (imm8[7:0]) OF -0: OP := _MM_CMPINT_EQ -1: OP := _MM_CMPINT_LT -2: OP := _MM_CMPINT_LE -3: OP := _MM_CMPINT_FALSE -4: OP := _MM_CMPINT_NE -5: OP := _MM_CMPINT_NLT -6: OP := _MM_CMPINT_NLE -7: OP := _MM_CMPINT_TRUE -ESAC + AVX512_BITALG + Bit Manipulation + + + + + Count the number of logical 1 bits in packed 16-bit integers in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +DEFINE POPCNT(a) { + count := 0 + DO WHILE a > 0 + count += a[0] + a >>= 1 + OD + RETURN count +} FOR j := 0 to 31 - i := j*8 - IF k1[j] - k[j] := ( a[i+7:i] OP b[i+7:i] ) ? 1 : 0 + i := j*16 + IF k[j] + dst[i+15:i] := POPCNT(a[i+15:i]) ELSE - k[j] := 0 + dst[i+15:i] := src[i+15:i] FI ENDFOR -k[MAX:32] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - Mask - AVX512VL - AVX512BW - Compare - - - - Compare packed unsigned 8-bit integers in "a" and "b" for equality, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). - + AVX512_BITALG + Bit Manipulation + + + + Count the number of logical 1 bits in packed 16-bit integers in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + +DEFINE POPCNT(a) { + count := 0 + DO WHILE a > 0 + count += a[0] + a >>= 1 + OD + RETURN count +} FOR j := 0 to 31 - i := j*8 - IF k1[j] - k[j] := ( a[i+7:i] == b[i+7:i] ) ? 1 : 0 + i := j*16 + IF k[j] + dst[i+15:i] := POPCNT(a[i+15:i]) ELSE - k[j] := 0 + dst[i+15:i] := 0 FI ENDFOR -k[MAX:32] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - Mask + AVX512_BITALG AVX512VL - AVX512BW - Compare - - - - Compare packed unsigned 8-bit integers in "a" and "b" for greater-than-or-equal, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). - -FOR j := 0 to 31 - i := j*8 - IF k1[j] - k[j] := ( a[i+7:i] >= b[i+7:i] ) ? 1 : 0 - ELSE - k[j] := 0 - FI + Bit Manipulation + + + Count the number of logical 1 bits in packed 16-bit integers in "a", and store the results in "dst". + +DEFINE POPCNT(a) { + count := 0 + DO WHILE a > 0 + count += a[0] + a >>= 1 + OD + RETURN count +} +FOR j := 0 to 15 + i := j*16 + dst[i+15:i] := POPCNT(a[i+15:i]) ENDFOR -k[MAX:32] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - Mask + AVX512_BITALG AVX512VL - AVX512BW - Compare - - - - Compare packed unsigned 8-bit integers in "a" and "b" for greater-than, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). - -FOR j := 0 to 31 - i := j*8 - IF k1[j] - k[j] := ( a[i+7:i] > b[i+7:i] ) ? 1 : 0 + Bit Manipulation + + + + + Count the number of logical 1 bits in packed 16-bit integers in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +DEFINE POPCNT(a) { + count := 0 + DO WHILE a > 0 + count += a[0] + a >>= 1 + OD + RETURN count +} +FOR j := 0 to 15 + i := j*16 + IF k[j] + dst[i+15:i] := POPCNT(a[i+15:i]) ELSE - k[j] := 0 + dst[i+15:i] := src[i+15:i] FI ENDFOR -k[MAX:32] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - Mask + AVX512_BITALG AVX512VL - AVX512BW - Compare - - - - Compare packed unsigned 8-bit integers in "a" and "b" for less-than-or-equal, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). - -FOR j := 0 to 31 - i := j*8 - IF k1[j] - k[j] := ( a[i+7:i] <= b[i+7:i] ) ? 1 : 0 + Bit Manipulation + + + + Count the number of logical 1 bits in packed 16-bit integers in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + +DEFINE POPCNT(a) { + count := 0 + DO WHILE a > 0 + count += a[0] + a >>= 1 + OD + RETURN count +} +FOR j := 0 to 15 + i := j*16 + IF k[j] + dst[i+15:i] := POPCNT(a[i+15:i]) ELSE - k[j] := 0 + dst[i+15:i] := 0 FI ENDFOR -k[MAX:32] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - Mask + AVX512_BITALG AVX512VL - AVX512BW - Compare - - - - Compare packed unsigned 8-bit integers in "a" and "b" for less-than, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). - -FOR j := 0 to 31 - i := j*8 - IF k1[j] - k[j] := ( a[i+7:i] < b[i+7:i] ) ? 1 : 0 - ELSE - k[j] := 0 - FI + Bit Manipulation + + + Count the number of logical 1 bits in packed 16-bit integers in "a", and store the results in "dst". + +DEFINE POPCNT(a) { + count := 0 + DO WHILE a > 0 + count += a[0] + a >>= 1 + OD + RETURN count +} +FOR j := 0 to 7 + i := j*16 + dst[i+15:i] := POPCNT(a[i+15:i]) ENDFOR -k[MAX:32] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - Mask + AVX512_BITALG AVX512VL - AVX512BW - Compare - - - - Compare packed unsigned 8-bit integers in "a" and "b" for not-equal, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). - -FOR j := 0 to 31 - i := j*8 - IF k1[j] - k[j] := ( a[i+7:i] != b[i+7:i] ) ? 1 : 0 + Bit Manipulation + + + + + Count the number of logical 1 bits in packed 16-bit integers in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +DEFINE POPCNT(a) { + count := 0 + DO WHILE a > 0 + count += a[0] + a >>= 1 + OD + RETURN count +} +FOR j := 0 to 7 + i := j*16 + IF k[j] + dst[i+15:i] := POPCNT(a[i+15:i]) ELSE - k[j] := 0 + dst[i+15:i] := src[i+15:i] FI ENDFOR -k[MAX:32] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - Mask - AVX512BW - Compare - - - - Compare packed unsigned 8-bit integers in "a" and "b" based on the comparison operand specified by "imm8", and store the results in mask vector "k". - -CASE (imm8[7:0]) OF -0: OP := _MM_CMPINT_EQ -1: OP := _MM_CMPINT_LT -2: OP := _MM_CMPINT_LE -3: OP := _MM_CMPINT_FALSE -4: OP := _MM_CMPINT_NE -5: OP := _MM_CMPINT_NLT -6: OP := _MM_CMPINT_NLE -7: OP := _MM_CMPINT_TRUE -ESAC -FOR j := 0 to 63 - i := j*8 - k[j] := ( a[i+7:i] OP b[i+7:i] ) ? 1 : 0 + AVX512_BITALG + AVX512VL + Bit Manipulation + + + + Count the number of logical 1 bits in packed 16-bit integers in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + +DEFINE POPCNT(a) { + count := 0 + DO WHILE a > 0 + count += a[0] + a >>= 1 + OD + RETURN count +} +FOR j := 0 to 7 + i := j*16 + IF k[j] + dst[i+15:i] := POPCNT(a[i+15:i]) + ELSE + dst[i+15:i] := 0 + FI ENDFOR -k[MAX:64] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - Mask - AVX512BW - Compare - - - Compare packed unsigned 8-bit integers in "a" and "b" for equality, and store the results in mask vector "k". - + AVX512_BITALG + Bit Manipulation + + + Count the number of logical 1 bits in packed 8-bit integers in "a", and store the results in "dst". + +DEFINE POPCNT(a) { + count := 0 + DO WHILE a > 0 + count += a[0] + a >>= 1 + OD + RETURN count +} FOR j := 0 to 63 i := j*8 - k[j] := ( a[i+7:i] == b[i+7:i] ) ? 1 : 0 + dst[i+7:i] := POPCNT(a[i+7:i]) ENDFOR -k[MAX:64] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - Mask - AVX512BW - Compare - - - Compare packed unsigned 8-bit integers in "a" and "b" for greater-than-or-equal, and store the results in mask vector "k". - + AVX512_BITALG + Bit Manipulation + + + + + Count the number of logical 1 bits in packed 8-bit integers in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +DEFINE POPCNT(a) { + count := 0 + DO WHILE a > 0 + count += a[0] + a >>= 1 + OD + RETURN count +} FOR j := 0 to 63 i := j*8 - k[j] := ( a[i+7:i] >= b[i+7:i] ) ? 1 : 0 + IF k[j] + dst[i+7:i] := POPCNT(a[i+7:i]) + ELSE + dst[i+7:i] := src[i+7:i] + FI ENDFOR -k[MAX:64] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - Mask - AVX512BW - Compare - - - Compare packed unsigned 8-bit integers in "a" and "b" for greater-than, and store the results in mask vector "k". - + AVX512_BITALG + Bit Manipulation + + + + Count the number of logical 1 bits in packed 8-bit integers in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + +DEFINE POPCNT(a) { + count := 0 + DO WHILE a > 0 + count += a[0] + a >>= 1 + OD + RETURN count +} FOR j := 0 to 63 i := j*8 - k[j] := ( a[i+7:i] > b[i+7:i] ) ? 1 : 0 + IF k[j] + dst[i+7:i] := POPCNT(a[i+7:i]) + ELSE + dst[i+7:i] := 0 + FI ENDFOR -k[MAX:64] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - Mask - AVX512BW - Compare - - - Compare packed unsigned 8-bit integers in "a" and "b" for less-than-or-equal, and store the results in mask vector "k". - -FOR j := 0 to 63 + AVX512_BITALG + AVX512VL + Bit Manipulation + + + Count the number of logical 1 bits in packed 8-bit integers in "a", and store the results in "dst". + +DEFINE POPCNT(a) { + count := 0 + DO WHILE a > 0 + count += a[0] + a >>= 1 + OD + RETURN count +} +FOR j := 0 to 31 i := j*8 - k[j] := ( a[i+7:i] <= b[i+7:i] ) ? 1 : 0 + dst[i+7:i] := POPCNT(a[i+7:i]) ENDFOR -k[MAX:64] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - Mask - AVX512BW - Compare - - - Compare packed unsigned 8-bit integers in "a" and "b" for less-than, and store the results in mask vector "k". - -FOR j := 0 to 63 + AVX512_BITALG + AVX512VL + Bit Manipulation + + + + + Count the number of logical 1 bits in packed 8-bit integers in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +DEFINE POPCNT(a) { + count := 0 + DO WHILE a > 0 + count += a[0] + a >>= 1 + OD + RETURN count +} +FOR j := 0 to 31 i := j*8 - k[j] := ( a[i+7:i] < b[i+7:i] ) ? 1 : 0 + IF k[j] + dst[i+7:i] := POPCNT(a[i+7:i]) + ELSE + dst[i+7:i] := src[i+7:i] + FI ENDFOR -k[MAX:64] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - Mask - AVX512BW - Compare - - - Compare packed unsigned 8-bit integers in "a" and "b" for not-equal, and store the results in mask vector "k". - -FOR j := 0 to 63 + AVX512_BITALG + AVX512VL + Bit Manipulation + + + + Count the number of logical 1 bits in packed 8-bit integers in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + +DEFINE POPCNT(a) { + count := 0 + DO WHILE a > 0 + count += a[0] + a >>= 1 + OD + RETURN count +} +FOR j := 0 to 31 i := j*8 - k[j] := ( a[i+7:i] != b[i+7:i] ) ? 1 : 0 + IF k[j] + dst[i+7:i] := POPCNT(a[i+7:i]) + ELSE + dst[i+7:i] := 0 + FI ENDFOR -k[MAX:64] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - Mask - AVX512BW - Compare - - - - - Compare packed unsigned 8-bit integers in "a" and "b" based on the comparison operand specified by "imm8", and store the results in mask vector "k1" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - -CASE (imm8[7:0]) OF -0: OP := _MM_CMPINT_EQ -1: OP := _MM_CMPINT_LT -2: OP := _MM_CMPINT_LE -3: OP := _MM_CMPINT_FALSE -4: OP := _MM_CMPINT_NE -5: OP := _MM_CMPINT_NLT -6: OP := _MM_CMPINT_NLE -7: OP := _MM_CMPINT_TRUE -ESAC -FOR j := 0 to 63 + AVX512_BITALG + AVX512VL + Bit Manipulation + + + Count the number of logical 1 bits in packed 8-bit integers in "a", and store the results in "dst". + +DEFINE POPCNT(a) { + count := 0 + DO WHILE a > 0 + count += a[0] + a >>= 1 + OD + RETURN count +} +FOR j := 0 to 15 i := j*8 - IF k1[j] - k[j] := ( a[i+7:i] OP b[i+7:i] ) ? 1 : 0 - ELSE - k[j] := 0 - FI + dst[i+7:i] := POPCNT(a[i+7:i]) ENDFOR -k[MAX:64] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - Mask - AVX512BW - Compare - - - - Compare packed unsigned 8-bit integers in "a" and "b" for equality, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). - -FOR j := 0 to 63 + AVX512_BITALG + AVX512VL + Bit Manipulation + + + + + Count the number of logical 1 bits in packed 8-bit integers in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +DEFINE POPCNT(a) { + count := 0 + DO WHILE a > 0 + count += a[0] + a >>= 1 + OD + RETURN count +} +FOR j := 0 to 15 i := j*8 - IF k1[j] - k[j] := ( a[i+7:i] == b[i+7:i] ) ? 1 : 0 + IF k[j] + dst[i+7:i] := POPCNT(a[i+7:i]) ELSE - k[j] := 0 + dst[i+7:i] := src[i+7:i] FI ENDFOR -k[MAX:64] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - Mask - AVX512BW - Compare - - - - Compare packed unsigned 8-bit integers in "a" and "b" for greater-than-or-equal, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). - -FOR j := 0 to 63 + AVX512_BITALG + AVX512VL + Bit Manipulation + + + + Count the number of logical 1 bits in packed 8-bit integers in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + +DEFINE POPCNT(a) { + count := 0 + DO WHILE a > 0 + count += a[0] + a >>= 1 + OD + RETURN count +} +FOR j := 0 to 15 i := j*8 - IF k1[j] - k[j] := ( a[i+7:i] >= b[i+7:i] ) ? 1 : 0 + IF k[j] + dst[i+7:i] := POPCNT(a[i+7:i]) ELSE - k[j] := 0 + dst[i+7:i] := 0 FI ENDFOR -k[MAX:64] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - - Integer - Mask - AVX512BW - Compare - - - - Compare packed unsigned 8-bit integers in "a" and "b" for greater-than, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). + + AVX512_VBMI + Bit Manipulation + + + + For each 64-bit element in "b", select 8 unaligned bytes using a byte-granular shift control within the corresponding 64-bit element of "a", and store the 8 assembled bytes to the corresponding 64-bit element of "dst". -FOR j := 0 to 63 - i := j*8 - IF k1[j] - k[j] := ( a[i+7:i] > b[i+7:i] ) ? 1 : 0 - ELSE - k[j] := 0 - FI +FOR i := 0 to 7 + q := i * 64 + FOR j := 0 to 7 + tmp8 := 0 + ctrl := a[q+j*8+7:q+j*8] & 63 + FOR l := 0 to 7 + tmp8[l] := b[q+((ctrl+l) & 63)] + ENDFOR + dst[q+j*8+7:q+j*8] := tmp8[7:0] + ENDFOR ENDFOR -k[MAX:64] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - Mask - AVX512BW - Compare - - - - Compare packed unsigned 8-bit integers in "a" and "b" for less-than-or-equal, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). + + AVX512_VBMI + Bit Manipulation + + + + + + For each 64-bit element in "b", select 8 unaligned bytes using a byte-granular shift control within the corresponding 64-bit element of "a", and store the 8 assembled bytes to the corresponding 64-bit element of "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 63 - i := j*8 - IF k1[j] - k[j] := ( a[i+7:i] <= b[i+7:i] ) ? 1 : 0 - ELSE - k[j] := 0 - FI +FOR i := 0 to 7 + q := i * 64 + FOR j := 0 to 7 + tmp8 := 0 + ctrl := a[q+j*8+7:q+j*8] & 63 + FOR l := 0 to 7 + tmp8[l] := b[q+((ctrl+l) & 63)] + ENDFOR + IF k[i*8+j] + dst[q+j*8+7:q+j*8] := tmp8[7:0] + ELSE + dst[q+j*8+7:q+j*8] := src[q+j*8+7:q+j*8] + FI + ENDFOR ENDFOR -k[MAX:64] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - Mask - AVX512BW - Compare - - - - Compare packed unsigned 8-bit integers in "a" and "b" for less-than, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). + + AVX512_VBMI + Bit Manipulation + + + + + For each 64-bit element in "b", select 8 unaligned bytes using a byte-granular shift control within the corresponding 64-bit element of "a", and store the 8 assembled bytes to the corresponding 64-bit element of "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 63 - i := j*8 - IF k1[j] - k[j] := ( a[i+7:i] < b[i+7:i] ) ? 1 : 0 - ELSE - k[j] := 0 - FI +FOR i := 0 to 7 + q := i * 64 + FOR j := 0 to 7 + tmp8 := 0 + ctrl := a[q+j*8+7:q+j*8] & 63 + FOR l := 0 to 7 + tmp8[l] := b[q+((ctrl+l) & 63)] + ENDFOR + IF k[i*8+j] + dst[q+j*8+7:q+j*8] := tmp8[7:0] + ELSE + dst[q+j*8+7:q+j*8] := 0 + FI + ENDFOR ENDFOR -k[MAX:64] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - Mask - AVX512BW - Compare - - - - Compare packed unsigned 8-bit integers in "a" and "b" for not-equal, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). + + AVX512_VBMI + AVX512VL + Bit Manipulation + + + + For each 64-bit element in "b", select 8 unaligned bytes using a byte-granular shift control within the corresponding 64-bit element of "a", and store the 8 assembled bytes to the corresponding 64-bit element of "dst". -FOR j := 0 to 63 - i := j*8 - IF k1[j] - k[j] := ( a[i+7:i] != b[i+7:i] ) ? 1 : 0 - ELSE - k[j] := 0 - FI +FOR i := 0 to 3 + q := i * 64 + FOR j := 0 to 7 + tmp8 := 0 + ctrl := a[q+j*8+7:q+j*8] & 63 + FOR l := 0 to 7 + tmp8[l] := b[q+((ctrl+l) & 63)] + ENDFOR + dst[q+j*8+7:q+j*8] := tmp8[7:0] + ENDFOR ENDFOR -k[MAX:64] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - - Integer - Mask + + AVX512_VBMI AVX512VL - AVX512BW - Compare - - - - Compare packed unsigned 8-bit integers in "a" and "b" based on the comparison operand specified by "imm8", and store the results in mask vector "k". + Bit Manipulation + + + + + + For each 64-bit element in "b", select 8 unaligned bytes using a byte-granular shift control within the corresponding 64-bit element of "a", and store the 8 assembled bytes to the corresponding 64-bit element of "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -CASE (imm8[7:0]) OF -0: OP := _MM_CMPINT_EQ -1: OP := _MM_CMPINT_LT -2: OP := _MM_CMPINT_LE -3: OP := _MM_CMPINT_FALSE -4: OP := _MM_CMPINT_NE -5: OP := _MM_CMPINT_NLT -6: OP := _MM_CMPINT_NLE -7: OP := _MM_CMPINT_TRUE -ESAC -FOR j := 0 to 15 - i := j*8 - k[j] := ( a[i+7:i] OP b[i+7:i] ) ? 1 : 0 +FOR i := 0 to 3 + q := i * 64 + FOR j := 0 to 7 + tmp8 := 0 + ctrl := a[q+j*8+7:q+j*8] & 63 + FOR l := 0 to 7 + tmp8[l] := b[q+((ctrl+l) & 63)] + ENDFOR + IF k[i*8+j] + dst[q+j*8+7:q+j*8] := tmp8[7:0] + ELSE + dst[q+j*8+7:q+j*8] := src[q+j*8+7:q+j*8] + FI + ENDFOR ENDFOR -k[MAX:16] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - - Integer - Mask + + AVX512_VBMI AVX512VL - AVX512BW - Compare - - - Compare packed unsigned 8-bit integers in "a" and "b" for equality, and store the results in mask vector "k". + Bit Manipulation + + + + + For each 64-bit element in "b", select 8 unaligned bytes using a byte-granular shift control within the corresponding 64-bit element of "a", and store the 8 assembled bytes to the corresponding 64-bit element of "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*8 - k[j] := ( a[i+7:i] == b[i+7:i] ) ? 1 : 0 +FOR i := 0 to 3 + q := i * 64 + FOR j := 0 to 7 + tmp8 := 0 + ctrl := a[q+j*8+7:q+j*8] & 63 + FOR l := 0 to 7 + tmp8[l] := b[q+((ctrl+l) & 63)] + ENDFOR + IF k[i*8+j] + dst[q+j*8+7:q+j*8] := tmp8[7:0] + ELSE + dst[q+j*8+7:q+j*8] := 0 + FI + ENDFOR ENDFOR -k[MAX:16] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - - Integer - Mask + + AVX512_VBMI AVX512VL - AVX512BW - Compare - - - Compare packed unsigned 8-bit integers in "a" and "b" for greater-than-or-equal, and store the results in mask vector "k". + Bit Manipulation + + + + For each 64-bit element in "b", select 8 unaligned bytes using a byte-granular shift control within the corresponding 64-bit element of "a", and store the 8 assembled bytes to the corresponding 64-bit element of "dst". -FOR j := 0 to 15 - i := j*8 - k[j] := ( a[i+7:i] >= b[i+7:i] ) ? 1 : 0 +FOR i := 0 to 1 + q := i * 64 + FOR j := 0 to 7 + tmp8 := 0 + ctrl := a[q+j*8+7:q+j*8] & 63 + FOR l := 0 to 7 + tmp8[l] := b[q+((ctrl+l) & 63)] + ENDFOR + dst[q+j*8+7:q+j*8] := tmp8[7:0] + ENDFOR ENDFOR -k[MAX:16] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - - Integer - Mask + + AVX512_VBMI AVX512VL - AVX512BW - Compare - - - Compare packed unsigned 8-bit integers in "a" and "b" for greater-than, and store the results in mask vector "k". + Bit Manipulation + + + + + + For each 64-bit element in "b", select 8 unaligned bytes using a byte-granular shift control within the corresponding 64-bit element of "a", and store the 8 assembled bytes to the corresponding 64-bit element of "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*8 - k[j] := ( a[i+7:i] > b[i+7:i] ) ? 1 : 0 +FOR i := 0 to 1 + q := i * 64 + FOR j := 0 to 7 + tmp8 := 0 + ctrl := a[q+j*8+7:q+j*8] & 63 + FOR l := 0 to 7 + tmp8[l] := b[q+((ctrl+l) & 63)] + ENDFOR + IF k[i*8+j] + dst[q+j*8+7:q+j*8] := tmp8[7:0] + ELSE + dst[q+j*8+7:q+j*8] := src[q+j*8+7:q+j*8] + FI + ENDFOR ENDFOR -k[MAX:16] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - - Integer - Mask + + AVX512_VBMI AVX512VL - AVX512BW - Compare - - - Compare packed unsigned 8-bit integers in "a" and "b" for less-than-or-equal, and store the results in mask vector "k". + Bit Manipulation + + + + + For each 64-bit element in "b", select 8 unaligned bytes using a byte-granular shift control within the corresponding 64-bit element of "a", and store the 8 assembled bytes to the corresponding 64-bit element of "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*8 - k[j] := ( a[i+7:i] <= b[i+7:i] ) ? 1 : 0 +FOR i := 0 to 1 + q := i * 64 + FOR j := 0 to 7 + tmp8 := 0 + ctrl := a[q+j*8+7:q+j*8] & 63 + FOR l := 0 to 7 + tmp8[l] := b[q+((ctrl+l) & 63)] + ENDFOR + IF k[i*8+j] + dst[q+j*8+7:q+j*8] := tmp8[7:0] + ELSE + dst[q+j*8+7:q+j*8] := 0 + FI + ENDFOR ENDFOR -k[MAX:16] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - - Integer - Mask - AVX512VL - AVX512BW - Compare - - - Compare packed unsigned 8-bit integers in "a" and "b" for less-than, and store the results in mask vector "k". + + AVX512_VBMI + Swizzle + + + + Shuffle 8-bit integers in "a" across lanes using the corresponding index in "idx", and store the results in "dst". -FOR j := 0 to 15 +FOR j := 0 to 63 i := j*8 - k[j] := ( a[i+7:i] < b[i+7:i] ) ? 1 : 0 + id := idx[i+5:i]*8 + dst[i+7:i] := a[id+7:id] ENDFOR -k[MAX:16] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - Mask - AVX512VL - AVX512BW - Compare - - - Compare packed unsigned 8-bit integers in "a" and "b" for not-equal, and store the results in mask vector "k". + + AVX512_VBMI + Swizzle + + + + + + Shuffle 8-bit integers in "a" across lanes using the corresponding index in "idx", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 15 +FOR j := 0 to 63 i := j*8 - k[j] := ( a[i+7:i] != b[i+7:i] ) ? 1 : 0 + id := idx[i+5:i]*8 + IF k[j] + dst[i+7:i] := a[id+7:id] + ELSE + dst[i+7:i] := src[i+7:i] + FI ENDFOR -k[MAX:16] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - Mask - AVX512VL - AVX512BW - Compare - - - - - Compare packed unsigned 8-bit integers in "a" and "b" based on the comparison operand specified by "imm8", and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). + + AVX512_VBMI + Swizzle + + + + + Shuffle 8-bit integers in "a" across lanes using the corresponding index in "idx", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -CASE (imm8[7:0]) OF -0: OP := _MM_CMPINT_EQ -1: OP := _MM_CMPINT_LT -2: OP := _MM_CMPINT_LE -3: OP := _MM_CMPINT_FALSE -4: OP := _MM_CMPINT_NE -5: OP := _MM_CMPINT_NLT -6: OP := _MM_CMPINT_NLE -7: OP := _MM_CMPINT_TRUE -ESAC -FOR j := 0 to 15 +FOR j := 0 to 63 i := j*8 - IF k1[j] - k[j] := ( a[i+7:i] OP b[i+7:i] ) ? 1 : 0 + id := idx[i+5:i]*8 + IF k[j] + dst[i+7:i] := a[id+7:id] ELSE - k[j] := 0 + dst[i+7:i] := 0 FI ENDFOR -k[MAX:16] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - Mask + + AVX512_VBMI AVX512VL - AVX512BW - Compare - - - - Compare packed unsigned 8-bit integers in "a" and "b" for equality, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). + Swizzle + + + + Shuffle 8-bit integers in "a" across lanes using the corresponding index in "idx", and store the results in "dst". -FOR j := 0 to 15 +FOR j := 0 to 31 i := j*8 - IF k1[j] - k[j] := ( a[i+7:i] == b[i+7:i] ) ? 1 : 0 + id := idx[i+4:i]*8 + dst[i+7:i] := a[id+7:id] +ENDFOR +dst[MAX:256] := 0 + + +
immintrin.h
+ + + AVX512_VBMI + AVX512VL + Swizzle + + + + + + Shuffle 8-bit integers in "a" across lanes using the corresponding index in "idx", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +FOR j := 0 to 31 + i := j*8 + id := idx[i+4:i]*8 + IF k[j] + dst[i+7:i] := a[id+7:id] ELSE - k[j] := 0 + dst[i+7:i] := src[i+7:i] FI ENDFOR -k[MAX:16] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - - Integer - Mask + + AVX512_VBMI AVX512VL - AVX512BW - Compare - - - - Compare packed unsigned 8-bit integers in "a" and "b" for greater-than-or-equal, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). + Swizzle + + + + + Shuffle 8-bit integers in "a" across lanes using the corresponding index in "idx", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 15 +FOR j := 0 to 31 i := j*8 - IF k1[j] - k[j] := ( a[i+7:i] >= b[i+7:i] ) ? 1 : 0 + id := idx[i+4:i]*8 + IF k[j] + dst[i+7:i] := a[id+7:id] ELSE - k[j] := 0 + dst[i+7:i] := 0 FI ENDFOR -k[MAX:16] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - - Integer - Mask + + AVX512_VBMI AVX512VL - AVX512BW - Compare - - - - Compare packed unsigned 8-bit integers in "a" and "b" for greater-than, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). + Swizzle + + + + Shuffle 8-bit integers in "a" using the corresponding index in "idx", and store the results in "dst". FOR j := 0 to 15 i := j*8 - IF k1[j] - k[j] := ( a[i+7:i] > b[i+7:i] ) ? 1 : 0 - ELSE - k[j] := 0 - FI + id := idx[i+3:i]*8 + dst[i+7:i] := a[id+7:id] ENDFOR -k[MAX:16] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - - Integer - Mask + + AVX512_VBMI AVX512VL - AVX512BW - Compare - - - - Compare packed unsigned 8-bit integers in "a" and "b" for less-than-or-equal, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). + Swizzle + + + + + + Shuffle 8-bit integers in "a" using the corresponding index in "idx", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). FOR j := 0 to 15 i := j*8 - IF k1[j] - k[j] := ( a[i+7:i] <= b[i+7:i] ) ? 1 : 0 + id := idx[i+3:i]*8 + IF k[j] + dst[i+7:i] := a[id+7:id] ELSE - k[j] := 0 + dst[i+7:i] := src[i+7:i] FI ENDFOR -k[MAX:16] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - - Integer - Mask + + AVX512_VBMI AVX512VL - AVX512BW - Compare - - - - Compare packed unsigned 8-bit integers in "a" and "b" for less-than, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). + Swizzle + + + + + Shuffle 8-bit integers in "a" using the corresponding index in "idx", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). FOR j := 0 to 15 i := j*8 - IF k1[j] - k[j] := ( a[i+7:i] < b[i+7:i] ) ? 1 : 0 + id := idx[i+3:i]*8 + IF k[j] + dst[i+7:i] := a[id+7:id] ELSE - k[j] := 0 + dst[i+7:i] := 0 FI ENDFOR -k[MAX:16] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - - Integer - Mask - AVX512VL - AVX512BW - Compare - - - - Compare packed unsigned 8-bit integers in "a" and "b" for not-equal, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). + + AVX512_VBMI + Swizzle + + + + + Shuffle 8-bit integers in "a" and "b" across lanes using the corresponding selector and index in "idx", and store the results in "dst". -FOR j := 0 to 15 +FOR j := 0 to 63 i := j*8 - IF k1[j] - k[j] := ( a[i+7:i] != b[i+7:i] ) ? 1 : 0 + off := 8*idx[i+5:i] + dst[i+7:i] := idx[i+6] ? b[off+7:off] : a[off+7:off] +ENDFOR +dst[MAX:512] := 0 + + +
immintrin.h
+ + + AVX512_VBMI + Swizzle + + + + + + Shuffle 8-bit integers in "a" and "b" across lanes using the corresponding selector and index in "idx", and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). + +FOR j := 0 to 63 + i := j*8 + IF k[j] + off := 8*idx[i+5:i] + dst[i+7:i] := idx[i+6] ? b[off+7:off] : a[off+7:off] ELSE - k[j] := 0 + dst[i+7:i] := a[i+7:i] FI ENDFOR -k[MAX:16] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - Mask - AVX512VL - AVX512F - Compare - - - - Compare packed unsigned 32-bit integers in "a" and "b" based on the comparison operand specified by "imm8", and store the results in mask vector "k". + + AVX512_VBMI + Swizzle + + + + + + Shuffle 8-bit integers in "a" and "b" across lanes using the corresponding selector and index in "idx", and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). -CASE (imm8[7:0]) OF -0: OP := _MM_CMPINT_EQ -1: OP := _MM_CMPINT_LT -2: OP := _MM_CMPINT_LE -3: OP := _MM_CMPINT_FALSE -4: OP := _MM_CMPINT_NE -5: OP := _MM_CMPINT_NLT -6: OP := _MM_CMPINT_NLE -7: OP := _MM_CMPINT_TRUE -ESAC -FOR j := 0 to 7 - i := j*32 - k[j] := ( a[i+31:i] OP b[i+31:i] ) ? 1 : 0 +FOR j := 0 to 63 + i := j*8 + IF k[j] + off := 8*idx[i+5:i] + dst[i+7:i] := idx[i+6] ? b[off+7:off] : a[off+7:off] + ELSE + dst[i+7:i] := idx[i+7:i] + FI ENDFOR -k[MAX:8] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - Mask - AVX512VL - AVX512F - Compare - - - Compare packed unsigned 32-bit integers in "a" and "b" for equality, and store the results in mask vector "k". + + AVX512_VBMI + Swizzle + + + + + + Shuffle 8-bit integers in "a" and "b" across lanes using the corresponding selector and index in "idx", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := j*32 - k[j] := ( a[i+31:i] == b[i+31:i] ) ? 1 : 0 +FOR j := 0 to 63 + i := j*8 + IF k[j] + off := 8*idx[i+5:i] + dst[i+7:i] := idx[i+6] ? b[off+7:off] : a[off+7:off] + ELSE + dst[i+7:i] := 0 + FI ENDFOR -k[MAX:8] := 0 +dst[MAX:512] := 0 - + +
immintrin.h
 - - Integer - Mask + + AVX512_VBMI AVX512VL - AVX512F - Compare - - - Compare packed unsigned 32-bit integers in "a" and "b" for greater-than-or-equal, and store the results in mask vector "k". + Swizzle + + + + + Shuffle 8-bit integers in "a" and "b" across lanes using the corresponding selector and index in "idx", and store the results in "dst". -FOR j := 0 to 7 - i := j*32 - k[j] := ( a[i+31:i] >= b[i+31:i] ) ? 1 : 0 +FOR j := 0 to 31 + i := j*8 + off := 8*idx[i+4:i] + dst[i+7:i] := idx[i+5] ? b[off+7:off] : a[off+7:off] ENDFOR -k[MAX:8] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - - Integer - Mask + + AVX512_VBMI AVX512VL - AVX512F - Compare - - - Compare packed unsigned 32-bit integers in "a" and "b" for greater-than, and store the results in mask vector "k". + Swizzle + + + + + + Shuffle 8-bit integers in "a" and "b" across lanes using the corresponding selector and index in "idx", and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := j*32 - k[j] := ( a[i+31:i] > b[i+31:i] ) ? 1 : 0 +FOR j := 0 to 31 + i := j*8 + IF k[j] + off := 8*idx[i+4:i] + dst[i+7:i] := idx[i+5] ? b[off+7:off] : a[off+7:off] + ELSE + dst[i+7:i] := a[i+7:i] + FI ENDFOR -k[MAX:8] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - - Integer - Mask + + AVX512_VBMI AVX512VL - AVX512F - Compare - - - Compare packed unsigned 32-bit integers in "a" and "b" for less-than-or-equal, and store the results in mask vector "k". + Swizzle + + + + + + Shuffle 8-bit integers in "a" and "b" across lanes using the corresponding selector and index in "idx", and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := j*32 - k[j] := ( a[i+31:i] <= b[i+31:i] ) ? 1 : 0 +FOR j := 0 to 31 + i := j*8 + IF k[j] + off := 8*idx[i+4:i] + dst[i+7:i] := idx[i+5] ? b[off+7:off] : a[off+7:off] + ELSE + dst[i+7:i] := idx[i+7:i] + FI ENDFOR -k[MAX:8] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - - Integer - Mask + + AVX512_VBMI AVX512VL - AVX512F - Compare - - - Compare packed unsigned 32-bit integers in "a" and "b" for less-than, and store the results in mask vector "k". + Swizzle + + + + + + Shuffle 8-bit integers in "a" and "b" across lanes using the corresponding selector and index in "idx", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := j*32 - k[j] := ( a[i+31:i] < b[i+31:i] ) ? 1 : 0 +FOR j := 0 to 31 + i := j*8 + IF k[j] + off := 8*idx[i+4:i] + dst[i+7:i] := idx[i+5] ? b[off+7:off] : a[off+7:off] + ELSE + dst[i+7:i] := 0 + FI ENDFOR -k[MAX:8] := 0 +dst[MAX:256] := 0 - + +
immintrin.h
 - - Integer - Mask + + AVX512_VBMI AVX512VL - AVX512F - Compare - - - Compare packed unsigned 32-bit integers in "a" and "b" for not-equal, and store the results in mask vector "k". + Swizzle + + + + + Shuffle 8-bit integers in "a" and "b" using the corresponding selector and index in "idx", and store the results in "dst". -FOR j := 0 to 7 - i := j*32 - k[j] := ( a[i+31:i] != b[i+31:i] ) ? 1 : 0 +FOR j := 0 to 15 + i := j*8 + off := 8*idx[i+3:i] + dst[i+7:i] := idx[i+4] ? b[off+7:off] : a[off+7:off] ENDFOR -k[MAX:8] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - - Integer - Mask + + AVX512_VBMI AVX512VL - AVX512F - Compare - - - - - Compare packed unsigned 32-bit integers in "a" and "b" based on the comparison operand specified by "imm8", and store the results in mask vector "k1" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Swizzle + + + + + + Shuffle 8-bit integers in "a" and "b" using the corresponding selector and index in "idx", and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). -CASE (imm8[7:0]) OF -0: OP := _MM_CMPINT_EQ -1: OP := _MM_CMPINT_LT -2: OP := _MM_CMPINT_LE -3: OP := _MM_CMPINT_FALSE -4: OP := _MM_CMPINT_NE -5: OP := _MM_CMPINT_NLT -6: OP := _MM_CMPINT_NLE -7: OP := _MM_CMPINT_TRUE -ESAC -FOR j := 0 to 7 - i := j*32 - IF k1[j] - k[j] := ( a[i+31:i] OP b[i+31:i] ) ? 1 : 0 - ELSE - k[j] := 0 +FOR j := 0 to 15 + i := j*8 + IF k[j] + off := 8*idx[i+3:i] + dst[i+7:i] := idx[i+4] ? b[off+7:off] : a[off+7:off] + ELSE + dst[i+7:i] := a[i+7:i] FI ENDFOR -k[MAX:8] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - - Integer - Mask + + AVX512_VBMI AVX512VL - AVX512F - Compare - - - - Compare packed unsigned 32-bit integers in "a" and "b" for equality, and store the results in mask vector "k1" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Swizzle + + + + + + Shuffle 8-bit integers in "a" and "b" using the corresponding selector and index in "idx", and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := j*32 - IF k1[j] - k[j] := ( a[i+31:i] == b[i+31:i] ) ? 1 : 0 - ELSE - k[j] := 0 +FOR j := 0 to 15 + i := j*8 + IF k[j] + off := 8*idx[i+3:i] + dst[i+7:i] := idx[i+4] ? b[off+7:off] : a[off+7:off] + ELSE + dst[i+7:i] := idx[i+7:i] FI ENDFOR -k[MAX:8] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - - Integer - Mask + + AVX512_VBMI AVX512VL - AVX512F - Compare - - - - Compare packed unsigned 32-bit integers in "a" and "b" for greater-than-or-equal, and store the results in mask vector "k1" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Swizzle + + + + + + Shuffle 8-bit integers in "a" and "b" using the corresponding selector and index in "idx", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := j*32 - IF k1[j] - k[j] := ( a[i+31:i] >= b[i+31:i] ) ? 1 : 0 - ELSE - k[j] := 0 +FOR j := 0 to 15 + i := j*8 + IF k[j] + off := 8*idx[i+3:i] + dst[i+7:i] := idx[i+4] ? b[off+7:off] : a[off+7:off] + ELSE + dst[i+7:i] := 0 FI ENDFOR -k[MAX:8] := 0 +dst[MAX:128] := 0 - + +
immintrin.h
 - + Integer - Mask - AVX512VL - AVX512F - Compare - - - - Compare packed unsigned 32-bit integers in "a" and "b" for greater-than, and store the results in mask vector "k1" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + AVX512_VBMI2 + Shift + + + + + + Concatenate packed 64-bit integers in "b" and "a" producing an intermediate 128-bit result. Shift the result right by the amount specified in the corresponding element of "c", and store the lower 64-bits in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). FOR j := 0 to 7 - i := j*32 - IF k1[j] - k[j] := ( a[i+31:i] > b[i+31:i] ) ? 1 : 0 - ELSE - k[j] := 0 + i := j*64 + IF k[j] + dst[i+63:i] := ((b[i+63:i] << 64)[127:0] | a[i+63:i]) >> (c[i+63:i] & 63) + ELSE + dst[i+63:i] := 0 FI ENDFOR -k[MAX:8] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - Mask - AVX512VL - AVX512F - Compare - - - - Compare packed unsigned 32-bit integers in "a" and "b" for less-than-or-equal, and store the results in mask vector "k1" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + AVX512_VBMI2 + Shift + + + + + + Concatenate packed 64-bit integers in "b" and "a" producing an intermediate 128-bit result. Shift the result right by the amount specified in the corresponding element of "c", and store the lower 64-bits in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). FOR j := 0 to 7 - i := j*32 - IF k1[j] - k[j] := ( a[i+31:i] <= b[i+31:i] ) ? 1 : 0 - ELSE - k[j] := 0 + i := j*64 + IF k[j] + dst[i+63:i] := ((b[i+63:i] << 64)[127:0] | a[i+63:i]) >> (c[i+63:i] & 63) + ELSE + dst[i+63:i] := a[i+63:i] FI ENDFOR -k[MAX:8] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - Mask - AVX512VL - AVX512F - Compare - - - - Compare packed unsigned 32-bit integers in "a" and "b" for less-than, and store the results in mask vector "k1" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + AVX512_VBMI2 + Shift + + + + + Concatenate packed 64-bit integers in "b" and "a" producing an intermediate 128-bit result. Shift the result right by the amount specified in the corresponding element of "c", and store the lower 64-bits in "dst". FOR j := 0 to 7 - i := j*32 - IF k1[j] - k[j] := ( a[i+31:i] < b[i+31:i] ) ? 1 : 0 - ELSE - k[j] := 0 - FI + i := j*64 + dst[i+63:i] := ((b[i+63:i] << 64)[127:0] | a[i+63:i]) >> (c[i+63:i] & 63) ENDFOR -k[MAX:8] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - Mask + AVX512_VBMI2 AVX512VL - AVX512F - Compare - - - - Compare packed unsigned 32-bit integers in "a" and "b" for not-equal, and store the results in mask vector "k1" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Shift + + + + + + Concatenate packed 64-bit integers in "b" and "a" producing an intermediate 128-bit result. Shift the result right by the amount specified in the corresponding element of "c", and store the lower 64-bits in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := j*32 - IF k1[j] - k[j] := ( a[i+31:i] != b[i+31:i] ) ? 1 : 0 - ELSE - k[j] := 0 +FOR j := 0 to 3 + i := j*64 + IF k[j] + dst[i+63:i] := ((b[i+63:i] << 64)[127:0] | a[i+63:i]) >> (c[i+63:i] & 63) + ELSE + dst[i+63:i] := 0 FI ENDFOR -k[MAX:8] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - Mask + AVX512_VBMI2 AVX512VL - AVX512F - Compare - - - - Compare packed unsigned 32-bit integers in "a" and "b" based on the comparison operand specified by "imm8", and store the results in mask vector "k". + Shift + + + + + + Concatenate packed 64-bit integers in "b" and "a" producing an intermediate 128-bit result. Shift the result right by the amount specified in the corresponding element of "c", and store the lower 64-bits in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). -CASE (imm8[7:0]) OF -0: OP := _MM_CMPINT_EQ -1: OP := _MM_CMPINT_LT -2: OP := _MM_CMPINT_LE -3: OP := _MM_CMPINT_FALSE -4: OP := _MM_CMPINT_NE -5: OP := _MM_CMPINT_NLT -6: OP := _MM_CMPINT_NLE -7: OP := _MM_CMPINT_TRUE -ESAC FOR j := 0 to 3 - i := j*32 - k[j] := ( a[i+31:i] OP b[i+31:i] ) ? 1 : 0 + i := j*64 + IF k[j] + dst[i+63:i] := ((b[i+63:i] << 64)[127:0] | a[i+63:i]) >> (c[i+63:i] & 63) + ELSE + dst[i+63:i] := a[i+63:i] + FI ENDFOR -k[MAX:4] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - Mask + AVX512_VBMI2 AVX512VL - AVX512F - Compare - - - Compare packed unsigned 32-bit integers in "a" and "b" for equality, and store the results in mask vector "k". + Shift + + + + + Concatenate packed 64-bit integers in "b" and "a" producing an intermediate 128-bit result. Shift the result right by the amount specified in the corresponding element of "c", and store the lower 64-bits in "dst". FOR j := 0 to 3 - i := j*32 - k[j] := ( a[i+31:i] == b[i+31:i] ) ? 1 : 0 + i := j*64 + dst[i+63:i] := ((b[i+63:i] << 64)[127:0] | a[i+63:i]) >> (c[i+63:i] & 63) ENDFOR -k[MAX:4] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - Mask + AVX512_VBMI2 AVX512VL - AVX512F - Compare - - - Compare packed unsigned 32-bit integers in "a" and "b" for greater-than-or-equal, and store the results in mask vector "k". + Shift + + + + + + Concatenate packed 64-bit integers in "b" and "a" producing an intermediate 128-bit result. Shift the result right by the amount specified in the corresponding element of "c", and store the lower 64-bits in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*32 - k[j] := ( a[i+31:i] >= b[i+31:i] ) ? 1 : 0 +FOR j := 0 to 1 + i := j*64 + IF k[j] + dst[i+63:i] := ((b[i+63:i] << 64)[127:0] | a[i+63:i]) >> (c[i+63:i] & 63) + ELSE + dst[i+63:i] := 0 + FI ENDFOR -k[MAX:4] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - Mask + AVX512_VBMI2 AVX512VL - AVX512F - Compare - - - Compare packed unsigned 32-bit integers in "a" and "b" for greater-than, and store the results in mask vector "k". + Shift + + + + + + Concatenate packed 64-bit integers in "b" and "a" producing an intermediate 128-bit result. Shift the result right by the amount specified in the corresponding element of "c", and store the lower 64-bits in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*32 - k[j] := ( a[i+31:i] > b[i+31:i] ) ? 1 : 0 +FOR j := 0 to 1 + i := j*64 + IF k[j] + dst[i+63:i] := ((b[i+63:i] << 64)[127:0] | a[i+63:i]) >> (c[i+63:i] & 63) + ELSE + dst[i+63:i] := a[i+63:i] + FI ENDFOR -k[MAX:4] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - Mask + AVX512_VBMI2 AVX512VL - AVX512F - Compare - - - Compare packed unsigned 32-bit integers in "a" and "b" for less-than-or-equal, and store the results in mask vector "k". + Shift + + + + + Concatenate packed 64-bit integers in "b" and "a" producing an intermediate 128-bit result. Shift the result right by the amount specified in the corresponding element of "c", and store the lower 64-bits in "dst". -FOR j := 0 to 3 - i := j*32 - k[j] := ( a[i+31:i] <= b[i+31:i] ) ? 1 : 0 +FOR j := 0 to 1 + i := j*64 + dst[i+63:i] := ((b[i+63:i] << 64)[127:0] | a[i+63:i]) >> (c[i+63:i] & 63) ENDFOR -k[MAX:4] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - Mask - AVX512VL - AVX512F - Compare - - - Compare packed unsigned 32-bit integers in "a" and "b" for less-than, and store the results in mask vector "k". + AVX512_VBMI2 + Shift + + + + + + Concatenate packed 32-bit integers in "b" and "a" producing an intermediate 64-bit result. Shift the result right by the amount specified in the corresponding element of "c", and store the lower 32-bits in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 3 +FOR j := 0 to 15 i := j*32 - k[j] := ( a[i+31:i] < b[i+31:i] ) ? 1 : 0 + IF k[j] + dst[i+31:i] := ((b[i+31:i] << 32)[63:0] | a[i+31:i]) >> (c[i+31:i] & 31) + ELSE + dst[i+31:i] := 0 + FI ENDFOR -k[MAX:4] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - Mask - AVX512VL - AVX512F - Compare - - - Compare packed unsigned 32-bit integers in "a" and "b" for not-equal, and store the results in mask vector "k". + AVX512_VBMI2 + Shift + + + + + + Concatenate packed 32-bit integers in "b" and "a" producing an intermediate 64-bit result. Shift the result right by the amount specified in the corresponding element of "c", and store the lower 32-bits in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). -FOR j := 0 to 3 +FOR j := 0 to 15 i := j*32 - k[j] := ( a[i+31:i] != b[i+31:i] ) ? 1 : 0 + IF k[j] + dst[i+31:i] := ((b[i+31:i] << 32)[63:0] | a[i+31:i]) >> (c[i+31:i] & 31) + ELSE + dst[i+31:i] := a[i+31:i] + FI ENDFOR -k[MAX:4] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - Mask - AVX512VL - AVX512F - Compare - - - - - Compare packed unsigned 32-bit integers in "a" and "b" based on the comparison operand specified by "imm8", and store the results in mask vector "k1" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + AVX512_VBMI2 + Shift + + + + + Concatenate packed 32-bit integers in "b" and "a" producing an intermediate 64-bit result. Shift the result right by the amount specified in the corresponding element of "c", and store the lower 32-bits in "dst". -CASE (imm8[7:0]) OF -0: OP := _MM_CMPINT_EQ -1: OP := _MM_CMPINT_LT -2: OP := _MM_CMPINT_LE -3: OP := _MM_CMPINT_FALSE -4: OP := _MM_CMPINT_NE -5: OP := _MM_CMPINT_NLT -6: OP := _MM_CMPINT_NLE -7: OP := _MM_CMPINT_TRUE -ESAC -FOR j := 0 to 3 +FOR j := 0 to 15 i := j*32 - IF k1[j] - k[j] := ( a[i+31:i] OP b[i+31:i] ) ? 1 : 0 - ELSE - k[j] := 0 - FI + dst[i+31:i] := ((b[i+31:i] << 32)[63:0] | a[i+31:i]) >> (c[i+31:i] & 31) ENDFOR -k[MAX:4] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - Mask + AVX512_VBMI2 AVX512VL - AVX512F - Compare - - - - Compare packed unsigned 32-bit integers in "a" and "b" for equality, and store the results in mask vector "k1" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Shift + + + + + + Concatenate packed 32-bit integers in "b" and "a" producing an intermediate 64-bit result. Shift the result right by the amount specified in the corresponding element of "c", and store the lower 32-bits in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 3 +FOR j := 0 to 7 i := j*32 - IF k1[j] - k[j] := ( a[i+31:i] == b[i+31:i] ) ? 1 : 0 - ELSE - k[j] := 0 + IF k[j] + dst[i+31:i] := ((b[i+31:i] << 32)[63:0] | a[i+31:i]) >> (c[i+31:i] & 31) + ELSE + dst[i+31:i] := 0 FI ENDFOR -k[MAX:4] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - Mask + AVX512_VBMI2 AVX512VL - AVX512F - Compare - - - - Compare packed unsigned 32-bit integers in "a" and "b" for greater-than-or-equal, and store the results in mask vector "k1" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - -FOR j := 0 to 3 - i := j*32 - IF k1[j] - k[j] := ( a[i+31:i] >= b[i+31:i] ) ? 1 : 0 - ELSE - k[j] := 0 + Shift + + + + + + Concatenate packed 32-bit integers in "b" and "a" producing an intermediate 64-bit result. Shift the result right by the amount specified in the corresponding element of "c", and store the lower 32-bits in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). + +FOR j := 0 to 7 + i := j*32 + IF k[j] + dst[i+31:i] := ((b[i+31:i] << 32)[63:0] | a[i+31:i]) >> (c[i+31:i] & 31) + ELSE + dst[i+31:i] := a[i+31:i] FI ENDFOR -k[MAX:4] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - Mask + AVX512_VBMI2 AVX512VL - AVX512F - Compare - - - - parameter varname='b' type='__m128i'/> - Compare packed unsigned 32-bit integers in "a" and "b" for greater-than, and store the results in mask vector "k1" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Shift + + + + + Concatenate packed 32-bit integers in "b" and "a" producing an intermediate 64-bit result. Shift the result right by the amount specified in the corresponding element of "c", and store the lower 32-bits in "dst". -FOR j := 0 to 3 +FOR j := 0 to 7 i := j*32 - IF k1[j] - k[j] := ( a[i+31:i] > b[i+31:i] ) ? 1 : 0 - ELSE - k[j] := 0 - FI + dst[i+31:i] := ((b[i+31:i] << 32)[63:0] | a[i+31:i]) >> (c[i+31:i] & 31) ENDFOR -k[MAX:4] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - Mask + AVX512_VBMI2 AVX512VL - AVX512F - Compare - - - - parameter varname='b' type='__m128i'/> - Compare packed unsigned 32-bit integers in "a" and "b" for less-than-or-equal, and store the results in mask vector "k1" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Shift + + + + + + Concatenate packed 32-bit integers in "b" and "a" producing an intermediate 64-bit result. Shift the result right by the amount specified in the corresponding element of "c", and store the lower 32-bits in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). FOR j := 0 to 3 i := j*32 - IF k1[j] - k[j] := ( a[i+31:i] <= b[i+31:i] ) ? 1 : 0 - ELSE - k[j] := 0 + IF k[j] + dst[i+31:i] := ((b[i+31:i] << 32)[63:0] | a[i+31:i]) >> (c[i+31:i] & 31) + ELSE + dst[i+31:i] := 0 FI ENDFOR -k[MAX:4] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - Mask + AVX512_VBMI2 AVX512VL - AVX512F - Compare - - - - parameter varname='b' type='__m128i'/> - Compare packed unsigned 32-bit integers in "a" and "b" for less-than, and store the results in mask vector "k1" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Shift + + + + + + Concatenate packed 32-bit integers in "b" and "a" producing an intermediate 64-bit result. Shift the result right by the amount specified in the corresponding element of "c", and store the lower 32-bits in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). FOR j := 0 to 3 i := j*32 - IF k1[j] - k[j] := ( a[i+31:i] < b[i+31:i] ) ? 1 : 0 - ELSE - k[j] := 0 + IF k[j] + dst[i+31:i] := ((b[i+31:i] << 32)[63:0] | a[i+31:i]) >> (c[i+31:i] & 31) + ELSE + dst[i+31:i] := a[i+31:i] FI ENDFOR -k[MAX:4] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - Mask + AVX512_VBMI2 AVX512VL - AVX512F - Compare - - - - parameter varname='b' type='__m128i'/> - Compare packed unsigned 32-bit integers in "a" and "b" for not-equal, and store the results in mask vector "k1" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Shift + + + + + Concatenate packed 32-bit integers in "b" and "a" producing an intermediate 64-bit result. Shift the result right by the amount specified in the corresponding element of "c", and store the lower 32-bits in "dst". FOR j := 0 to 3 i := j*32 - IF k1[j] - k[j] := ( a[i+31:i] != b[i+31:i] ) ? 1 : 0 - ELSE - k[j] := 0 - FI + dst[i+31:i] := ((b[i+31:i] << 32)[63:0] | a[i+31:i]) >> (c[i+31:i] & 31) ENDFOR -k[MAX:4] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - Mask - AVX512VL - AVX512F - Compare - - - - Compare packed unsigned 64-bit integers in "a" and "b" based on the comparison operand specified by "imm8", and store the results in mask vector "k". + AVX512_VBMI2 + Shift + + + + + + Concatenate packed 16-bit integers in "b" and "a" producing an intermediate 32-bit result. Shift the result right by the amount specified in the corresponding element of "c", and store the lower 16-bits in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -CASE (imm8[7:0]) OF -0: OP := _MM_CMPINT_EQ -1: OP := _MM_CMPINT_LT -2: OP := _MM_CMPINT_LE -3: OP := _MM_CMPINT_FALSE -4: OP := _MM_CMPINT_NE -5: OP := _MM_CMPINT_NLT -6: OP := _MM_CMPINT_NLE -7: OP := _MM_CMPINT_TRUE -ESAC -FOR j := 0 to 3 - i := j*64 - k[j] := ( a[i+63:i] OP b[i+63:i] ) ? 1 : 0 +FOR j := 0 to 31 + i := j*16 + IF k[j] + dst[i+15:i] := ((b[i+15:i] << 16)[31:0] | a[i+15:i]) >> (c[i+15:i] & 15) + ELSE + dst[i+15:i] := 0 + FI ENDFOR -k[MAX:4] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - Mask - AVX512VL - AVX512F - Compare - - - Compare packed unsigned 64-bit integers in "a" and "b" for equality, and store the results in mask vector "k". + AVX512_VBMI2 + Shift + + + + + + Concatenate packed 16-bit integers in "b" and "a" producing an intermediate 32-bit result. Shift the result right by the amount specified in the corresponding element of "c", and store the lower 16-bits in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*64 - k[j] := ( a[i+63:i] == b[i+63:i] ) ? 1 : 0 +FOR j := 0 to 31 + i := j*16 + IF k[j] + dst[i+15:i] := ((b[i+15:i] << 16)[31:0] | a[i+15:i]) >> (c[i+15:i] & 15) + ELSE + dst[i+15:i] := a[i+15:i] + FI ENDFOR -k[MAX:4] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - Mask - AVX512VL - AVX512F - Compare - - - Compare packed unsigned 64-bit integers in "a" and "b" for greater-than-or-equal, and store the results in mask vector "k". + AVX512_VBMI2 + Shift + + + + + Concatenate packed 16-bit integers in "b" and "a" producing an intermediate 32-bit result. Shift the result right by the amount specified in the corresponding element of "c", and store the lower 16-bits in "dst". -FOR j := 0 to 3 - i := j*64 - k[j] := ( a[i+63:i] >= b[i+63:i] ) ? 1 : 0 +FOR j := 0 to 31 + i := j*16 + dst[i+15:i] := ((b[i+15:i] << 16)[31:0] | a[i+15:i]) >> (c[i+15:i] & 15) ENDFOR -k[MAX:4] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - Mask + AVX512_VBMI2 AVX512VL - AVX512F - Compare - - - Compare packed unsigned 64-bit integers in "a" and "b" for greater-than, and store the results in mask vector "k". + Shift + + + + + + Concatenate packed 16-bit integers in "b" and "a" producing an intermediate 32-bit result. Shift the result right by the amount specified in the corresponding element of "c", and store the lower 16-bits in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*64 - k[j] := ( a[i+63:i] > b[i+63:i] ) ? 1 : 0 +FOR j := 0 to 15 + i := j*16 + IF k[j] + dst[i+15:i] := ((b[i+15:i] << 16)[31:0] | a[i+15:i]) >> (c[i+15:i] & 15) + ELSE + dst[i+15:i] := 0 + FI ENDFOR -k[MAX:4] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - Mask + AVX512_VBMI2 AVX512VL - AVX512F - Compare - - - Compare packed unsigned 64-bit integers in "a" and "b" for less-than-or-equal, and store the results in mask vector "k". + Shift + + + + + + Concatenate packed 16-bit integers in "b" and "a" producing an intermediate 32-bit result. Shift the result right by the amount specified in the corresponding element of "c", and store the lower 16-bits in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*64 - k[j] := ( a[i+63:i] <= b[i+63:i] ) ? 1 : 0 +FOR j := 0 to 15 + i := j*16 + IF k[j] + dst[i+15:i] := ((b[i+15:i] << 16)[31:0] | a[i+15:i]) >> (c[i+15:i] & 15) + ELSE + dst[i+15:i] := a[i+15:i] + FI ENDFOR -k[MAX:4] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - Mask + AVX512_VBMI2 AVX512VL - AVX512F - Compare - - - Compare packed unsigned 64-bit integers in "a" and "b" for less-than, and store the results in mask vector "k". + Shift + + + + + Concatenate packed 16-bit integers in "b" and "a" producing an intermediate 32-bit result. Shift the result right by the amount specified in the corresponding element of "c", and store the lower 16-bits in "dst". -FOR j := 0 to 3 - i := j*64 - k[j] := ( a[i+63:i] < b[i+63:i] ) ? 1 : 0 +FOR j := 0 to 15 + i := j*16 + dst[i+15:i] := ((b[i+15:i] << 16)[31:0] | a[i+15:i]) >> (c[i+15:i] & 15) ENDFOR -k[MAX:4] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - Mask + AVX512_VBMI2 AVX512VL - AVX512F - Compare - - - Compare packed unsigned 64-bit integers in "a" and "b" for not-equal, and store the results in mask vector "k". + Shift + + + + + + Concatenate packed 16-bit integers in "b" and "a" producing an intermediate 32-bit result. Shift the result right by the amount specified in the corresponding element of "c", and store the lower 16-bits in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*64 - k[j] := ( a[i+63:i] != b[i+63:i] ) ? 1 : 0 +FOR j := 0 to 7 + i := j*16 + IF k[j] + dst[i+15:i] := ((b[i+15:i] << 16)[31:0] | a[i+15:i]) >> (c[i+15:i] & 15) + ELSE + dst[i+15:i] := 0 + FI ENDFOR -k[MAX:4] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - Mask + AVX512_VBMI2 AVX512VL - AVX512F - Compare - - - - - Compare packed unsigned 64-bit integers in "a" and "b" based on the comparison operand specified by "imm8", and store the results in mask vector "k1" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Shift + + + + + + Concatenate packed 16-bit integers in "b" and "a" producing an intermediate 32-bit result. Shift the result right by the amount specified in the corresponding element of "c", and store the lower 16-bits in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). -CASE (imm8[7:0]) OF -0: OP := _MM_CMPINT_EQ -1: OP := _MM_CMPINT_LT -2: OP := _MM_CMPINT_LE -3: OP := _MM_CMPINT_FALSE -4: OP := _MM_CMPINT_NE -5: OP := _MM_CMPINT_NLT -6: OP := _MM_CMPINT_NLE -7: OP := _MM_CMPINT_TRUE -ESAC -FOR j := 0 to 3 - i := j*64 - IF k1[j] - k[j] := ( a[i+63:i] OP b[i+63:i] ) ? 1 : 0 - ELSE - k[j] := 0 +FOR j := 0 to 7 + i := j*16 + IF k[j] + dst[i+15:i] := ((b[i+15:i] << 16)[31:0] | a[i+15:i]) >> (c[i+15:i] & 15) + ELSE + dst[i+15:i] := a[i+15:i] FI ENDFOR -k[MAX:4] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - Mask + AVX512_VBMI2 AVX512VL - AVX512F - Compare - - - - Compare packed unsigned 64-bit integers in "a" and "b" for equality, and store the results in mask vector "k1" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Shift + + + + + Concatenate packed 16-bit integers in "b" and "a" producing an intermediate 32-bit result. Shift the result right by the amount specified in the corresponding element of "c", and store the lower 16-bits in "dst". -FOR j := 0 to 3 - i := j*64 - IF k1[j] - k[j] := ( a[i+63:i] == b[i+63:i] ) ? 1 : 0 - ELSE - k[j] := 0 - FI +FOR j := 0 to 7 + i := j*16 + dst[i+15:i] := ((b[i+15:i] << 16)[31:0] | a[i+15:i]) >> (c[i+15:i] & 15) ENDFOR -k[MAX:4] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - Mask - AVX512VL - AVX512F - Compare - - - - Compare packed unsigned 64-bit integers in "a" and "b" for greater-than-or-equal, and store the results in mask vector "k1" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + AVX512_VBMI2 + Shift + + + + + + Concatenate packed 64-bit integers in "b" and "a" producing an intermediate 128-bit result. Shift the result right by "imm8" bits, and store the lower 64-bits in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 3 +FOR j := 0 to 7 i := j*64 - IF k1[j] - k[j] := ( a[i+63:i] >= b[i+63:i] ) ? 1 : 0 - ELSE - k[j] := 0 + IF k[j] + dst[i+63:i] := ((b[i+63:i] << 64)[127:0] | a[i+63:i]) >> imm8[5:0] + ELSE + dst[i+63:i] := 0 FI ENDFOR -k[MAX:4] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - Mask - AVX512VL - AVX512F - Compare - - - - Compare packed unsigned 64-bit integers in "a" and "b" for greater-than, and store the results in mask vector "k1" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + AVX512_VBMI2 + Shift + + + + + + + Concatenate packed 64-bit integers in "b" and "a" producing an intermediate 128-bit result. Shift the result right by "imm8" bits, and store the lower 64-bits in "dst" using writemask "k" (elements are copied from "src"" when the corresponding mask bit is not set). -FOR j := 0 to 3 +FOR j := 0 to 7 i := j*64 - IF k1[j] - k[j] := ( a[i+63:i] > b[i+63:i] ) ? 1 : 0 - ELSE - k[j] := 0 + IF k[j] + dst[i+63:i] := ((b[i+63:i] << 64)[127:0] | a[i+63:i]) >> imm8[5:0] + ELSE + dst[i+63:i] := src[i+63:i] FI ENDFOR -k[MAX:4] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - Mask - AVX512VL - AVX512F - Compare - - - - Compare packed unsigned 64-bit integers in "a" and "b" for less-than-or-equal, and store the results in mask vector "k1" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + AVX512_VBMI2 + Shift + + + + + Concatenate packed 64-bit integers in "b" and "a" producing an intermediate 128-bit result. Shift the result right by "imm8" bits, and store the lower 64-bits in "dst". -FOR j := 0 to 3 +FOR j := 0 to 7 i := j*64 - IF k1[j] - k[j] := ( a[i+63:i] <= b[i+63:i] ) ? 1 : 0 - ELSE - k[j] := 0 - FI + dst[i+63:i] := ((b[i+63:i] << 64)[127:0] | a[i+63:i]) >> imm8[5:0] ENDFOR -k[MAX:4] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - Mask + AVX512_VBMI2 AVX512VL - AVX512F - Compare - - - - Compare packed unsigned 64-bit integers in "a" and "b" for less-than, and store the results in mask vector "k1" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Shift + + + + + + Concatenate packed 64-bit integers in "b" and "a" producing an intermediate 128-bit result. Shift the result right by "imm8" bits, and store the lower 64-bits in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). FOR j := 0 to 3 i := j*64 - IF k1[j] - k[j] := ( a[i+63:i] < b[i+63:i] ) ? 1 : 0 - ELSE - k[j] := 0 + IF k[j] + dst[i+63:i] := ((b[i+63:i] << 64)[127:0] | a[i+63:i]) >> imm8[5:0] + ELSE + dst[i+63:i] := 0 FI ENDFOR -k[MAX:4] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - Mask + AVX512_VBMI2 AVX512VL - AVX512F - Compare - - - - Compare packed unsigned 64-bit integers in "a" and "b" for not-equal, and store the results in mask vector "k1" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Shift + + + + + + + Concatenate packed 64-bit integers in "b" and "a" producing an intermediate 128-bit result. Shift the result right by "imm8" bits, and store the lower 64-bits in "dst" using writemask "k" (elements are copied from "src"" when the corresponding mask bit is not set). FOR j := 0 to 3 i := j*64 - IF k1[j] - k[j] := ( a[i+63:i] != b[i+63:i] ) ? 1 : 0 - ELSE - k[j] := 0 + IF k[j] + dst[i+63:i] := ((b[i+63:i] << 64)[127:0] | a[i+63:i]) >> imm8[5:0] + ELSE + dst[i+63:i] := src[i+63:i] FI ENDFOR -k[MAX:4] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - Mask + AVX512_VBMI2 AVX512VL - AVX512F - Compare - - - - Compare packed unsigned 64-bit integers in "a" and "b" based on the comparison operand specified by "imm8", and store the results in mask vector "k". + Shift + + + + + Concatenate packed 64-bit integers in "b" and "a" producing an intermediate 128-bit result. Shift the result right by "imm8" bits, and store the lower 64-bits in "dst". -CASE (imm8[7:0]) OF -0: OP := _MM_CMPINT_EQ -1: OP := _MM_CMPINT_LT -2: OP := _MM_CMPINT_LE -3: OP := _MM_CMPINT_FALSE -4: OP := _MM_CMPINT_NE -5: OP := _MM_CMPINT_NLT -6: OP := _MM_CMPINT_NLE -7: OP := _MM_CMPINT_TRUE -ESAC -FOR j := 0 to 1 +FOR j := 0 to 3 i := j*64 - k[j] := ( a[i+63:i] OP b[i+63:i] ) ? 1 : 0 + dst[i+63:i] := ((b[i+63:i] << 64)[127:0] | a[i+63:i]) >> imm8[5:0] ENDFOR -k[MAX:2] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - Mask + AVX512_VBMI2 AVX512VL - AVX512F - Compare - - - Compare packed unsigned 64-bit integers in "a" and "b" for equality, and store the results in mask vector "k". + Shift + + + + + + Concatenate packed 64-bit integers in "b" and "a" producing an intermediate 128-bit result. Shift the result right by "imm8" bits, and store the lower 64-bits in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). FOR j := 0 to 1 i := j*64 - k[j] := ( a[i+63:i] == b[i+63:i] ) ? 1 : 0 + IF k[j] + dst[i+63:i] := ((b[i+63:i] << 64)[127:0] | a[i+63:i]) >> imm8[5:0] + ELSE + dst[i+63:i] := 0 + FI ENDFOR -k[MAX:2] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - Mask + AVX512_VBMI2 AVX512VL - AVX512F - Compare - - - Compare packed unsigned 64-bit integers in "a" and "b" for greater-than-or-equal, and store the results in mask vector "k". + Shift + + + + + + + Concatenate packed 64-bit integers in "b" and "a" producing an intermediate 128-bit result. Shift the result right by "imm8" bits, and store the lower 64-bits in "dst" using writemask "k" (elements are copied from "src"" when the corresponding mask bit is not set). FOR j := 0 to 1 i := j*64 - k[j] := ( a[i+63:i] >= b[i+63:i] ) ? 1 : 0 + IF k[j] + dst[i+63:i] := ((b[i+63:i] << 64)[127:0] | a[i+63:i]) >> imm8[5:0] + ELSE + dst[i+63:i] := src[i+63:i] + FI ENDFOR -k[MAX:2] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - Mask + AVX512_VBMI2 AVX512VL - AVX512F - Compare - - - Compare packed unsigned 64-bit integers in "a" and "b" for greater-than, and store the results in mask vector "k". + Shift + + + + + Concatenate packed 64-bit integers in "b" and "a" producing an intermediate 128-bit result. Shift the result right by "imm8" bits, and store the lower 64-bits in "dst". FOR j := 0 to 1 i := j*64 - k[j] := ( a[i+63:i] > b[i+63:i] ) ? 1 : 0 + dst[i+63:i] := ((b[i+63:i] << 64)[127:0] | a[i+63:i]) >> imm8[5:0] ENDFOR -k[MAX:2] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - Mask - AVX512VL - AVX512F - Compare - - - Compare packed unsigned 64-bit integers in "a" and "b" for less-than-or-equal, and store the results in mask vector "k". + AVX512_VBMI2 + Shift + + + + + + Concatenate packed 32-bit integers in "b" and "a" producing an intermediate 64-bit result. Shift the result right by "imm8" bits, and store the lower 32-bits in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 1 - i := j*64 - k[j] := ( a[i+63:i] <= b[i+63:i] ) ? 1 : 0 +FOR j := 0 to 15 + i := j*32 + IF k[j] + dst[i+31:i] := ((b[i+31:i] << 32)[63:0] | a[i+31:i]) >> imm8[4:0] + ELSE + dst[i+31:i] := 0 + FI ENDFOR -k[MAX:2] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - Mask - AVX512VL - AVX512F - Compare - - - Compare packed unsigned 64-bit integers in "a" and "b" for less-than, and store the results in mask vector "k". + AVX512_VBMI2 + Shift + + + + + + + Concatenate packed 32-bit integers in "b" and "a" producing an intermediate 64-bit result. Shift the result right by "imm8" bits, and store the lower 32-bits in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 1 - i := j*64 - k[j] := ( a[i+63:i] < b[i+63:i] ) ? 1 : 0 +FOR j := 0 to 15 + i := j*32 + IF k[j] + dst[i+31:i] := ((b[i+31:i] << 32)[63:0] | a[i+31:i]) >> imm8[4:0] + ELSE + dst[i+31:i] := src[i+31:i] + FI ENDFOR -k[MAX:2] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - Mask - AVX512VL - AVX512F - Compare - - - Compare packed unsigned 64-bit integers in "a" and "b" for not-equal, and store the results in mask vector "k". + AVX512_VBMI2 + Shift + + + + + Concatenate packed 32-bit integers in "b" and "a" producing an intermediate 64-bit result. Shift the result right by "imm8" bits, and store the lower 32-bits in "dst". -FOR j := 0 to 1 - i := j*64 - k[j] := ( a[i+63:i] != b[i+63:i] ) ? 1 : 0 +FOR j := 0 to 15 + i := j*32 + dst[i+31:i] := ((b[i+31:i] << 32)[63:0] | a[i+31:i]) >> imm8[4:0] ENDFOR -k[MAX:2] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - Mask + AVX512_VBMI2 AVX512VL - AVX512F - Compare - - - - - Compare packed unsigned 64-bit integers in "a" and "b" based on the comparison operand specified by "imm8", and store the results in mask vector "k1" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Shift + + + + + + Concatenate packed 32-bit integers in "b" and "a" producing an intermediate 64-bit result. Shift the result right by "imm8" bits, and store the lower 32-bits in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -CASE (imm8[7:0]) OF -0: OP := _MM_CMPINT_EQ -1: OP := _MM_CMPINT_LT -2: OP := _MM_CMPINT_LE -3: OP := _MM_CMPINT_FALSE -4: OP := _MM_CMPINT_NE -5: OP := _MM_CMPINT_NLT -6: OP := _MM_CMPINT_NLE -7: OP := _MM_CMPINT_TRUE -ESAC -FOR j := 0 to 1 - i := j*64 - IF k1[j] - k[j] := ( a[i+63:i] OP b[i+63:i] ) ? 1 : 0 - ELSE - k[j] := 0 +FOR j := 0 to 7 + i := j*32 + IF k[j] + dst[i+31:i] := ((b[i+31:i] << 32)[63:0] | a[i+31:i]) >> imm8[4:0] + ELSE + dst[i+31:i] := 0 FI ENDFOR -k[MAX:2] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - Mask + AVX512_VBMI2 AVX512VL - AVX512F - Compare - - - - Compare packed unsigned 64-bit integers in "a" and "b" for equality, and store the results in mask vector "k1" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Shift + + + + + + + Concatenate packed 32-bit integers in "b" and "a" producing an intermediate 64-bit result. Shift the result right by "imm8" bits, and store the lower 32-bits in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 1 - i := j*64 - IF k1[j] - k[j] := ( a[i+63:i] == b[i+63:i] ) ? 1 : 0 - ELSE - k[j] := 0 +FOR j := 0 to 7 + i := j*32 + IF k[j] + dst[i+31:i] := ((b[i+31:i] << 32)[63:0] | a[i+31:i]) >> imm8[4:0] + ELSE + dst[i+31:i] := src[i+31:i] FI ENDFOR -k[MAX:2] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - Mask + AVX512_VBMI2 AVX512VL - AVX512F - Compare - - - - Compare packed unsigned 64-bit integers in "a" and "b" for greater-than-or-equal, and store the results in mask vector "k1" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Shift + + + + + Concatenate packed 32-bit integers in "b" and "a" producing an intermediate 64-bit result. Shift the result right by "imm8" bits, and store the lower 32-bits in "dst". -FOR j := 0 to 1 - i := j*64 - IF k1[j] - k[j] := ( a[i+63:i] >= b[i+63:i] ) ? 1 : 0 - ELSE - k[j] := 0 - FI +FOR j := 0 to 7 + i := j*32 + dst[i+31:i] := ((b[i+31:i] << 32)[63:0] | a[i+31:i]) >> imm8[4:0] ENDFOR -k[MAX:2] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - Mask + AVX512_VBMI2 AVX512VL - AVX512F - Compare - - - - Compare packed unsigned 64-bit integers in "a" and "b" for greater-than, and store the results in mask vector "k1" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Shift + + + + + + Concatenate packed 32-bit integers in "b" and "a" producing an intermediate 64-bit result. Shift the result right by "imm8" bits, and store the lower 32-bits in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 1 - i := j*64 - IF k1[j] - k[j] := ( a[i+63:i] > b[i+63:i] ) ? 1 : 0 - ELSE - k[j] := 0 +FOR j := 0 to 3 + i := j*32 + IF k[j] + dst[i+31:i] := ((b[i+31:i] << 32)[63:0] | a[i+31:i]) >> imm8[4:0] + ELSE + dst[i+31:i] := 0 FI ENDFOR -k[MAX:2] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - Mask + AVX512_VBMI2 AVX512VL - AVX512F - Compare - - - - Compare packed unsigned 64-bit integers in "a" and "b" for less-than-or-equal, and store the results in mask vector "k1" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Shift + + + + + + + Concatenate packed 32-bit integers in "b" and "a" producing an intermediate 64-bit result. Shift the result right by "imm8" bits, and store the lower 32-bits in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 1 - i := j*64 - IF k1[j] - k[j] := ( a[i+63:i] <= b[i+63:i] ) ? 1 : 0 - ELSE - k[j] := 0 +FOR j := 0 to 3 + i := j*32 + IF k[j] + dst[i+31:i] := ((b[i+31:i] << 32)[63:0] | a[i+31:i]) >> imm8[4:0] + ELSE + dst[i+31:i] := src[i+31:i] FI ENDFOR -k[MAX:2] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - Mask + AVX512_VBMI2 AVX512VL - AVX512F - Compare - - - - Compare packed unsigned 64-bit integers in "a" and "b" for less-than, and store the results in mask vector "k1" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Shift + + + + + Concatenate packed 32-bit integers in "b" and "a" producing an intermediate 64-bit result. Shift the result right by "imm8" bits, and store the lower 32-bits in "dst". -FOR j := 0 to 1 - i := j*64 - IF k1[j] - k[j] := ( a[i+63:i] < b[i+63:i] ) ? 1 : 0 - ELSE - k[j] := 0 - FI +FOR j := 0 to 3 + i := j*32 + dst[i+31:i] := ((b[i+31:i] << 32)[63:0] | a[i+31:i]) >> imm8[4:0] ENDFOR -k[MAX:2] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - Mask - AVX512VL - AVX512F - Compare - - - - Compare packed unsigned 64-bit integers in "a" and "b" for not-equal, and store the results in mask vector "k1" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + AVX512_VBMI2 + Shift + + + + + + Concatenate packed 16-bit integers in "b" and "a" producing an intermediate 32-bit result. Shift the result right by "imm8" bits, and store the lower 16-bits in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 1 - i := j*64 - IF k1[j] - k[j] := ( a[i+63:i] != b[i+63:i] ) ? 1 : 0 - ELSE - k[j] := 0 +FOR j := 0 to 31 + i := j*16 + IF k[j] + dst[i+15:i] := ((b[i+15:i] << 16)[31:0] | a[i+15:i]) >> imm8[3:0] + ELSE + dst[i+15:i] := 0 + FI +ENDFOR +dst[MAX:512] := 0 + + +
immintrin.h
+ + + Integer + AVX512_VBMI2 + Shift + + + + + + + Concatenate packed 16-bit integers in "b" and "a" producing an intermediate 32-bit result. Shift the result right by "imm8" bits, and store the lower 16-bits in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +FOR j := 0 to 31 + i := j*16 + IF k[j] + dst[i+15:i] := ((b[i+15:i] << 16)[31:0] | a[i+15:i]) >> imm8[3:0] + ELSE + dst[i+15:i] := src[i+15:i] FI ENDFOR -k[MAX:2] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - Mask - AVX512VL - AVX512BW - Compare - - - - Compare packed unsigned 16-bit integers in "a" and "b" based on the comparison operand specified by "imm8", and store the results in mask vector "k". + AVX512_VBMI2 + Shift + + + + + Concatenate packed 16-bit integers in "b" and "a" producing an intermediate 32-bit result. Shift the result right by "imm8" bits, and store the lower 16-bits in "dst". -CASE (imm8[7:0]) OF -0: OP := _MM_CMPINT_EQ -1: OP := _MM_CMPINT_LT -2: OP := _MM_CMPINT_LE -3: OP := _MM_CMPINT_FALSE -4: OP := _MM_CMPINT_NE -5: OP := _MM_CMPINT_NLT -6: OP := _MM_CMPINT_NLE -7: OP := _MM_CMPINT_TRUE -ESAC -FOR j := 0 to 15 +FOR j := 0 to 31 i := j*16 - k[j] := ( a[i+15:i] OP b[i+15:i] ) ? 1 : 0 + dst[i+15:i] := ((b[i+15:i] << 16)[31:0] | a[i+15:i]) >> imm8[3:0] ENDFOR -k[MAX:16] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - Mask + AVX512_VBMI2 AVX512VL - AVX512BW - Compare - - - Compare packed unsigned 16-bit integers in "a" and "b" for equality, and store the results in mask vector "k". + Shift + + + + + + Concatenate packed 16-bit integers in "b" and "a" producing an intermediate 32-bit result. Shift the result right by "imm8" bits, and store the lower 16-bits in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). FOR j := 0 to 15 i := j*16 - k[j] := ( a[i+15:i] == b[i+15:i] ) ? 1 : 0 + IF k[j] + dst[i+15:i] := ((b[i+15:i] << 16)[31:0] | a[i+15:i]) >> imm8[3:0] + ELSE + dst[i+15:i] := 0 + FI ENDFOR -k[MAX:16] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - Mask + AVX512_VBMI2 AVX512VL - AVX512BW - Compare - - - Compare packed unsigned 16-bit integers in "a" and "b" for greater-than-or-equal, and store the results in mask vector "k". + Shift + + + + + + + Concatenate packed 16-bit integers in "b" and "a" producing an intermediate 32-bit result. Shift the result right by "imm8" bits, and store the lower 16-bits in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). FOR j := 0 to 15 i := j*16 - k[j] := ( a[i+15:i] >= b[i+15:i] ) ? 1 : 0 + IF k[j] + dst[i+15:i] := ((b[i+15:i] << 16)[31:0] | a[i+15:i]) >> imm8[3:0] + ELSE + dst[i+15:i] := src[i+15:i] + FI ENDFOR -k[MAX:16] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - Mask + AVX512_VBMI2 AVX512VL - AVX512BW - Compare - - - Compare packed unsigned 16-bit integers in "a" and "b" for greater-than, and store the results in mask vector "k". + Shift + + + + + Concatenate packed 16-bit integers in "b" and "a" producing an intermediate 32-bit result. Shift the result right by "imm8" bits, and store the lower 16-bits in "dst". FOR j := 0 to 15 i := j*16 - k[j] := ( a[i+15:i] > b[i+15:i] ) ? 1 : 0 + dst[i+15:i] := ((b[i+15:i] << 16)[31:0] | a[i+15:i]) >> imm8[3:0] ENDFOR -k[MAX:16] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - Mask + AVX512_VBMI2 AVX512VL - AVX512BW - Compare - - - Compare packed unsigned 16-bit integers in "a" and "b" for less-than-or-equal, and store the results in mask vector "k". + Shift + + + + + + Concatenate packed 16-bit integers in "b" and "a" producing an intermediate 32-bit result. Shift the result right by "imm8" bits, and store the lower 16-bits in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 15 +FOR j := 0 to 7 i := j*16 - k[j] := ( a[i+15:i] <= b[i+15:i] ) ? 1 : 0 + IF k[j] + dst[i+15:i] := ((b[i+15:i] << 16)[31:0] | a[i+15:i]) >> imm8[3:0] + ELSE + dst[i+15:i] := 0 + FI ENDFOR -k[MAX:16] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - Mask + AVX512_VBMI2 AVX512VL - AVX512BW - Compare - - - Compare packed unsigned 16-bit integers in "a" and "b" for less-than, and store the results in mask vector "k". + Shift + + + + + + + Concatenate packed 16-bit integers in "b" and "a" producing an intermediate 32-bit result. Shift the result right by "imm8" bits, and store the lower 16-bits in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 15 +FOR j := 0 to 7 i := j*16 - k[j] := ( a[i+15:i] < b[i+15:i] ) ? 1 : 0 + IF k[j] + dst[i+15:i] := ((b[i+15:i] << 16)[31:0] | a[i+15:i]) >> imm8[3:0] + ELSE + dst[i+15:i] := src[i+15:i] + FI ENDFOR -k[MAX:16] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - Mask + AVX512_VBMI2 AVX512VL - AVX512BW - Compare - - - Compare packed unsigned 16-bit integers in "a" and "b" for not-equal, and store the results in mask vector "k". + Shift + + + + + Concatenate packed 16-bit integers in "b" and "a" producing an intermediate 32-bit result. Shift the result right by "imm8" bits, and store the lower 16-bits in "dst". -FOR j := 0 to 15 +FOR j := 0 to 7 i := j*16 - k[j] := ( a[i+15:i] != b[i+15:i] ) ? 1 : 0 + dst[i+15:i] := ((b[i+15:i] << 16)[31:0] | a[i+15:i]) >> imm8[3:0] ENDFOR -k[MAX:16] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - Mask - AVX512VL - AVX512BW - Compare - - - - - Compare packed unsigned 16-bit integers in "a" and "b" based on the comparison operand specified by "imm8", and store the results in mask vector "k1" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + AVX512_VBMI2 + Shift + + + + + + Concatenate packed 64-bit integers in "a" and "b" producing an intermediate 128-bit result. Shift the result left by the amount specified in the corresponding element of "c", and store the upper 64-bits in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -CASE (imm8[7:0]) OF -0: OP := _MM_CMPINT_EQ -1: OP := _MM_CMPINT_LT -2: OP := _MM_CMPINT_LE -3: OP := _MM_CMPINT_FALSE -4: OP := _MM_CMPINT_NE -5: OP := _MM_CMPINT_NLT -6: OP := _MM_CMPINT_NLE -7: OP := _MM_CMPINT_TRUE -ESAC -FOR j := 0 to 15 - i := j*16 - IF k1[j] - k[j] := ( a[i+15:i] OP b[i+15:i] ) ? 1 : 0 - ELSE - k[j] := 0 +FOR j := 0 to 7 + i := j*64 + IF k[j] + tmp[127:0] := ((a[i+63:i] << 64)[127:0] | b[i+63:i]) << (c[i+63:i] & 63) + dst[i+63:i] := tmp[127:64] + ELSE + dst[i+63:i] := 0 FI ENDFOR -k[MAX:16] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - Mask - AVX512VL - AVX512BW - Compare - - - - Compare packed unsigned 16-bit integers in "a" and "b" for equality, and store the results in mask vector "k1" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + AVX512_VBMI2 + Shift + + + + + + Concatenate packed 64-bit integers in "a" and "b" producing an intermediate 128-bit result. Shift the result left by the amount specified in the corresponding element of "c", and store the upper 64-bits in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*16 - IF k1[j] - k[j] := ( a[i+15:i] == b[i+15:i] ) ? 1 : 0 - ELSE - k[j] := 0 +FOR j := 0 to 7 + i := j*64 + IF k[j] + tmp[127:0] := ((a[i+63:i] << 64)[127:0] | b[i+63:i]) << (c[i+63:i] & 63) + dst[i+63:i] := tmp[127:64] + ELSE + dst[i+63:i] := a[i+63:i] FI ENDFOR -k[MAX:16] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - Mask - AVX512VL - AVX512BW - Compare - - - - Compare packed unsigned 16-bit integers in "a" and "b" for greater-than-or-equal, and store the results in mask vector "k1" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + AVX512_VBMI2 + Shift + + + + + Concatenate packed 64-bit integers in "a" and "b" producing an intermediate 128-bit result. Shift the result left by the amount specified in the corresponding element of "c", and store the upper 64-bits in "dst". -FOR j := 0 to 15 - i := j*16 - IF k1[j] - k[j] := ( a[i+15:i] >= b[i+15:i] ) ? 1 : 0 - ELSE - k[j] := 0 - FI +FOR j := 0 to 7 + i := j*64 + tmp[127:0] := ((a[i+63:i] << 64)[127:0] | b[i+63:i]) << (c[i+63:i] & 63) + dst[i+63:i] := tmp[127:64] ENDFOR -k[MAX:16] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - Mask + AVX512_VBMI2 AVX512VL - AVX512BW - Compare - - - - Compare packed unsigned 16-bit integers in "a" and "b" for greater-than, and store the results in mask vector "k1" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Shift + + + + + + Concatenate packed 64-bit integers in "a" and "b" producing an intermediate 128-bit result. Shift the result left by the amount specified in the corresponding element of "c", and store the upper 64-bits in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*16 - IF k1[j] - k[j] := ( a[i+15:i] > b[i+15:i] ) ? 1 : 0 - ELSE - k[j] := 0 +FOR j := 0 to 3 + i := j*64 + IF k[j] + tmp[127:0] := ((a[i+63:i] << 64)[127:0] | b[i+63:i]) << (c[i+63:i] & 63) + dst[i+63:i] := tmp[127:64] + ELSE + dst[i+63:i] := 0 FI ENDFOR -k[MAX:16] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - Mask + AVX512_VBMI2 AVX512VL - AVX512BW - Compare - - - - Compare packed unsigned 16-bit integers in "a" and "b" for less-than-or-equal, and store the results in mask vector "k1" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Shift + + + + + + Concatenate packed 64-bit integers in "a" and "b" producing an intermediate 128-bit result. Shift the result left by the amount specified in the corresponding element of "c", and store the upper 64-bits in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*16 - IF k1[j] - k[j] := ( a[i+15:i] <= b[i+15:i] ) ? 1 : 0 - ELSE - k[j] := 0 +FOR j := 0 to 3 + i := j*64 + IF k[j] + tmp[127:0] := ((a[i+63:i] << 64)[127:0] | b[i+63:i]) << (c[i+63:i] & 63) + dst[i+63:i] := tmp[127:64] + ELSE + dst[i+63:i] := a[i+63:i] FI ENDFOR -k[MAX:16] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - Mask + AVX512_VBMI2 AVX512VL - AVX512BW - Compare - - - - Compare packed unsigned 16-bit integers in "a" and "b" for less-than, and store the results in mask vector "k1" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Shift + + + + + Concatenate packed 64-bit integers in "a" and "b" producing an intermediate 128-bit result. Shift the result left by the amount specified in the corresponding element of "c", and store the upper 64-bits in "dst". -FOR j := 0 to 15 - i := j*16 - IF k1[j] - k[j] := ( a[i+15:i] < b[i+15:i] ) ? 1 : 0 - ELSE - k[j] := 0 - FI +FOR j := 0 to 3 + i := j*64 + tmp[127:0] := ((a[i+63:i] << 64)[127:0] | b[i+63:i]) << (c[i+63:i] & 63) + dst[i+63:i] := tmp[127:64] ENDFOR -k[MAX:16] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - Mask + AVX512_VBMI2 AVX512VL - AVX512BW - Compare - - - - Compare packed unsigned 16-bit integers in "a" and "b" for not-equal, and store the results in mask vector "k1" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Shift + + + + + + Concatenate packed 64-bit integers in "a" and "b" producing an intermediate 128-bit result. Shift the result left by the amount specified in the corresponding element of "c", and store the upper 64-bits in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*16 - IF k1[j] - k[j] := ( a[i+15:i] != b[i+15:i] ) ? 1 : 0 - ELSE - k[j] := 0 +FOR j := 0 to 1 + i := j*64 + IF k[j] + tmp[127:0] := ((a[i+63:i] << 64)[127:0] | b[i+63:i]) << (c[i+63:i] & 63) + dst[i+63:i] := tmp[127:64] + ELSE + dst[i+63:i] := 0 FI ENDFOR -k[MAX:16] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - Mask - AVX512BW - Compare - - - - Compare packed unsigned 16-bit integers in "a" and "b" based on the comparison operand specified by "imm8", and store the results in mask vector "k". + AVX512_VBMI2 + AVX512VL + Shift + + + + + + Concatenate packed 64-bit integers in "a" and "b" producing an intermediate 128-bit result. Shift the result left by the amount specified in the corresponding element of "c", and store the upper 64-bits in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). -CASE (imm8[7:0]) OF -0: OP := _MM_CMPINT_EQ -1: OP := _MM_CMPINT_LT -2: OP := _MM_CMPINT_LE -3: OP := _MM_CMPINT_FALSE -4: OP := _MM_CMPINT_NE -5: OP := _MM_CMPINT_NLT -6: OP := _MM_CMPINT_NLE -7: OP := _MM_CMPINT_TRUE -ESAC -FOR j := 0 to 31 - i := j*16 - k[j] := ( a[i+15:i] OP b[i+15:i] ) ? 1 : 0 +FOR j := 0 to 1 + i := j*64 + IF k[j] + tmp[127:0] := ((a[i+63:i] << 64)[127:0] | b[i+63:i]) << (c[i+63:i] & 63) + dst[i+63:i] := tmp[127:64] + ELSE + dst[i+63:i] := a[i+63:i] + FI ENDFOR -k[MAX:32] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - Mask - AVX512BW - Compare - - - Compare packed unsigned 16-bit integers in "a" and "b" for equality, and store the results in mask vector "k". + AVX512_VBMI2 + AVX512VL + Shift + + + + + Concatenate packed 64-bit integers in "a" and "b" producing an intermediate 128-bit result. Shift the result left by the amount specified in the corresponding element of "c", and store the upper 64-bits in "dst". -FOR j := 0 to 31 - i := j*16 - k[j] := ( a[i+15:i] == b[i+15:i] ) ? 1 : 0 +FOR j := 0 to 1 + i := j*64 + tmp[127:0] := ((a[i+63:i] << 64)[127:0] | b[i+63:i]) << (c[i+63:i] & 63) + dst[i+63:i] := tmp[127:64] ENDFOR -k[MAX:32] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - Mask - AVX512BW - Compare - - - Compare packed unsigned 16-bit integers in "a" and "b" for greater-than-or-equal, and store the results in mask vector "k". + AVX512_VBMI2 + Shift + + + + + + Concatenate packed 32-bit integers in "a" and "b" producing an intermediate 64-bit result. Shift the result left by the amount specified in the corresponding element of "c", and store the upper 32-bits in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 31 - i := j*16 - k[j] := ( a[i+15:i] >= b[i+15:i] ) ? 1 : 0 +FOR j := 0 to 15 + i := j*32 + IF k[j] + tmp[63:0] := ((a[i+31:i] << 32)[63:0] | b[i+31:i]) << (c[i+31:i] & 31) + dst[i+31:i] := tmp[63:32] + ELSE + dst[i+31:i] := 0 + FI ENDFOR -k[MAX:32] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - Mask - AVX512BW - Compare - - - Compare packed unsigned 16-bit integers in "a" and "b" for greater-than, and store the results in mask vector "k". + AVX512_VBMI2 + Shift + + + + + + Concatenate packed 32-bit integers in "a" and "b" producing an intermediate 64-bit result. Shift the result left by the amount specified in the corresponding element of "c", and store the upper 32-bits in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). -FOR j := 0 to 31 - i := j*16 - k[j] := ( a[i+15:i] > b[i+15:i] ) ? 1 : 0 +FOR j := 0 to 15 + i := j*32 + IF k[j] + tmp[63:0] := ((a[i+31:i] << 32)[63:0] | b[i+31:i]) << (c[i+31:i] & 31) + dst[i+31:i] := tmp[63:32] + ELSE + dst[i+31:i] := a[i+31:i] + FI ENDFOR -k[MAX:32] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - Mask - AVX512BW - Compare - - - Compare packed unsigned 16-bit integers in "a" and "b" for less-than-or-equal, and store the results in mask vector "k". + AVX512_VBMI2 + Shift + + + + + Concatenate packed 32-bit integers in "a" and "b" producing an intermediate 64-bit result. Shift the result left by the amount specified in the corresponding element of "c", and store the upper 32-bits in "dst". -FOR j := 0 to 31 - i := j*16 - k[j] := ( a[i+15:i] <= b[i+15:i] ) ? 1 : 0 +FOR j := 0 to 15 + i := j*32 + tmp[63:0] := ((a[i+31:i] << 32)[63:0] | b[i+31:i]) << (c[i+31:i] & 31) + dst[i+31:i] := tmp[63:32] ENDFOR -k[MAX:32] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - Mask - AVX512BW - Compare - - - Compare packed unsigned 16-bit integers in "a" and "b" for less-than, and store the results in mask vector "k". + AVX512_VBMI2 + AVX512VL + Shift + + + + + + Concatenate packed 32-bit integers in "a" and "b" producing an intermediate 64-bit result. Shift the result left by the amount specified in the corresponding element of "c", and store the upper 32-bits in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 31 - i := j*16 - k[j] := ( a[i+15:i] < b[i+15:i] ) ? 1 : 0 +FOR j := 0 to 7 + i := j*32 + IF k[j] + tmp[63:0] := ((a[i+31:i] << 32)[63:0] | b[i+31:i]) << (c[i+31:i] & 31) + dst[i+31:i] := tmp[63:32] + ELSE + dst[i+31:i] := 0 + FI ENDFOR -k[MAX:32] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - Mask - AVX512BW - Compare - - - Compare packed unsigned 16-bit integers in "a" and "b" for not-equal, and store the results in mask vector "k". + AVX512_VBMI2 + AVX512VL + Shift + + + + + + Concatenate packed 32-bit integers in "a" and "b" producing an intermediate 64-bit result. Shift the result left by the amount specified in the corresponding element of "c", and store the upper 32-bits in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). -FOR j := 0 to 31 - i := j*16 - k[j] := ( a[i+15:i] != b[i+15:i] ) ? 1 : 0 +FOR j := 0 to 7 + i := j*32 + IF k[j] + tmp[63:0] := ((a[i+31:i] << 32)[63:0] | b[i+31:i]) << (c[i+31:i] & 31) + dst[i+31:i] := tmp[63:32] + ELSE + dst[i+31:i] := a[i+31:i] + FI ENDFOR -k[MAX:32] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - Mask - AVX512BW - Compare - - - - - Compare packed unsigned 16-bit integers in "a" and "b" based on the comparison operand specified by "imm8", and store the results in mask vector "k1" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + AVX512_VBMI2 + AVX512VL + Shift + + + + + Concatenate packed 32-bit integers in "a" and "b" producing an intermediate 64-bit result. Shift the result left by the amount specified in the corresponding element of "c", and store the upper 32-bits in "dst". -CASE (imm8[7:0]) OF -0: OP := _MM_CMPINT_EQ -1: OP := _MM_CMPINT_LT -2: OP := _MM_CMPINT_LE -3: OP := _MM_CMPINT_FALSE -4: OP := _MM_CMPINT_NE -5: OP := _MM_CMPINT_NLT -6: OP := _MM_CMPINT_NLE -7: OP := _MM_CMPINT_TRUE -ESAC -FOR j := 0 to 31 - i := j*16 - IF k1[j] - k[j] := ( a[i+15:i] OP b[i+15:i] ) ? 1 : 0 - ELSE - k[j] := 0 - FI +FOR j := 0 to 7 + i := j*32 + tmp[63:0] := ((a[i+31:i] << 32)[63:0] | b[i+31:i]) << (c[i+31:i] & 31) + dst[i+31:i] := tmp[63:32] ENDFOR -k[MAX:32] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - Mask - AVX512BW - Compare - - - - Compare packed unsigned 16-bit integers in "a" and "b" for equality, and store the results in mask vector "k1" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + AVX512_VBMI2 + AVX512VL + Shift + + + + + + Concatenate packed 32-bit integers in "a" and "b" producing an intermediate 64-bit result. Shift the result left by the amount specified in the corresponding element of "c", and store the upper 32-bits in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 31 - i := j*16 - IF k1[j] - k[j] := ( a[i+15:i] == b[i+15:i] ) ? 1 : 0 - ELSE - k[j] := 0 +FOR j := 0 to 3 + i := j*32 + IF k[j] + tmp[63:0] := ((a[i+31:i] << 32)[63:0] | b[i+31:i]) << (c[i+31:i] & 31) + dst[i+31:i] := tmp[63:32] + ELSE + dst[i+31:i] := 0 FI ENDFOR -k[MAX:32] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - Mask - AVX512BW - Compare - - - - Compare packed unsigned 16-bit integers in "a" and "b" for greater-than-or-equal, and store the results in mask vector "k1" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + AVX512_VBMI2 + AVX512VL + Shift + + + + + + Concatenate packed 32-bit integers in "a" and "b" producing an intermediate 64-bit result. Shift the result left by the amount specified in the corresponding element of "c", and store the upper 32-bits in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). -FOR j := 0 to 31 - i := j*16 - IF k1[j] - k[j] := ( a[i+15:i] >= b[i+15:i] ) ? 1 : 0 - ELSE - k[j] := 0 +FOR j := 0 to 3 + i := j*32 + IF k[j] + tmp[63:0] := ((a[i+31:i] << 32)[63:0] | b[i+31:i]) << (c[i+31:i] & 31) + dst[i+31:i] := tmp[63:32] + ELSE + dst[i+31:i] := a[i+31:i] FI ENDFOR -k[MAX:32] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - Mask - AVX512BW - Compare - - - - Compare packed unsigned 16-bit integers in "a" and "b" for greater-than, and store the results in mask vector "k1" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + AVX512_VBMI2 + AVX512VL + Shift + + + + + Concatenate packed 32-bit integers in "a" and "b" producing an intermediate 64-bit result. Shift the result left by the amount specified in the corresponding element of "c", and store the upper 32-bits in "dst". -FOR j := 0 to 31 - i := j*16 - IF k1[j] - k[j] := ( a[i+15:i] > b[i+15:i] ) ? 1 : 0 - ELSE - k[j] := 0 - FI +FOR j := 0 to 3 + i := j*32 + tmp[63:0] := ((a[i+31:i] << 32)[63:0] | b[i+31:i]) << (c[i+31:i] & 31) + dst[i+31:i] := tmp[63:32] ENDFOR -k[MAX:32] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - Mask - AVX512BW - Compare - - - - Compare packed unsigned 16-bit integers in "a" and "b" for less-than-or-equal, and store the results in mask vector "k1" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + AVX512_VBMI2 + Shift + + + + + + Concatenate packed 16-bit integers in "a" and "b" producing an intermediate 32-bit result. Shift the result left by the amount specified in the corresponding element of "c", and store the upper 16-bits in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). FOR j := 0 to 31 i := j*16 - IF k1[j] - k[j] := ( a[i+15:i] <= b[i+15:i] ) ? 1 : 0 - ELSE - k[j] := 0 + IF k[j] + tmp[31:0] := ((a[i+15:i] << 16)[31:0] | b[i+15:i]) << (c[i+15:i] & 15) + dst[i+15:i] := tmp[31:16] + ELSE + dst[i+15:i] := 0 FI ENDFOR -k[MAX:32] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - Mask - AVX512BW - Compare - - - - Compare packed unsigned 16-bit integers in "a" and "b" for less-than, and store the results in mask vector "k1" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + AVX512_VBMI2 + Shift + + + + + + Concatenate packed 16-bit integers in "a" and "b" producing an intermediate 32-bit result. Shift the result left by the amount specified in the corresponding element of "c", and store the upper 16-bits in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). FOR j := 0 to 31 i := j*16 - IF k1[j] - k[j] := ( a[i+15:i] < b[i+15:i] ) ? 1 : 0 - ELSE - k[j] := 0 + IF k[j] + tmp[31:0] := ((a[i+15:i] << 16)[31:0] | b[i+15:i]) << (c[i+15:i] & 15) + dst[i+15:i] := tmp[31:16] + ELSE + dst[i+15:i] := a[i+15:i] FI ENDFOR -k[MAX:32] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - Mask - AVX512BW - Compare - - - - Compare packed unsigned 16-bit integers in "a" and "b" for not-equal, and store the results in mask vector "k1" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + AVX512_VBMI2 + Shift + + + + + Concatenate packed 16-bit integers in "a" and "b" producing an intermediate 32-bit result. Shift the result left by the amount specified in the corresponding element of "c", and store the upper 16-bits in "dst". FOR j := 0 to 31 i := j*16 - IF k1[j] - k[j] := ( a[i+15:i] != b[i+15:i] ) ? 1 : 0 - ELSE - k[j] := 0 - FI + tmp[31:0] := ((a[i+15:i] << 16)[31:0] | b[i+15:i]) << (c[i+15:i] & 15) + dst[i+15:i] := tmp[31:16] ENDFOR -k[MAX:32] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - Mask + AVX512_VBMI2 AVX512VL - AVX512BW - Compare - - - - Compare packed unsigned 16-bit integers in "a" and "b" based on the comparison operand specified by "imm8", and store the results in mask vector "k". + Shift + + + + + + Concatenate packed 16-bit integers in "a" and "b" producing an intermediate 32-bit result. Shift the result left by the amount specified in the corresponding element of "c", and store the upper 16-bits in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -CASE (imm8[7:0]) OF -0: OP := _MM_CMPINT_EQ -1: OP := _MM_CMPINT_LT -2: OP := _MM_CMPINT_LE -3: OP := _MM_CMPINT_FALSE -4: OP := _MM_CMPINT_NE -5: OP := _MM_CMPINT_NLT -6: OP := _MM_CMPINT_NLE -7: OP := _MM_CMPINT_TRUE -ESAC -FOR j := 0 to 7 +FOR j := 0 to 15 i := j*16 - k[j] := ( a[i+15:i] OP b[i+15:i] ) ? 1 : 0 + IF k[j] + tmp[31:0] := ((a[i+15:i] << 16)[31:0] | b[i+15:i]) << (c[i+15:i] & 15) + dst[i+15:i] := tmp[31:16] + ELSE + dst[i+15:i] := 0 + FI ENDFOR -k[MAX:8] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - Mask + AVX512_VBMI2 AVX512VL - AVX512BW - Compare - - - Compare packed unsigned 16-bit integers in "a" and "b" for equality, and store the results in mask vector "k". + Shift + + + + + + Concatenate packed 16-bit integers in "a" and "b" producing an intermediate 32-bit result. Shift the result left by the amount specified in the corresponding element of "c", and store the upper 16-bits in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). -FOR j := 0 to 7 +FOR j := 0 to 15 i := j*16 - k[j] := ( a[i+15:i] == b[i+15:i] ) ? 1 : 0 + IF k[j] + tmp[31:0] := ((a[i+15:i] << 16)[31:0] | b[i+15:i]) << (c[i+15:i] & 15) + dst[i+15:i] := tmp[31:16] + ELSE + dst[i+15:i] := a[i+15:i] + FI ENDFOR -k[MAX:8] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - Mask + AVX512_VBMI2 AVX512VL - AVX512BW - Compare - - - Compare packed unsigned 16-bit integers in "a" and "b" for greater-than-or-equal, and store the results in mask vector "k". + Shift + + + + + Concatenate packed 16-bit integers in "a" and "b" producing an intermediate 32-bit result. Shift the result left by the amount specified in the corresponding element of "c", and store the upper 16-bits in "dst". -FOR j := 0 to 7 +FOR j := 0 to 15 i := j*16 - k[j] := ( a[i+15:i] >= b[i+15:i] ) ? 1 : 0 + tmp[31:0] := ((a[i+15:i] << 16)[31:0] | b[i+15:i]) << (c[i+15:i] & 15) + dst[i+15:i] := tmp[31:16] ENDFOR -k[MAX:8] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - Mask + AVX512_VBMI2 AVX512VL - AVX512BW - Compare - - - Compare packed unsigned 16-bit integers in "a" and "b" for greater-than, and store the results in mask vector "k". + Shift + + + + + + Concatenate packed 16-bit integers in "a" and "b" producing an intermediate 32-bit result. Shift the result left by the amount specified in the corresponding element of "c", and store the upper 16-bits in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). FOR j := 0 to 7 i := j*16 - k[j] := ( a[i+15:i] > b[i+15:i] ) ? 1 : 0 + IF k[j] + tmp[31:0] := ((a[i+15:i] << 16)[31:0] | b[i+15:i]) << (c[i+15:i] & 15) + dst[i+15:i] := tmp[31:16] + ELSE + dst[i+15:i] := 0 + FI ENDFOR -k[MAX:8] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - Mask + AVX512_VBMI2 AVX512VL - AVX512BW - Compare - - - Compare packed unsigned 16-bit integers in "a" and "b" for less-than-or-equal, and store the results in mask vector "k". + Shift + + + + + + Concatenate packed 16-bit integers in "a" and "b" producing an intermediate 32-bit result. Shift the result left by the amount specified in the corresponding element of "c", and store the upper 16-bits in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). FOR j := 0 to 7 i := j*16 - k[j] := ( a[i+15:i] <= b[i+15:i] ) ? 1 : 0 + IF k[j] + tmp[31:0] := ((a[i+15:i] << 16)[31:0] | b[i+15:i]) << (c[i+15:i] & 15) + dst[i+15:i] := tmp[31:16] + ELSE + dst[i+15:i] := a[i+15:i] + FI ENDFOR -k[MAX:8] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - Mask + AVX512_VBMI2 AVX512VL - AVX512BW - Compare - - - Compare packed unsigned 16-bit integers in "a" and "b" for less-than, and store the results in mask vector "k". + Shift + + + + + Concatenate packed 16-bit integers in "a" and "b" producing an intermediate 32-bit result. Shift the result left by the amount specified in the corresponding element of "c", and store the upper 16-bits in "dst". FOR j := 0 to 7 i := j*16 - k[j] := ( a[i+15:i] < b[i+15:i] ) ? 1 : 0 + tmp[31:0] := ((a[i+15:i] << 16)[31:0] | b[i+15:i]) << (c[i+15:i] & 15) + dst[i+15:i] := tmp[31:16] ENDFOR -k[MAX:8] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - Mask - AVX512VL - AVX512BW - Compare - - - Compare packed unsigned 16-bit integers in "a" and "b" for not-equal, and store the results in mask vector "k". + AVX512_VBMI2 + Shift + + + + + + Concatenate packed 64-bit integers in "a" and "b" producing an intermediate 128-bit result. Shift the result left by "imm8" bits, and store the upper 64-bits in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). FOR j := 0 to 7 - i := j*16 - k[j] := ( a[i+15:i] != b[i+15:i] ) ? 1 : 0 + i := j*64 + IF k[j] + tmp[127:0] := ((a[i+63:i] << 64)[127:0] | b[i+63:i]) << imm8[5:0] + dst[i+63:i] := tmp[127:64] + ELSE + dst[i+63:i] := 0 + FI ENDFOR -k[MAX:8] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - Mask - AVX512VL - AVX512BW - Compare - - - - - Compare packed unsigned 16-bit integers in "a" and "b" based on the comparison operand specified by "imm8", and store the results in mask vector "k1" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + AVX512_VBMI2 + Shift + + + + + + + Concatenate packed 64-bit integers in "a" and "b" producing an intermediate 128-bit result. Shift the result left by "imm8" bits, and store the upper 64-bits in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -CASE (imm8[7:0]) OF -0: OP := _MM_CMPINT_EQ -1: OP := _MM_CMPINT_LT -2: OP := _MM_CMPINT_LE -3: OP := _MM_CMPINT_FALSE -4: OP := _MM_CMPINT_NE -5: OP := _MM_CMPINT_NLT -6: OP := _MM_CMPINT_NLE -7: OP := _MM_CMPINT_TRUE -ESAC FOR j := 0 to 7 - i := j*16 - IF k1[j] - k[j] := ( a[i+15:i] OP b[i+15:i] ) ? 1 : 0 - ELSE - k[j] := 0 + i := j*64 + IF k[j] + tmp[127:0] := ((a[i+63:i] << 64)[127:0] | b[i+63:i]) << imm8[5:0] + dst[i+63:i] := tmp[127:64] + ELSE + dst[i+63:i] := src[i+63:i] FI ENDFOR -k[MAX:8] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - Mask - AVX512VL - AVX512BW - Compare - - - - Compare packed unsigned 16-bit integers in "a" and "b" for equality, and store the results in mask vector "k1" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + AVX512_VBMI2 + Shift + + + + + Concatenate packed 64-bit integers in "a" and "b" producing an intermediate 128-bit result. Shift the result left by "imm8" bits, and store the upper 64-bits in "dst"). FOR j := 0 to 7 - i := j*16 - IF k1[j] - k[j] := ( a[i+15:i] == b[i+15:i] ) ? 1 : 0 - ELSE - k[j] := 0 - FI + i := j*64 + tmp[127:0] := ((a[i+63:i] << 64)[127:0] | b[i+63:i]) << imm8[5:0] + dst[i+63:i] := tmp[127:64] ENDFOR -k[MAX:8] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - Mask + AVX512_VBMI2 AVX512VL - AVX512BW - Compare - - - - Compare packed unsigned 16-bit integers in "a" and "b" for greater-than-or-equal, and store the results in mask vector "k1" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Shift + + + + + + Concatenate packed 64-bit integers in "a" and "b" producing an intermediate 128-bit result. Shift the result left by "imm8" bits, and store the upper 64-bits in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := j*16 - IF k1[j] - k[j] := ( a[i+15:i] >= b[i+15:i] ) ? 1 : 0 - ELSE - k[j] := 0 +FOR j := 0 to 3 + i := j*64 + IF k[j] + tmp[127:0] := ((a[i+63:i] << 64)[127:0] | b[i+63:i]) << imm8[5:0] + dst[i+63:i] := tmp[127:64] + ELSE + dst[i+63:i] := 0 FI ENDFOR -k[MAX:8] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - Mask + AVX512_VBMI2 AVX512VL - AVX512BW - Compare - - - - Compare packed unsigned 16-bit integers in "a" and "b" for greater-than, and store the results in mask vector "k1" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Shift + + + + + + + Concatenate packed 64-bit integers in "a" and "b" producing an intermediate 128-bit result. Shift the result left by "imm8" bits, and store the upper 64-bits in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := j*16 - IF k1[j] - k[j] := ( a[i+15:i] >== b[i+15:i] ) ? 1 : 0 - ELSE - k[j] := 0 +FOR j := 0 to 3 + i := j*64 + IF k[j] + tmp[127:0] := ((a[i+63:i] << 64)[127:0] | b[i+63:i]) << imm8[5:0] + dst[i+63:i] := tmp[127:64] + ELSE + dst[i+63:i] := src[i+63:i] FI ENDFOR -k[MAX:8] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - Mask + AVX512_VBMI2 AVX512VL - AVX512BW - Compare - - - - Compare packed unsigned 16-bit integers in "a" and "b" for less-than-or-equal, and store the results in mask vector "k1" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Shift + + + + + Concatenate packed 64-bit integers in "a" and "b" producing an intermediate 128-bit result. Shift the result left by "imm8" bits, and store the upper 64-bits in "dst"). -FOR j := 0 to 7 - i := j*16 - IF k1[j] - k[j] := ( a[i+15:i] <= b[i+15:i] ) ? 1 : 0 - ELSE - k[j] := 0 - FI +FOR j := 0 to 3 + i := j*64 + tmp[127:0] := ((a[i+63:i] << 64)[127:0] | b[i+63:i]) << imm8[5:0] + dst[i+63:i] := tmp[127:64] ENDFOR -k[MAX:8] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - Mask + AVX512_VBMI2 AVX512VL - AVX512BW - Compare - - - - Compare packed unsigned 16-bit integers in "a" and "b" for less-than, and store the results in mask vector "k1" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Shift + + + + + + Concatenate packed 64-bit integers in "a" and "b" producing an intermediate 128-bit result. Shift the result left by "imm8" bits, and store the upper 64-bits in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := j*16 - IF k1[j] - k[j] := ( a[i+15:i] < b[i+15:i] ) ? 1 : 0 - ELSE - k[j] := 0 +FOR j := 0 to 1 + i := j*64 + IF k[j] + tmp[127:0] := ((a[i+63:i] << 64)[127:0] | b[i+63:i]) << imm8[5:0] + dst[i+63:i] := tmp[127:64] + ELSE + dst[i+63:i] := 0 FI ENDFOR -k[MAX:8] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - Mask + AVX512_VBMI2 AVX512VL - AVX512BW - Compare - - - - Compare packed unsigned 16-bit integers in "a" and "b" for not-equal, and store the results in mask vector "k1" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Shift + + + + + + + Concatenate packed 64-bit integers in "a" and "b" producing an intermediate 128-bit result. Shift the result left by "imm8" bits, and store the upper 64-bits in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := j*16 - IF k1[j] - k[j] := ( a[i+15:i] != b[i+15:i] ) ? 1 : 0 - ELSE - k[j] := 0 +FOR j := 0 to 1 + i := j*64 + IF k[j] + tmp[127:0] := ((a[i+63:i] << 64)[127:0] | b[i+63:i]) << imm8[5:0] + dst[i+63:i] := tmp[127:64] + ELSE + dst[i+63:i] := src[i+63:i] FI ENDFOR -k[MAX:8] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - Mask + AVX512_VBMI2 AVX512VL - AVX512BW - Compare - - - - Compare packed 16-bit integers in "a" and "b" based on the comparison operand specified by "imm8", and store the results in mask vector "k". + Shift + + + + + Concatenate packed 64-bit integers in "a" and "b" producing an intermediate 128-bit result. Shift the result left by "imm8" bits, and store the upper 64-bits in "dst"). -CASE (imm8[7:0]) OF -0: OP := _MM_CMPINT_EQ -1: OP := _MM_CMPINT_LT -2: OP := _MM_CMPINT_LE -3: OP := _MM_CMPINT_FALSE -4: OP := _MM_CMPINT_NE -5: OP := _MM_CMPINT_NLT -6: OP := _MM_CMPINT_NLE -7: OP := _MM_CMPINT_TRUE -ESAC -FOR j := 0 to 15 - i := j*16 - k[j] := ( a[i+15:i] OP b[i+15:i] ) ? 1 : 0 +FOR j := 0 to 1 + i := j*64 + tmp[127:0] := ((a[i+63:i] << 64)[127:0] | b[i+63:i]) << imm8[5:0] + dst[i+63:i] := tmp[127:64] ENDFOR -k[MAX:16] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - Mask - AVX512VL - AVX512BW - Compare - - - Compare packed 16-bit integers in "a" and "b" for equality, and store the results in mask vector "k". + AVX512_VBMI2 + Shift + + + + + + Concatenate packed 32-bit integers in "a" and "b" producing an intermediate 64-bit result. Shift the result left by "imm8" bits, and store the upper 32-bits in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). FOR j := 0 to 15 - i := j*16 - k[j] := ( a[i+15:i] == b[i+15:i] ) ? 1 : 0 + i := j*32 + IF k[j] + tmp[63:0] := ((a[i+31:i] << 32)[63:0] | b[i+31:i]) << imm8[4:0] + dst[i+31:i] := tmp[63:32] + ELSE + dst[i+31:i] := 0 + FI ENDFOR -k[MAX:16] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - Mask - AVX512VL - AVX512BW - Compare - - - Compare packed 16-bit integers in "a" and "b" for greater-than-or-equal, and store the results in mask vector "k". + AVX512_VBMI2 + Shift + + + + + + + Concatenate packed 32-bit integers in "a" and "b" producing an intermediate 64-bit result. Shift the result left by "imm8" bits, and store the upper 32-bits in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). FOR j := 0 to 15 - i := j*16 - k[j] := ( a[i+15:i] >= b[i+15:i] ) ? 1 : 0 + i := j*32 + IF k[j] + tmp[63:0] := ((a[i+31:i] << 32)[63:0] | b[i+31:i]) << imm8[4:0] + dst[i+31:i] := tmp[63:32] + ELSE + dst[i+31:i] := src[i+31:i] + FI ENDFOR -k[MAX:16] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - Mask - AVX512VL - AVX512BW - Compare - - - Compare packed 16-bit integers in "a" and "b" for greater-than, and store the results in mask vector "k". + AVX512_VBMI2 + Shift + + + + + Concatenate packed 32-bit integers in "a" and "b" producing an intermediate 64-bit result. Shift the result left by "imm8" bits, and store the upper 32-bits in "dst". FOR j := 0 to 15 - i := j*16 - k[j] := ( a[i+15:i] > b[i+15:i] ) ? 1 : 0 + i := j*32 + tmp[63:0] := ((a[i+31:i] << 32)[63:0] | b[i+31:i]) << imm8[4:0] + dst[i+31:i] := tmp[63:32] ENDFOR -k[MAX:16] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - Mask + AVX512_VBMI2 AVX512VL - AVX512BW - Compare - - - Compare packed 16-bit integers in "a" and "b" for less-than-or-equal, and store the results in mask vector "k". + Shift + + + + + + Concatenate packed 32-bit integers in "a" and "b" producing an intermediate 64-bit result. Shift the result left by "imm8" bits, and store the upper 32-bits in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*16 - k[j] := ( a[i+15:i] <= b[i+15:i] ) ? 1 : 0 +FOR j := 0 to 7 + i := j*32 + IF k[j] + tmp[63:0] := ((a[i+31:i] << 32)[63:0] | b[i+31:i]) << imm8[4:0] + dst[i+31:i] := tmp[63:32] + ELSE + dst[i+31:i] := 0 + FI ENDFOR -k[MAX:16] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - Mask + AVX512_VBMI2 AVX512VL - AVX512BW - Compare - - - Compare packed 16-bit integers in "a" and "b" for less-than, and store the results in mask vector "k". + Shift + + + + + + + Concatenate packed 32-bit integers in "a" and "b" producing an intermediate 64-bit result. Shift the result left by "imm8" bits, and store the upper 32-bits in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*16 - k[j] := ( a[i+15:i] < b[i+15:i] ) ? 1 : 0 +FOR j := 0 to 7 + i := j*32 + IF k[j] + tmp[63:0] := ((a[i+31:i] << 32)[63:0] | b[i+31:i]) << imm8[4:0] + dst[i+31:i] := tmp[63:32] + ELSE + dst[i+31:i] := src[i+31:i] + FI ENDFOR -k[MAX:16] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - Mask + AVX512_VBMI2 AVX512VL - AVX512BW - Compare - - - Compare packed 16-bit integers in "a" and "b" for not-equal, and store the results in mask vector "k". + Shift + + + + + Concatenate packed 32-bit integers in "a" and "b" producing an intermediate 64-bit result. Shift the result left by "imm8" bits, and store the upper 32-bits in "dst". -FOR j := 0 to 15 - i := j*16 - k[j] := ( a[i+15:i] != b[i+15:i] ) ? 1 : 0 +FOR j := 0 to 7 + i := j*32 + tmp[63:0] := ((a[i+31:i] << 32)[63:0] | b[i+31:i]) << imm8[4:0] + dst[i+31:i] := tmp[63:32] ENDFOR -k[MAX:16] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - Mask + AVX512_VBMI2 AVX512VL - AVX512BW - Compare - - - - - Compare packed 16-bit integers in "a" and "b" based on the comparison operand specified by "imm8", and store the results in mask vector "k1" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Shift + + + + + + Concatenate packed 32-bit integers in "a" and "b" producing an intermediate 64-bit result. Shift the result left by "imm8" bits, and store the upper 32-bits in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -CASE (imm8[7:0]) OF -0: OP := _MM_CMPINT_EQ -1: OP := _MM_CMPINT_LT -2: OP := _MM_CMPINT_LE -3: OP := _MM_CMPINT_FALSE -4: OP := _MM_CMPINT_NE -5: OP := _MM_CMPINT_NLT -6: OP := _MM_CMPINT_NLE -7: OP := _MM_CMPINT_TRUE -ESAC -FOR j := 0 to 15 - i := j*16 - IF k1[j] - k[j] := ( a[i+15:i] OP b[i+15:i] ) ? 1 : 0 - ELSE - k[j] := 0 +FOR j := 0 to 3 + i := j*32 + IF k[j] + tmp[63:0] := ((a[i+31:i] << 32)[63:0] | b[i+31:i]) << imm8[4:0] + dst[i+31:i] := tmp[63:32] + ELSE + dst[i+31:i] := 0 FI ENDFOR -k[MAX:16] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - Mask + AVX512_VBMI2 AVX512VL - AVX512BW - Compare - - - - Compare packed 16-bit integers in "a" and "b" for equality, and store the results in mask vector "k1" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Shift + + + + + + + Concatenate packed 32-bit integers in "a" and "b" producing an intermediate 64-bit result. Shift the result left by "imm8" bits, and store the upper 32-bits in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 15 - i := j*16 - IF k1[j] - k[j] := ( a[i+15:i] == b[i+15:i] ) ? 1 : 0 - ELSE - k[j] := 0 +FOR j := 0 to 3 + i := j*32 + IF k[j] + tmp[63:0] := ((a[i+31:i] << 32)[63:0] | b[i+31:i]) << imm8[4:0] + dst[i+31:i] := tmp[63:32] + ELSE + dst[i+31:i] := src[i+31:i] FI ENDFOR -k[MAX:16] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - Mask + AVX512_VBMI2 AVX512VL - AVX512BW - Compare - - - - Compare packed 16-bit integers in "a" and "b" for greater-than-or-equal, and store the results in mask vector "k1" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Shift + + + + + Concatenate packed 32-bit integers in "a" and "b" producing an intermediate 64-bit result. Shift the result left by "imm8" bits, and store the upper 32-bits in "dst". -FOR j := 0 to 15 - i := j*16 - IF k1[j] - k[j] := ( a[i+15:i] >= b[i+15:i] ) ? 1 : 0 - ELSE - k[j] := 0 - FI +FOR j := 0 to 3 + i := j*32 + tmp[63:0] := ((a[i+31:i] << 32)[63:0] | b[i+31:i]) << imm8[4:0] + dst[i+31:i] := tmp[63:32] ENDFOR -k[MAX:16] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - Mask - AVX512VL - AVX512BW - Compare - - - - Compare packed 16-bit integers in "a" and "b" for greater-than, and store the results in mask vector "k1" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + AVX512_VBMI2 + Shift + + + + + + Concatenate packed 16-bit integers in "a" and "b" producing an intermediate 32-bit result. Shift the result left by "imm8" bits, and store the upper 16-bits in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 15 +FOR j := 0 to 31 i := j*16 - IF k1[j] - k[j] := ( a[i+15:i] > b[i+15:i] ) ? 1 : 0 - ELSE - k[j] := 0 + IF k[j] + tmp[31:0] := ((a[i+15:i] << 16)[31:0] | b[i+15:i]) << imm8[3:0] + dst[i+15:i] := tmp[31:16] + ELSE + dst[i+15:i] := 0 FI ENDFOR -k[MAX:16] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - Mask - AVX512VL - AVX512BW - Compare - - - - Compare packed 16-bit integers in "a" and "b" for less-than-or-equal, and store the results in mask vector "k1" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + AVX512_VBMI2 + Shift + + + + + + + Concatenate packed 16-bit integers in "a" and "b" producing an intermediate 32-bit result. Shift the result left by "imm8" bits, and store the upper 16-bits in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 15 +FOR j := 0 to 31 i := j*16 - IF k1[j] - k[j] := ( a[i+15:i] <= b[i+15:i] ) ? 1 : 0 - ELSE - k[j] := 0 + IF k[j] + tmp[31:0] := ((a[i+15:i] << 16)[31:0] | b[i+15:i]) << imm8[3:0] + dst[i+15:i] := tmp[31:16] + ELSE + dst[i+15:i] := src[i+15:i] FI ENDFOR -k[MAX:16] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - Mask - AVX512VL - AVX512BW - Compare - - - - Compare packed 16-bit integers in "a" and "b" for less-than, and store the results in mask vector "k1" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + AVX512_VBMI2 + Shift + + + + + Concatenate packed 16-bit integers in "a" and "b" producing an intermediate 32-bit result. Shift the result left by "imm8" bits, and store the upper 16-bits in "dst"). -FOR j := 0 to 15 +FOR j := 0 to 31 i := j*16 - IF k1[j] - k[j] := ( a[i+15:i] < b[i+15:i] ) ? 1 : 0 - ELSE - k[j] := 0 - FI + tmp[31:0] := ((a[i+15:i] << 16)[31:0] | b[i+15:i]) << imm8[3:0] + dst[i+15:i] := tmp[31:16] ENDFOR -k[MAX:16] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - Mask + AVX512_VBMI2 AVX512VL - AVX512BW - Compare - - - - Compare packed 16-bit integers in "a" and "b" for not-equal, and store the results in mask vector "k1" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Shift + + + + + + Concatenate packed 16-bit integers in "a" and "b" producing an intermediate 32-bit result. Shift the result left by "imm8" bits, and store the upper 16-bits in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). FOR j := 0 to 15 i := j*16 - IF k1[j] - k[j] := ( a[i+15:i] != b[i+15:i] ) ? 1 : 0 - ELSE - k[j] := 0 + IF k[j] + tmp[31:0] := ((a[i+15:i] << 16)[31:0] | b[i+15:i]) << imm8[3:0] + dst[i+15:i] := tmp[31:16] + ELSE + dst[i+15:i] := 0 FI ENDFOR -k[MAX:16] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - Mask - AVX512BW - Compare - - - - Compare packed 16-bit integers in "a" and "b" based on the comparison operand specified by "imm8", and store the results in mask vector "k". + AVX512_VBMI2 + AVX512VL + Shift + + + + + + + Concatenate packed 16-bit integers in "a" and "b" producing an intermediate 32-bit result. Shift the result left by "imm8" bits, and store the upper 16-bits in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -CASE (imm8[7:0]) OF -0: OP := _MM_CMPINT_EQ -1: OP := _MM_CMPINT_LT -2: OP := _MM_CMPINT_LE -3: OP := _MM_CMPINT_FALSE -4: OP := _MM_CMPINT_NE -5: OP := _MM_CMPINT_NLT -6: OP := _MM_CMPINT_NLE -7: OP := _MM_CMPINT_TRUE -ESAC -FOR j := 0 to 31 +FOR j := 0 to 15 i := j*16 - k[j] := ( a[i+15:i] OP b[i+15:i] ) ? 1 : 0 + IF k[j] + tmp[31:0] := ((a[i+15:i] << 16)[31:0] | b[i+15:i]) << imm8[3:0] + dst[i+15:i] := tmp[31:16] + ELSE + dst[i+15:i] := src[i+15:i] + FI ENDFOR -k[MAX:32] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - Mask - AVX512BW - Compare - - - Compare packed 16-bit integers in "a" and "b" for equality, and store the results in mask vector "k". + AVX512_VBMI2 + AVX512VL + Shift + + + + + Concatenate packed 16-bit integers in "a" and "b" producing an intermediate 32-bit result. Shift the result left by "imm8" bits, and store the upper 16-bits in "dst"). -FOR j := 0 to 31 +FOR j := 0 to 15 i := j*16 - k[j] := ( a[i+15:i] == b[i+15:i] ) ? 1 : 0 + tmp[31:0] := ((a[i+15:i] << 16)[31:0] | b[i+15:i]) << imm8[3:0] + dst[i+15:i] := tmp[31:16] ENDFOR -k[MAX:32] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - Mask - AVX512BW - Compare - - - Compare packed 16-bit integers in "a" and "b" for greater-than-or-equal, and store the results in mask vector "k". + AVX512_VBMI2 + AVX512VL + Shift + + + + + + Concatenate packed 16-bit integers in "a" and "b" producing an intermediate 32-bit result. Shift the result left by "imm8" bits, and store the upper 16-bits in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 31 +FOR j := 0 to 7 i := j*16 - k[j] := ( a[i+15:i] >= b[i+15:i] ) ? 1 : 0 + IF k[j] + tmp[31:0] := ((a[i+15:i] << 16)[31:0] | b[i+15:i]) << imm8[3:0] + dst[i+15:i] := tmp[31:16] + ELSE + dst[i+15:i] := 0 + FI ENDFOR -k[MAX:32] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - Mask - AVX512BW - Compare - - - Compare packed 16-bit integers in "a" and "b" for greater-than, and store the results in mask vector "k". + AVX512_VBMI2 + AVX512VL + Shift + + + + + + + Concatenate packed 16-bit integers in "a" and "b" producing an intermediate 32-bit result. Shift the result left by "imm8" bits, and store the upper 16-bits in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 31 +FOR j := 0 to 7 i := j*16 - k[j] := ( a[i+15:i] > b[i+15:i] ) ? 1 : 0 + IF k[j] + tmp[31:0] := ((a[i+15:i] << 16)[31:0] | b[i+15:i]) << imm8[3:0] + dst[i+15:i] := tmp[31:16] + ELSE + dst[i+15:i] := src[i+15:i] + FI ENDFOR -k[MAX:32] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - Mask - AVX512BW - Compare - - - Compare packed 16-bit integers in "a" and "b" for less-than-or-equal, and store the results in mask vector "k". + AVX512_VBMI2 + AVX512VL + Shift + + + + + Concatenate packed 16-bit integers in "a" and "b" producing an intermediate 32-bit result. Shift the result left by "imm8" bits, and store the upper 16-bits in "dst"). -FOR j := 0 to 31 +FOR j := 0 to 7 i := j*16 - k[j] := ( a[i+15:i] <= b[i+15:i] ) ? 1 : 0 + tmp[31:0] := ((a[i+15:i] << 16)[31:0] | b[i+15:i]) << imm8[3:0] + dst[i+15:i] := tmp[31:16] ENDFOR -k[MAX:32] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - Mask - AVX512BW - Compare - - - Compare packed 16-bit integers in "a" and "b" for less-than, and store the results in mask vector "k". + AVX512_VBMI2 + Load + Swizzle + + + + Load contiguous active 16-bit integers from unaligned memory at "mem_addr" (those with their respective bit set in mask "k"), and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). +m := 0 FOR j := 0 to 31 i := j*16 - k[j] := ( a[i+15:i] < b[i+15:i] ) ? 1 : 0 + IF k[j] + dst[i+15:i] := MEM[mem_addr+m+15:mem_addr+m] + m := m + 16 + ELSE + dst[i+15:i] := 0 + FI ENDFOR -k[MAX:32] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - Mask - AVX512BW - Compare - - - Compare packed 16-bit integers in "a" and "b" for not-equal, and store the results in mask vector "k". + AVX512_VBMI2 + Load + Swizzle + + + + + Load contiguous active 16-bit integers from unaligned memory at "mem_addr" (those with their respective bit set in mask "k"), and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). +m := 0 FOR j := 0 to 31 i := j*16 - k[j] := ( a[i+15:i] != b[i+15:i] ) ? 1 : 0 + IF k[j] + dst[i+15:i] := MEM[mem_addr+m+15:mem_addr+m] + m := m + 16 + ELSE + dst[i+15:i] := src[i+15:i] + FI ENDFOR -k[MAX:32] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - Mask - AVX512BW - Compare - - - - - Compare packed 16-bit integers in "a" and "b" based on the comparison operand specified by "imm8", and store the results in mask vector "k1" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + AVX512_VBMI2 + Swizzle + + + + Load contiguous active 16-bit integers from "a" (those with their respective bit set in mask "k"), and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -CASE (imm8[7:0]) OF -0: OP := _MM_CMPINT_EQ -1: OP := _MM_CMPINT_LT -2: OP := _MM_CMPINT_LE -3: OP := _MM_CMPINT_FALSE -4: OP := _MM_CMPINT_NE -5: OP := _MM_CMPINT_NLT -6: OP := _MM_CMPINT_NLE -7: OP := _MM_CMPINT_TRUE -ESAC +m := 0 FOR j := 0 to 31 - i := j*16 - IF k1[j] - k[j] := ( a[i+15:i] OP b[i+15:i] ) ? 1 : 0 - ELSE - k[j] := 0 + i := j*16 + IF k[j] + dst[i+15:i] := a[m+15:m] + m := m + 16 + ELSE + dst[i+15:i] := 0 FI ENDFOR -k[MAX:32] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - Mask - AVX512BW - Compare - - - - Compare packed 16-bit integers in "a" and "b" for equality, and store the results in mask vector "k1" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + AVX512_VBMI2 + Swizzle + + + + + Load contiguous active 16-bit integers from "a" (those with their respective bit set in mask "k"), and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). +m := 0 FOR j := 0 to 31 i := j*16 - IF k1[j] - k[j] := ( a[i+15:i] == b[i+15:i] ) ? 1 : 0 - ELSE - k[j] := 0 + IF k[j] + dst[i+15:i] := a[m+15:m] + m := m + 16 + ELSE + dst[i+15:i] := src[i+15:i] FI ENDFOR -k[MAX:32] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - Mask - AVX512BW - Compare - - - - Compare packed 16-bit integers in "a" and "b" for greater-than-or-equal, and store the results in mask vector "k1" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + AVX512_VBMI2 + AVX512VL + Load + Swizzle + + + + Load contiguous active 16-bit integers from unaligned memory at "mem_addr" (those with their respective bit set in mask "k"), and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 31 +m := 0 +FOR j := 0 to 15 i := j*16 - IF k1[j] - k[j] := ( a[i+15:i] >= b[i+15:i] ) ? 1 : 0 - ELSE - k[j] := 0 + IF k[j] + dst[i+15:i] := MEM[mem_addr+m+15:mem_addr+m] + m := m + 16 + ELSE + dst[i+15:i] := 0 FI ENDFOR -k[MAX:32] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - Mask - AVX512BW - Compare - - - - Compare packed 16-bit integers in "a" and "b" for greater-than, and store the results in mask vector "k1" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + AVX512_VBMI2 + AVX512VL + Load + Swizzle + + + + + Load contiguous active 16-bit integers from unaligned memory at "mem_addr" (those with their respective bit set in mask "k"), and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 31 +m := 0 +FOR j := 0 to 15 i := j*16 - IF k1[j] - k[j] := ( a[i+15:i] > b[i+15:i] ) ? 1 : 0 - ELSE - k[j] := 0 + IF k[j] + dst[i+15:i] := MEM[mem_addr+m+15:mem_addr+m] + m := m + 16 + ELSE + dst[i+15:i] := src[i+15:i] FI ENDFOR -k[MAX:32] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - Mask - AVX512BW - Compare - - - - Compare packed 16-bit integers in "a" and "b" for less-than-or-equal, and store the results in mask vector "k1" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + AVX512_VBMI2 + AVX512VL + Swizzle + + + + Load contiguous active 16-bit integers from "a" (those with their respective bit set in mask "k"), and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 31 +m := 0 +FOR j := 0 to 15 i := j*16 - IF k1[j] - k[j] := ( a[i+15:i] <= b[i+15:i] ) ? 1 : 0 - ELSE - k[j] := 0 + IF k[j] + dst[i+15:i] := a[m+15:m] + m := m + 16 + ELSE + dst[i+15:i] := 0 FI ENDFOR -k[MAX:32] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - Mask - AVX512BW - Compare - - - - Compare packed 16-bit integers in "a" and "b" for less-than, and store the results in mask vector "k1" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + AVX512_VBMI2 + AVX512VL + Swizzle + + + + + Load contiguous active 16-bit integers from "a" (those with their respective bit set in mask "k"), and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 31 +m := 0 +FOR j := 0 to 15 i := j*16 - IF k1[j] - k[j] := ( a[i+15:i] < b[i+15:i] ) ? 1 : 0 - ELSE - k[j] := 0 + IF k[j] + dst[i+15:i] := a[m+15:m] + m := m + 16 + ELSE + dst[i+15:i] := src[i+15:i] FI ENDFOR -k[MAX:32] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - Mask - AVX512BW - Compare - - - - Compare packed 16-bit integers in "a" and "b" for not-equal, and store the results in mask vector "k1" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + AVX512_VBMI2 + AVX512VL + Load + Swizzle + + + + Load contiguous active 16-bit integers from unaligned memory at "mem_addr" (those with their respective bit set in mask "k"), and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 31 +m := 0 +FOR j := 0 to 7 i := j*16 - IF k1[j] - k[j] := ( a[i+15:i] != b[i+15:i] ) ? 1 : 0 - ELSE - k[j] := 0 + IF k[j] + dst[i+15:i] := MEM[mem_addr+m+15:mem_addr+m] + m := m + 16 + ELSE + dst[i+15:i] := 0 FI ENDFOR -k[MAX:32] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - Mask + AVX512_VBMI2 AVX512VL - AVX512BW - Compare - - - - Compare packed 16-bit integers in "a" and "b" based on the comparison operand specified by "imm8", and store the results in mask vector "k". + Load + Swizzle + + + + + Load contiguous active 16-bit integers from unaligned memory at "mem_addr" (those with their respective bit set in mask "k"), and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -CASE (imm8[7:0]) OF -0: OP := _MM_CMPINT_EQ -1: OP := _MM_CMPINT_LT -2: OP := _MM_CMPINT_LE -3: OP := _MM_CMPINT_FALSE -4: OP := _MM_CMPINT_NE -5: OP := _MM_CMPINT_NLT -6: OP := _MM_CMPINT_NLE -7: OP := _MM_CMPINT_TRUE -ESAC +m := 0 FOR j := 0 to 7 i := j*16 - k[j] := ( a[i+15:i] OP b[i+15:i] ) ? 1 : 0 + IF k[j] + dst[i+15:i] := MEM[mem_addr+m+15:mem_addr+m] + m := m + 16 + ELSE + dst[i+15:i] := src[i+15:i] + FI ENDFOR -k[MAX:8] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - Mask + AVX512_VBMI2 AVX512VL - AVX512BW - Compare - - - Compare packed 16-bit integers in "a" and "b" for equality, and store the results in mask vector "k". + Swizzle + + + + Load contiguous active 16-bit integers from "a" (those with their respective bit set in mask "k"), and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). +m := 0 FOR j := 0 to 7 i := j*16 - k[j] := ( a[i+15:i] == b[i+15:i] ) ? 1 : 0 + IF k[j] + dst[i+15:i] := a[m+15:m] + m := m + 16 + ELSE + dst[i+15:i] := 0 + FI ENDFOR -k[MAX:8] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - Mask + AVX512_VBMI2 AVX512VL - AVX512BW - Compare - - - Compare packed 16-bit integers in "a" and "b" for greater-than-or-equal, and store the results in mask vector "k". + Swizzle + + + + + Load contiguous active 16-bit integers from "a" (those with their respective bit set in mask "k"), and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). +m := 0 FOR j := 0 to 7 i := j*16 - k[j] := ( a[i+15:i] >= b[i+15:i] ) ? 1 : 0 + IF k[j] + dst[i+15:i] := a[m+15:m] + m := m + 16 + ELSE + dst[i+15:i] := src[i+15:i] + FI ENDFOR -k[MAX:8] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - Mask - AVX512VL - AVX512BW - Compare - - - Compare packed 16-bit integers in "a" and "b" for greater-than, and store the results in mask vector "k". + AVX512_VBMI2 + Load + Swizzle + + + + Load contiguous active 8-bit integers from unaligned memory at "mem_addr" (those with their respective bit set in mask "k"), and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := j*16 - k[j] := ( a[i+15:i] > b[i+15:i] ) ? 1 : 0 +m := 0 +FOR j := 0 to 63 + i := j*8 + IF k[j] + dst[i+7:i] := MEM[mem_addr+m+7:mem_addr+m] + m := m + 8 + ELSE + dst[i+7:i] := 0 + FI ENDFOR -k[MAX:8] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - Mask - AVX512VL - AVX512BW - Compare - - - Compare packed 16-bit integers in "a" and "b" for less-than-or-equal, and store the results in mask vector "k". + AVX512_VBMI2 + Load + Swizzle + + + + + Load contiguous active 8-bit integers from unaligned memory at "mem_addr" (those with their respective bit set in mask "k"), and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := j*16 - k[j] := ( a[i+15:i] <= b[i+15:i] ) ? 1 : 0 +m := 0 +FOR j := 0 to 63 + i := j*8 + IF k[j] + dst[i+7:i] := MEM[mem_addr+m+7:mem_addr+m] + m := m + 8 + ELSE + dst[i+7:i] := src[i+7:i] + FI ENDFOR -k[MAX:8] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - Mask + AVX512_VBMI2 AVX512VL - AVX512BW - Compare - - - Compare packed 16-bit integers in "a" and "b" for less-than, and store the results in mask vector "k". + Load + Swizzle + + + + Load contiguous active 8-bit integers from unaligned memory at "mem_addr" (those with their respective bit set in mask "k"), and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := j*16 - k[j] := ( a[i+15:i] < b[i+15:i] ) ? 1 : 0 +m := 0 +FOR j := 0 to 31 + i := j*8 + IF k[j] + dst[i+7:i] := MEM[mem_addr+m+7:mem_addr+m] + m := m + 8 + ELSE + dst[i+7:i] := 0 + FI ENDFOR -k[MAX:8] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - Mask + AVX512_VBMI2 AVX512VL - AVX512BW - Compare - - - Compare packed 16-bit integers in "a" and "b" for not-equal, and store the results in mask vector "k". + Load + Swizzle + + + + + Load contiguous active 8-bit integers from unaligned memory at "mem_addr" (those with their respective bit set in mask "k"), and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := j*16 - k[j] := ( a[i+15:i] != b[i+15:i] ) ? 1 : 0 +m := 0 +FOR j := 0 to 31 + i := j*8 + IF k[j] + dst[i+7:i] := MEM[mem_addr+m+7:mem_addr+m] + m := m + 8 + ELSE + dst[i+7:i] := src[i+7:i] + FI ENDFOR -k[MAX:8] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - Mask + AVX512_VBMI2 AVX512VL - AVX512BW - Compare - - - - - Compare packed 16-bit integers in "a" and "b" based on the comparison operand specified by "imm8", and store the results in mask vector "k1" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Load + Swizzle + + + + Load contiguous active 8-bit integers from unaligned memory at "mem_addr" (those with their respective bit set in mask "k"), and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -CASE (imm8[7:0]) OF -0: OP := _MM_CMPINT_EQ -1: OP := _MM_CMPINT_LT -2: OP := _MM_CMPINT_LE -3: OP := _MM_CMPINT_FALSE -4: OP := _MM_CMPINT_NE -5: OP := _MM_CMPINT_NLT -6: OP := _MM_CMPINT_NLE -7: OP := _MM_CMPINT_TRUE -ESAC -FOR j := 0 to 7 - i := j*16 - IF k1[j] - k[j] := ( a[i+15:i] OP b[i+15:i] ) ? 1 : 0 - ELSE - k[j] := 0 +m := 0 +FOR j := 0 to 15 + i := j*8 + IF k[j] + dst[i+7:i] := MEM[mem_addr+m+7:mem_addr+m] + m := m + 8 + ELSE + dst[i+7:i] := 0 FI ENDFOR -k[MAX:8] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - Mask + AVX512_VBMI2 AVX512VL - AVX512BW - Compare - - - - Compare packed 16-bit integers in "a" and "b" for equality, and store the results in mask vector "k1" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Load + Swizzle + + + + + Load contiguous active 8-bit integers from unaligned memory at "mem_addr" (those with their respective bit set in mask "k"), and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := j*16 - IF k1[j] - k[j] := ( a[i+15:i] == b[i+15:i] ) ? 1 : 0 - ELSE - k[j] := 0 +m := 0 +FOR j := 0 to 15 + i := j*8 + IF k[j] + dst[i+7:i] := MEM[mem_addr+m+7:mem_addr+m] + m := m + 8 + ELSE + dst[i+7:i] := src[i+7:i] FI ENDFOR -k[MAX:8] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - Mask - AVX512VL - AVX512BW - Compare - - - - Compare packed 16-bit integers in "a" and "b" for greater-than-or-equal, and store the results in mask vector "k1" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + AVX512_VBMI2 + Swizzle + + + + Load contiguous active 8-bit integers from "a" (those with their respective bit set in mask "k"), and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := j*16 - IF k1[j] - k[j] := ( a[i+15:i] >= b[i+15:i] ) ? 1 : 0 - ELSE - k[j] := 0 +m := 0 +FOR j := 0 to 63 + i := j*8 + IF k[j] + dst[i+7:i] := a[m+7:m] + m := m + 8 + ELSE + dst[i+7:i] := 0 FI ENDFOR -k[MAX:8] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - Mask + AVX512_VBMI2 + Swizzle + + + + + Load contiguous active 8-bit integers from "a" (those with their respective bit set in mask "k"), and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + +m := 0 +FOR j := 0 to 63 + i := j*8 + IF k[j] + dst[i+7:i] := a[m+7:m] + m := m + 8 + ELSE + dst[i+7:i] := src[i+7:i] + FI +ENDFOR +dst[MAX:512] := 0 + + +
immintrin.h
+ + + Integer + AVX512_VBMI2 AVX512VL - AVX512BW - Compare - - - - Compare packed 16-bit integers in "a" and "b" for greater-than, and store the results in mask vector "k1" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Swizzle + + + + Load contiguous active 8-bit integers from "a" (those with their respective bit set in mask "k"), and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := j*16 - IF k1[j] - k[j] := ( a[i+15:i] >== b[i+15:i] ) ? 1 : 0 - ELSE - k[j] := 0 +m := 0 +FOR j := 0 to 31 + i := j*8 + IF k[j] + dst[i+7:i] := a[m+7:m] + m := m + 8 + ELSE + dst[i+7:i] := 0 FI ENDFOR -k[MAX:8] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - Mask + AVX512_VBMI2 AVX512VL - AVX512BW - Compare - - - - Compare packed 16-bit integers in "a" and "b" for less-than-or-equal, and store the results in mask vector "k1" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Swizzle + + + + + Load contiguous active 8-bit integers from "a" (those with their respective bit set in mask "k"), and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := j*16 - IF k1[j] - k[j] := ( a[i+15:i] <= b[i+15:i] ) ? 1 : 0 - ELSE - k[j] := 0 +m := 0 +FOR j := 0 to 31 + i := j*8 + IF k[j] + dst[i+7:i] := a[m+7:m] + m := m + 8 + ELSE + dst[i+7:i] := src[i+7:i] FI ENDFOR -k[MAX:8] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - Mask + AVX512_VBMI2 AVX512VL - AVX512BW - Compare - - - - Compare packed 16-bit integers in "a" and "b" for less-than, and store the results in mask vector "k1" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Swizzle + + + + Load contiguous active 8-bit integers from "a" (those with their respective bit set in mask "k"), and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := j*16 - IF k1[j] - k[j] := ( a[i+15:i] < b[i+15:i] ) ? 1 : 0 - ELSE - k[j] := 0 +m := 0 +FOR j := 0 to 15 + i := j*8 + IF k[j] + dst[i+7:i] := a[m+7:m] + m := m + 8 + ELSE + dst[i+7:i] := 0 FI ENDFOR -k[MAX:8] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - Mask + AVX512_VBMI2 AVX512VL - AVX512BW - Compare - - - - Compare packed 16-bit integers in "a" and "b" for not-equal, and store the results in mask vector "k1" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Swizzle + + + + + Load contiguous active 8-bit integers from "a" (those with their respective bit set in mask "k"), and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 7 +m := 0 +FOR j := 0 to 15 + i := j*8 + IF k[j] + dst[i+7:i] := a[m+7:m] + m := m + 8 + ELSE + dst[i+7:i] := src[i+7:i] + FI +ENDFOR +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Integer + AVX512_VBMI2 + Store + Swizzle + + + + + Contiguously store the active 16-bit integers in "a" (those with their respective bit set in writemask "k") to unaligned memory at "base_addr". + +size := 16 +m := base_addr +FOR j := 0 to 31 i := j*16 - IF k1[j] - k[j] := ( a[i+15:i] != b[i+15:i] ) ? 1 : 0 - ELSE - k[j] := 0 + IF k[j] + MEM[m+size-1:m] := a[i+15:i] + m := m + size FI ENDFOR -k[MAX:8] := 0 - +
immintrin.h
 - + Integer + AVX512_VBMI2 AVX512VL - AVX512F - Miscellaneous - - - - Contiguously store the active 32-bit integers in "a" (those with their respective bit set in writemask "k") to "dst", and pass through the remaining elements from "src". + Store + Swizzle + + + + + Contiguously store the active 16-bit integers in "a" (those with their respective bit set in writemask "k") to unaligned memory at "base_addr". -size := 32 -m := 0 -FOR j := 0 to 7 - i := j*32 +size := 16 +m := base_addr +FOR j := 0 to 15 + i := j*16 IF k[j] - dst[m+size-1:m] := a[i+31:i] + MEM[m+size-1:m] := a[i+15:i] m := m + size FI ENDFOR -dst[255:m] := src[255:m] -dst[MAX:256] := 0 - +
immintrin.h
 - + Integer + AVX512_VBMI2 AVX512VL - AVX512F Store - - - - Contiguously store the active 32-bit integers in "a" (those with their respective bit set in writemask "k") to unaligned memory at "base_addr". + Swizzle + + + + + Contiguously store the active 16-bit integers in "a" (those with their respective bit set in writemask "k") to unaligned memory at "base_addr". -size := 32 +size := 16 m := base_addr FOR j := 0 to 7 - i := j*32 + i := j*16 IF k[j] - MEM[m+size-1:m] := a[i+31:i] + MEM[m+size-1:m] := a[i+15:i] m := m + size FI ENDFOR - +
immintrin.h
 - + Integer - AVX512VL - AVX512F - Miscellaneous - - - Contiguously store the active 32-bit integers in "a" (those with their respective bit set in zeromask "k") to "dst", and set the remaining elements to zero. + AVX512_VBMI2 + Swizzle + + + + Contiguously store the active 16-bit integers in "a" (those with their respective bit set in zeromask "k") to "dst", and set the remaining elements to zero. -size := 32 +size := 16 m := 0 -FOR j := 0 to 7 - i := j*32 +FOR j := 0 to 31 + i := j*16 IF k[j] - dst[m+size-1:m] := a[i+31:i] + dst[m+size-1:m] := a[i+15:i] m := m + size FI ENDFOR -dst[255:m] := 0 -dst[MAX:256] := 0 +dst[511:m] := 0 +dst[MAX:512] := 0 + + +
immintrin.h
+ + + Integer + AVX512_VBMI2 + Swizzle + + + + + Contiguously store the active 16-bit integers in "a" (those with their respective bit set in writemask "k") to "dst", and pass through the remaining elements from "src". + +size := 16 +m := 0 +FOR j := 0 to 31 + i := j*16 + IF k[j] + dst[m+size-1:m] := a[i+15:i] + m := m + size + FI +ENDFOR +dst[511:m] := src[511:m] +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer + AVX512_VBMI2 AVX512VL - AVX512F - Miscellaneous - - - - Contiguously store the active 32-bit integers in "a" (those with their respective bit set in writemask "k") to "dst", and pass through the remaining elements from "src". + Swizzle + + + + Contiguously store the active 16-bit integers in "a" (those with their respective bit set in zeromask "k") to "dst", and set the remaining elements to zero. -size := 32 +size := 16 m := 0 -FOR j := 0 to 3 - i := j*32 +FOR j := 0 to 15 + i := j*16 IF k[j] - dst[m+size-1:m] := a[i+31:i] + dst[m+size-1:m] := a[i+15:i] m := m + size FI ENDFOR -dst[127:m] := src[127:m] -dst[MAX:128] := 0 +dst[255:m] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer + AVX512_VBMI2 AVX512VL - AVX512F - Store - - - - Contiguously store the active 32-bit integers in "a" (those with their respective bit set in writemask "k") to unaligned memory at "base_addr". + Swizzle + + + + + Contiguously store the active 16-bit integers in "a" (those with their respective bit set in writemask "k") to "dst", and pass through the remaining elements from "src". -size := 32 -m := base_addr -FOR j := 0 to 3 - i := j*32 +size := 16 +m := 0 +FOR j := 0 to 15 + i := j*16 IF k[j] - MEM[m+size-1:m] := a[i+31:i] + dst[m+size-1:m] := a[i+15:i] m := m + size FI ENDFOR +dst[255:m] := src[255:m] +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer + AVX512_VBMI2 AVX512VL - AVX512F - Miscellaneous - - - Contiguously store the active 32-bit integers in "a" (those with their respective bit set in zeromask "k") to "dst", and set the remaining elements to zero. + Swizzle + + + + Contiguously store the active 16-bit integers in "a" (those with their respective bit set in zeromask "k") to "dst", and set the remaining elements to zero. -size := 32 +size := 16 m := 0 -FOR j := 0 to 3 - i := j*32 +FOR j := 0 to 7 + i := j*16 IF k[j] - dst[m+size-1:m] := a[i+31:i] + dst[m+size-1:m] := a[i+15:i] m := m + size FI ENDFOR dst[127:m] := 0 dst[MAX:128] := 0 - +
immintrin.h
 - + Integer + AVX512_VBMI2 AVX512VL - AVX512F - Miscellaneous - - - - Contiguously store the active 64-bit integers in "a" (those with their respective bit set in writemask "k") to "dst", and pass through the remaining elements from "src". + Swizzle + + + + + Contiguously store the active 16-bit integers in "a" (those with their respective bit set in writemask "k") to "dst", and pass through the remaining elements from "src". -size := 64 +size := 16 m := 0 -FOR j := 0 to 3 - i := j*64 +FOR j := 0 to 7 + i := j*16 IF k[j] - dst[m+size-1:m] := a[i+63:i] + dst[m+size-1:m] := a[i+15:i] m := m + size FI ENDFOR -dst[255:m] := src[255:m] -dst[MAX:256] := 0 +dst[127:m] := src[127:m] +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - AVX512VL - AVX512F + AVX512_VBMI2 Store - - - - Contiguously store the active 64-bit integers in "a" (those with their respective bit set in writemask "k") to unaligned memory at "base_addr". + Swizzle + + + + + Contiguously store the active 8-bit integers in "a" (those with their respective bit set in writemask "k") to unaligned memory at "base_addr". -size := 64 +size := 8 m := base_addr -FOR j := 0 to 3 - i := j*64 +FOR j := 0 to 63 + i := j*8 IF k[j] - MEM[m+size-1:m] := a[i+63:i] + MEM[m+size-1:m] := a[i+7:i] m := m + size FI ENDFOR - +
immintrin.h
 - + Integer + AVX512_VBMI2 AVX512VL - AVX512F - Miscellaneous - - - Contiguously store the active 64-bit integers in "a" (those with their respective bit set in zeromask "k") to "dst", and set the remaining elements to zero. + Store + Swizzle + + + + + Contiguously store the active 8-bit integers in "a" (those with their respective bit set in writemask "k") to unaligned memory at "base_addr". -size := 64 -m := 0 -FOR j := 0 to 3 - i := j*64 +size := 8 +m := base_addr +FOR j := 0 to 31 + i := j*8 IF k[j] - dst[m+size-1:m] := a[i+63:i] + MEM[m+size-1:m] := a[i+7:i] m := m + size FI ENDFOR -dst[255:m] := 0 -dst[MAX:256] := 0 - +
immintrin.h
 - + Integer + AVX512_VBMI2 AVX512VL - AVX512F - Miscellaneous - - - - Contiguously store the active 64-bit integers in "a" (those with their respective bit set in writemask "k") to "dst", and pass through the remaining elements from "src". + Store + Swizzle + + + + + Contiguously store the active 8-bit integers in "a" (those with their respective bit set in writemask "k") to unaligned memory at "base_addr". -size := 64 -m := 0 -FOR j := 0 to 1 - i := j*64 +size := 8 +m := base_addr +FOR j := 0 to 15 + i := j*8 IF k[j] - dst[m+size-1:m] := a[i+63:i] + MEM[m+size-1:m] := a[i+7:i] m := m + size FI ENDFOR -dst[127:m] := src[127:m] -dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - AVX512VL - AVX512F - Store - - - - Contiguously store the active 64-bit integers in "a" (those with their respective bit set in writemask "k") to unaligned memory at "base_addr". + AVX512_VBMI2 + Swizzle + + + + Contiguously store the active 8-bit integers in "a" (those with their respective bit set in zeromask "k") to "dst", and set the remaining elements to zero. -size := 64 -m := base_addr -FOR j := 0 to 1 - i := j*64 +size := 8 +m := 0 +FOR j := 0 to 63 + i := j*8 IF k[j] - MEM[m+size-1:m] := a[i+63:i] + dst[m+size-1:m] := a[i+7:i] m := m + size FI ENDFOR +dst[511:m] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX512VL - AVX512F - Miscellaneous - - - Contiguously store the active 64-bit integers in "a" (those with their respective bit set in zeromask "k") to "dst", and set the remaining elements to zero. + AVX512_VBMI2 + Swizzle + + + + + Contiguously store the active 8-bit integers in "a" (those with their respective bit set in writemask "k") to "dst", and pass through the remaining elements from "src". -size := 64 +size := 8 m := 0 -FOR j := 0 to 1 - i := j*64 +FOR j := 0 to 63 + i := j*8 IF k[j] - dst[m+size-1:m] := a[i+63:i] + dst[m+size-1:m] := a[i+7:i] m := m + size FI ENDFOR -dst[127:m] := 0 -dst[MAX:128] := 0 +dst[511:m] := src[511:m] +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer + AVX512_VBMI2 AVX512VL - AVX512CD - Compare - - Test each 32-bit element of "a" for equality with all other elements in "a" closer to the least significant bit. Each element's comparison forms a zero extended bit vector in "dst". + Swizzle + + + + Contiguously store the active 8-bit integers in "a" (those with their respective bit set in zeromask "k") to "dst", and set the remaining elements to zero. -FOR j := 0 to 7 - i := j*32 - FOR k := 0 to j-1 - m := k*32 - dst[i+k] := (a[i+31:i] == a[m+31:m]) ? 1 : 0 - ENDFOR - dst[i+31:i+j] := 0 +size := 8 +m := 0 +FOR j := 0 to 31 + i := j*8 + IF k[j] + dst[m+size-1:m] := a[i+7:i] + m := m + size + FI ENDFOR +dst[255:m] := 0 dst[MAX:256] := 0 - +
immintrin.h
 - + Integer + AVX512_VBMI2 AVX512VL - AVX512CD - Compare - - - - Test each 32-bit element of "a" for equality with all other elements in "a" closer to the least significant bit using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). Each element's comparison forms a zero extended bit vector in "dst". + Swizzle + + + + + Contiguously store the active 8-bit integers in "a" (those with their respective bit set in writemask "k") to "dst", and pass through the remaining elements from "src". -FOR j := 0 to 7 - i := j*32 - IF k[i] - FOR l := 0 to j-1 - m := l*32 - dst[i+l] := (a[i+31:i] == a[m+31:m]) ? 1 : 0 - ENDFOR - dst[i+31:i+j] := 0 - ELSE - dst[i+31:i] := src[i+31:i] +size := 8 +m := 0 +FOR j := 0 to 31 + i := j*8 + IF k[j] + dst[m+size-1:m] := a[i+7:i] + m := m + size FI ENDFOR +dst[255:m] := src[255:m] dst[MAX:256] := 0 - +
immintrin.h
 - + Integer + AVX512_VBMI2 AVX512VL - AVX512CD - Compare - - - Test each 32-bit element of "a" for equality with all other elements in "a" closer to the least significant bit using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). Each element's comparison forms a zero extended bit vector in "dst". + Swizzle + + + + Contiguously store the active 8-bit integers in "a" (those with their respective bit set in zeromask "k") to "dst", and set the remaining elements to zero. -FOR j := 0 to 7 - i := j*32 - IF k[i] - FOR l := 0 to j-1 - m := l*32 - dst[i+l] := (a[i+31:i] == a[m+31:m]) ? 1 : 0 - ENDFOR - dst[i+31:i+j] := 0 - ELSE - dst[i+31:i] := 0 +size := 8 +m := 0 +FOR j := 0 to 15 + i := j*8 + IF k[j] + dst[m+size-1:m] := a[i+7:i] + m := m + size FI ENDFOR -dst[MAX:256] := 0 +dst[127:m] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer + AVX512_VBMI2 AVX512VL - AVX512CD - Compare - - Test each 32-bit element of "a" for equality with all other elements in "a" closer to the least significant bit. Each element's comparison forms a zero extended bit vector in "dst". + Swizzle + + + + + Contiguously store the active 8-bit integers in "a" (those with their respective bit set in writemask "k") to "dst", and pass through the remaining elements from "src". -FOR j := 0 to 3 - i := j*32 - FOR k := 0 to j-1 - m := k*32 - dst[i+k] := (a[i+31:i] == a[m+31:m]) ? 1 : 0 - ENDFOR - dst[i+31:i+j] := 0 +size := 8 +m := 0 +FOR j := 0 to 15 + i := j*8 + IF k[j] + dst[m+size-1:m] := a[i+7:i] + m := m + size + FI ENDFOR +dst[127:m] := src[127:m] dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - AVX512VL - AVX512CD - Compare - - - - Test each 32-bit element of "a" for equality with all other elements in "a" closer to the least significant bit using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). Each element's comparison forms a zero extended bit vector in "dst". + AVX512_VNNI + Arithmetic + + + + + + Multiply groups of 2 adjacent pairs of signed 16-bit integers in "a" with corresponding 16-bit integers in "b", producing 2 intermediate signed 32-bit results. Sum these 2 results with the corresponding 32-bit integer in "src" using signed saturation, and store the packed 32-bit results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*32 - IF k[i] - FOR l := 0 to j-1 - m := l*32 - dst[i+l] := (a[i+31:i] == a[m+31:m]) ? 1 : 0 - ENDFOR - dst[i+31:i+j] := 0 +FOR j := 0 to 15 + IF k[j] + tmp1.dword := SignExtend32(a.word[2*j]) * SignExtend32(b.word[2*j]) + tmp2.dword := SignExtend32(a.word[2*j+1]) * SignExtend32(b.word[2*j+1]) + dst.dword[j] := Saturate32(src.dword[j] + tmp1 + tmp2) ELSE - dst[i+31:i] := src[i+31:i] + dst.dword[j] := 0 FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX512VL - AVX512CD - Compare - - - Test each 32-bit element of "a" for equality with all other elements in "a" closer to the least significant bit using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). Each element's comparison forms a zero extended bit vector in "dst". + AVX512_VNNI + Arithmetic + + + + + + Multiply groups of 2 adjacent pairs of signed 16-bit integers in "a" with corresponding 16-bit integers in "b", producing 2 intermediate signed 32-bit results. Sum these 2 results with the corresponding 32-bit integer in "src" using signed saturation, and store the packed 32-bit results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*32 - IF k[i] - FOR l := 0 to j-1 - m := l*32 - dst[i+l] := (a[i+31:i] == a[m+31:m]) ? 1 : 0 - ENDFOR - dst[i+31:i+j] := 0 +FOR j := 0 to 15 + IF k[j] + tmp1.dword := SignExtend32(a.word[2*j]) * SignExtend32(b.word[2*j]) + tmp2.dword := SignExtend32(a.word[2*j+1]) * SignExtend32(b.word[2*j+1]) + dst.dword[j] := Saturate32(src.dword[j] + tmp1 + tmp2) ELSE - dst[i+31:i] := 0 + dst.dword[j] := src.dword[j] FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX512VL - AVX512CD - Compare - - Test each 64-bit element of "a" for equality with all other elements in "a" closer to the least significant bit. Each element's comparison forms a zero extended bit vector in "dst". + AVX512_VNNI + Arithmetic + + + + + Multiply groups of 2 adjacent pairs of signed 16-bit integers in "a" with corresponding 16-bit integers in "b", producing 2 intermediate signed 32-bit results. Sum these 2 results with the corresponding 32-bit integer in "src" using signed saturation, and store the packed 32-bit results in "dst". -FOR j := 0 to 3 - i := j*64 - FOR k := 0 to j-1 - m := k*64 - dst[i+k] := (a[i+63:i] == a[m+63:m]) ? 1 : 0 - ENDFOR - dst[i+63:i+j] := 0 +FOR j := 0 to 15 + tmp1.dword := SignExtend32(a.word[2*j]) * SignExtend32(b.word[2*j]) + tmp2.dword := SignExtend32(a.word[2*j+1]) * SignExtend32(b.word[2*j+1]) + dst.dword[j] := Saturate32(src.dword[j] + tmp1 + tmp2) ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer + AVX512_VNNI AVX512VL - AVX512CD - Compare - - - - Test each 64-bit element of "a" for equality with all other elements in "a" closer to the least significant bit using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). Each element's comparison forms a zero extended bit vector in "dst". + Arithmetic + + + + + + Multiply groups of 2 adjacent pairs of signed 16-bit integers in "a" with corresponding 16-bit integers in "b", producing 2 intermediate signed 32-bit results. Sum these 2 results with the corresponding 32-bit integer in "src" using signed saturation, and store the packed 32-bit results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*64 +FOR j := 0 to 7 IF k[j] - FOR l := 0 to j-1 - m := l*64 - dst[i+l] := (a[i+63:i] == a[m+63:m]) ? 1 : 0 - ENDFOR - dst[i+63:i+j] := 0 + tmp1.dword := SignExtend32(a.word[2*j]) * SignExtend32(b.word[2*j]) + tmp2.dword := SignExtend32(a.word[2*j+1]) * SignExtend32(b.word[2*j+1]) + dst.dword[j] := Saturate32(src.dword[j] + tmp1 + tmp2) ELSE - dst[i+63:i] := src[i+63:i] + dst.dword[j] := 0 FI ENDFOR dst[MAX:256] := 0 - +
immintrin.h
 - + Integer + AVX512_VNNI AVX512VL - AVX512CD - Compare - - - Test each 64-bit element of "a" for equality with all other elements in "a" closer to the least significant bit using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). Each element's comparison forms a zero extended bit vector in "dst". + Arithmetic + + + + + + Multiply groups of 2 adjacent pairs of signed 16-bit integers in "a" with corresponding 16-bit integers in "b", producing 2 intermediate signed 32-bit results. Sum these 2 results with the corresponding 32-bit integer in "src" using signed saturation, and store the packed 32-bit results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*64 +FOR j := 0 to 7 IF k[j] - FOR l := 0 to j-1 - m := l*64 - dst[i+l] := (a[i+63:i] == a[m+63:m]) ? 1 : 0 - ENDFOR - dst[i+63:i+j] := 0 + tmp1.dword := SignExtend32(a.word[2*j]) * SignExtend32(b.word[2*j]) + tmp2.dword := SignExtend32(a.word[2*j+1]) * SignExtend32(b.word[2*j+1]) + dst.dword[j] := Saturate32(src.dword[j] + tmp1 + tmp2) ELSE - dst[i+63:i] := 0 + dst.dword[j] := src.dword[j] FI ENDFOR dst[MAX:256] := 0 - +
immintrin.h
 - + Integer + AVX512_VNNI AVX512VL - AVX512CD - Compare - - Test each 64-bit element of "a" for equality with all other elements in "a" closer to the least significant bit. Each element's comparison forms a zero extended bit vector in "dst". + Arithmetic + + + + + Multiply groups of 2 adjacent pairs of signed 16-bit integers in "a" with corresponding 16-bit integers in "b", producing 2 intermediate signed 32-bit results. Sum these 2 results with the corresponding 32-bit integer in "src" using signed saturation, and store the packed 32-bit results in "dst". -FOR j := 0 to 1 - i := j*64 - FOR k := 0 to j-1 - m := k*64 - dst[i+k] := (a[i+63:i] == a[m+63:m]) ? 1 : 0 - ENDFOR - dst[i+63:i+j] := 0 +FOR j := 0 to 7 + tmp1.dword := SignExtend32(a.word[2*j]) * SignExtend32(b.word[2*j]) + tmp2.dword := SignExtend32(a.word[2*j+1]) * SignExtend32(b.word[2*j+1]) + dst.dword[j] := Saturate32(src.dword[j] + tmp1 + tmp2) ENDFOR -dst[MAX:128] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer + AVX512_VNNI AVX512VL - AVX512CD - Compare - - - - Test each 64-bit element of "a" for equality with all other elements in "a" closer to the least significant bit using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). Each element's comparison forms a zero extended bit vector in "dst". + Arithmetic + + + + + + Multiply groups of 2 adjacent pairs of signed 16-bit integers in "a" with corresponding 16-bit integers in "b", producing 2 intermediate signed 32-bit results. Sum these 2 results with the corresponding 32-bit integer in "src" using signed saturation, and store the packed 32-bit results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 1 - i := j*64 +FOR j := 0 to 3 IF k[j] - FOR l := 0 to j-1 - m := l*64 - dst[i+l] := (a[i+63:i] == a[m+63:m]) ? 1 : 0 - ENDFOR - dst[i+63:i+j] := 0 + tmp1.dword := SignExtend32(a.word[2*j]) * SignExtend32(b.word[2*j]) + tmp2.dword := SignExtend32(a.word[2*j+1]) * SignExtend32(b.word[2*j+1]) + dst.dword[j] := Saturate32(src.dword[j] + tmp1 + tmp2) ELSE - dst[i+63:i] := src[i+63:i] + dst.dword[j] := 0 FI ENDFOR dst[MAX:128] := 0 - +
immintrin.h
 - + Integer + AVX512_VNNI AVX512VL - AVX512CD - Compare - - - Test each 64-bit element of "a" for equality with all other elements in "a" closer to the least significant bit using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). Each element's comparison forms a zero extended bit vector in "dst". + Arithmetic + + + + + + Multiply groups of 2 adjacent pairs of signed 16-bit integers in "a" with corresponding 16-bit integers in "b", producing 2 intermediate signed 32-bit results. Sum these 2 results with the corresponding 32-bit integer in "src" using signed saturation, and store the packed 32-bit results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 1 - i := j*64 +FOR j := 0 to 3 IF k[j] - FOR l := 0 to j-1 - m := l*64 - dst[i+l] := (a[i+63:i] == a[m+63:m]) ? 1 : 0 - ENDFOR - dst[i+63:i+j] := 0 + tmp1.dword := SignExtend32(a.word[2*j]) * SignExtend32(b.word[2*j]) + tmp2.dword := SignExtend32(a.word[2*j+1]) * SignExtend32(b.word[2*j+1]) + dst.dword[j] := Saturate32(src.dword[j] + tmp1 + tmp2) ELSE - dst[i+63:i] := 0 + dst.dword[j] := src.dword[j] FI ENDFOR dst[MAX:128] := 0 - +
immintrin.h
 - + Integer + AVX512_VNNI AVX512VL - AVX512F - Miscellaneous - - - - - Shuffle 32-bit integers in "a" across lanes using the corresponding index in "idx", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Arithmetic + + + + + Multiply groups of 2 adjacent pairs of signed 16-bit integers in "a" with corresponding 16-bit integers in "b", producing 2 intermediate signed 32-bit results. Sum these 2 results with the corresponding 32-bit integer in "src" using signed saturation, and store the packed 32-bit results in "dst". -FOR j := 0 to 7 - i := j*32 - id := idx[i+2:i]*32 - IF k[j] - dst[i+31:i] := a[id+31:id] - ELSE - dst[i+31:i] := src[i+31:i] - FI +FOR j := 0 to 3 + tmp1.dword := SignExtend32(a.word[2*j]) * SignExtend32(b.word[2*j]) + tmp2.dword := SignExtend32(a.word[2*j+1]) * SignExtend32(b.word[2*j+1]) + dst.dword[j] := Saturate32(src.dword[j] + tmp1 + tmp2) ENDFOR -dst[MAX:256] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - AVX512VL - AVX512F - Miscellaneous - - - - Shuffle 32-bit integers in "a" across lanes using the corresponding index in "idx", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + AVX512_VNNI + Arithmetic + + + + + + Multiply groups of 2 adjacent pairs of signed 16-bit integers in "a" with corresponding 16-bit integers in "b", producing 2 intermediate signed 32-bit results. Sum these 2 results with the corresponding 32-bit integer in "src", and store the packed 32-bit results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := j*32 - id := idx[i+2:i]*32 +FOR j := 0 to 15 IF k[j] - dst[i+31:i] := a[id+31:id] + tmp1.dword := SignExtend32(a.word[2*j]) * SignExtend32(b.word[2*j]) + tmp2.dword := SignExtend32(a.word[2*j+1]) * SignExtend32(b.word[2*j+1]) + dst.dword[j] := src.dword[j] + tmp1 + tmp2 ELSE - dst[i+31:i] := 0 + dst.dword[j] := 0 FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX512VL - AVX512F - Miscellaneous - - - Shuffle 32-bit integers in "a" across lanes using the corresponding index in "idx", and store the results in "dst". - + AVX512_VNNI + Arithmetic + + + + + + Multiply groups of 2 adjacent pairs of signed 16-bit integers in "a" with corresponding 16-bit integers in "b", producing 2 intermediate signed 32-bit results. Sum these 2 results with the corresponding 32-bit integer in "src", and store the packed 32-bit results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := j*32 - id := idx[i+2:i]*32 - dst[i+31:i] := a[id+31:id] +FOR j := 0 to 15 + IF k[j] + tmp1.dword := SignExtend32(a.word[2*j]) * SignExtend32(b.word[2*j]) + tmp2.dword := SignExtend32(a.word[2*j+1]) * SignExtend32(b.word[2*j+1]) + dst.dword[j] := src.dword[j] + tmp1 + tmp2 + ELSE + dst.dword[j] := src.dword[j] + FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX512VL - AVX512F - Miscellaneous - - - - - Shuffle 32-bit integers in "a" and "b" across lanes using the corresponding selector and index in "idx", and store the results in "dst" using writemask "k" (elements are copied from "idx" when the corresponding mask bit is not set). + AVX512_VNNI + Arithmetic + + + + + Multiply groups of 2 adjacent pairs of signed 16-bit integers in "a" with corresponding 16-bit integers in "b", producing 2 intermediate signed 32-bit results. Sum these 2 results with the corresponding 32-bit integer in "src", and store the packed 32-bit results in "dst". -FOR j := 0 to 7 - i := j*32 - off := idx[i+2:i]*32 - IF k[j] - dst[i+31:i] := idx[i+3] ? b[off+31:off] : a[off+31:off] - ELSE - dst[i+31:i] := idx[i+31:i] - FI +FOR j := 0 to 15 + tmp1.dword := SignExtend32(a.word[2*j]) * SignExtend32(b.word[2*j]) + tmp2.dword := SignExtend32(a.word[2*j+1]) * SignExtend32(b.word[2*j+1]) + dst.dword[j] := src.dword[j] + tmp1 + tmp2 ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer + AVX512_VNNI AVX512VL - AVX512F - Miscellaneous - - - - - Shuffle 32-bit integers in "a" and "b" across lanes using the corresponding selector and index in "idx", and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). + Arithmetic + + + + + + Multiply groups of 2 adjacent pairs of signed 16-bit integers in "a" with corresponding 16-bit integers in "b", producing 2 intermediate signed 32-bit results. Sum these 2 results with the corresponding 32-bit integer in "src", and store the packed 32-bit results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). FOR j := 0 to 7 - i := j*32 - off := idx[i+3:i]*32 IF k[j] - dst[i+31:i] := idx[i+4] ? b[off+31:off] : a[off+31:off] + tmp1.dword := SignExtend32(a.word[2*j]) * SignExtend32(b.word[2*j]) + tmp2.dword := SignExtend32(a.word[2*j+1]) * SignExtend32(b.word[2*j+1]) + dst.dword[j] := src.dword[j] + tmp1 + tmp2 ELSE - dst[i+31:i] := a[i+31:i] + dst.dword[j] := 0 FI ENDFOR dst[MAX:256] := 0 - +
immintrin.h
 - + Integer + AVX512_VNNI AVX512VL - AVX512F - Miscellaneous - - - - - Shuffle 32-bit integers in "a" and "b" across lanes using the corresponding selector and index in "idx", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Arithmetic + + + + + + Multiply groups of 2 adjacent pairs of signed 16-bit integers in "a" with corresponding 16-bit integers in "b", producing 2 intermediate signed 32-bit results. Sum these 2 results with the corresponding 32-bit integer in "src", and store the packed 32-bit results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). FOR j := 0 to 7 - i := j*32 - off := idx[i+2:i]*32 IF k[j] - dst[i+31:i] := (idx[i+3]) ? b[off+31:off] : a[off+31:off] + tmp1.dword := SignExtend32(a.word[2*j]) * SignExtend32(b.word[2*j]) + tmp2.dword := SignExtend32(a.word[2*j+1]) * SignExtend32(b.word[2*j+1]) + dst.dword[j] := src.dword[j] + tmp1 + tmp2 ELSE - dst[i+31:i] := 0 + dst.dword[j] := src.dword[j] FI ENDFOR dst[MAX:256] := 0 - - +
immintrin.h
 - + Integer + AVX512_VNNI AVX512VL - AVX512F - Miscellaneous - - - - Shuffle 32-bit integers in "a" and "b" across lanes using the corresponding selector and index in "idx", and store the results in "dst". - + Arithmetic + + + + + Multiply groups of 2 adjacent pairs of signed 16-bit integers in "a" with corresponding 16-bit integers in "b", producing 2 intermediate signed 32-bit results. Sum these 2 results with the corresponding 32-bit integer in "src", and store the packed 32-bit results in "dst". FOR j := 0 to 7 - i := j*32 - off := idx[i+2:i]*32 - dst[i+31:i] := idx[i+3] ? b[off+31:off] : a[off+31:off] + tmp1.dword := SignExtend32(a.word[2*j]) * SignExtend32(b.word[2*j]) + tmp2.dword := SignExtend32(a.word[2*j+1]) * SignExtend32(b.word[2*j+1]) + dst.dword[j] := src.dword[j] + tmp1 + tmp2 ENDFOR dst[MAX:256] := 0 - - +
immintrin.h
 - + Integer + AVX512_VNNI AVX512VL - AVX512F - Miscellaneous - - - - - Shuffle 32-bit integers in "a" and "b" using the corresponding selector and index in "idx", and store the results in "dst" using writemask "k" (elements are copied from "idx" when the corresponding mask bit is not set). + Arithmetic + + + + + + Multiply groups of 2 adjacent pairs of signed 16-bit integers in "a" with corresponding 16-bit integers in "b", producing 2 intermediate signed 32-bit results. Sum these 2 results with the corresponding 32-bit integer in "src", and store the packed 32-bit results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). FOR j := 0 to 3 - i := j*32 - off := idx[i+1:i]*32 IF k[j] - dst[i+31:i] := idx[i+2] ? b[off+31:off] : a[off+31:off] + tmp1.dword := SignExtend32(a.word[2*j]) * SignExtend32(b.word[2*j]) + tmp2.dword := SignExtend32(a.word[2*j+1]) * SignExtend32(b.word[2*j+1]) + dst.dword[j] := src.dword[j] + tmp1 + tmp2 ELSE - dst[i+31:i] := idx[i+31:i] + dst.dword[j] := 0 FI ENDFOR dst[MAX:128] := 0 - +
immintrin.h
 - + Integer + AVX512_VNNI AVX512VL - AVX512F - Miscellaneous - - - - - Shuffle 32-bit integers in "a" and "b" using the corresponding selector and index in "idx", and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). + Arithmetic + + + + + + Multiply groups of 2 adjacent pairs of signed 16-bit integers in "a" with corresponding 16-bit integers in "b", producing 2 intermediate signed 32-bit results. Sum these 2 results with the corresponding 32-bit integer in "src", and store the packed 32-bit results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). FOR j := 0 to 3 - i := j*32 - off := idx[i+1:i]*32 IF k[j] - dst[i+31:i] := idx[i+2] ? b[off+31:off] : a[off+31:off] + tmp1.dword := SignExtend32(a.word[2*j]) * SignExtend32(b.word[2*j]) + tmp2.dword := SignExtend32(a.word[2*j+1]) * SignExtend32(b.word[2*j+1]) + dst.dword[j] := src.dword[j] + tmp1 + tmp2 ELSE - dst[i+31:i] := a[i+31:i] + dst.dword[j] := src.dword[j] FI ENDFOR dst[MAX:128] := 0 - +
immintrin.h
 - + Integer + AVX512_VNNI AVX512VL - AVX512F - Miscellaneous - - - - - Shuffle 32-bit integers in "a" and "b" using the corresponding selector and index in "idx", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Arithmetic + + + + + Multiply groups of 2 adjacent pairs of signed 16-bit integers in "a" with corresponding 16-bit integers in "b", producing 2 intermediate signed 32-bit results. Sum these 2 results with the corresponding 32-bit integer in "src", and store the packed 32-bit results in "dst". FOR j := 0 to 3 - i := j*32 - off := idx[i+1:i]*32 - IF k[j] - dst[i+31:i] := (idx[i+2]) ? b[off+31:off] : a[off+31:off] - ELSE - dst[i+31:i] := 0 - FI + tmp1.dword := SignExtend32(a.word[2*j]) * SignExtend32(b.word[2*j]) + tmp2.dword := SignExtend32(a.word[2*j+1]) * SignExtend32(b.word[2*j+1]) + dst.dword[j] := src.dword[j] + tmp1 + tmp2 ENDFOR dst[MAX:128] := 0 - - +
immintrin.h
 - + Integer - AVX512VL - AVX512F - Miscellaneous - - - - Shuffle 32-bit integers in "a" and "b" using the corresponding selector and index in "idx", and store the results in "dst". - + AVX512_VNNI + Arithmetic + + + + + + Multiply groups of 4 adjacent pairs of unsigned 8-bit integers in "a" with corresponding signed 8-bit integers in "b", producing 4 intermediate signed 16-bit results. Sum these 4 results with the corresponding 32-bit integer in "src" using signed saturation, and store the packed 32-bit results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*32 - off := idx[i+2:i]*32 - dst[i+31:i] := idx[i+3] ? b[off+31:off] : a[off+31:off] +FOR j := 0 to 15 + IF k[j] + tmp1.word := Signed(ZeroExtend16(a.byte[4*j]) * SignExtend16(b.byte[4*j])) + tmp2.word := Signed(ZeroExtend16(a.byte[4*j+1]) * SignExtend16(b.byte[4*j+1])) + tmp3.word := Signed(ZeroExtend16(a.byte[4*j+2]) * SignExtend16(b.byte[4*j+2])) + tmp4.word := Signed(ZeroExtend16(a.byte[4*j+3]) * SignExtend16(b.byte[4*j+3])) + dst.dword[j] := Saturate32(src.dword[j] + tmp1 + tmp2 + tmp3 + tmp4) + ELSE + dst.dword[j] := 0 + FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - - +
immintrin.h
 - - Floating Point - AVX512VL - AVX512F - Miscellaneous - - - - - Shuffle double-precision (64-bit) floating-point elements in "a" and "b" across lanes using the corresponding selector and index in "idx", and store the results in "dst" using writemask "k" (elements are copied from "idx" when the corresponding mask bit is not set). + + Integer + AVX512_VNNI + Arithmetic + + + + + + Multiply groups of 4 adjacent pairs of unsigned 8-bit integers in "a" with corresponding signed 8-bit integers in "b", producing 4 intermediate signed 16-bit results. Sum these 4 results with the corresponding 32-bit integer in "src" using signed saturation, and store the packed 32-bit results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*64 - off := idx[i+1:i]*64 +FOR j := 0 to 15 IF k[j] - dst[i+63:i] := idx[i+2] ? b[off+63:off] : a[off+63:off] + tmp1.word := Signed(ZeroExtend16(a.byte[4*j]) * SignExtend16(b.byte[4*j])) + tmp2.word := Signed(ZeroExtend16(a.byte[4*j+1]) * SignExtend16(b.byte[4*j+1])) + tmp3.word := Signed(ZeroExtend16(a.byte[4*j+2]) * SignExtend16(b.byte[4*j+2])) + tmp4.word := Signed(ZeroExtend16(a.byte[4*j+3]) * SignExtend16(b.byte[4*j+3])) + dst.dword[j] := Saturate32(src.dword[j] + tmp1 + tmp2 + tmp3 + tmp4) ELSE - dst[i+63:i] := idx[i+63:i] + dst.dword[j] := src.dword[j] FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512_VNNI + Arithmetic + + + + + Multiply groups of 4 adjacent pairs of unsigned 8-bit integers in "a" with corresponding signed 8-bit integers in "b", producing 4 intermediate signed 16-bit results. Sum these 4 results with the corresponding 32-bit integer in "src" using signed saturation, and store the packed 32-bit results in "dst". + +FOR j := 0 to 15 + tmp1.word := Signed(ZeroExtend16(a.byte[4*j]) * SignExtend16(b.byte[4*j])) + tmp2.word := Signed(ZeroExtend16(a.byte[4*j+1]) * SignExtend16(b.byte[4*j+1])) + tmp3.word := Signed(ZeroExtend16(a.byte[4*j+2]) * SignExtend16(b.byte[4*j+2])) + tmp4.word := Signed(ZeroExtend16(a.byte[4*j+3]) * SignExtend16(b.byte[4*j+3])) + dst.dword[j] := Saturate32(src.dword[j] + tmp1 + tmp2 + tmp3 + tmp4) +ENDFOR +dst[MAX:512] := 0 + + +
immintrin.h
+ + + Integer + AVX512_VNNI AVX512VL - AVX512F - Miscellaneous - - - - - Shuffle double-precision (64-bit) floating-point elements in "a" and "b" across lanes using the corresponding selector and index in "idx", and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). + Arithmetic + + + + + + Multiply groups of 4 adjacent pairs of unsigned 8-bit integers in "a" with corresponding signed 8-bit integers in "b", producing 4 intermediate signed 16-bit results. Sum these 4 results with the corresponding 32-bit integer in "src" using signed saturation, and store the packed 32-bit results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*64 - off := idx[i+1:i]*64 +FOR j := 0 to 7 IF k[j] - dst[i+63:i] := idx[i+2] ? b[off+63:off] : a[off+63:off] + tmp1.word := Signed(ZeroExtend16(a.byte[4*j]) * SignExtend16(b.byte[4*j])) + tmp2.word := Signed(ZeroExtend16(a.byte[4*j+1]) * SignExtend16(b.byte[4*j+1])) + tmp3.word := Signed(ZeroExtend16(a.byte[4*j+2]) * SignExtend16(b.byte[4*j+2])) + tmp4.word := Signed(ZeroExtend16(a.byte[4*j+3]) * SignExtend16(b.byte[4*j+3])) + dst.dword[j] := Saturate32(src.dword[j] + tmp1 + tmp2 + tmp3 + tmp4) ELSE - dst[i+63:i] := a[i+63:i] + dst.dword[j] := 0 FI ENDFOR dst[MAX:256] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512_VNNI AVX512VL - AVX512F - Miscellaneous - - - - - Shuffle double-precision (64-bit) floating-point elements in "a" and "b" across lanes using the corresponding selector and index in "idx", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Arithmetic + + + + + + Multiply groups of 4 adjacent pairs of unsigned 8-bit integers in "a" with corresponding signed 8-bit integers in "b", producing 4 intermediate signed 16-bit results. Sum these 4 results with the corresponding 32-bit integer in "src" using signed saturation, and store the packed 32-bit results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*64 - off := idx[i+1:i]*64 +FOR j := 0 to 7 IF k[j] - dst[i+63:i] := (idx[i+2]) ? b[off+63:off] : a[off+63:off] + tmp1.word := Signed(ZeroExtend16(a.byte[4*j]) * SignExtend16(b.byte[4*j])) + tmp2.word := Signed(ZeroExtend16(a.byte[4*j+1]) * SignExtend16(b.byte[4*j+1])) + tmp3.word := Signed(ZeroExtend16(a.byte[4*j+2]) * SignExtend16(b.byte[4*j+2])) + tmp4.word := Signed(ZeroExtend16(a.byte[4*j+3]) * SignExtend16(b.byte[4*j+3])) + dst.dword[j] := Saturate32(src.dword[j] + tmp1 + tmp2 + tmp3 + tmp4) ELSE - dst[i+63:i] := 0 + dst.dword[j] := src.dword[j] FI ENDFOR dst[MAX:256] := 0 - - +
immintrin.h
 - - Floating Point + + Integer + AVX512_VNNI AVX512VL - AVX512F - Miscellaneous - - - - Shuffle double-precision (64-bit) floating-point elements in "a" and "b" across lanes using the corresponding selector and index in "idx", and store the results in "dst". + Arithmetic + + + + + Multiply groups of 4 adjacent pairs of unsigned 8-bit integers in "a" with corresponding signed 8-bit integers in "b", producing 4 intermediate signed 16-bit results. Sum these 4 results with the corresponding 32-bit integer in "src" using signed saturation, and store the packed 32-bit results in "dst". -FOR j := 0 to 3 - i := j*64 - off := idx[i+1:i]*64 - dst[i+63:i] := idx[i+2] ? b[off+63:off] : a[off+63:off] +FOR j := 0 to 7 + tmp1.word := Signed(ZeroExtend16(a.byte[4*j]) * SignExtend16(b.byte[4*j])) + tmp2.word := Signed(ZeroExtend16(a.byte[4*j+1]) * SignExtend16(b.byte[4*j+1])) + tmp3.word := Signed(ZeroExtend16(a.byte[4*j+2]) * SignExtend16(b.byte[4*j+2])) + tmp4.word := Signed(ZeroExtend16(a.byte[4*j+3]) * SignExtend16(b.byte[4*j+3])) + dst.dword[j] := Saturate32(src.dword[j] + tmp1 + tmp2 + tmp3 + tmp4) ENDFOR dst[MAX:256] := 0 - - +
immintrin.h
 - - Floating Point + + Integer + AVX512_VNNI AVX512VL - AVX512F - Miscellaneous - - - - - Shuffle double-precision (64-bit) floating-point elements in "a" and "b" using the corresponding selector and index in "idx", and store the results in "dst" using writemask "k" (elements are copied from "idx" when the corresponding mask bit is not set) + Arithmetic + + + + + + Multiply groups of 4 adjacent pairs of unsigned 8-bit integers in "a" with corresponding signed 8-bit integers in "b", producing 4 intermediate signed 16-bit results. Sum these 4 results with the corresponding 32-bit integer in "src" using signed saturation, and store the packed 32-bit results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 1 - i := j*64 - off := idx[i]*64 +FOR j := 0 to 3 IF k[j] - dst[i+63:i] := idx[i+1] ? b[off+63:off] : a[off+63:off] + tmp1.word := Signed(ZeroExtend16(a.byte[4*j]) * SignExtend16(b.byte[4*j])) + tmp2.word := Signed(ZeroExtend16(a.byte[4*j+1]) * SignExtend16(b.byte[4*j+1])) + tmp3.word := Signed(ZeroExtend16(a.byte[4*j+2]) * SignExtend16(b.byte[4*j+2])) + tmp4.word := Signed(ZeroExtend16(a.byte[4*j+3]) * SignExtend16(b.byte[4*j+3])) + dst.dword[j] := Saturate32(src.dword[j] + tmp1 + tmp2 + tmp3 + tmp4) ELSE - dst[i+63:i] := idx[i+63:i] + dst.dword[j] := 0 FI ENDFOR dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512_VNNI AVX512VL - AVX512F - Miscellaneous - - - - - Shuffle double-precision (64-bit) floating-point elements in "a" and "b" using the corresponding selector and index in "idx", and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). + Arithmetic + + + + + + Multiply groups of 4 adjacent pairs of unsigned 8-bit integers in "a" with corresponding signed 8-bit integers in "b", producing 4 intermediate signed 16-bit results. Sum these 4 results with the corresponding 32-bit integer in "src" using signed saturation, and store the packed 32-bit results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 1 - i := j*64 - off := idx[i]*64 +FOR j := 0 to 3 IF k[j] - dst[i+63:i] := idx[i+1] ? b[off+63:off] : a[off+63:off] + tmp1.word := Signed(ZeroExtend16(a.byte[4*j]) * SignExtend16(b.byte[4*j])) + tmp2.word := Signed(ZeroExtend16(a.byte[4*j+1]) * SignExtend16(b.byte[4*j+1])) + tmp3.word := Signed(ZeroExtend16(a.byte[4*j+2]) * SignExtend16(b.byte[4*j+2])) + tmp4.word := Signed(ZeroExtend16(a.byte[4*j+3]) * SignExtend16(b.byte[4*j+3])) + dst.dword[j] := Saturate32(src.dword[j] + tmp1 + tmp2 + tmp3 + tmp4) ELSE - dst[i+63:i] := a[i+63:i] + dst.dword[j] := src.dword[j] FI ENDFOR dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512_VNNI AVX512VL - AVX512F - Miscellaneous - - - - - Shuffle double-precision (64-bit) floating-point elements in "a" and "b" using the corresponding selector and index in "idx", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Arithmetic + + + + + Multiply groups of 4 adjacent pairs of unsigned 8-bit integers in "a" with corresponding signed 8-bit integers in "b", producing 4 intermediate signed 16-bit results. Sum these 4 results with the corresponding 32-bit integer in "src" using signed saturation, and store the packed 32-bit results in "dst". -FOR j := 0 to 1 - i := j*64 - off := idx[i]*64 - IF k[j] - dst[i+63:i] := (idx[i+1]) ? b[off+63:off] : a[off+63:off] - ELSE - dst[i+63:i] := 0 - FI +FOR j := 0 to 3 + tmp1.word := Signed(ZeroExtend16(a.byte[4*j]) * SignExtend16(b.byte[4*j])) + tmp2.word := Signed(ZeroExtend16(a.byte[4*j+1]) * SignExtend16(b.byte[4*j+1])) + tmp3.word := Signed(ZeroExtend16(a.byte[4*j+2]) * SignExtend16(b.byte[4*j+2])) + tmp4.word := Signed(ZeroExtend16(a.byte[4*j+3]) * SignExtend16(b.byte[4*j+3])) + dst.dword[j] := Saturate32(src.dword[j] + tmp1 + tmp2 + tmp3 + tmp4) ENDFOR dst[MAX:128] := 0 - - +
immintrin.h
 - - Floating Point - AVX512VL - AVX512F - Miscellaneous - - - - Shuffle double-precision (64-bit) floating-point elements in "a" and "b" using the corresponding selector and index in "idx", and store the results in "dst". + + Integer + AVX512_VNNI + Arithmetic + + + + + + Multiply groups of 4 adjacent pairs of unsigned 8-bit integers in "a" with corresponding signed 8-bit integers in "b", producing 4 intermediate signed 16-bit results. Sum these 4 results with the corresponding 32-bit integer in "src", and store the packed 32-bit results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 1 - i := j*64 - off := idx[i]*64 - dst[i+63:i] := idx[i+1] ? b[off+63:off] : a[off+63:off] +FOR j := 0 to 15 + IF k[j] + tmp1.word := Signed(ZeroExtend16(a.byte[4*j]) * SignExtend16(b.byte[4*j])) + tmp2.word := Signed(ZeroExtend16(a.byte[4*j+1]) * SignExtend16(b.byte[4*j+1])) + tmp3.word := Signed(ZeroExtend16(a.byte[4*j+2]) * SignExtend16(b.byte[4*j+2])) + tmp4.word := Signed(ZeroExtend16(a.byte[4*j+3]) * SignExtend16(b.byte[4*j+3])) + dst.dword[j] := src.dword[j] + tmp1 + tmp2 + tmp3 + tmp4 + ELSE + dst.dword[j] := 0 + FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - - +
immintrin.h
 - - Floating Point - AVX512VL - AVX512F - Miscellaneous - - - - - Shuffle single-precision (32-bit) floating-point elements in "a" and "b" across lanes using the corresponding selector and index in "idx", and store the results in "dst" using writemask "k" (elements are copied from "idx" when the corresponding mask bit is not set). + + Integer + AVX512_VNNI + Arithmetic + + + + + + Multiply groups of 4 adjacent pairs of unsigned 8-bit integers in "a" with corresponding signed 8-bit integers in "b", producing 4 intermediate signed 16-bit results. Sum these 4 results with the corresponding 32-bit integer in "src", and store the packed 32-bit results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := j*32 - off := idx[i+2:i]*32 +FOR j := 0 to 15 IF k[j] - dst[i+31:i] := idx[i+3] ? b[off+31:off] : a[off+31:off] + tmp1.word := Signed(ZeroExtend16(a.byte[4*j]) * SignExtend16(b.byte[4*j])) + tmp2.word := Signed(ZeroExtend16(a.byte[4*j+1]) * SignExtend16(b.byte[4*j+1])) + tmp3.word := Signed(ZeroExtend16(a.byte[4*j+2]) * SignExtend16(b.byte[4*j+2])) + tmp4.word := Signed(ZeroExtend16(a.byte[4*j+3]) * SignExtend16(b.byte[4*j+3])) + dst.dword[j] := src.dword[j] + tmp1 + tmp2 + tmp3 + tmp4 ELSE - dst[i+31:i] := idx[i+31:i] + dst.dword[j] := src.dword[j] FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512_VNNI + Arithmetic + + + + + Multiply groups of 4 adjacent pairs of unsigned 8-bit integers in "a" with corresponding signed 8-bit integers in "b", producing 4 intermediate signed 16-bit results. Sum these 4 results with the corresponding 32-bit integer in "src", and store the packed 32-bit results in "dst". + +FOR j := 0 to 15 + tmp1.word := Signed(ZeroExtend16(a.byte[4*j]) * SignExtend16(b.byte[4*j])) + tmp2.word := Signed(ZeroExtend16(a.byte[4*j+1]) * SignExtend16(b.byte[4*j+1])) + tmp3.word := Signed(ZeroExtend16(a.byte[4*j+2]) * SignExtend16(b.byte[4*j+2])) + tmp4.word := Signed(ZeroExtend16(a.byte[4*j+3]) * SignExtend16(b.byte[4*j+3])) + dst.dword[j] := src.dword[j] + tmp1 + tmp2 + tmp3 + tmp4 +ENDFOR +dst[MAX:512] := 0 + + +
immintrin.h
+ + + Integer + AVX512_VNNI AVX512VL - AVX512F - Miscellaneous - - - - - Shuffle single-precision (32-bit) floating-point elements in "a" and "b" across lanes using the corresponding selector and index in "idx", and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). + Arithmetic + + + + + + Multiply groups of 4 adjacent pairs of unsigned 8-bit integers in "a" with corresponding signed 8-bit integers in "b", producing 4 intermediate signed 16-bit results. Sum these 4 results with the corresponding 32-bit integer in "src", and store the packed 32-bit results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). FOR j := 0 to 7 - i := j*32 - off := idx[i+2:i]*32 IF k[j] - dst[i+31:i] := idx[i+3] ? b[off+31:off] : a[off+31:off] + tmp1.word := Signed(ZeroExtend16(a.byte[4*j]) * SignExtend16(b.byte[4*j])) + tmp2.word := Signed(ZeroExtend16(a.byte[4*j+1]) * SignExtend16(b.byte[4*j+1])) + tmp3.word := Signed(ZeroExtend16(a.byte[4*j+2]) * SignExtend16(b.byte[4*j+2])) + tmp4.word := Signed(ZeroExtend16(a.byte[4*j+3]) * SignExtend16(b.byte[4*j+3])) + dst.dword[j] := src.dword[j] + tmp1 + tmp2 + tmp3 + tmp4 ELSE - dst[i+31:i] := a[i+31:i] + dst.dword[j] := 0 FI ENDFOR dst[MAX:256] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512_VNNI AVX512VL - AVX512F - Miscellaneous - - - - - Shuffle single-precision (32-bit) floating-point elements in "a" and "b" across lanes using the corresponding selector and index in "idx", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Arithmetic + + + + + + Multiply groups of 4 adjacent pairs of unsigned 8-bit integers in "a" with corresponding signed 8-bit integers in "b", producing 4 intermediate signed 16-bit results. Sum these 4 results with the corresponding 32-bit integer in "src", and store the packed 32-bit results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). FOR j := 0 to 7 - i := j*32 - off := idx[i+2:i]*32 IF k[j] - dst[i+31:i] := (idx[i+3]) ? b[off+31:off] : a[off+31:off] + tmp1.word := Signed(ZeroExtend16(a.byte[4*j]) * SignExtend16(b.byte[4*j])) + tmp2.word := Signed(ZeroExtend16(a.byte[4*j+1]) * SignExtend16(b.byte[4*j+1])) + tmp3.word := Signed(ZeroExtend16(a.byte[4*j+2]) * SignExtend16(b.byte[4*j+2])) + tmp4.word := Signed(ZeroExtend16(a.byte[4*j+3]) * SignExtend16(b.byte[4*j+3])) + dst.dword[j] := src.dword[j] + tmp1 + tmp2 + tmp3 + tmp4 ELSE - dst[i+31:i] := 0 + dst.dword[j] := src.dword[j] FI ENDFOR dst[MAX:256] := 0 - - +
immintrin.h
 - - Floating Point + + Integer + AVX512_VNNI AVX512VL - AVX512F - Miscellaneous - - - - Shuffle single-precision (32-bit) floating-point elements in "a" and "b" across lanes using the corresponding selector and index in "idx", and store the results in "dst". + Arithmetic + + + + + Multiply groups of 4 adjacent pairs of unsigned 8-bit integers in "a" with corresponding signed 8-bit integers in "b", producing 4 intermediate signed 16-bit results. Sum these 4 results with the corresponding 32-bit integer in "src", and store the packed 32-bit results in "dst". FOR j := 0 to 7 - i := j*32 - off := idx[i+2:i]*32 - dst[i+31:i] := idx[i+3] ? b[off+31:off] : a[off+31:off] + tmp1.word := Signed(ZeroExtend16(a.byte[4*j]) * SignExtend16(b.byte[4*j])) + tmp2.word := Signed(ZeroExtend16(a.byte[4*j+1]) * SignExtend16(b.byte[4*j+1])) + tmp3.word := Signed(ZeroExtend16(a.byte[4*j+2]) * SignExtend16(b.byte[4*j+2])) + tmp4.word := Signed(ZeroExtend16(a.byte[4*j+3]) * SignExtend16(b.byte[4*j+3])) + dst.dword[j] := src.dword[j] + tmp1 + tmp2 + tmp3 + tmp4 ENDFOR dst[MAX:256] := 0 - - +
immintrin.h
 - - Floating Point + + Integer + AVX512_VNNI AVX512VL - AVX512F - Miscellaneous - - - - - Shuffle single-precision (32-bit) floating-point elements in "a" and "b" using the corresponding selector and index in "idx", and store the results in "dst" using writemask "k" (elements are copied from "idx" when the corresponding mask bit is not set). + Arithmetic + + + + + + Multiply groups of 4 adjacent pairs of unsigned 8-bit integers in "a" with corresponding signed 8-bit integers in "b", producing 4 intermediate signed 16-bit results. Sum these 4 results with the corresponding 32-bit integer in "src", and store the packed 32-bit results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). FOR j := 0 to 3 - i := j*32 - off := idx[i+1:i]*32 IF k[j] - dst[i+31:i] := idx[i+2] ? b[off+31:off] : a[off+31:off] + tmp1.word := Signed(ZeroExtend16(a.byte[4*j]) * SignExtend16(b.byte[4*j])) + tmp2.word := Signed(ZeroExtend16(a.byte[4*j+1]) * SignExtend16(b.byte[4*j+1])) + tmp3.word := Signed(ZeroExtend16(a.byte[4*j+2]) * SignExtend16(b.byte[4*j+2])) + tmp4.word := Signed(ZeroExtend16(a.byte[4*j+3]) * SignExtend16(b.byte[4*j+3])) + dst.dword[j] := src.dword[j] + tmp1 + tmp2 + tmp3 + tmp4 ELSE - dst[i+31:i] := idx[i+31:i] + dst.dword[j] := 0 FI ENDFOR dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512_VNNI AVX512VL - AVX512F - Miscellaneous - - - - - Shuffle single-precision (32-bit) floating-point elements in "a" and "b" using the corresponding selector and index in "idx", and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). + Arithmetic + + + + + + Multiply groups of 4 adjacent pairs of unsigned 8-bit integers in "a" with corresponding signed 8-bit integers in "b", producing 4 intermediate signed 16-bit results. Sum these 4 results with the corresponding 32-bit integer in "src", and store the packed 32-bit results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). FOR j := 0 to 3 - i := j*32 - off := idx[i+1:i]*32 IF k[j] - dst[i+31:i] := idx[i+2] ? b[off+31:off] : a[off+31:off] + tmp1.word := Signed(ZeroExtend16(a.byte[4*j]) * SignExtend16(b.byte[4*j])) + tmp2.word := Signed(ZeroExtend16(a.byte[4*j+1]) * SignExtend16(b.byte[4*j+1])) + tmp3.word := Signed(ZeroExtend16(a.byte[4*j+2]) * SignExtend16(b.byte[4*j+2])) + tmp4.word := Signed(ZeroExtend16(a.byte[4*j+3]) * SignExtend16(b.byte[4*j+3])) + dst.dword[j] := src.dword[j] + tmp1 + tmp2 + tmp3 + tmp4 ELSE - dst[i+31:i] := a[i+31:i] + dst.dword[j] := src.dword[j] FI ENDFOR dst[MAX:128] := 0 - +
immintrin.h
 - - Floating Point + + Integer + AVX512_VNNI AVX512VL - AVX512F - Miscellaneous - - - - - Shuffle single-precision (32-bit) floating-point elements in "a" and "b" using the corresponding selector and index in "idx", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Arithmetic + + + + + Multiply groups of 4 adjacent pairs of unsigned 8-bit integers in "a" with corresponding signed 8-bit integers in "b", producing 4 intermediate signed 16-bit results. Sum these 4 results with the corresponding 32-bit integer in "src", and store the packed 32-bit results in "dst". FOR j := 0 to 3 - i := j*32 - off := idx[i+1:i]*32 - IF k[j] - dst[i+31:i] := (idx[i+2]) ? b[off+31:off] : a[off+31:off] - ELSE - dst[i+31:i] := 0 - FI + tmp1.word := Signed(ZeroExtend16(a.byte[4*j]) * SignExtend16(b.byte[4*j])) + tmp2.word := Signed(ZeroExtend16(a.byte[4*j+1]) * SignExtend16(b.byte[4*j+1])) + tmp3.word := Signed(ZeroExtend16(a.byte[4*j+2]) * SignExtend16(b.byte[4*j+2])) + tmp4.word := Signed(ZeroExtend16(a.byte[4*j+3]) * SignExtend16(b.byte[4*j+3])) + dst.dword[j] := src.dword[j] + tmp1 + tmp2 + tmp3 + tmp4 ENDFOR dst[MAX:128] := 0 - - +
immintrin.h
 - - Floating Point + + Integer + AVX512_VP2INTERSECT AVX512VL - AVX512F - Miscellaneous - - - - Shuffle single-precision (32-bit) floating-point elements in "a" and "b" using the corresponding selector and index in "idx", and store the results in "dst". - -FOR j := 0 to 3 - i := j*32 - off := idx[i+1:i]*32 - dst[i+31:i] := idx[i+2] ? b[off+31:off] : a[off+31:off] + Mask + + + + + + Compute intersection of packed 32-bit integer vectors "a" and "b", and store indication of match in the corresponding bit of two mask registers specified by "k1" and "k2". A match in corresponding elements of "a" and "b" is indicated by a set bit in the corresponding bit of the mask registers. + +MEM[k1+7:k1] := 0 +MEM[k2+7:k2] := 0 +FOR i := 0 TO 3 + FOR j := 0 TO 3 + match := (a.dword[i] == b.dword[j] ? 1 : 0) + MEM[k1+7:k1].bit[i] |= match + MEM[k2+7:k2].bit[j] |= match + ENDFOR ENDFOR -dst[MAX:128] := 0 - - +
immintrin.h
 - + Integer + AVX512_VP2INTERSECT AVX512VL - AVX512F - Miscellaneous - - - - - Shuffle 64-bit integers in "a" and "b" across lanes using the corresponding selector and index in "idx", and store the results in "dst" using writemask "k" (elements are copied from "idx" when the corresponding mask bit is not set). - -FOR j := 0 to 3 - i := j*64 - off := idx[i+1:i]*64 - IF k[j] - dst[i+63:i] := idx[i+2] ? b[off+63:off] : a[off+63:off] - ELSE - dst[i+63:i] := idx[i+63:i] - FI + Mask + + + + + + Compute intersection of packed 32-bit integer vectors "a" and "b", and store indication of match in the corresponding bit of two mask registers specified by "k1" and "k2". A match in corresponding elements of "a" and "b" is indicated by a set bit in the corresponding bit of the mask registers. + +MEM[k1+7:k1] := 0 +MEM[k2+7:k2] := 0 +FOR i := 0 TO 7 + FOR j := 0 TO 7 + match := (a.dword[i] == b.dword[j] ? 1 : 0) + MEM[k1+7:k1].bit[i] |= match + MEM[k2+7:k2].bit[j] |= match + ENDFOR ENDFOR -dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - AVX512VL + AVX512_VP2INTERSECT AVX512F - Miscellaneous - - - - - Shuffle 64-bit integers in "a" and "b" across lanes using the corresponding selector and index in "idx", and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). - -FOR j := 0 to 3 - i := j*64 - off := idx[i+1:i]*64 - IF k[j] - dst[i+63:i] := idx[i+2] ? b[off+63:off] : a[off+63:off] - ELSE - dst[i+63:i] := a[i+63:i] - FI + Mask + + + + + + Compute intersection of packed 32-bit integer vectors "a" and "b", and store indication of match in the corresponding bit of two mask registers specified by "k1" and "k2". A match in corresponding elements of "a" and "b" is indicated by a set bit in the corresponding bit of the mask registers. + +MEM[k1+15:k1] := 0 +MEM[k2+15:k2] := 0 +FOR i := 0 TO 15 + FOR j := 0 TO 15 + match := (a.dword[i] == b.dword[j] ? 1 : 0) + MEM[k1+15:k1].bit[i] |= match + MEM[k2+15:k2].bit[j] |= match + ENDFOR ENDFOR -dst[MAX:256] := 0 - +
immintrin.h
 - + Integer + AVX512_VP2INTERSECT AVX512VL - AVX512F - Miscellaneous - - - - - Shuffle 64-bit integers in "a" and "b" across lanes using the corresponding selector and index in "idx", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - -FOR j := 0 to 3 - i := j*64 - off := idx[i+1:i]*64 - IF k[j] - dst[i+63:i] := (idx[i+2]) ? b[off+63:off] : a[off+63:off] - ELSE - dst[i+63:i] := 0 - FI + Mask + + + + + + Compute intersection of packed 64-bit integer vectors "a" and "b", and store indication of match in the corresponding bit of two mask registers specified by "k1" and "k2". A match in corresponding elements of "a" and "b" is indicated by a set bit in the corresponding bit of the mask registers. + +MEM[k1+7:k1] := 0 +MEM[k2+7:k2] := 0 +FOR i := 0 TO 1 + FOR j := 0 TO 1 + match := (a.qword[i] == b.qword[j] ? 1 : 0) + MEM[k1+7:k1].bit[i] |= match + MEM[k2+7:k2].bit[j] |= match + ENDFOR ENDFOR -dst[MAX:256] := 0 - - +
immintrin.h
 - + Integer + AVX512_VP2INTERSECT AVX512VL - AVX512F - Miscellaneous - - - - Shuffle 64-bit integers in "a" and "b" across lanes using the corresponding selector and index in "idx", and store the results in "dst". - - -FOR j := 0 to 3 - i := j*64 - off := idx[i+1:i]*64 - dst[i+63:i] := idx[i+2] ? b[off+63:off] : a[off+63:off] + Mask + + + + + + Compute intersection of packed 64-bit integer vectors "a" and "b", and store indication of match in the corresponding bit of two mask registers specified by "k1" and "k2". A match in corresponding elements of "a" and "b" is indicated by a set bit in the corresponding bit of the mask registers. + +MEM[k1+7:k1] := 0 +MEM[k2+7:k2] := 0 +FOR i := 0 TO 3 + FOR j := 0 TO 3 + match := (a.qword[i] == b.qword[j] ? 1 : 0) + MEM[k1+7:k1].bit[i] |= match + MEM[k2+7:k2].bit[j] |= match + ENDFOR ENDFOR -dst[MAX:256] := 0 - - +
immintrin.h
 - + Integer - AVX512VL + AVX512_VP2INTERSECT AVX512F - Miscellaneous - - - - - Shuffle 64-bit integers in "a" and "b" using the corresponding selector and index in "idx", and store the results in "dst" using writemask "k" (elements are copied from "idx" when the corresponding mask bit is not set). - -FOR j := 0 to 1 - i := j*64 - off := idx[i]*64 - IF k[j] - dst[i+63:i] := idx[i+1] ? b[off+63:off] : a[off+63:off] - ELSE - dst[i+63:i] := idx[i+63:i] - FI + Mask + + + + + + Compute intersection of packed 64-bit integer vectors "a" and "b", and store indication of match in the corresponding bit of two mask registers specified by "k1" and "k2". A match in corresponding elements of "a" and "b" is indicated by a set bit in the corresponding bit of the mask registers. + +MEM[k1+7:k1] := 0 +MEM[k2+7:k2] := 0 +FOR i := 0 TO 7 + FOR j := 0 TO 7 + match := (a.qword[i] == b.qword[j] ? 1 : 0) + MEM[k1+7:k1].bit[i] |= match + MEM[k2+7:k2].bit[j] |= match + ENDFOR ENDFOR -dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - AVX512VL - AVX512F - Miscellaneous - - - - - Shuffle 64-bit integers in "a" and "b" using the corresponding selector and index in "idx", and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). + BMI1 + Bit Manipulation + + + + + Extract contiguous bits from unsigned 32-bit integer "a", and store the result in "dst". Extract the number of bits specified by "len", starting at the bit specified by "start". -FOR j := 0 to 1 - i := j*64 - off := idx[i]*64 - IF k[j] - dst[i+63:i] := idx[i+1] ? b[off+63:off] : a[off+63:off] - ELSE - dst[i+63:i] := a[i+63:i] - FI -ENDFOR -dst[MAX:128] := 0 +tmp[511:0] := a +dst[31:0] := ZeroExtend32(tmp[(start[7:0] + len[7:0] - 1):start[7:0]]) - +
immintrin.h
 - + Integer - AVX512VL - AVX512F - Miscellaneous - - - - - Shuffle 64-bit integers in "a" and "b" using the corresponding selector and index in "idx", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + BMI1 + Bit Manipulation + + + + Extract contiguous bits from unsigned 32-bit integer "a", and store the result in "dst". Extract the number of bits specified by bits 15:8 of "control", starting at the bit specified by bits 0:7 of "control". -FOR j := 0 to 1 - i := j*64 - off := idx[i]*64 - IF k[j] - dst[i+63:i] := (idx[i+1]) ? b[off+63:off] : a[off+63:off] - ELSE - dst[i+63:i] := 0 - FI -ENDFOR -dst[MAX:128] := 0 +start := control[7:0] +len := control[15:8] +tmp[511:0] := a +dst[31:0] := ZeroExtend32(tmp[(start[7:0] + len[7:0] - 1):start[7:0]]) - - +
immintrin.h
 - + Integer - AVX512VL - AVX512F - Miscellaneous - - - - Shuffle 64-bit integers in "a" and "b" using the corresponding selector and index in "idx", and store the results in "dst". - + BMI1 + Bit Manipulation + + + + + Extract contiguous bits from unsigned 64-bit integer "a", and store the result in "dst". Extract the number of bits specified by "len", starting at the bit specified by "start". -FOR j := 0 to 1 - i := j*64 - off := idx[i]*64 - dst[i+63:i] := idx[i+1] ? b[off+63:off] : a[off+63:off] -ENDFOR -dst[MAX:128] := 0 +tmp[511:0] := a +dst[63:0] := ZeroExtend64(tmp[(start[7:0] + len[7:0] - 1):start[7:0]]) - - +
immintrin.h
 - + Integer - AVX512VL - AVX512BW - Miscellaneous - - - - - Shuffle 16-bit integers in "a" and "b" across lanes using the corresponding selector and index in "idx", and store the results in "dst" using writemask "k" (elements are copied from "idx" when the corresponding mask bit is not set). + BMI1 + Bit Manipulation + + + + Extract contiguous bits from unsigned 64-bit integer "a", and store the result in "dst". Extract the number of bits specified by bits 15:8 of "control", starting at the bit specified by bits 0:7 of "control".. -FOR j := 0 to 15 - i := j*16 - IF k[j] - off := 16*idx[i+3:i] - dst[i+15:i] := idx[i+4] ? b[off+15:off] : a[off+15:off] - ELSE - dst[i+15:i] := idx[i+15:i] - FI -ENDFOR -dst[MAX:256] := 0 +start := control[7:0] +len := control[15:8] +tmp[511:0] := a +dst[63:0] := ZeroExtend64(tmp[(start[7:0] + len[7:0] - 1):start[7:0]]) - +
immintrin.h
 - + Integer - AVX512VL - AVX512BW - Miscellaneous - - - - - Shuffle 16-bit integers in "a" and "b" across lanes using the corresponding selector and index in "idx", and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). + BMI1 + Bit Manipulation + + + Extract the lowest set bit from unsigned 32-bit integer "a" and set the corresponding bit in "dst". All other bits in "dst" are zeroed, and all bits are zeroed if no bits are set in "a". -FOR j := 0 to 15 - i := j*16 - IF k[j] - off := 16*idx[i+3:i] - dst[i+15:i] := idx[i+4] ? b[off+15:off] : a[off+15:off] - ELSE - dst[i+15:i] := a[i+15:i] - FI -ENDFOR -dst[MAX:256] := 0 +dst := (-a) AND a - +
immintrin.h
 - + Integer - AVX512VL - AVX512BW - Miscellaneous - - - - - Shuffle 16-bit integers in "a" and "b" across lanes using the corresponding selector and index in "idx", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + BMI1 + Bit Manipulation + + + Extract the lowest set bit from unsigned 64-bit integer "a" and set the corresponding bit in "dst". All other bits in "dst" are zeroed, and all bits are zeroed if no bits are set in "a". -FOR j := 0 to 15 - i := j*16 - IF k[j] - off := 16*idx[i+3:i] - dst[i+15:i] := idx[i+4] ? b[off+15:off] : a[off+15:off] - ELSE - dst[i+15:i] := 0 - FI -ENDFOR -dst[MAX:256] := 0 +dst := (-a) AND a - - +
immintrin.h
 - + Integer - AVX512VL - AVX512BW - Miscellaneous - - - - Shuffle 16-bit integers in "a" and "b" across lanes using the corresponding selector and index in "idx", and store the results in "dst". + BMI1 + Bit Manipulation + + + Set all the lower bits of "dst" up to and including the lowest set bit in unsigned 32-bit integer "a". -FOR j := 0 to 15 - i := j*16 - off := 16*idx[i+3:i] - dst[i+15:i] := idx[i+4] ? b[off+15:off] : a[off+15:off] -ENDFOR -dst[MAX:256] := 0 +dst := (a - 1) XOR a - - +
immintrin.h
 - + Integer - AVX512BW - Miscellaneous - - - - - Shuffle 16-bit integers in "a" and "b" across lanes using the corresponding selector and index in "idx", and store the results in "dst" using writemask "k" (elements are copied from "idx" when the corresponding mask bit is not set). + BMI1 + Bit Manipulation + + + Set all the lower bits of "dst" up to and including the lowest set bit in unsigned 64-bit integer "a". -FOR j := 0 to 31 - i := j*16 - IF k[j] - off := 16*idx[i+4:i] - dst[i+15:i] := idx[i+5] ? b[off+15:off] : a[off+15:off] - ELSE - dst[i+15:i] := idx[i+15:i] +dst := (a - 1) XOR a + + +
immintrin.h
+ + + Integer + BMI1 + Bit Manipulation + + + Copy all bits from unsigned 32-bit integer "a" to "dst", and reset (set to 0) the bit in "dst" that corresponds to the lowest set bit in "a". + +dst := (a - 1) AND a + + +
immintrin.h
+ + + Integer + BMI1 + Bit Manipulation + + + Copy all bits from unsigned 64-bit integer "a" to "dst", and reset (set to 0) the bit in "dst" that corresponds to the lowest set bit in "a". + +dst := (a - 1) AND a + + +
immintrin.h
+ + + Integer + BMI1 + Bit Manipulation + + + + Compute the bitwise NOT of 32-bit integer "a" and then AND with b, and store the results in dst. + +dst[31:0] := ((NOT a[31:0]) AND b[31:0]) + + +
immintrin.h
+ + + Integer + BMI1 + Bit Manipulation + + + + Compute the bitwise NOT of 64-bit integer "a" and then AND with b, and store the results in dst. + +dst[63:0] := ((NOT a[63:0]) AND b[63:0]) + + +
immintrin.h
+ + + Integer + BMI1 + Bit Manipulation + + + Count the number of trailing zero bits in unsigned 32-bit integer "a", and return that count in "dst". + +tmp := 0 +dst := 0 +DO WHILE ((tmp < 32) AND a[tmp] == 0) + tmp := tmp + 1 + dst := dst + 1 +OD + + +
immintrin.h
+ + + Integer + BMI1 + Bit Manipulation + + + Count the number of trailing zero bits in unsigned 64-bit integer "a", and return that count in "dst". + +tmp := 0 +dst := 0 +DO WHILE ((tmp < 64) AND a[tmp] == 0) + tmp := tmp + 1 + dst := dst + 1 +OD + + +
immintrin.h
+ + + Integer + BMI1 + Bit Manipulation + + + Count the number of trailing zero bits in unsigned 32-bit integer "a", and return that count in "dst". + +tmp := 0 +dst := 0 +DO WHILE ((tmp < 32) AND a[tmp] == 0) + tmp := tmp + 1 + dst := dst + 1 +OD + + +
immintrin.h
+ + + Integer + BMI1 + Bit Manipulation + + + Count the number of trailing zero bits in unsigned 64-bit integer "a", and return that count in "dst". + +tmp := 0 +dst := 0 +DO WHILE ((tmp < 64) AND a[tmp] == 0) + tmp := tmp + 1 + dst := dst + 1 +OD + + +
immintrin.h
+ + + Integer + BMI2 + Bit Manipulation + + + + Copy all bits from unsigned 32-bit integer "a" to "dst", and reset (set to 0) the high bits in "dst" starting at "index". + +n := index[7:0] +dst := a +IF (n < 32) + dst[31:n] := 0 +FI + + +
immintrin.h
+ + + Integer + BMI2 + Bit Manipulation + + + + Copy all bits from unsigned 64-bit integer "a" to "dst", and reset (set to 0) the high bits in "dst" starting at "index". + +n := index[7:0] +dst := a +IF (n < 64) + dst[63:n] := 0 +FI + + +
immintrin.h
+ + + Integer + BMI2 + Bit Manipulation + + + + Deposit contiguous low bits from unsigned 32-bit integer "a" to "dst" at the corresponding bit locations specified by "mask"; all other bits in "dst" are set to zero. + +tmp := a +dst := 0 +m := 0 +k := 0 +DO WHILE m < 32 + IF mask[m] == 1 + dst[m] := tmp[k] + k := k + 1 FI -ENDFOR -dst[MAX:512] := 0 + m := m + 1 +OD - +
immintrin.h
 - + Integer - AVX512BW - Miscellaneous - - - - - Shuffle 16-bit integers in "a" and "b" across lanes using the corresponding selector and index in "idx", and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). + BMI2 + Bit Manipulation + + + + Deposit contiguous low bits from unsigned 64-bit integer "a" to "dst" at the corresponding bit locations specified by "mask"; all other bits in "dst" are set to zero. -FOR j := 0 to 31 - i := j*16 - IF k[j] - off := 16*idx[i+4:i] - dst[i+15:i] := idx[i+5] ? b[off+15:off] : a[off+15:off] - ELSE - dst[i+15:i] := a[i+15:i] +tmp := a +dst := 0 +m := 0 +k := 0 +DO WHILE m < 64 + IF mask[m] == 1 + dst[m] := tmp[k] + k := k + 1 FI -ENDFOR -dst[MAX:512] := 0 + m := m + 1 +OD - +
immintrin.h
 - + Integer - AVX512BW - Miscellaneous - - - - - Shuffle 16-bit integers in "a" and "b" across lanes using the corresponding selector and index in "idx", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + BMI2 + Bit Manipulation + + + + Extract bits from unsigned 32-bit integer "a" at the corresponding bit locations specified by "mask" to contiguous low bits in "dst"; the remaining upper bits in "dst" are set to zero. -FOR j := 0 to 31 - i := j*16 - IF k[j] - off := 16*idx[i+4:i] - dst[i+15:i] := idx[i+5] ? b[off+15:off] : a[off+15:off] - ELSE - dst[i+15:i] := 0 +tmp := a +dst := 0 +m := 0 +k := 0 +DO WHILE m < 32 + IF mask[m] == 1 + dst[k] := tmp[m] + k := k + 1 FI -ENDFOR -dst[MAX:512] := 0 + m := m + 1 +OD - - +
immintrin.h
 - + Integer - AVX512BW + BMI2 + Bit Manipulation + + + + Extract bits from unsigned 64-bit integer "a" at the corresponding bit locations specified by "mask" to contiguous low bits in "dst"; the remaining upper bits in "dst" are set to zero. + +tmp := a +dst := 0 +m := 0 +k := 0 +DO WHILE m < 64 + IF mask[m] == 1 + dst[k] := tmp[m] + k := k + 1 + FI + m := m + 1 +OD + + +
immintrin.h
+ + + Integer + BMI2 + Arithmetic + + + + + Multiply unsigned 32-bit integers "a" and "b", store the low 32-bits of the result in "dst", and store the high 32-bits in "hi". This does not read or write arithmetic flags. + +dst[31:0] := (a * b)[31:0] +MEM[hi+31:hi] := (a * b)[63:32] + + +
immintrin.h
+ + + Integer + BMI2 + Arithmetic + + + + + Multiply unsigned 64-bit integers "a" and "b", store the low 64-bits of the result in "dst", and store the high 64-bits in "hi". This does not read or write arithmetic flags. + +dst[63:0] := (a * b)[63:0] +MEM[hi+63:hi] := (a * b)[127:64] + + +
immintrin.h
+ + + CET_SS Miscellaneous - - - - Shuffle 16-bit integers in "a" and "b" across lanes using the corresponding selector and index in "idx", and store the results in "dst". + + + Increment the shadow stack pointer by 4 times the value specified in bits [7:0] of "a". -FOR j := 0 to 31 - i := j*16 - off := 16*idx[i+4:i] - dst[i+15:i] := idx[i+5] ? b[off+15:off] : a[off+15:off] -ENDFOR -dst[MAX:512] := 0 +SSP := SSP + a[7:0] * 4 + + +
immintrin.h
+ + + CET_SS + Miscellaneous + + + Increment the shadow stack pointer by 8 times the value specified in bits [7:0] of "a". + +SSP := SSP + a[7:0] * 8 + + +
immintrin.h
+ + + CET_SS + Miscellaneous + + + Read the low 32-bits of the current shadow stack pointer, and store the result in "dst". + dst := SSP[31:0] - - + +
immintrin.h
+ + + CET_SS + Miscellaneous + + + Read the current shadow stack pointer, and store the result in "dst". + dst := SSP[63:0] + + +
immintrin.h
+ + + CET_SS + Miscellaneous + + + Save the previous shadow stack pointer context. + +
immintrin.h
+ + + CET_SS + Miscellaneous + + + Restore the saved shadow stack pointer from the shadow stack restore token previously created on shadow stack by saveprevssp. + +
immintrin.h
+ + + CET_SS + Miscellaneous + + + + Write 32-bit value in "val" to a shadow stack page in memory specified by "p". + +
immintrin.h
+ + + CET_SS + Miscellaneous + + + + Write 64-bit value in "val" to a shadow stack page in memory specified by "p". + +
immintrin.h
+ + + CET_SS + Miscellaneous + + + + Write 32-bit value in "val" to a user shadow stack page in memory specified by "p". + +
immintrin.h
+ + + CET_SS + Miscellaneous + + + + Write 64-bit value in "val" to a user shadow stack page in memory specified by "p". + +
immintrin.h
+ + + CET_SS + Miscellaneous + + + Mark shadow stack pointed to by IA32_PL0_SSP as busy. +
immintrin.h
 - - Integer - AVX512VL - AVX512BW + + CET_SS Miscellaneous - - - - - Shuffle 16-bit integers in "a" and "b" using the corresponding selector and index in "idx", and store the results in "dst" using writemask "k" (elements are copied from "idx" when the corresponding mask bit is not set). - -FOR j := 0 to 7 - i := j*16 - IF k[j] - off := 16*idx[i+2:i] - dst[i+15:i] := idx[i+3] ? b[off+15:off] : a[off+15:off] - ELSE - dst[i+15:i] := idx[i+15:i] - FI -ENDFOR -dst[MAX:128] := 0 - - + + + Mark shadow stack pointed to by "p" as not busy. +
immintrin.h
 - - Integer - AVX512VL - AVX512BW + + CET_SS Miscellaneous - - - - - Shuffle 16-bit integers in "a" and "b" using the corresponding selector and index in "idx", and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). - -FOR j := 0 to 7 - i := j*16 - IF k[j] - off := 16*idx[i+2:i] - dst[i+15:i] := idx[i+3] ? b[off+15:off] : a[off+15:off] - ELSE - dst[i+15:i] := a[i+15:i] - FI -ENDFOR -dst[MAX:128] := 0 + + + If CET is enabled, read the low 32-bits of the current shadow stack pointer, and store the result in "dst". Otherwise return 0. + dst := SSP[31:0] - +
immintrin.h
 - - Integer - AVX512VL - AVX512BW + + CET_SS Miscellaneous - - - - - Shuffle 16-bit integers in "a" and "b" using the corresponding selector and index in "idx", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - -FOR j := 0 to 7 - i := j*16 - IF k[j] - off := 16*idx[i+2:i] - dst[i+15:i] := idx[i+3] ? b[off+15:off] : a[off+15:off] - ELSE - dst[i+15:i] := 0 - FI -ENDFOR -dst[MAX:128] := 0 + + + If CET is enabled, read the current shadow stack pointer, and store the result in "dst". Otherwise return 0. + dst := SSP[63:0] - - +
immintrin.h
 - - Integer - AVX512VL - AVX512BW + + CET_SS Miscellaneous - - - - Shuffle 16-bit integers in "a" and "b" using the corresponding selector and index in "idx", and store the results in "dst". + + + Increment the shadow stack pointer by 4 times the value specified in bits [7:0] of "a". -FOR j := 0 to 7 - i := j*16 - off := 16*idx[i+2:i] - dst[i+15:i] := idx[i+4] ? b[off+15:off] : a[off+15:off] -ENDFOR -dst[MAX:128] := 0 +SSP := SSP + a[7:0] * 4 - - +
immintrin.h
 - - Floating Point - AVX512VL - AVX512F + + CLDEMOTE Miscellaneous - - - - - Shuffle double-precision (64-bit) floating-point elements in "a" within 128-bit lanes using the control in "imm8", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + + Hint to hardware that the cache line that contains "p" should be demoted from the cache closest to the processor core to a level more distant from the processor core. + +
immintrin.h
+ + + CLFLUSHOPT + General Support + + + Invalidate and flush the cache line that contains "p" from all levels of the cache hierarchy. + +
immintrin.h
+ + + CLWB + General Support + + + Write back to memory the cache line that contains "p" from any level of the cache hierarchy in the cache coherence domain. + +
immintrin.h
+ + + Floating Point + FMA + Arithmetic + + + + + Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", add the intermediate result to packed elements in "c", and store the results in "dst". -IF (imm8[0] == 0) tmp_dst[63:0] := a[63:0] -IF (imm8[0] == 1) tmp_dst[63:0] := a[127:64] -IF (imm8[1] == 0) tmp_dst[127:64] := a[63:0] -IF (imm8[1] == 1) tmp_dst[127:64] := a[127:64] -IF (imm8[2] == 0) tmp_dst[191:128] := a[191:128] -IF (imm8[2] == 1) tmp_dst[191:128] := a[255:192] -IF (imm8[3] == 0) tmp_dst[255:192] := a[191:128] -IF (imm8[3] == 1) tmp_dst[255:192] := a[255:192] -FOR j := 0 to 3 +FOR j := 0 to 1 i := j*64 - IF k[j] - dst[i+63:i] := tmp_dst[i+63:i] - ELSE - dst[i+63:i] := src[i+63:i] - FI + dst[i+63:i] := (a[i+63:i] * b[i+63:i]) + c[i+63:i] ENDFOR -dst[MAX:256] := 0 +dst[MAX:128] := 0 - + + +
immintrin.h
 - + Floating Point - AVX512VL - AVX512F - Miscellaneous - - - - - Shuffle double-precision (64-bit) floating-point elements in "a" within 128-bit lanes using the control in "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + FMA + Arithmetic + + + + + Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", add the intermediate result to packed elements in "c", and store the results in "dst". -IF (b[1] == 0) tmp_dst[63:0] := a[63:0] -IF (b[1] == 1) tmp_dst[63:0] := a[127:64] -IF (b[65] == 0) tmp_dst[127:64] := a[63:0] -IF (b[65] == 1) tmp_dst[127:64] := a[127:64] -IF (b[129] == 0) tmp_dst[191:128] := a[191:128] -IF (b[129] == 1) tmp_dst[191:128] := a[255:192] -IF (b[193] == 0) tmp_dst[255:192] := a[191:128] -IF (b[193] == 1) tmp_dst[255:192] := a[255:192] FOR j := 0 to 3 i := j*64 - IF k[j] - dst[i+63:i] := tmp_dst[i+63:i] - ELSE - dst[i+63:i] := src[i+63:i] - FI + dst[i+63:i] := (a[i+63:i] * b[i+63:i]) + c[i+63:i] ENDFOR dst[MAX:256] := 0 - + + +
immintrin.h
 - + Floating Point - AVX512VL - AVX512F - Miscellaneous - - - - Shuffle double-precision (64-bit) floating-point elements in "a" within 128-bit lanes using the control in "imm8", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + FMA + Arithmetic + + + + + Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", add the intermediate result to packed elements in "c", and store the results in "dst". -IF (imm8[0] == 0) tmp_dst[63:0] := a[63:0] -IF (imm8[0] == 1) tmp_dst[63:0] := a[127:64] -IF (imm8[1] == 0) tmp_dst[127:64] := a[63:0] -IF (imm8[1] == 1) tmp_dst[127:64] := a[127:64] -IF (imm8[2] == 0) tmp_dst[191:128] := a[191:128] -IF (imm8[2] == 1) tmp_dst[191:128] := a[255:192] -IF (imm8[3] == 0) tmp_dst[255:192] := a[191:128] -IF (imm8[3] == 1) tmp_dst[255:192] := a[255:192] FOR j := 0 to 3 - i := j*64 - IF k[j] - dst[i+63:i] := tmp_dst[i+63:i] - ELSE - dst[i+63:i] := 0 - FI + i := j*32 + dst[i+31:i] := (a[i+31:i] * b[i+31:i]) + c[i+31:i] ENDFOR -dst[MAX:256] := 0 +dst[MAX:128] := 0 - + + +
immintrin.h
 - + Floating Point - AVX512VL - AVX512F - Miscellaneous - - - - Shuffle double-precision (64-bit) floating-point elements in "a" within 128-bit lanes using the control in "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + FMA + Arithmetic + + + + + Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", add the intermediate result to packed elements in "c", and store the results in "dst". -IF (b[1] == 0) tmp_dst[63:0] := a[63:0] -IF (b[1] == 1) tmp_dst[63:0] := a[127:64] -IF (b[65] == 0) tmp_dst[127:64] := a[63:0] -IF (b[65] == 1) tmp_dst[127:64] := a[127:64] -IF (b[129] == 0) tmp_dst[191:128] := a[191:128] -IF (b[129] == 1) tmp_dst[191:128] := a[255:192] -IF (b[193] == 0) tmp_dst[255:192] := a[191:128] -IF (b[193] == 1) tmp_dst[255:192] := a[255:192] -FOR j := 0 to 3 - i := j*64 - IF k[j] - dst[i+63:i] := tmp_dst[i+63:i] - ELSE - dst[i+63:i] := 0 - FI +FOR j := 0 to 7 + i := j*32 + dst[i+31:i] := (a[i+31:i] * b[i+31:i]) + c[i+31:i] ENDFOR dst[MAX:256] := 0 - + + +
immintrin.h
 - + Floating Point - AVX512VL - AVX512F - Miscellaneous - - - - - Shuffle double-precision (64-bit) floating-point elements in "a" using the control in "imm8", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + FMA + Arithmetic + + + + + Multiply the lower double-precision (64-bit) floating-point elements in "a" and "b", and add the intermediate result to the lower element in "c". Store the result in the lower element of "dst", and copy the upper element from "a" to the upper element of "dst". -IF (imm8[0] == 0) tmp_dst[63:0] := a[63:0] -IF (imm8[0] == 1) tmp_dst[63:0] := a[127:64] -IF (imm8[1] == 0) tmp_dst[127:64] := a[63:0] -IF (imm8[1] == 1) tmp_dst[127:64] := a[127:64] -FOR j := 0 to 1 - i := j*64 - IF k[j] - dst[i+63:i] := tmp_dst[i+63:i] - ELSE - dst[i+63:i] := src[i+63:i] - FI -ENDFOR +dst[63:0] := (a[63:0] * b[63:0]) + c[63:0] +dst[127:64] := a[127:64] dst[MAX:128] := 0 - + + +
immintrin.h
 - + Floating Point - AVX512VL - AVX512F - Miscellaneous - - - - - Shuffle double-precision (64-bit) floating-point elements in "a" using the control in "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + FMA + Arithmetic + + + + + Multiply the lower single-precision (32-bit) floating-point elements in "a" and "b", and add the intermediate result to the lower element in "c". Store the result in the lower element of "dst", and copy the upper 3 packed elements from "a" to the upper elements of "dst". -IF (b[1] == 0) tmp_dst[63:0] := a[63:0] -IF (b[1] == 1) tmp_dst[63:0] := a[127:64] -IF (b[65] == 0) tmp_dst[127:64] := a[63:0] -IF (b[65] == 1) tmp_dst[127:64] := a[127:64] -FOR j := 0 to 1 - i := j*64 - IF k[j] - dst[i+63:i] := tmp_dst[i+63:i] - ELSE - dst[i+63:i] := src[i+63:i] - FI -ENDFOR +dst[31:0] := (a[31:0] * b[31:0]) + c[31:0] +dst[127:32] := a[127:32] dst[MAX:128] := 0 - + + +
immintrin.h
 - + Floating Point - AVX512VL - AVX512F - Miscellaneous - - - - Shuffle double-precision (64-bit) floating-point elements in "a" using the control in "imm8", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + FMA + Arithmetic + + + + + Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", alternatively add and subtract packed elements in "c" to/from the intermediate result, and store the results in "dst". -IF (imm8[0] == 0) tmp_dst[63:0] := a[63:0] -IF (imm8[0] == 1) tmp_dst[63:0] := a[127:64] -IF (imm8[1] == 0) tmp_dst[127:64] := a[63:0] -IF (imm8[1] == 1) tmp_dst[127:64] := a[127:64] FOR j := 0 to 1 i := j*64 - IF k[j] - dst[i+63:i] := tmp_dst[i+63:i] + IF ((j & 1) == 0) + dst[i+63:i] := (a[i+63:i] * b[i+63:i]) - c[i+63:i] ELSE - dst[i+63:i] := 0 + dst[i+63:i] := (a[i+63:i] * b[i+63:i]) + c[i+63:i] FI ENDFOR dst[MAX:128] := 0 - + + +
immintrin.h
 - + Floating Point - AVX512VL - AVX512F - Miscellaneous - - - - Shuffle double-precision (64-bit) floating-point elements in "a" using the control in "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + FMA + Arithmetic + + + + + Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", alternatively add and subtract packed elements in "c" to/from the intermediate result, and store the results in "dst". -IF (b[1] == 0) tmp_dst[63:0] := a[63:0] -IF (b[1] == 1) tmp_dst[63:0] := a[127:64] -IF (b[65] == 0) tmp_dst[127:64] := a[63:0] -IF (b[65] == 1) tmp_dst[127:64] := a[127:64] -FOR j := 0 to 1 +FOR j := 0 to 3 i := j*64 - IF k[j] - dst[i+63:i] := tmp_dst[i+63:i] + IF ((j & 1) == 0) + dst[i+63:i] := (a[i+63:i] * b[i+63:i]) - c[i+63:i] ELSE - dst[i+63:i] := 0 + dst[i+63:i] := (a[i+63:i] * b[i+63:i]) + c[i+63:i] FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:256] := 0 - + + +
immintrin.h
 - + Floating Point - AVX512VL - AVX512F - Miscellaneous - - - - - Shuffle single-precision (32-bit) floating-point elements in "a" within 128-bit lanes using the control in "imm8", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + FMA + Arithmetic + + + + + Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", alternatively add and subtract packed elements in "c" to/from the intermediate result, and store the results in "dst". -SELECT4(src, control){ - CASE(control[1:0]) - 0: tmp[31:0] := src[31:0] - 1: tmp[31:0] := src[63:32] - 2: tmp[31:0] := src[95:64] - 3: tmp[31:0] := src[127:96] - ESAC - RETURN tmp[31:0] -} - -tmp_dst[31:0] := SELECT4(a[127:0], imm8[1:0]) -tmp_dst[63:32] := SELECT4(a[127:0], imm8[3:2]) -tmp_dst[95:64] := SELECT4(a[127:0], imm8[5:4]) -tmp_dst[127:96] := SELECT4(a[127:0], imm8[7:6]) -tmp_dst[159:128] := SELECT4(a[255:128], imm8[1:0]) -tmp_dst[191:160] := SELECT4(a[255:128], imm8[3:2]) -tmp_dst[223:192] := SELECT4(a[255:128], imm8[5:4]) -tmp_dst[255:224] := SELECT4(a[255:128], imm8[7:6]) -FOR j := 0 to 7 +FOR j := 0 to 3 i := j*32 - IF k[j] - dst[i+31:i] := tmp_dst[i+31:i] + IF ((j & 1) == 0) + dst[i+31:i] := (a[i+31:i] * b[i+31:i]) - c[i+31:i] ELSE - dst[i+31:i] := src[i+31:i] + dst[i+31:i] := (a[i+31:i] * b[i+31:i]) + c[i+31:i] FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:128] := 0 - + + +
immintrin.h
 - + Floating Point - AVX512VL - AVX512F - Miscellaneous - - - - - Shuffle single-precision (32-bit) floating-point elements in "a" within 128-bit lanes using the control in "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + FMA + Arithmetic + + + + + Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", alternatively add and subtract packed elements in "c" to/from the intermediate result, and store the results in "dst". -SELECT4(src, control){ - CASE(control[1:0]) - 0: tmp[31:0] := src[31:0] - 1: tmp[31:0] := src[63:32] - 2: tmp[31:0] := src[95:64] - 3: tmp[31:0] := src[127:96] - ESAC - RETURN tmp[31:0] -} - -tmp_dst[31:0] := SELECT4(a[127:0], b[1:0]) -tmp_dst[63:32] := SELECT4(a[127:0], b[33:32]) -tmp_dst[95:64] := SELECT4(a[127:0], b[65:64]) -tmp_dst[127:96] := SELECT4(a[127:0], b[97:96]) -tmp_dst[159:128] := SELECT4(a[255:128], b[129:128]) -tmp_dst[191:160] := SELECT4(a[255:128], b[161:160]) -tmp_dst[223:192] := SELECT4(a[255:128], b[193:192]) -tmp_dst[255:224] := SELECT4(a[255:128], b[225:224]) FOR j := 0 to 7 i := j*32 - IF k[j] - dst[i+31:i] := tmp_dst[i+31:i] + IF ((j & 1) == 0) + dst[i+31:i] := (a[i+31:i] * b[i+31:i]) - c[i+31:i] ELSE - dst[i+31:i] := src[i+31:i] + dst[i+31:i] := (a[i+31:i] * b[i+31:i]) + c[i+31:i] FI ENDFOR dst[MAX:256] := 0 - + + +
immintrin.h
 - + Floating Point - AVX512VL - AVX512F - Miscellaneous - - - - Shuffle single-precision (32-bit) floating-point elements in "a" within 128-bit lanes using the control in "imm8", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + FMA + Arithmetic + + + + + Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", subtract packed elements in "c" from the intermediate result, and store the results in "dst". -SELECT4(src, control){ - CASE(control[1:0]) - 0: tmp[31:0] := src[31:0] - 1: tmp[31:0] := src[63:32] - 2: tmp[31:0] := src[95:64] - 3: tmp[31:0] := src[127:96] - ESAC - RETURN tmp[31:0] -} - -tmp_dst[31:0] := SELECT4(a[127:0], imm8[1:0]) -tmp_dst[63:32] := SELECT4(a[127:0], imm8[3:2]) -tmp_dst[95:64] := SELECT4(a[127:0], imm8[5:4]) -tmp_dst[127:96] := SELECT4(a[127:0], imm8[7:6]) -tmp_dst[159:128] := SELECT4(a[255:128], imm8[1:0]) -tmp_dst[191:160] := SELECT4(a[255:128], imm8[3:2]) -tmp_dst[223:192] := SELECT4(a[255:128], imm8[5:4]) -tmp_dst[255:224] := SELECT4(a[255:128], imm8[7:6]) -FOR j := 0 to 7 - i := j*32 - IF k[j] - dst[i+31:i] := tmp_dst[i+31:i] - ELSE - dst[i+31:i] := 0 - FI +FOR j := 0 to 1 + i := j*64 + dst[i+63:i] := (a[i+63:i] * b[i+63:i]) - c[i+63:i] ENDFOR -dst[MAX:256] := 0 +dst[MAX:128] := 0 - + + +
immintrin.h
 - + Floating Point - AVX512VL - AVX512F - Miscellaneous - - - - Shuffle single-precision (32-bit) floating-point elements in "a" within 128-bit lanes using the control in "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + FMA + Arithmetic + + + + + Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", subtract packed elements in "c" from the intermediate result, and store the results in "dst". -SELECT4(src, control){ - CASE(control[1:0]) - 0: tmp[31:0] := src[31:0] - 1: tmp[31:0] := src[63:32] - 2: tmp[31:0] := src[95:64] - 3: tmp[31:0] := src[127:96] - ESAC - RETURN tmp[31:0] -} - -tmp_dst[31:0] := SELECT4(a[127:0], b[1:0]) -tmp_dst[63:32] := SELECT4(a[127:0], b[33:32]) -tmp_dst[95:64] := SELECT4(a[127:0], b[65:64]) -tmp_dst[127:96] := SELECT4(a[127:0], b[97:96]) -tmp_dst[159:128] := SELECT4(a[255:128], b[129:128]) -tmp_dst[191:160] := SELECT4(a[255:128], b[161:160]) -tmp_dst[223:192] := SELECT4(a[255:128], b[193:192]) -tmp_dst[255:224] := SELECT4(a[255:128], b[225:224]) -FOR j := 0 to 7 - i := j*32 - IF k[j] - dst[i+31:i] := tmp_dst[i+31:i] - ELSE - dst[i+31:i] := 0 - FI +FOR j := 0 to 3 + i := j*64 + dst[i+63:i] := (a[i+63:i] * b[i+63:i]) - c[i+63:i] ENDFOR dst[MAX:256] := 0 - + + +
immintrin.h
 - + Floating Point - AVX512VL - AVX512F - Miscellaneous - - - - - Shuffle single-precision (32-bit) floating-point elements in "a" using the control in "imm8", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + FMA + Arithmetic + + + + + Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", subtract packed elements in "c" from the intermediate result, and store the results in "dst". -SELECT4(src, control){ - CASE(control[1:0]) - 0: tmp[31:0] := src[31:0] - 1: tmp[31:0] := src[63:32] - 2: tmp[31:0] := src[95:64] - 3: tmp[31:0] := src[127:96] - ESAC - RETURN tmp[31:0] -} - -tmp_dst[31:0] := SELECT4(a[127:0], imm8[1:0]) -tmp_dst[63:32] := SELECT4(a[127:0], imm8[3:2]) -tmp_dst[95:64] := SELECT4(a[127:0], imm8[5:4]) -tmp_dst[127:96] := SELECT4(a[127:0], imm8[7:6]) FOR j := 0 to 3 i := j*32 - IF k[j] - dst[i+31:i] := tmp_dst[i+31:i] - ELSE - dst[i+31:i] := src[i+31:i] - FI + dst[i+31:i] := (a[i+31:i] * b[i+31:i]) - c[i+31:i] ENDFOR dst[MAX:128] := 0 - + + +
immintrin.h
 - + Floating Point - AVX512VL - AVX512F - Miscellaneous - - - - - Shuffle single-precision (32-bit) floating-point elements in "a" using the control in "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + FMA + Arithmetic + + + + + Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", subtract packed elements in "c" from the intermediate result, and store the results in "dst". -SELECT4(src, control){ - CASE(control[1:0]) - 0: tmp[31:0] := src[31:0] - 1: tmp[31:0] := src[63:32] - 2: tmp[31:0] := src[95:64] - 3: tmp[31:0] := src[127:96] - ESAC - RETURN tmp[31:0] -} - -tmp_dst[31:0] := SELECT4(a[127:0], b[1:0]) -tmp_dst[63:32] := SELECT4(a[127:0], b[33:32]) -tmp_dst[95:64] := SELECT4(a[127:0], b[65:64]) -tmp_dst[127:96] := SELECT4(a[127:0], b[97:96]) -FOR j := 0 to 3 +FOR j := 0 to 7 i := j*32 - IF k[j] - dst[i+31:i] := tmp_dst[i+31:i] - ELSE - dst[i+31:i] := src[i+31:i] - FI + dst[i+31:i] := (a[i+31:i] * b[i+31:i]) - c[i+31:i] ENDFOR -dst[MAX:128] := 0 +dst[MAX:256] := 0 - + + +
immintrin.h
 - + Floating Point - AVX512VL - AVX512F - Miscellaneous - - - - Shuffle single-precision (32-bit) floating-point elements in "a" using the control in "imm8", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + FMA + Arithmetic + + + + + Multiply the lower double-precision (64-bit) floating-point elements in "a" and "b", and subtract the lower element in "c" from the intermediate result. Store the result in the lower element of "dst", and copy the upper element from "a" to the upper element of "dst". -SELECT4(src, control){ - CASE(control[1:0]) - 0: tmp[31:0] := src[31:0] - 1: tmp[31:0] := src[63:32] - 2: tmp[31:0] := src[95:64] - 3: tmp[31:0] := src[127:96] - ESAC - RETURN tmp[31:0] -} - -tmp_dst[31:0] := SELECT4(a[127:0], imm8[1:0]) -tmp_dst[63:32] := SELECT4(a[127:0], imm8[3:2]) -tmp_dst[95:64] := SELECT4(a[127:0], imm8[5:4]) -tmp_dst[127:96] := SELECT4(a[127:0], imm8[7:6]) -FOR j := 0 to 3 - i := j*32 - IF k[j] - dst[i+31:i] := tmp_dst[i+31:i] - ELSE - dst[i+31:i] := 0 - FI -ENDFOR +dst[63:0] := (a[63:0] * b[63:0]) - c[63:0] +dst[127:64] := a[127:64] dst[MAX:128] := 0 - + + +
immintrin.h
 - + Floating Point - AVX512VL - AVX512F - Miscellaneous - - - - Shuffle single-precision (32-bit) floating-point elements in "a" using the control in "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + FMA + Arithmetic + + + + + Multiply the lower single-precision (32-bit) floating-point elements in "a" and "b", and subtract the lower element in "c" from the intermediate result. Store the result in the lower element of "dst", and copy the upper 3 packed elements from "a" to the upper elements of "dst". -SELECT4(src, control){ - CASE(control[1:0]) - 0: tmp[31:0] := src[31:0] - 1: tmp[31:0] := src[63:32] - 2: tmp[31:0] := src[95:64] - 3: tmp[31:0] := src[127:96] - ESAC - RETURN tmp[31:0] -} - -tmp_dst[31:0] := SELECT4(a[127:0], b[1:0]) -tmp_dst[63:32] := SELECT4(a[127:0], b[33:32]) -tmp_dst[95:64] := SELECT4(a[127:0], b[65:64]) -tmp_dst[127:96] := SELECT4(a[127:0], b[97:96]) -FOR j := 0 to 3 - i := j*32 - IF k[j] - dst[i+31:i] := tmp_dst[i+31:i] - ELSE - dst[i+31:i] := 0 - FI -ENDFOR +dst[31:0] := (a[31:0] * b[31:0]) - c[31:0] +dst[127:32] := a[127:32] dst[MAX:128] := 0 - + + +
immintrin.h
 - + Floating Point - AVX512VL - AVX512F - Miscellaneous - - - - - Shuffle double-precision (64-bit) floating-point elements in "a" across lanes using the control in "imm8", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + FMA + Arithmetic + + + + + Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", alternatively subtract and add packed elements in "c" from/to the intermediate result, and store the results in "dst". -SELECT4(src, control){ - CASE(control[1:0]) - 0: tmp[63:0] := src[63:0] - 1: tmp[63:0] := src[127:64] - 2: tmp[63:0] := src[191:128] - 3: tmp[63:0] := src[255:192] - ESAC - RETURN tmp[63:0] -} - -tmp_dst[63:0] := SELECT4(a[255:0], imm8[1:0]) -tmp_dst[127:64] := SELECT4(a[255:0], imm8[3:2]) -tmp_dst[191:128] := SELECT4(a[255:0], imm8[5:4]) -tmp_dst[255:192] := SELECT4(a[255:0], imm8[7:6]) -FOR j := 0 to 3 +FOR j := 0 to 1 i := j*64 - IF k[j] - dst[i+63:i] := tmp_dst[i+63:i] + IF ((j & 1) == 0) + dst[i+63:i] := (a[i+63:i] * b[i+63:i]) + c[i+63:i] ELSE - dst[i+63:i] := src[i+63:i] + dst[i+63:i] := (a[i+63:i] * b[i+63:i]) - c[i+63:i] FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:128] := 0 - + + +
immintrin.h
 - + Floating Point - AVX512VL - AVX512F - Miscellaneous - - - - - Shuffle double-precision (64-bit) floating-point elements in "a" across lanes using the corresponding index in "idx", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + FMA + Arithmetic + + + + + Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", alternatively subtract and add packed elements in "c" from/to the intermediate result, and store the results in "dst". FOR j := 0 to 3 i := j*64 - id := idx[i+1:i]*64 - IF k[j] - dst[i+63:i] := a[id+63:id] + IF ((j & 1) == 0) + dst[i+63:i] := (a[i+63:i] * b[i+63:i]) + c[i+63:i] ELSE - dst[i+63:i] := src[i+63:i] + dst[i+63:i] := (a[i+63:i] * b[i+63:i]) - c[i+63:i] FI ENDFOR dst[MAX:256] := 0 - + + +
immintrin.h
 - + Floating Point - AVX512VL - AVX512F - Miscellaneous - - - - Shuffle double-precision (64-bit) floating-point elements in "a" across lanes using the control in "imm8", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + FMA + Arithmetic + + + + + Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", alternatively subtract and add packed elements in "c" from/to the intermediate result, and store the results in "dst". -SELECT4(src, control){ - CASE(control[1:0]) - 0: tmp[63:0] := src[63:0] - 1: tmp[63:0] := src[127:64] - 2: tmp[63:0] := src[191:128] - 3: tmp[63:0] := src[255:192] - ESAC - RETURN tmp[63:0] -} - -tmp_dst[63:0] := SELECT4(a[255:0], imm8[1:0]) -tmp_dst[127:64] := SELECT4(a[255:0], imm8[3:2]) -tmp_dst[191:128] := SELECT4(a[255:0], imm8[5:4]) -tmp_dst[255:192] := SELECT4(a[255:0], imm8[7:6]) FOR j := 0 to 3 - i := j*64 - IF k[j] - dst[i+63:i] := tmp_dst[i+63:i] + i := j*32 + IF ((j & 1) == 0) + dst[i+31:i] := (a[i+31:i] * b[i+31:i]) + c[i+31:i] ELSE - dst[i+63:i] := 0 + dst[i+31:i] := (a[i+31:i] * b[i+31:i]) - c[i+31:i] FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:128] := 0 - + + +
immintrin.h
 - + Floating Point - AVX512VL - AVX512F - Miscellaneous - - - - Shuffle double-precision (64-bit) floating-point elements in "a" across lanes using the corresponding index in "idx", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + FMA + Arithmetic + + + + + Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", alternatively subtract and add packed elements in "c" from/to the intermediate result, and store the results in "dst". -FOR j := 0 to 3 - i := j*64 - id := idx[i+1:i]*64 - IF k[j] - dst[i+63:i] := a[id+63:id] +FOR j := 0 to 7 + i := j*32 + IF ((j & 1) == 0) + dst[i+31:i] := (a[i+31:i] * b[i+31:i]) + c[i+31:i] ELSE - dst[i+63:i] := 0 + dst[i+31:i] := (a[i+31:i] * b[i+31:i]) - c[i+31:i] FI ENDFOR dst[MAX:256] := 0 - + + +
immintrin.h
 - - Floating Point - AVX512VL - AVX512F - Miscellaneous - - - Shuffle double-precision (64-bit) floating-point elements in "a" across lanes using the control in "imm8", and store the results in "dst". - -SELECT4(src, control){ - CASE(control[1:0]) - 0: tmp[63:0] := src[63:0] - 1: tmp[63:0] := src[127:64] - 2: tmp[63:0] := src[191:128] - 3: tmp[63:0] := src[255:192] - ESAC - RETURN tmp[63:0] -} - -dst[63:0] := SELECT4(a[255:0], imm8[1:0]) -dst[127:64] := SELECT4(a[255:0], imm8[3:2]) -dst[191:128] := SELECT4(a[255:0], imm8[5:4]) -dst[255:192] := SELECT4(a[255:0], imm8[7:6]) -dst[MAX:256] := 0 + + Floating Point + FMA + Arithmetic + + + + + Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", add the negated intermediate result to packed elements in "c", and store the results in "dst". + +FOR j := 0 to 1 + i := j*64 + dst[i+63:i] := -(a[i+63:i] * b[i+63:i]) + c[i+63:i] +ENDFOR +dst[MAX:128] := 0 - + + +
immintrin.h
 - + Floating Point - AVX512VL - AVX512F - Miscellaneous - - - Shuffle double-precision (64-bit) floating-point elements in "a" across lanes using the corresponding index in "idx", and store the results in "dst". + FMA + Arithmetic + + + + + Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", add the negated intermediate result to packed elements in "c", and store the results in "dst". FOR j := 0 to 3 i := j*64 - id := idx[i+1:i]*64 - dst[i+63:i] := a[id+63:id] -ENDFOR + dst[i+63:i] := -(a[i+63:i] * b[i+63:i]) + c[i+63:i] +ENDFOR dst[MAX:256] := 0 - + + +
immintrin.h
 - + Floating Point - AVX512VL - AVX512F - Miscellaneous - - - - - Shuffle single-precision (32-bit) floating-point elements in "a" across lanes using the corresponding index in "idx", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + FMA + Arithmetic + + + + + Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", add the negated intermediate result to packed elements in "c", and store the results in "dst". -FOR j := 0 to 7 +FOR j := 0 to 3 i := j*32 - id := idx[i+2:i]*32 - IF k[j] - dst[i+31:i] := a[id+31:id] - ELSE - dst[i+31:i] := src[i+31:i] - FI -ENDFOR -dst[MAX:256] := 0 + dst[i+31:i] := -(a[i+31:i] * b[i+31:i]) + c[i+31:i] +ENDFOR +dst[MAX:128] := 0 - + + +
immintrin.h
 - + Floating Point - AVX512VL - AVX512F - Miscellaneous - - - - Shuffle single-precision (32-bit) floating-point elements in "a" across lanes using the corresponding index in "idx", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + FMA + Arithmetic + + + + + Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", add the negated intermediate result to packed elements in "c", and store the results in "dst". FOR j := 0 to 7 i := j*32 - id := idx[i+2:i]*32 - IF k[j] - dst[i+31:i] := a[id+31:id] - ELSE - dst[i+31:i] := 0 - FI -ENDFOR + dst[i+31:i] := -(a[i+31:i] * b[i+31:i]) + c[i+31:i] +ENDFOR dst[MAX:256] := 0 - + + +
immintrin.h
 - + Floating Point - AVX512VL - AVX512F - Miscellaneous - - - Shuffle single-precision (32-bit) floating-point elements in "a" across lanes using the corresponding index in "idx". + FMA + Arithmetic + + + + + Multiply the lower double-precision (64-bit) floating-point elements in "a" and "b", and add the negated intermediate result to the lower element in "c". Store the result in the lower element of "dst", and copy the upper element from "a" to the upper element of "dst". -FOR j := 0 to 7 - i := j*32 - id := idx[i+2:i]*32 - dst[i+31:i] := a[id+31:id] -ENDFOR -dst[MAX:256] := 0 +dst[63:0] := -(a[63:0] * b[63:0]) + c[63:0] +dst[127:64] := a[127:64] +dst[MAX:128] := 0 - + + +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Miscellaneous - - - - - Shuffle 64-bit integers in "a" across lanes lanes using the control in "imm8", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Floating Point + FMA + Arithmetic + + + + + Multiply the lower single-precision (32-bit) floating-point elements in "a" and "b", and add the negated intermediate result to the lower element in "c". Store the result in the lower element of "dst", and copy the upper 3 packed elements from "a" to the upper elements of "dst". -SELECT4(src, control){ - CASE(control[1:0]) - 0: tmp[63:0] := src[63:0] - 1: tmp[63:0] := src[127:64] - 2: tmp[63:0] := src[191:128] - 3: tmp[63:0] := src[255:192] - ESAC - RETURN tmp[63:0] -} - -tmp_dst[63:0] := SELECT4(a[255:0], imm8[1:0]) -tmp_dst[127:64] := SELECT4(a[255:0], imm8[3:2]) -tmp_dst[191:128] := SELECT4(a[255:0], imm8[5:4]) -tmp_dst[255:192] := SELECT4(a[255:0], imm8[7:6]) -FOR j := 0 to 3 - i := j*64 - IF k[j] - dst[i+63:i] := tmp_dst[i+63:i] - ELSE - dst[i+63:i] := src[i+63:i] - FI -ENDFOR -dst[MAX:256] := 0 +dst[31:0] := -(a[31:0] * b[31:0]) + c[31:0] +dst[127:32] := a[127:32] +dst[MAX:128] := 0 - + + +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Miscellaneous - - - - - Shuffle 64-bit integers in "a" across lanes using the corresponding index in "idx", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Floating Point + FMA + Arithmetic + + + + + Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", subtract packed elements in "c" from the negated intermediate result, and store the results in "dst". -FOR j := 0 to 3 +FOR j := 0 to 1 i := j*64 - id := idx[i+1:i]*64 - IF k[j] - dst[i+63:i] := a[id+63:id] - ELSE - dst[i+63:i] := src[i+63:i] - FI -ENDFOR -dst[MAX:256] := 0 + dst[i+63:i] := -(a[i+63:i] * b[i+63:i]) - c[i+63:i] +ENDFOR +dst[MAX:128] := 0 - + + +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Miscellaneous - - - - Shuffle 64-bit integers in "a" across lanes using the control in "imm8", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Floating Point + FMA + Arithmetic + + + + + Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", subtract packed elements in "c" from the negated intermediate result, and store the results in "dst". -SELECT4(src, control){ - CASE(control[1:0]) - 0: tmp[63:0] := src[63:0] - 1: tmp[63:0] := src[127:64] - 2: tmp[63:0] := src[191:128] - 3: tmp[63:0] := src[255:192] - ESAC - RETURN tmp[63:0] -} - -tmp_dst[63:0] := SELECT4(a[255:0], imm8[1:0]) -tmp_dst[127:64] := SELECT4(a[255:0], imm8[3:2]) -tmp_dst[191:128] := SELECT4(a[255:0], imm8[5:4]) -tmp_dst[255:192] := SELECT4(a[255:0], imm8[7:6]) FOR j := 0 to 3 i := j*64 - IF k[j] - dst[i+63:i] := tmp_dst[i+63:i] - ELSE - dst[i+63:i] := 0 - FI -ENDFOR + dst[i+63:i] := -(a[i+63:i] * b[i+63:i]) - c[i+63:i] +ENDFOR dst[MAX:256] := 0 - + + +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Miscellaneous - - - - Shuffle 64-bit integers in "a" across lanes using the corresponding index in "idx", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Floating Point + FMA + Arithmetic + + + + + Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", subtract packed elements in "c" from the negated intermediate result, and store the results in "dst". FOR j := 0 to 3 - i := j*64 - id := idx[i+1:i]*64 - IF k[j] - dst[i+63:i] := a[id+63:id] - ELSE - dst[i+63:i] := 0 - FI -ENDFOR -dst[MAX:256] := 0 + i := j*32 + dst[i+31:i] := -(a[i+31:i] * b[i+31:i]) - c[i+31:i] +ENDFOR +dst[MAX:128] := 0 - + + +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Miscellaneous - - - Shuffle 64-bit integers in "a" across lanes using the control in "imm8", and store the results in "dst". + + Floating Point + FMA + Arithmetic + + + + + Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", subtract packed elements in "c" from the negated intermediate result, and store the results in "dst". -SELECT4(src, control){ - CASE(control[1:0]) - 0: tmp[63:0] := src[63:0] - 1: tmp[63:0] := src[127:64] - 2: tmp[63:0] := src[191:128] - 3: tmp[63:0] := src[255:192] - ESAC - RETURN tmp[63:0] -} - -dst[63:0] := SELECT4(a[255:0], imm8[1:0]) -dst[127:64] := SELECT4(a[255:0], imm8[3:2]) -dst[191:128] := SELECT4(a[255:0], imm8[5:4]) -dst[255:192] := SELECT4(a[255:0], imm8[7:6]) +FOR j := 0 to 7 + i := j*32 + dst[i+31:i] := -(a[i+31:i] * b[i+31:i]) - c[i+31:i] +ENDFOR dst[MAX:256] := 0 - + + +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Miscellaneous - - - Shuffle 64-bit integers in "a" across lanes using the corresponding index in "idx", and store the results in "dst". + + Floating Point + FMA + Arithmetic + + + + + Multiply the lower double-precision (64-bit) floating-point elements in "a" and "b", and subtract the lower element in "c" from the negated intermediate result. Store the result in the lower element of "dst", and copy the upper element from "a" to the upper element of "dst". -FOR j := 0 to 3 - i := j*64 - id := idx[i+1:i]*64 - dst[i+63:i] := a[id+63:id] -ENDFOR -dst[MAX:256] := 0 +dst[63:0] := -(a[63:0] * b[63:0]) - c[63:0] +dst[127:64] := a[127:64] +dst[MAX:128] := 0 - + + +
immintrin.h
 - - Integer - AVX512VL - AVX512BW - Miscellaneous - - - - - Shuffle 16-bit integers in "a" across lanes using the corresponding index in "idx", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Floating Point + FMA + Arithmetic + + + + + Multiply the lower single-precision (32-bit) floating-point elements in "a" and "b", and subtract the lower element in "c" from the negated intermediate result. Store the result in the lower element of "dst", and copy the upper 3 packed elements from "a" to the upper elements of "dst". -FOR j := 0 to 15 - i := j*16 - id := idx[i+3:i]*16 - IF k[j] - dst[i+15:i] := a[id+15:id] - ELSE - dst[i+15:i] := src[i+15:i] - FI -ENDFOR -dst[MAX:256] := 0 +dst[31:0] := -(a[31:0] * b[31:0]) - c[31:0] +dst[127:32] := a[127:32] +dst[MAX:128] := 0 - + + +
immintrin.h
 - - Integer - AVX512VL - AVX512BW - Miscellaneous - - - - Shuffle 16-bit integers in "a" across lanes using the corresponding index in "idx", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Floating Point + FP16C + Convert + + + Convert packed half-precision (16-bit) floating-point elements in "a" to packed single-precision (32-bit) floating-point elements, and store the results in "dst". -FOR j := 0 to 15 - i := j*16 - id := idx[i+3:i]*16 - IF k[j] - dst[i+15:i] := a[id+15:id] - ELSE - dst[i+15:i] := 0 - FI +FOR j := 0 to 7 + i := j*32 + m := j*16 + dst[i+31:i] := Convert_FP16_To_FP32(a[m+15:m]) ENDFOR dst[MAX:256] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512BW - Miscellaneous - - - Shuffle 16-bit integers in "a" across lanes using the corresponding index in "idx", and store the results in "dst". + + Floating Point + FP16C + Convert + + + + Convert packed single-precision (32-bit) floating-point elements in "a" to packed half-precision (16-bit) floating-point elements, and store the results in "dst". + [sae_note] -FOR j := 0 to 15 - i := j*16 - id := idx[i+3:i]*16 - dst[i+15:i] := a[id+15:id] +FOR j := 0 to 7 + i := 16*j + l := 32*j + dst[i+15:i] := Convert_FP32_To_FP16(a[l+31:l]) ENDFOR -dst[MAX:256] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - - Integer - AVX512BW - Miscellaneous - - - - - Shuffle 16-bit integers in "a" across lanes using the corresponding index in "idx", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Floating Point + FP16C + Convert + + + Convert packed half-precision (16-bit) floating-point elements in "a" to packed single-precision (32-bit) floating-point elements, and store the results in "dst". -FOR j := 0 to 31 - i := j*16 - id := idx[i+4:i]*16 - IF k[j] - dst[i+15:i] := a[id+15:id] - ELSE - dst[i+15:i] := src[i+15:i] - FI +FOR j := 0 to 3 + i := j*32 + m := j*16 + dst[i+31:i] := Convert_FP16_To_FP32(a[m+15:m]) ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 - -
immintrin.h
+ +
emmintrin.h
 - - Integer - AVX512BW - Miscellaneous - - - - Shuffle 16-bit integers in "a" across lanes using the corresponding index in "idx", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Floating Point + FP16C + Convert + + + + Convert packed single-precision (32-bit) floating-point elements in "a" to packed half-precision (16-bit) floating-point elements, and store the results in "dst". + [sae_note] -FOR j := 0 to 31 - i := j*16 - id := idx[i+4:i]*16 - IF k[j] - dst[i+15:i] := a[id+15:id] - ELSE - dst[i+15:i] := 0 - FI +FOR j := 0 to 3 + i := 16*j + l := 32*j + dst[i+15:i] := Convert_FP32_To_FP16(a[l+31:l]) ENDFOR -dst[MAX:512] := 0 +dst[MAX:64] := 0 - -
immintrin.h
+ +
emmintrin.h
 - + Integer - AVX512BW - Miscellaneous - - - Shuffle 16-bit integers in "a" across lanes using the corresponding index in "idx", and store the results in "dst". - -FOR j := 0 to 31 - i := j*16 - id := idx[i+4:i]*16 - dst[i+15:i] := a[id+15:id] -ENDFOR -dst[MAX:512] := 0 + FSGSBASE + General Support + + Read the FS segment base register and store the 32-bit result in "dst". + dst[31:0] := FS_Segment_Base_Register +dst[63:32] := 0 - +
immintrin.h
 - + Integer - AVX512VL - AVX512BW - Miscellaneous - - - - - Shuffle 16-bit integers in "a" using the corresponding index in "idx", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -FOR j := 0 to 7 - i := j*16 - id := idx[i+2:i]*16 - IF k[j] - dst[i+15:i] := a[id+15:id] - ELSE - dst[i+15:i] := src[i+15:i] - FI -ENDFOR -dst[MAX:128] := 0 + FSGSBASE + General Support + + Read the FS segment base register and store the 64-bit result in "dst". + dst[63:0] := FS_Segment_Base_Register - +
immintrin.h
 - + Integer - AVX512VL - AVX512BW - Miscellaneous - - - - Shuffle 16-bit integers in "a" using the corresponding index in "idx", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - -FOR j := 0 to 7 - i := j*16 - id := idx[i+2:i]*16 - IF k[j] - dst[i+15:i] := a[id+15:id] - ELSE - dst[i+15:i] := 0 - FI -ENDFOR -dst[MAX:128] := 0 + FSGSBASE + General Support + + Read the GS segment base register and store the 32-bit result in "dst". + dst[31:0] := GS_Segment_Base_Register +dst[63:32] := 0 - +
immintrin.h
 - + Integer - AVX512VL - AVX512BW - Miscellaneous - - - Shuffle 16-bit integers in "a" using the corresponding index in "idx", and store the results in "dst". - -FOR j := 0 to 7 - i := j*16 - id := idx[i+2:i]*16 - dst[i+15:i] := a[id+15:id] -ENDFOR -dst[MAX:128] := 0 + FSGSBASE + General Support + + Read the GS segment base register and store the 64-bit result in "dst". + dst[63:0] := GS_Segment_Base_Register - +
immintrin.h
 - + Integer - AVX512VL - AVX512F - Miscellaneous - - - - Load contiguous active 32-bit integers from "a" (those with their respective bit set in mask "k"), and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + FSGSBASE + General Support + + + Write the unsigned 32-bit integer "a" to the FS segment base register. -m := 0 -FOR j := 0 to 7 - i := j*32 - IF k[j] - dst[i+31:i] := a[m+31:m] - m := m + 32 - ELSE - dst[i+31:i] := src[i+31:i] - FI -ENDFOR -dst[MAX:256] := 0 +FS_Segment_Base_Register[31:0] := a[31:0] +FS_Segment_Base_Register[63:32] := 0 - +
immintrin.h
 - + Integer - AVX512VL - AVX512F - Load - - - - Load contiguous active 32-bit integers from unaligned memory at "mem_addr" (those with their respective bit set in mask "k"), and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + FSGSBASE + General Support + + + Write the unsigned 64-bit integer "a" to the FS segment base register. -m := 0 -FOR j := 0 to 7 - i := j*32 - IF k[j] - dst[i+31:i] := MEM[mem_addr+m+31:mem_addr+m] - m := m + 32 - ELSE - dst[i+31:i] := src[i+31:i] - FI -ENDFOR -dst[MAX:256] := 0 +FS_Segment_Base_Register[63:0] := a[63:0] - +
immintrin.h
 - + Integer - AVX512VL - AVX512F - Miscellaneous - - - Load contiguous active 32-bit integers from "a" (those with their respective bit set in mask "k"), and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + FSGSBASE + General Support + + + Write the unsigned 32-bit integer "a" to the GS segment base register. -m := 0 -FOR j := 0 to 7 - i := j*32 - IF k[j] - dst[i+31:i] := a[m+31:m] - m := m + 32 - ELSE - dst[i+31:i] := 0 - FI -ENDFOR -dst[MAX:256] := 0 +GS_Segment_Base_Register[31:0] := a[31:0] +GS_Segment_Base_Register[63:32] := 0 - +
immintrin.h
 - + Integer - AVX512VL - AVX512F - Load - - - Load contiguous active 32-bit integers from unaligned memory at "mem_addr" (those with their respective bit set in mask "k"), and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + FSGSBASE + General Support + + + Write the unsigned 64-bit integer "a" to the GS segment base register. -m := 0 -FOR j := 0 to 7 - i := j*32 - IF k[j] - dst[i+31:i] := MEM[mem_addr+m+31:mem_addr+m] - m := m + 32 - ELSE - dst[i+31:i] := 0 - FI -ENDFOR -dst[MAX:256] := 0 +GS_Segment_Base_Register[63:0] := a[63:0] - +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Miscellaneous - - - - Load contiguous active 32-bit integers from "a" (those with their respective bit set in mask "k"), and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -m := 0 -FOR j := 0 to 3 - i := j*32 - IF k[j] - dst[i+31:i] := a[m+31:m] - m := m + 32 - ELSE - dst[i+31:i] := src[i+31:i] - FI -ENDFOR -dst[MAX:128] := 0 + + FXSR + OS-Targeted + + + Reload the x87 FPU, MMX technology, XMM, and MXCSR registers from the 512-byte memory image at "mem_addr". This data should have been written to memory previously using the FXSAVE instruction, and in the same format as required by the operating mode. "mem_addr" must be aligned on a 16-byte boundary. + state_x87_fpu_mmx_sse := fxrstor(MEM[mem_addr+512*8:mem_addr]) + + +
immintrin.h
+ + + FXSR + OS-Targeted + + + Reload the x87 FPU, MMX technology, XMM, and MXCSR registers from the 512-byte memory image at "mem_addr". This data should have been written to memory previously using the FXSAVE64 instruction, and in the same format as required by the operating mode. "mem_addr" must be aligned on a 16-byte boundary. + state_x87_fpu_mmx_sse := fxrstor64(MEM[mem_addr+512*8:mem_addr]) + + +
immintrin.h
+ + + FXSR + OS-Targeted + + + Save the current state of the x87 FPU, MMX technology, XMM, and MXCSR registers to a 512-byte memory location at "mem_addr". The layout of the 512-byte region depends on the operating mode. Bytes [511:464] are available for software use and will not be overwritten by the processor. + MEM[mem_addr+512*8:mem_addr] := fxsave(state_x87_fpu_mmx_sse) + + +
immintrin.h
+ + + FXSR + OS-Targeted + + + Save the current state of the x87 FPU, MMX technology, XMM, and MXCSR registers to a 512-byte memory location at "mem_addr". The layout of the 512-byte region depends on the operating mode. Bytes [511:464] are available for software use and will not be overwritten by the processor. + MEM[mem_addr+512*8:mem_addr] := fxsave64(state_x87_fpu_mmx_sse) - +
immintrin.h
 - + Integer - AVX512VL + GFNI AVX512F - Load - - - - Load contiguous active 32-bit integers from unaligned memory at "mem_addr" (those with their respective bit set in mask "k"), and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Arithmetic + + + + + Multiply the packed 8-bit integers in "a" and "b" in the finite field GF(2^8), and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). The field GF(2^8) is represented in polynomial representation with the reduction polynomial x^8 + x^4 + x^3 + x + 1. -m := 0 -FOR j := 0 to 3 - i := j*32 - IF k[j] - dst[i+31:i] := MEM[mem_addr+m+31:mem_addr+m] - m := m + 32 +DEFINE gf2p8mul_byte(src1byte, src2byte) { + tword := 0 + FOR i := 0 to 7 + IF src2byte.bit[i] + tword := tword XOR (src1byte << i) + FI + ENDFOR + FOR i := 14 downto 8 + p := 0x11B << (i-8) + IF tword.bit[i] + tword := tword XOR p + FI + ENDFOR + RETURN tword.byte[0] +} +FOR j := 0 TO 63 + IF k[j] + dst.byte[j] := gf2p8mul_byte(a.byte[j], b.byte[j]) ELSE - dst[i+31:i] := src[i+31:i] + dst.byte[j] := 0 FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX512VL + GFNI AVX512F - Miscellaneous - - - Load contiguous active 32-bit integers from "a" (those with their respective bit set in mask "k"), and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Arithmetic + + + + + + Multiply the packed 8-bit integers in "a" and "b" in the finite field GF(2^8), and store the results in "dst" using writemask "k" (elements are copied from "src"" when the corresponding mask bit is not set). The field GF(2^8) is represented in polynomial representation with the reduction polynomial x^8 + x^4 + x^3 + x + 1. -m := 0 -FOR j := 0 to 3 - i := j*32 +DEFINE gf2p8mul_byte(src1byte, src2byte) { + tword := 0 + FOR i := 0 to 7 + IF src2byte.bit[i] + tword := tword XOR (src1byte << i) + FI + ENDFOR + FOR i := 14 downto 8 + p := 0x11B << (i-8) + IF tword.bit[i] + tword := tword XOR p + FI + ENDFOR + RETURN tword.byte[0] +} +FOR j := 0 TO 63 IF k[j] - dst[i+31:i] := a[m+31:m] - m := m + 32 + dst.byte[j] := gf2p8mul_byte(a.byte[j], b.byte[j]) ELSE - dst[i+31:i] := 0 + dst.byte[j] := src.byte[j] FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX512VL + GFNI AVX512F - Load - - - Load contiguous active 32-bit integers from unaligned memory at "mem_addr" (those with their respective bit set in mask "k"), and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Arithmetic + + + + Multiply the packed 8-bit integers in "a" and "b" in the finite field GF(2^8), and store the results in "dst". The field GF(2^8) is represented in polynomial representation with the reduction polynomial x^8 + x^4 + x^3 + x + 1. -m := 0 -FOR j := 0 to 3 - i := j*32 - IF k[j] - dst[i+31:i] := MEM[mem_addr+m+31:mem_addr+m] - m := m + 32 - ELSE - dst[i+31:i] := 0 - FI +DEFINE gf2p8mul_byte(src1byte, src2byte) { + tword := 0 + FOR i := 0 to 7 + IF src2byte.bit[i] + tword := tword XOR (src1byte << i) + FI + ENDFOR + FOR i := 14 downto 8 + p := 0x11B << (i-8) + IF tword.bit[i] + tword := tword XOR p + FI + ENDFOR + RETURN tword.byte[0] +} +FOR j := 0 TO 63 + dst.byte[j] := gf2p8mul_byte(a.byte[j], b.byte[j]) ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer + GFNI AVX512VL - AVX512F - Miscellaneous - - - - Load contiguous active 64-bit integers from "a" (those with their respective bit set in mask "k"), and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Arithmetic + + + + + Multiply the packed 8-bit integers in "a" and "b" in the finite field GF(2^8), and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). The field GF(2^8) is represented in polynomial representation with the reduction polynomial x^8 + x^4 + x^3 + x + 1. -m := 0 -FOR j := 0 to 3 - i := j*64 +DEFINE gf2p8mul_byte(src1byte, src2byte) { + tword := 0 + FOR i := 0 to 7 + IF src2byte.bit[i] + tword := tword XOR (src1byte << i) + FI + ENDFOR + FOR i := 14 downto 8 + p := 0x11B << (i-8) + IF tword.bit[i] + tword := tword XOR p + FI + ENDFOR + RETURN tword.byte[0] +} +FOR j := 0 TO 31 IF k[j] - dst[i+63:i] := a[m+63:m] - m := m + 64 + dst.byte[j] := gf2p8mul_byte(a.byte[j], b.byte[j]) ELSE - dst[i+63:i] := src[i+63:i] + dst.byte[j] := 0 FI ENDFOR dst[MAX:256] := 0 - +
immintrin.h
 - + Integer + GFNI AVX512VL - AVX512F - Load - - - - Load contiguous active 64-bit integers from unaligned memory at "mem_addr" (those with their respective bit set in mask "k"), and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Arithmetic + + + + + + Multiply the packed 8-bit integers in "a" and "b" in the finite field GF(2^8), and store the results in "dst" using writemask "k" (elements are copied from "src"" when the corresponding mask bit is not set). The field GF(2^8) is represented in polynomial representation with the reduction polynomial x^8 + x^4 + x^3 + x + 1. -m := 0 -FOR j := 0 to 3 - i := j*64 +DEFINE gf2p8mul_byte(src1byte, src2byte) { + tword := 0 + FOR i := 0 to 7 + IF src2byte.bit[i] + tword := tword XOR (src1byte << i) + FI + ENDFOR + FOR i := 14 downto 8 + p := 0x11B << (i-8) + IF tword.bit[i] + tword := tword XOR p + FI + ENDFOR + RETURN tword.byte[0] +} +FOR j := 0 TO 31 IF k[j] - dst[i+63:i] := MEM[mem_addr+m+63:mem_addr+m] - m := m + 64 + dst.byte[j] := gf2p8mul_byte(a.byte[j], b.byte[j]) ELSE - dst[i+63:i] := src[i+63:i] + dst.byte[j] := src.byte[j] FI ENDFOR dst[MAX:256] := 0 - +
immintrin.h
 - + Integer + GFNI AVX512VL - AVX512F - Miscellaneous - - - Load contiguous active 64-bit integers from "a" (those with their respective bit set in mask "k"), and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Arithmetic + + + + Multiply the packed 8-bit integers in "a" and "b" in the finite field GF(2^8), and store the results in "dst". The field GF(2^8) is represented in polynomial representation with the reduction polynomial x^8 + x^4 + x^3 + x + 1. -m := 0 -FOR j := 0 to 3 - i := j*64 - IF k[j] - dst[i+63:i] := a[m+63:m] - m := m + 64 - ELSE - dst[i+63:i] := 0 - FI +DEFINE gf2p8mul_byte(src1byte, src2byte) { + tword := 0 + FOR i := 0 to 7 + IF src2byte.bit[i] + tword := tword XOR (src1byte << i) + FI + ENDFOR + FOR i := 14 downto 8 + p := 0x11B << (i-8) + IF tword.bit[i] + tword := tword XOR p + FI + ENDFOR + RETURN tword.byte[0] +} +FOR j := 0 TO 31 + dst.byte[j] := gf2p8mul_byte(a.byte[j], b.byte[j]) ENDFOR dst[MAX:256] := 0 - +
immintrin.h
 - + Integer + GFNI AVX512VL - AVX512F - Load - - - Load contiguous active 64-bit integers from unaligned memory at "mem_addr" (those with their respective bit set in mask "k"), and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Arithmetic + + + + + Multiply the packed 8-bit integers in "a" and "b" in the finite field GF(2^8), and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). The field GF(2^8) is represented in polynomial representation with the reduction polynomial x^8 + x^4 + x^3 + x + 1. -m := 0 -FOR j := 0 to 3 - i := j*64 +DEFINE gf2p8mul_byte(src1byte, src2byte) { + tword := 0 + FOR i := 0 to 7 + IF src2byte.bit[i] + tword := tword XOR (src1byte << i) + FI + ENDFOR + FOR i := 14 downto 8 + p := 0x11B << (i-8) + IF tword.bit[i] + tword := tword XOR p + FI + ENDFOR + RETURN tword.byte[0] +} +FOR j := 0 TO 15 IF k[j] - dst[i+63:i] := MEM[mem_addr+m+63:mem_addr+m] - m := m + 64 + dst.byte[j] := gf2p8mul_byte(a.byte[j], b.byte[j]) ELSE - dst[i+63:i] := 0 + dst.byte[j] := 0 FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer + GFNI AVX512VL - AVX512F - Miscellaneous - - - - Load contiguous active 64-bit integers from "a" (those with their respective bit set in mask "k"), and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Arithmetic + + + + + + Multiply the packed 8-bit integers in "a" and "b" in the finite field GF(2^8), and store the results in "dst" using writemask "k" (elements are copied from "src"" when the corresponding mask bit is not set). The field GF(2^8) is represented in polynomial representation with the reduction polynomial x^8 + x^4 + x^3 + x + 1. -m := 0 -FOR j := 0 to 1 - i := j*64 +DEFINE gf2p8mul_byte(src1byte, src2byte) { + tword := 0 + FOR i := 0 to 7 + IF src2byte.bit[i] + tword := tword XOR (src1byte << i) + FI + ENDFOR + FOR i := 14 downto 8 + p := 0x11B << (i-8) + IF tword.bit[i] + tword := tword XOR p + FI + ENDFOR + RETURN tword.byte[0] +} +FOR j := 0 TO 15 IF k[j] - dst[i+63:i] := a[m+63:m] - m := m + 64 + dst.byte[j] := gf2p8mul_byte(a.byte[j], b.byte[j]) ELSE - dst[i+63:i] := src[i+63:i] + dst.byte[j] := src.byte[j] FI ENDFOR dst[MAX:128] := 0 - +
immintrin.h
 - + Integer + GFNI AVX512VL - AVX512F - Load - - - - Load contiguous active 64-bit integers from unaligned memory at "mem_addr" (those with their respective bit set in mask "k"), and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Arithmetic + + + + Multiply the packed 8-bit integers in "a" and "b" in the finite field GF(2^8), and store the results in "dst". The field GF(2^8) is represented in polynomial representation with the reduction polynomial x^8 + x^4 + x^3 + x + 1. -m := 0 -FOR j := 0 to 1 - i := j*64 - IF k[j] - dst[i+63:i] := MEM[mem_addr+m+63:mem_addr+m] - m := m + 64 - ELSE - dst[i+63:i] := src[i+63:i] - FI +DEFINE gf2p8mul_byte(src1byte, src2byte) { + tword := 0 + FOR i := 0 to 7 + IF src2byte.bit[i] + tword := tword XOR (src1byte << i) + FI + ENDFOR + FOR i := 14 downto 8 + p := 0x11B << (i-8) + IF tword.bit[i] + tword := tword XOR p + FI + ENDFOR + RETURN tword.byte[0] +} +FOR j := 0 TO 15 + dst.byte[j] := gf2p8mul_byte(a.byte[j], b.byte[j]) ENDFOR dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - AVX512VL + GFNI AVX512F - Miscellaneous - - - Load contiguous active 64-bit integers from "a" (those with their respective bit set in mask "k"), and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Arithmetic + + + + + + Compute an affine transformation in the Galois Field 2^8. An affine transformation is defined by "A" * "x" + "b", where "A" represents an 8 by 8 bit matrix, "x" represents an 8-bit vector, and "b" is a constant immediate byte. Store the packed 8-bit results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -m := 0 -FOR j := 0 to 1 - i := j*64 - IF k[j] - dst[i+63:i] := a[m+63:m] - m := m + 64 - ELSE - dst[i+63:i] := 0 - FI +DEFINE parity(x) { + t := 0 + FOR i := 0 to 7 + t := t XOR x.bit[i] + ENDFOR + RETURN t +} +DEFINE affine_byte(tsrc2qw, src1byte, imm8) { + FOR i := 0 to 7 + retbyte.bit[i] := parity(tsrc2qw.byte[7-i] AND src1byte) XOR imm8.bit[i] + ENDFOR + RETURN retbyte +} +FOR j := 0 TO 7 + FOR i := 0 to 7 + IF k[j*8+i] + dst.qword[j].byte[i] := affine_byte(A.qword[j], x.qword[j].byte[i], b) + ELSE + dst.qword[j].byte[i] := 0 + FI + ENDFOR ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX512VL + GFNI AVX512F - Load - - - Load contiguous active 64-bit integers from unaligned memory at "mem_addr" (those with their respective bit set in mask "k"), and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Arithmetic + + + + + + + Compute an affine transformation in the Galois Field 2^8. An affine transformation is defined by "A" * "x" + "b", where "A" represents an 8 by 8 bit matrix, "x" represents an 8-bit vector, and "b" is a constant immediate byte. Store the packed 8-bit results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -m := 0 -FOR j := 0 to 1 - i := j*64 - IF k[j] - dst[i+63:i] := MEM[mem_addr+m+63:mem_addr+m] - m := m + 64 - ELSE - dst[i+63:i] := 0 - FI +DEFINE parity(x) { + t := 0 + FOR i := 0 to 7 + t := t XOR x.bit[i] + ENDFOR + RETURN t +} +DEFINE affine_byte(tsrc2qw, src1byte, imm8) { + FOR i := 0 to 7 + retbyte.bit[i] := parity(tsrc2qw.byte[7-i] AND src1byte) XOR imm8.bit[i] + ENDFOR + RETURN retbyte +} +FOR j := 0 TO 7 + FOR i := 0 to 7 + IF k[j*8+i] + dst.qword[j].byte[i] := affine_byte(A.qword[j], x.qword[j].byte[i], b) + ELSE + dst.qword[j].byte[i] := src.qword[j].byte[i] + FI + ENDFOR ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX512VL + GFNI AVX512F - Load - - - - - - - Gather 32-bit integers from memory using 32-bit indices. 32-bit elements are loaded from addresses starting at "base_addr" and offset by each 32-bit element in "vindex" (each index is scaled by the factor in "scale"). Gathered elements are merged into "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). "scale" should be 1, 2, 4 or 8. - + Arithmetic + + + + + Compute an affine transformation in the Galois Field 2^8. An affine transformation is defined by "A" * "x" + "b", where "A" represents an 8 by 8 bit matrix, "x" represents an 8-bit vector, and "b" is a constant immediate byte. Store the packed 8-bit results in "dst". -FOR j := 0 to 7 - i := j*32 - IF k[j] - dst[i+31:i] := MEM[base_addr + SignExtend(vindex[i+31:i])*scale] - k[j] := 0 - ELSE - dst[i+31:i] := src[i+31:i] - FI +DEFINE parity(x) { + t := 0 + FOR i := 0 to 7 + t := t XOR x.bit[i] + ENDFOR + RETURN t +} +DEFINE affine_byte(tsrc2qw, src1byte, imm8) { + FOR i := 0 to 7 + retbyte.bit[i] := parity(tsrc2qw.byte[7-i] AND src1byte) XOR imm8.bit[i] + ENDFOR + RETURN retbyte +} +FOR j := 0 TO 7 + FOR i := 0 to 7 + dst.qword[j].byte[i] := affine_byte(A.qword[j], x.qword[j].byte[i], b) + ENDFOR +ENDFOR +dst[MAX:512] := 0 + + +
immintrin.h
+ + + Integer + GFNI + AVX512VL + Arithmetic + + + + + + Compute an affine transformation in the Galois Field 2^8. An affine transformation is defined by "A" * "x" + "b", where "A" represents an 8 by 8 bit matrix, "x" represents an 8-bit vector, and "b" is a constant immediate byte. Store the packed 8-bit results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + +DEFINE parity(x) { + t := 0 + FOR i := 0 to 7 + t := t XOR x.bit[i] + ENDFOR + RETURN t +} +DEFINE affine_byte(tsrc2qw, src1byte, imm8) { + FOR i := 0 to 7 + retbyte.bit[i] := parity(tsrc2qw.byte[7-i] AND src1byte) XOR imm8.bit[i] + ENDFOR + RETURN retbyte +} +FOR j := 0 TO 3 + FOR i := 0 to 7 + IF k[j*8+i] + dst.qword[j].byte[i] := affine_byte(A.qword[j], x.qword[j].byte[i], b) + ELSE + dst.qword[j].byte[i] := 0 + FI + ENDFOR ENDFOR -k[MAX:8] := 0 dst[MAX:256] := 0 - +
immintrin.h
 - + Integer + GFNI AVX512VL - AVX512F - Load - - - - - - - Gather 32-bit integers from memory using 32-bit indices. 32-bit elements are loaded from addresses starting at "base_addr" and offset by each 32-bit element in "vindex" (each index is scaled by the factor in "scale"). Gathered elements are merged into "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). "scale" should be 1, 2, 4 or 8. - + Arithmetic + + + + + + + Compute an affine transformation in the Galois Field 2^8. An affine transformation is defined by "A" * "x" + "b", where "A" represents an 8 by 8 bit matrix, "x" represents an 8-bit vector, and "b" is a constant immediate byte. Store the packed 8-bit results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*32 - IF k[j] - dst[i+31:i] := MEM[base_addr + SignExtend(vindex[i+31:i])*scale] - k[j] := 0 - ELSE - dst[i+31:i] := src[i+31:i] - FI +DEFINE parity(x) { + t := 0 + FOR i := 0 to 7 + t := t XOR x.bit[i] + ENDFOR + RETURN t +} +DEFINE affine_byte(tsrc2qw, src1byte, imm8) { + FOR i := 0 to 7 + retbyte.bit[i] := parity(tsrc2qw.byte[7-i] AND src1byte) XOR imm8.bit[i] + ENDFOR + RETURN retbyte +} +FOR j := 0 TO 3 + FOR i := 0 to 7 + IF k[j*8+i] + dst.qword[j].byte[i] := affine_byte(A.qword[j], x.qword[j].byte[i], b) + ELSE + dst.qword[j].byte[i] := src.qword[j].byte[i] + FI + ENDFOR ENDFOR -k[MAX:4] := 0 -dst[MAX:128] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - + Integer + GFNI AVX512VL - AVX512F - Load - - - - - - Gather 64-bit integers from memory using 32-bit indices. 64-bit elements are loaded from addresses starting at "base_addr" and offset by each 32-bit element in "vindex" (each index is scaled by the factor in "scale"). Gathered elements are merged into "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). "scale" should be 1, 2, 4 or 8. - + Arithmetic + + + + + Compute an affine transformation in the Galois Field 2^8. An affine transformation is defined by "A" * "x" + "b", where "A" represents an 8 by 8 bit matrix, "x" represents an 8-bit vector, and "b" is a constant immediate byte. Store the packed 8-bit results in "dst". -FOR j := 0 to 3 - i := j*64 - m := j*32 - IF k[j] - dst[i+63:i] := MEM[base_addr + SignExtend(vindex[m+31:m])*scale] - k[j] := 0 - ELSE - dst[i+63:i] := src[i+63:i] - FI +DEFINE parity(x) { + t := 0 + FOR i := 0 to 7 + t := t XOR x.bit[i] + ENDFOR + RETURN t +} +DEFINE affine_byte(tsrc2qw, src1byte, imm8) { + FOR i := 0 to 7 + retbyte.bit[i] := parity(tsrc2qw.byte[7-i] AND src1byte) XOR imm8.bit[i] + ENDFOR + RETURN retbyte +} +FOR j := 0 TO 3 + FOR i := 0 to 7 + dst.qword[j].byte[i] := affine_byte(A.qword[j], x.qword[j].byte[i], b) + ENDFOR ENDFOR -k[MAX:4] := 0 dst[MAX:256] := 0 - +
immintrin.h
 - + Integer + GFNI AVX512VL - AVX512F - Load - - - - - - Gather 64-bit integers from memory using 32-bit indices. 64-bit elements are loaded from addresses starting at "base_addr" and offset by each 32-bit element in "vindex" (each index is scaled by the factor in "scale"). Gathered elements are merged into "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). "scale" should be 1, 2, 4 or 8. - + Arithmetic + + + + + + Compute an affine transformation in the Galois Field 2^8. An affine transformation is defined by "A" * "x" + "b", where "A" represents an 8 by 8 bit matrix, "x" represents an 8-bit vector, and "b" is a constant immediate byte. Store the packed 8-bit results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). -FOR j := 0 to 1 - i := j*64 - m := j*32 - IF k[j] - dst[i+63:i] := MEM[base_addr + SignExtend(vindex[m+31:m])*scale] - k[j] := 0 - ELSE - dst[i+63:i] := src[i+63:i] - FI +DEFINE parity(x) { + t := 0 + FOR i := 0 to 7 + t := t XOR x.bit[i] + ENDFOR + RETURN t +} +DEFINE affine_byte(tsrc2qw, src1byte, imm8) { + FOR i := 0 to 7 + retbyte.bit[i] := parity(tsrc2qw.byte[7-i] AND src1byte) XOR imm8.bit[i] + ENDFOR + RETURN retbyte +} +FOR j := 0 TO 1 + FOR i := 0 to 7 + IF k[j*8+i] + dst.qword[j].byte[i] := affine_byte(A.qword[j], x.qword[j].byte[i], b) + ELSE + dst.qword[j].byte[i] := 0 + FI + ENDFOR ENDFOR -k[MAX:2] := 0 dst[MAX:128] := 0 - +
immintrin.h
 - + Integer + GFNI AVX512VL - AVX512F - Load - - - - - - Gather 32-bit integers from memory using 64-bit indices. 32-bit elements are loaded from addresses starting at "base_addr" and offset by each 64-bit element in "vindex" (each index is scaled by the factor in "scale"). Gathered elements are merged into "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). "scale" should be 1, 2, 4 or 8. - + Arithmetic + + + + + + + Compute an affine transformation in the Galois Field 2^8. An affine transformation is defined by "A" * "x" + "b", where "A" represents an 8 by 8 bit matrix, "x" represents an 8-bit vector, and "b" is a constant immediate byte. Store the packed 8-bit results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := j*32 - m := j*64 - IF k[j] - dst[i+31:i] := MEM[base_addr + SignExtend(vindex[i+63:i])*scale] - k[j] := 0 - ELSE - dst[i+31:i] := src[i+31:i] - FI +DEFINE parity(x) { + t := 0 + FOR i := 0 to 7 + t := t XOR x.bit[i] + ENDFOR + RETURN t +} +DEFINE affine_byte(tsrc2qw, src1byte, imm8) { + FOR i := 0 to 7 + retbyte.bit[i] := parity(tsrc2qw.byte[7-i] AND src1byte) XOR imm8.bit[i] + ENDFOR + RETURN retbyte +} +FOR j := 0 TO 1 + FOR i := 0 to 7 + IF k[j*8+i] + dst.qword[j].byte[i] := affine_byte(A.qword[j], x.qword[j].byte[i], b) + ELSE + dst.qword[j].byte[i] := src.qword[j].byte[i] + FI + ENDFOR ENDFOR -k[MAX:4] := 0 dst[MAX:128] := 0 - +
immintrin.h
 - + Integer + GFNI AVX512VL - AVX512F - Load - - - - - - Gather 32-bit integers from memory using 64-bit indices. 32-bit elements are loaded from addresses starting at "base_addr" and offset by each 64-bit element in "vindex" (each index is scaled by the factor in "scale"). Gathered elements are merged into "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). "scale" should be 1, 2, 4 or 8. - + Arithmetic + + + + + Compute an affine transformation in the Galois Field 2^8. An affine transformation is defined by "A" * "x" + "b", where "A" represents an 8 by 8 bit matrix, "x" represents an 8-bit vector, and "b" is a constant immediate byte. Store the packed 8-bit results in "dst". -FOR j := 0 to 1 - i := j*32 - m := j*64 - IF k[j] - dst[i+31:i] := MEM[base_addr + SignExtend(vindex[i+63:i])*scale] - k[j] := 0 - ELSE - dst[i+31:i] := src[i+31:i] - FI +DEFINE parity(x) { + t := 0 + FOR i := 0 to 7 + t := t XOR x.bit[i] + ENDFOR + RETURN t +} +DEFINE affine_byte(tsrc2qw, src1byte, imm8) { + FOR i := 0 to 7 + retbyte.bit[i] := parity(tsrc2qw.byte[7-i] AND src1byte) XOR imm8.bit[i] + ENDFOR + RETURN retbyte +} +FOR j := 0 TO 1 + FOR i := 0 to 7 + dst.qword[j].byte[i] := affine_byte(A.qword[j], x.qword[j].byte[i], b) + ENDFOR ENDFOR -k[MAX:2] := 0 -dst[MAX:64] := 0 +dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - AVX512VL + GFNI AVX512F - Load - - - - - - Gather 64-bit integers from memory using 64-bit indices. 64-bit elements are loaded from addresses starting at "base_addr" and offset by each 64-bit element in "vindex" (each index is scaled by the factor in "scale"). Gathered elements are merged into "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). "scale" should be 1, 2, 4 or 8. - - -FOR j := 0 to 3 - i := j*64 - IF k[j] - dst[i+63:i] := MEM[base_addr + SignExtend(vindex[i+63:i])*scale] - k[j] := 0 - ELSE - dst[i+63:i] := src[i+63:i] - FI + Arithmetic + + + + + + Compute an inverse affine transformation in the Galois Field 2^8. An affine transformation is defined by "A" * "x" + "b", where "A" represents an 8 by 8 bit matrix, "x" represents an 8-bit vector, and "b" is a constant immediate byte. The inverse of the 8-bit values in "x" is defined with respect to the reduction polynomial x^8 + x^4 + x^3 + x + 1. Store the packed 8-bit results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + DEFINE parity(x) { + t := 0 + FOR i := 0 to 7 + t := t XOR x.bit[i] + ENDFOR + RETURN t +} +DEFINE affine_inverse_byte(tsrc2qw, src1byte, imm8) { + FOR i := 0 to 7 + retbyte.bit[i] := parity(tsrc2qw.byte[7-i] AND inverse(src1byte)) XOR imm8.bit[i] + ENDFOR + RETURN retbyte +} +FOR j := 0 TO 7 + FOR i := 0 to 7 + IF k[j*8+i] + dst.qword[j].byte[i] := affine_inverse_byte(A.qword[j], x.qword[j].byte[i], b) + ELSE + dst.qword[j].byte[i] := 0 + FI + ENDFOR ENDFOR -k[MAX:4] := 0 -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX512VL + GFNI AVX512F - Load - - - - - - Gather 64-bit integers from memory using 64-bit indices. 64-bit elements are loaded from addresses starting at "base_addr" and offset by each 64-bit element in "vindex" (each index is scaled by the factor in "scale"). Gathered elements are merged into "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). "scale" should be 1, 2, 4 or 8. - - -FOR j := 0 to 1 - i := j*64 - IF k[j] - dst[i+63:i] := MEM[base_addr + SignExtend(vindex[i+63:i])*scale] - k[j] := 0 - ELSE - dst[i+63:i] := src[i+63:i] - FI + Arithmetic + + + + + + + Compute an inverse affine transformation in the Galois Field 2^8. An affine transformation is defined by "A" * "x" + "b", where "A" represents an 8 by 8 bit matrix, "x" represents an 8-bit vector, and "b" is a constant immediate byte. The inverse of the 8-bit values in "x" is defined with respect to the reduction polynomial x^8 + x^4 + x^3 + x + 1. Store the packed 8-bit results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + DEFINE parity(x) { + t := 0 + FOR i := 0 to 7 + t := t XOR x.bit[i] + ENDFOR + RETURN t +} +DEFINE affine_inverse_byte(tsrc2qw, src1byte, imm8) { + FOR i := 0 to 7 + retbyte.bit[i] := parity(tsrc2qw.byte[7-i] AND inverse(src1byte)) XOR imm8.bit[i] + ENDFOR + RETURN retbyte +} +FOR j := 0 TO 7 + FOR i := 0 to 7 + IF k[j*8+i] + dst.qword[j].byte[i] := affine_inverse_byte(A.qword[j], x.qword[j].byte[i], b) + ELSE + dst.qword[j].byte[i] := src.qword[j].byte[b] + FI + ENDFOR ENDFOR -k[MAX:2] := 0 -dst[MAX:128] := 0 +dst[MAX:512] := 0 + + +
immintrin.h
+ + + Integer + GFNI + AVX512F + Arithmetic + + + + + Compute an inverse affine transformation in the Galois Field 2^8. An affine transformation is defined by "A" * "x" + "b", where "A" represents an 8 by 8 bit matrix, "x" represents an 8-bit vector, and "b" is a constant immediate byte. The inverse of the 8-bit values in "x" is defined with respect to the reduction polynomial x^8 + x^4 + x^3 + x + 1. Store the packed 8-bit results in "dst". + DEFINE parity(x) { + t := 0 + FOR i := 0 to 7 + t := t XOR x.bit[i] + ENDFOR + RETURN t +} +DEFINE affine_inverse_byte(tsrc2qw, src1byte, imm8) { + FOR i := 0 to 7 + retbyte.bit[i] := parity(tsrc2qw.byte[7-i] AND inverse(src1byte)) XOR imm8.bit[i] + ENDFOR + RETURN retbyte +} +FOR j := 0 TO 7 + FOR i := 0 to 7 + dst.qword[j].byte[i] := affine_inverse_byte(A.qword[j], x.qword[j].byte[i], b) + ENDFOR +ENDFOR +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer + GFNI AVX512VL - AVX512CD - Bit Manipulation - - Counts the number of leading zero bits in each packed 32-bit integer in "a", and store the results in "dst". - -FOR j := 0 to 7 - i := j*32 - tmp := 31 - dst[i+31:i] := 0 - DO WHILE (tmp >= 0 AND a[i+tmp] == 0) - tmp := tmp - 1 - dst[i+31:i] := dst[i+31:i] + 1 - OD + Arithmetic + + + + + + Compute an inverse affine transformation in the Galois Field 2^8. An affine transformation is defined by "A" * "x" + "b", where "A" represents an 8 by 8 bit matrix, "x" represents an 8-bit vector, and "b" is a constant immediate byte. The inverse of the 8-bit values in "x" is defined with respect to the reduction polynomial x^8 + x^4 + x^3 + x + 1. Store the packed 8-bit results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + DEFINE parity(x) { + t := 0 + FOR i := 0 to 7 + t := t XOR x.bit[i] + ENDFOR + RETURN t +} +DEFINE affine_inverse_byte(tsrc2qw, src1byte, imm8) { + FOR i := 0 to 7 + retbyte.bit[i] := parity(tsrc2qw.byte[7-i] AND inverse(src1byte)) XOR imm8.bit[i] + ENDFOR + RETURN retbyte +} +FOR j := 0 TO 3 + FOR i := 0 to 7 + IF k[j*8+i] + dst.qword[j].byte[i] := affine_inverse_byte(A.qword[j], x.qword[j].byte[i], b) + ELSE + dst.qword[j].byte[i] := 0 + FI + ENDFOR ENDFOR dst[MAX:256] := 0 - +
immintrin.h
 - + Integer + GFNI AVX512VL - AVX512CD - Bit Manipulation - - - - Counts the number of leading zero bits in each packed 32-bit integer in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -FOR j := 0 to 7 - i := j*32 - IF k[j] - tmp := 31 - dst[i+31:i] := 0 - DO WHILE (tmp >= 0 AND a[i+tmp] == 0) - tmp := tmp - 1 - dst[i+31:i] := dst[i+31:i] + 1 - OD - ELSE - dst[i+31:i] := src[i+31:i] - FI + Arithmetic + + + + + + + Compute an inverse affine transformation in the Galois Field 2^8. An affine transformation is defined by "A" * "x" + "b", where "A" represents an 8 by 8 bit matrix, "x" represents an 8-bit vector, and "b" is a constant immediate byte. The inverse of the 8-bit values in "x" is defined with respect to the reduction polynomial x^8 + x^4 + x^3 + x + 1. Store the packed 8-bit results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + DEFINE parity(x) { + t := 0 + FOR i := 0 to 7 + t := t XOR x.bit[i] + ENDFOR + RETURN t +} +DEFINE affine_inverse_byte(tsrc2qw, src1byte, imm8) { + FOR i := 0 to 7 + retbyte.bit[i] := parity(tsrc2qw.byte[7-i] AND inverse(src1byte)) XOR imm8.bit[i] + ENDFOR + RETURN retbyte +} +FOR j := 0 TO 3 + FOR i := 0 to 7 + IF k[j*8+i] + dst.qword[j].byte[i] := affine_inverse_byte(A.qword[j], x.qword[j].byte[i], b) + ELSE + dst.qword[j].byte[i] := src.qword[j].byte[i] + FI + ENDFOR ENDFOR dst[MAX:256] := 0 - +
immintrin.h
 - + Integer + GFNI AVX512VL - AVX512CD - Bit Manipulation - - - Counts the number of leading zero bits in each packed 32-bit integer in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - -FOR j := 0 to 7 - i := j*32 - IF k[j] - tmp := 31 - dst[i+31:i] := 0 - DO WHILE (tmp >= 0 AND a[i+tmp] == 0) - tmp := tmp - 1 - dst[i+31:i] := dst[i+31:i] + 1 - OD - ELSE - dst[i+31:i] := 0 - FI + Arithmetic + + + + + Compute an inverse affine transformation in the Galois Field 2^8. An affine transformation is defined by "A" * "x" + "b", where "A" represents an 8 by 8 bit matrix, "x" represents an 8-bit vector, and "b" is a constant immediate byte. The inverse of the 8-bit values in "x" is defined with respect to the reduction polynomial x^8 + x^4 + x^3 + x + 1. Store the packed 8-bit results in "dst". + DEFINE parity(x) { + t := 0 + FOR i := 0 to 7 + t := t XOR x.bit[i] + ENDFOR + RETURN t +} +DEFINE affine_inverse_byte(tsrc2qw, src1byte, imm8) { + FOR i := 0 to 7 + retbyte.bit[i] := parity(tsrc2qw.byte[7-i] AND inverse(src1byte)) XOR imm8.bit[i] + ENDFOR + RETURN retbyte +} +FOR j := 0 TO 3 + FOR i := 0 to 7 + dst.qword[j].byte[i] := affine_inverse_byte(A.qword[j], x.qword[j].byte[i], b) + ENDFOR ENDFOR dst[MAX:256] := 0 - +
immintrin.h
 - + Integer + GFNI AVX512VL - AVX512CD - Bit Manipulation - - Counts the number of leading zero bits in each packed 32-bit integer in "a", and store the results in "dst". - -FOR j := 0 to 3 - i := j*32 - tmp := 31 - dst[i+31:i] := 0 - DO WHILE (tmp >= 0 AND a[i+tmp] == 0) - tmp := tmp - 1 - dst[i+31:i] := dst[i+31:i] + 1 - OD + Arithmetic + + + + + + Compute an inverse affine transformation in the Galois Field 2^8. An affine transformation is defined by "A" * "x" + "b", where "A" represents an 8 by 8 bit matrix, "x" represents an 8-bit vector, and "b" is a constant immediate byte. The inverse of the 8-bit values in "x" is defined with respect to the reduction polynomial x^8 + x^4 + x^3 + x + 1. Store the packed 8-bit results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + DEFINE parity(x) { + t := 0 + FOR i := 0 to 7 + t := t XOR x.bit[i] + ENDFOR + RETURN t +} +DEFINE affine_inverse_byte(tsrc2qw, src1byte, imm8) { + FOR i := 0 to 7 + retbyte.bit[i] := parity(tsrc2qw.byte[7-i] AND inverse(src1byte)) XOR imm8.bit[i] + ENDFOR + RETURN retbyte +} +FOR j := 0 TO 1 + FOR i := 0 to 7 + IF k[j*8+i] + dst.qword[j].byte[i] := affine_inverse_byte(A.qword[j], x.qword[j].byte[i], b) + ELSE + dst.qword[j].byte[i] := 0 + FI + ENDFOR ENDFOR dst[MAX:128] := 0 - +
immintrin.h
 - + Integer + GFNI AVX512VL - AVX512CD - Bit Manipulation - - - - Counts the number of leading zero bits in each packed 32-bit integer in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -FOR j := 0 to 3 - i := j*32 - IF k[j] - tmp := 31 - dst[i+31:i] := 0 - DO WHILE (tmp >= 0 AND a[i+tmp] == 0) - tmp := tmp - 1 - dst[i+31:i] := dst[i+31:i] + 1 - OD - ELSE - dst[i+31:i] := src[i+31:i] - FI + Arithmetic + + + + + + + Compute an inverse affine transformation in the Galois Field 2^8. An affine transformation is defined by "A" * "x" + "b", where "A" represents an 8 by 8 bit matrix, "x" represents an 8-bit vector, and "b" is a constant immediate byte. The inverse of the 8-bit values in "x" is defined with respect to the reduction polynomial x^8 + x^4 + x^3 + x + 1. Store the packed 8-bit results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + DEFINE parity(x) { + t := 0 + FOR i := 0 to 7 + t := t XOR x.bit[i] + ENDFOR + RETURN t +} +DEFINE affine_inverse_byte(tsrc2qw, src1byte, imm8) { + FOR i := 0 to 7 + retbyte.bit[i] := parity(tsrc2qw.byte[7-i] AND inverse(src1byte)) XOR imm8.bit[i] + ENDFOR + RETURN retbyte +} +FOR j := 0 TO 1 + FOR i := 0 to 7 + IF k[j*8+i] + dst.qword[j].byte[i] := affine_inverse_byte(A.qword[j], x.qword[j].byte[i], b) + ELSE + dst.qword[j].byte[i] := src.qword[j].byte[i] + FI + ENDFOR +ENDFOR +dst[MAX:128] := 0 + + +
immintrin.h
+ + + Integer + GFNI + AVX512VL + Arithmetic + + + + + Compute an inverse affine transformation in the Galois Field 2^8. An affine transformation is defined by "A" * "x" + "b", where "A" represents an 8 by 8 bit matrix, "x" represents an 8-bit vector, and "b" is a constant immediate byte. The inverse of the 8-bit values in "x" is defined with respect to the reduction polynomial x^8 + x^4 + x^3 + x + 1. Store the packed 8-bit results in "dst". + DEFINE parity(x) { + t := 0 + FOR i := 0 to 7 + t := t XOR x.bit[i] + ENDFOR + RETURN t +} +DEFINE affine_inverse_byte(tsrc2qw, src1byte, imm8) { + FOR i := 0 to 7 + retbyte.bit[i] := parity(tsrc2qw.byte[7-i] AND inverse(src1byte)) XOR imm8.bit[i] + ENDFOR + RETURN retbyte +} +FOR j := 0 TO 1 + FOR i := 0 to 7 + dst.qword[j].byte[i] := affine_inverse_byte(A.qword[j], x.qword[j].byte[i], b) + ENDFOR ENDFOR dst[MAX:128] := 0 - + +
immintrin.h
+ + + INVPCID + OS-Targeted + + + + Invalidate mappings in the Translation Lookaside Buffers (TLBs) and paging-structure caches for the processor context identifier (PCID) specified by "descriptor" based on the invalidation type specified in "type". + The PCID "descriptor" is specified as a 16-byte memory operand (with no alignment restrictions) where bits [11:0] specify the PCID, and bits [127:64] specify the linear address; bits [63:12] are reserved. + The types supported are: + 0) Individual-address invalidation: If "type" is 0, the logical processor invalidates mappings for a single linear address and tagged with the PCID specified in "descriptor", except global translations. The instruction may also invalidate global translations, mappings for other linear addresses, or mappings tagged with other PCIDs. + 1) Single-context invalidation: If "type" is 1, the logical processor invalidates all mappings tagged with the PCID specified in "descriptor" except global translations. In some cases, it may invalidate mappings for other PCIDs as well. + 2) All-context invalidation: If "type" is 2, the logical processor invalidates all mappings tagged with any PCID. + 3) All-context invalidation, retaining global translations: If "type" is 3, the logical processor invalidates all mappings tagged with any PCID except global translations, ignoring "descriptor". The instruction may also invalidate global translations as well. + +CASE type[1:0] OF +0: // individual-address invalidation retaining global translations + OP_PCID := MEM[descriptor+11:descriptor] + ADDR := MEM[descriptor+127:descriptor+64] + BREAK +1: // single PCID invalidation retaining globals + OP_PCID := MEM[descriptor+11:descriptor] + // invalidate all mappings tagged with OP_PCID except global translations + BREAK +2: // all PCID invalidation + // invalidate all mappings tagged with any PCID + BREAK +3: // all PCID invalidation retaining global translations + // invalidate all mappings tagged with any PCID except global translations + BREAK +ESAC + +
immintrin.h
 - - Integer - AVX512VL - AVX512CD - Bit Manipulation - - - Counts the number of leading zero bits in each packed 32-bit integer in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + KNCNI + General Support + + + + Fetch the line of data from memory that contains address "p" to a location in the cache heirarchy specified by the locality hint "i". + + + + + + + + +
xmmintrin.h
+ + + Mask + KNCNI + Mask + + + + Compute the bitwise NOT of 16-bit masks "a" and then AND with "b", and store the result in "k". -FOR j := 0 to 3 - i := j*32 - IF k[j] - tmp := 31 - dst[i+31:i] := 0 - DO WHILE (tmp >= 0 AND a[i+tmp] == 0) - tmp := tmp - 1 - dst[i+31:i] := dst[i+31:i] + 1 - OD - ELSE - dst[i+31:i] := 0 - FI -ENDFOR -dst[MAX:128] := 0 +k[15:0] := (NOT a[15:0]) AND b[15:0] +k[MAX:16] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512CD - Bit Manipulation - - Counts the number of leading zero bits in each packed 64-bit integer in "a", and store the results in "dst". + + Mask + KNCNI + Mask + + + + Compute the bitwise AND of 16-bit masks "a" and "b", and store the result in "k". -FOR j := 0 to 3 - i := j*64 - tmp := 63 - dst[i+63:i] := 0 - DO WHILE (tmp >= 0 AND a[i+tmp] == 0) - tmp := tmp - 1 - dst[i+63:i] := dst[i+63:i] + 1 - OD -ENDFOR -dst[MAX:256] := 0 +k[15:0] := a[15:0] AND b[15:0] +k[MAX:16] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512CD - Bit Manipulation - - - - Counts the number of leading zero bits in each packed 64-bit integer in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Mask + KNCNI + Mask + + + Copy 16-bit mask "a" to "k". -FOR j := 0 to 3 - i := j*64 - IF k[j] - tmp := 63 - dst[i+63:i] := 0 - DO WHILE (tmp >= 0 AND a[i+tmp] == 0) - tmp := tmp - 1 - dst[i+63:i] := dst[i+63:i] + 1 - OD - ELSE - dst[i+63:i] := src[i+63:i] - FI -ENDFOR -dst[MAX:256] := 0 +k[15:0] := a[15:0] +k[MAX:16] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512CD - Bit Manipulation - - - Counts the number of leading zero bits in each packed 64-bit integer in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Mask + KNCNI + Mask + + + Compute the bitwise NOT of 16-bit mask "a", and store the result in "k". -FOR j := 0 to 3 - i := j*64 - IF k[j] - tmp := 63 - dst[i+63:i] := 0 - DO WHILE (tmp >= 0 AND a[i+tmp] == 0) - tmp := tmp - 1 - dst[i+63:i] := dst[i+63:i] + 1 - OD - ELSE - dst[i+63:i] := 0 - FI -ENDFOR -dst[MAX:256] := 0 +k[15:0] := NOT a[15:0] +k[MAX:16] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512CD - Bit Manipulation - - Counts the number of leading zero bits in each packed 64-bit integer in "a", and store the results in "dst". + + Mask + KNCNI + Mask + + + + Compute the bitwise OR of 16-bit masks "a" and "b", and store the result in "k". -FOR j := 0 to 1 - i := j*64 - tmp := 63 - dst[i+63:i] := 0 - DO WHILE (tmp >= 0 AND a[i+tmp] == 0) - tmp := tmp - 1 - dst[i+63:i] := dst[i+63:i] + 1 - OD -ENDFOR -dst[MAX:128] := 0 +k[15:0] := a[15:0] OR b[15:0] +k[MAX:16] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512CD - Bit Manipulation - - - - Counts the number of leading zero bits in each packed 64-bit integer in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Mask + KNCNI + Mask + + + + Compute the bitwise XNOR of 16-bit masks "a" and "b", and store the result in "k". -FOR j := 0 to 1 - i := j*64 - IF k[j] - tmp := 63 - dst[i+63:i] := 0 - DO WHILE (tmp >= 0 AND a[i+tmp] == 0) - tmp := tmp - 1 - dst[i+63:i] := dst[i+63:i] + 1 - OD - ELSE - dst[i+63:i] := src[i+63:i] - FI -ENDFOR -dst[MAX:128] := 0 +k[15:0] := NOT (a[15:0] XOR b[15:0]) +k[MAX:16] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512CD - Bit Manipulation - - - Counts the number of leading zero bits in each packed 64-bit integer in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Mask + KNCNI + Mask + + + + Compute the bitwise XOR of 16-bit masks "a" and "b", and store the result in "k". -FOR j := 0 to 1 - i := j*64 - IF k[j] - tmp := 63 - dst[i+63:i] := 0 - DO WHILE (tmp >= 0 AND a[i+tmp] == 0) - tmp := tmp - 1 - dst[i+63:i] := dst[i+63:i] + 1 - OD - ELSE - dst[i+63:i] := 0 - FI -ENDFOR -dst[MAX:128] := 0 +k[15:0] := a[15:0] XOR b[15:0] +k[MAX:16] := 0 - +
immintrin.h
 - + Integer - AVX512VL - AVX512BW - Arithmetic - - - - - Multiply packed unsigned 8-bit integers in "a" by packed signed 8-bit integers in "b", producing intermediate signed 16-bit integers. Horizontally add adjacent pairs of intermediate signed 16-bit integers, and pack the saturated results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - + Mask + KNCNI + Compare + + + + Compare packed signed 32-bit integers in "a" and "b" for less-than, and store the results in mask vector "k". FOR j := 0 to 15 - i := j*16 - IF k[j] - dst[i+15:i] := Saturate_To_Int16( a[i+15:i+8]*b[i+15:i+8] + a[i+7:i]*b[i+7:i] ) - ELSE - dst[i+15:i] := src[i+15:i] - FI + i := j*32 + k[j] := ( a[i+31:i] < b[i+31:i] ) ? 1 : 0 ENDFOR -dst[MAX:256] := 0 +k[MAX:16] := 0 - +
immintrin.h
 - + Integer - AVX512VL - AVX512BW - Arithmetic - - - - Multiply packed unsigned 8-bit integers in "a" by packed signed 8-bit integers in "b", producing intermediate signed 16-bit integers. Horizontally add adjacent pairs of intermediate signed 16-bit integers, and pack the saturated results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - + Mask + KNCNI + Compare + + + + + Compare packed signed 32-bit integers in "a" and "b" for less-than-or-equal, and store the results in mask vector "k" using zeromask "k1" (elements are zeroed out when the corresponding mask bit is not set). FOR j := 0 to 15 - i := j*16 - IF k[j] - dst[i+15:i] := Saturate_To_Int16( a[i+15:i+8]*b[i+15:i+8] + a[i+7:i]*b[i+7:i] ) - ELSE - dst[i+15:i] := 0 + i := j*32 + IF k1[j] + k[j] := ( a[i+31:i] < b[i+31:i] ) ? 1 : 0 + ELSE + k[j] := 0 FI ENDFOR -dst[MAX:256] := 0 +k[MAX:16] := 0 - +
immintrin.h
 - - Integer - AVX512BW - Arithmetic - - - Vertically multiply each unsigned 8-bit integer from "a" with the corresponding signed 8-bit integer from "b", producing intermediate signed 16-bit integers. Horizontally add adjacent pairs of intermediate signed 16-bit integers, and pack the saturated results in "dst". - - -FOR j := 0 to 31 - i := j*16 - dst[i+15:i] := Saturate_To_Int16( a[i+15:i+8]*b[i+15:i+8] + a[i+7:i]*b[i+7:i] ) + + Floating Point + KNCNI + Load + + + + + + Depending on "bc", loads 1, 4, or 16 elements of type and size determined by "conv" from memory address "mt" and converts all elements to single-precision (32-bit) floating-point elements, storing the results in "dst". "hint" indicates to the processor whether the data is non-temporal. + addr := MEM[mt] +FOR j := 0 to 15 + i := j*32 + CASE bc OF + _MM_BROADCAST32_NONE: + CASE conv OF + _MM_UPCONV_PS_NONE: + n := j*32 + dst[i+31:i] := addr[n+31:n] + _MM_UPCONV_PS_FLOAT16: + n := j*16 + dst[i+31:i] := Convert_FP16_To_FP32(addr[n+15:n]) + _MM_UPCONV_PS_UINT8: + n := j*8 + dst[i+31:i] := Convert_UInt8_To_FP32(addr[n+7:n]) + _MM_UPCONV_PS_SINT8: + n := j*8 + dst[i+31:i] := Convert_Int8_To_FP32(addr[n+7:n]) + _MM_UPCONV_PS_UINT16: + n := j*16 + dst[i+31:i] := Convert_UInt16_To_FP32(addr[n+15:n]) + _MM_UPCONV_PS_SINT16: + n := j*16 + dst[i+31:i] := Convert_Int16_To_FP32(addr[n+15:n]) + ESAC + _MM_BROADCAST_1X16: + CASE conv OF + _MM_UPCONV_PS_NONE: + n := j*32 + dst[i+31:i] := addr[31:0] + _MM_UPCONV_PS_FLOAT16: + n := j*16 + dst[i+31:i] := Convert_FP16_To_FP32(addr[15:0]) + _MM_UPCONV_PS_UINT8: + n := j*8 + dst[i+31:i] := Convert_UInt8_To_FP32(addr[7:0]) + _MM_UPCONV_PS_SINT8: + n := j*8 + dst[i+31:i] := Convert_Int8_To_FP32(addr[7:0]) + _MM_UPCONV_PS_UINT16: + n := j*16 + dst[i+31:i] := Convert_UInt16_To_FP32(addr[15:0]) + _MM_UPCONV_PS_SINT16: + n := j*16 + dst[i+31:i] := Convert_Int16_To_FP32(addr[15:0]) + ESAC + _MM_BROADCAST_4X16: + mod := j%4 + CASE conv OF + _MM_UPCONV_PS_NONE: + n := mod*32 + dst[i+31:i] := addr[n+31:n] + _MM_UPCONV_PS_FLOAT16: + n := mod*16 + dst[i+31:i] := Convert_FP16_To_FP32(addr[n+15:n]) + _MM_UPCONV_PS_UINT8: + n := mod*8 + dst[i+31:i] := Convert_UInt8_To_FP32(addr[n+7:n]) + _MM_UPCONV_PS_SINT8: + n := mod*8 + dst[i+31:i] := Convert_Int8_To_FP32(addr[n+7:n]) + _MM_UPCONV_PS_UINT16: + n := mod*16 + dst[i+31:i] := Convert_UInt16_To_FP32(addr[n+15:n]) + _MM_UPCONV_PS_SINT16: + n := mod*16 + dst[i+31:i] := Convert_Int16_To_FP32(addr[n+15:n]) + ESAC + ESAC ENDFOR dst[MAX:512] := 0 - + + +
immintrin.h
 - - Integer - AVX512BW - Arithmetic - - - - - Multiply packed unsigned 8-bit integers in "a" by packed signed 8-bit integers in "b", producing intermediate signed 16-bit integers. Horizontally add adjacent pairs of intermediate signed 16-bit integers, and pack the saturated results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - - -FOR j := 0 to 31 - i := j*16 + + Floating Point + KNCNI + Load + + + + + + + + Depending on "bc", loads 1, 4, or 16 elements of type and size determined by "conv" from memory address "mt" and converts all elements to single-precision (32-bit) floating-point elements, storing the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). "hint" indicates to the processor whether the data is non-temporal. + addr := MEM[mt] +FOR j := 0 to 15 + i := j*32 IF k[j] - dst[i+15:i] := Saturate_To_Int16( a[i+15:i+8]*b[i+15:i+8] + a[i+7:i]*b[i+7:i] ) + CASE bc OF + _MM_BROADCAST32_NONE: + CASE conv OF + _MM_UPCONV_PS_NONE: + n := j*32 + dst[i+31:i] := addr[n+31:n] + _MM_UPCONV_PS_FLOAT16: + n := j*16 + dst[i+31:i] := Convert_FP16_To_FP32(addr[n+15:n]) + _MM_UPCONV_PS_UINT8: + n := j*8 + dst[i+31:i] := Convert_UInt8_To_FP32(addr[n+7:n]) + _MM_UPCONV_PS_SINT8: + n := j*8 + dst[i+31:i] := Convert_Int8_To_FP32(addr[n+7:n]) + _MM_UPCONV_PS_UINT16: + n := j*16 + dst[i+31:i] := Convert_UInt16_To_FP32(addr[n+15:n]) + _MM_UPCONV_PS_SINT16: + n := j*16 + dst[i+31:i] := Convert_Int16_To_FP32(addr[n+15:n]) + ESAC + _MM_BROADCAST_1X16: + CASE conv OF + _MM_UPCONV_PS_NONE: + n := j*32 + dst[i+31:i] := addr[31:0] + _MM_UPCONV_PS_FLOAT16: + n := j*16 + dst[i+31:i] := Convert_FP16_To_FP32(addr[15:0]) + _MM_UPCONV_PS_UINT8: + n := j*8 + dst[i+31:i] := Convert_UInt8_To_FP32(addr[7:0]) + _MM_UPCONV_PS_SINT8: + n := j*8 + dst[i+31:i] := Convert_Int8_To_FP32(addr[7:0]) + _MM_UPCONV_PS_UINT16: + n := j*16 + dst[i+31:i] := Convert_UInt16_To_FP32(addr[15:0]) + _MM_UPCONV_PS_SINT16: + n := j*16 + dst[i+31:i] := Convert_Int16_To_FP32(addr[15:0]) + ESAC + _MM_BROADCAST_4X16: + mod := j%4 + CASE conv OF + _MM_UPCONV_PS_NONE: + n := mod*32 + dst[i+31:i] := addr[n+31:n] + _MM_UPCONV_PS_FLOAT16: + n := mod*16 + dst[i+31:i] := Convert_FP16_To_FP32(addr[n+15:n]) + _MM_UPCONV_PS_UINT8: + n := mod*8 + dst[i+31:i] := Convert_UInt8_To_FP32(addr[n+7:n]) + _MM_UPCONV_PS_SINT8: + n := mod*8 + dst[i+31:i] := Convert_Int8_To_FP32(addr[n+7:n]) + _MM_UPCONV_PS_UINT16: + n := mod*16 + dst[i+31:i] := Convert_UInt16_To_FP32(addr[n+15:n]) + _MM_UPCONV_PS_SINT16: + n := mod*16 + dst[i+31:i] := Convert_Int16_To_FP32(addr[n+15:n]) + ESAC + ESAC ELSE - dst[i+15:i] := src[i+15:i] + dst[i+31:i] := src[i+31:i] FI ENDFOR dst[MAX:512] := 0 - + + +
immintrin.h
 - + Integer - AVX512BW - Arithmetic - - - - Multiply packed unsigned 8-bit integers in "a" by packed signed 8-bit integers in "b", producing intermediate signed 16-bit integers. Horizontally add adjacent pairs of intermediate signed 16-bit integers, and pack the saturated results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - - -FOR j := 0 to 31 - i := j*16 - IF k[j] - dst[i+15:i] := Saturate_To_Int16( a[i+15:i+8]*b[i+15:i+8] + a[i+7:i]*b[i+7:i] ) - ELSE - dst[i+15:i] := 0 - FI + KNCNI + Load + + + + + + Depending on "bc", loads 1, 4, or 16 elements of type and size determined by "conv" from memory address "mt" and converts all elements to 32-bit integer elements, storing the results in "dst". "hint" indicates to the processor whether the data is non-temporal. + addr := MEM[mt] +FOR j := 0 to 15 + i := j*32 + CASE bc OF + _MM_BROADCAST32_NONE: + CASE conv OF + _MM_UPCONV_EPI32_NONE: + n := j*32 + dst[i+31:i] := addr[n+31:n] + _MM_UPCONV_EPI32_UINT8: + n := j*8 + dst[i+31:i] := ZeroExtend32(addr[n+7:n]) + _MM_UPCONV_EPI32_SINT8: + n := j*8 + dst[i+31:i] := SignExtend32(addr[n+7:n]) + _MM_UPCONV_EPI32_UINT16: + n := j*16 + dst[i+31:i] := ZeroExtend32(addr[n+15:n]) + _MM_UPCONV_EPI32_SINT16: + n := j*16 + dst[i+31:i] := SignExtend32(addr[n+15:n]) + ESAC + _MM_BROADCAST_1X16: + CASE conv OF + _MM_UPCONV_EPI32_NONE: + n := j*32 + dst[i+31:i] := addr[31:0] + _MM_UPCONV_EPI32_UINT8: + n := j*8 + dst[i+31:i] := ZeroExtend32(addr[7:0]) + _MM_UPCONV_EPI32_SINT8: + n := j*8 + dst[i+31:i] := SignExtend32(addr[7:0]) + _MM_UPCONV_EPI32_UINT16: + n := j*16 + dst[i+31:i] := ZeroExtend32(addr[15:0]) + _MM_UPCONV_EPI32_SINT16: + n := j*16 + dst[i+31:i] := SignExtend32(addr[15:0]) + ESAC + _MM_BROADCAST_4X16: + mod := j%4 + CASE conv OF + _MM_UPCONV_EPI32_NONE: + n := mod*32 + dst[i+31:i] := addr[n+31:n] + _MM_UPCONV_EPI32_UINT8: + n := mod*8 + dst[i+31:i] := ZeroExtend32(addr[n+7:n]) + _MM_UPCONV_EPI32_SINT8: + n := mod*8 + dst[i+31:i] := SignExtend32(addr[n+7:n]) + _MM_UPCONV_EPI32_UINT16: + n := mod*16 + dst[i+31:i] := ZeroExtend32(addr[n+15:n]) + _MM_UPCONV_EPI32_SINT16: + n := mod*16 + dst[i+31:i] := SignExtend32(addr[n+15:n]) + ESAC + ESAC ENDFOR dst[MAX:512] := 0 - + + +
immintrin.h
 - + Integer - AVX512VL - AVX512BW - Arithmetic - - - - - Multiply packed unsigned 8-bit integers in "a" by packed signed 8-bit integers in "b", producing intermediate signed 16-bit integers. Horizontally add adjacent pairs of intermediate signed 16-bit integers, and pack the saturated results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - - -FOR j := 0 to 7 - i := j*16 + KNCNI + Load + + + + + + + + Depending on "bc", loads 1, 4, or 16 elements of type and size determined by "conv" from memory address "mt" and converts all elements to 32-bit integer elements, storing the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). "hint" indicates to the processor whether the data is non-temporal. + addr := MEM[mt] +FOR j := 0 to 15 + i := j*32 IF k[j] - dst[i+15:i] := Saturate_To_Int16( a[i+15:i+8]*b[i+15:i+8] + a[i+7:i]*b[i+7:i] ) + CASE bc OF + _MM_BROADCAST32_NONE: + CASE conv OF + _MM_UPCONV_EPI32_NONE: + n := j*32 + dst[i+31:i] := addr[n+31:n] + _MM_UPCONV_EPI32_UINT8: + n := j*8 + dst[i+31:i] := ZeroExtend32(addr[n+7:n]) + _MM_UPCONV_EPI32_SINT8: + n := j*8 + dst[i+31:i] := SignExtend32(addr[n+7:n]) + _MM_UPCONV_EPI32_UINT16: + n := j*16 + dst[i+31:i] := ZeroExtend32(addr[n+15:n]) + _MM_UPCONV_EPI32_SINT16: + n := j*16 + dst[i+31:i] := SignExtend32(addr[n+15:n]) + ESAC + _MM_BROADCAST_1X16: + CASE conv OF + _MM_UPCONV_EPI32_NONE: + n := j*32 + dst[i+31:i] := addr[31:0] + _MM_UPCONV_EPI32_UINT8: + n := j*8 + dst[i+31:i] := ZeroExtend32(addr[7:0]) + _MM_UPCONV_EPI32_SINT8: + n := j*8 + dst[i+31:i] := SignExtend32(addr[7:0]) + _MM_UPCONV_EPI32_UINT16: + n := j*16 + dst[i+31:i] := ZeroExtend32(addr[15:0]) + _MM_UPCONV_EPI32_SINT16: + n := j*16 + dst[i+31:i] := SignExtend32(addr[15:0]) + ESAC + _MM_BROADCAST_4X16: + mod := j%4 + CASE conv OF + _MM_UPCONV_EPI32_NONE: + n := mod*32 + dst[i+31:i] := addr[n+31:n] + _MM_UPCONV_EPI32_UINT8: + n := mod*8 + dst[i+31:i] := ZeroExtend32(addr[n+7:n]) + _MM_UPCONV_EPI32_SINT8: + n := mod*8 + dst[i+31:i] := SignExtend32(addr[n+7:n]) + _MM_UPCONV_EPI32_UINT16: + n := mod*16 + dst[i+31:i] := ZeroExtend32(addr[n+15:n]) + _MM_UPCONV_EPI32_SINT16: + n := mod*16 + dst[i+31:i] := SignExtend32(addr[n+15:n]) + ESAC + ESAC ELSE - dst[i+15:i] := src[i+15:i] + dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - + + +
immintrin.h
 - - Integer - AVX512VL - AVX512BW - Arithmetic - - - - Multiply packed unsigned 8-bit integers in "a" by packed signed 8-bit integers in "b", producing intermediate signed 16-bit integers. Horizontally add adjacent pairs of intermediate signed 16-bit integers, and pack the saturated results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - - + + Floating Point + KNCNI + Load + + + + + + Depending on "bc", loads 1, 4, or 8 elements of type and size determined by "conv" from memory address "mt" and converts all elements to double-precision (64-bit) floating-point elements, storing the results in "dst". "hint" indicates to the processor whether the data is non-temporal. + addr := MEM[mt] FOR j := 0 to 7 - i := j*16 - IF k[j] - dst[i+15:i] := Saturate_To_Int16( a[i+15:i+8]*b[i+15:i+8] + a[i+7:i]*b[i+7:i] ) - ELSE - dst[i+15:i] := 0 - FI + i := j*64 + CASE bc OF + _MM_BROADCAST64_NONE: + CASE conv OF + _MM_UPCONV_PD_NONE: + n := j*64 + dst[i+63:i] := addr[n+63:n] + ESAC + _MM_BROADCAST_1X8: + CASE conv OF + _MM_UPCONV_PD_NONE: + n := j*64 + dst[i+63:i] := addr[63:0] + ESAC + _MM_BROADCAST_4X8: + mod := j%4 + CASE conv OF + _MM_UPCONV_PD_NONE: + n := mod*64 + dst[i+63:i] := addr[n+63:n] + ESAC + ESAC ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - + + +
immintrin.h
 - - Integer - AVX512VL - AVX512BW - Arithmetic - - - - - Multiply packed 16-bit integers in "a" and "b", producing intermediate 32-bit integers. Horizontally add adjacent pairs of intermediate 32-bit integers, and pack the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - - + + Floating Point + KNCNI + Load + + + + + + + + Depending on "bc", loads 1, 4, or 8 elements of type and size determined by "conv" from memory address "mt" and converts all elements to double-precision (64-bit) floating-point elements, storing the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). "hint" indicates to the processor whether the data is non-temporal. + addr := MEM[mt] FOR j := 0 to 7 - i := j*32 + i := j*64 IF k[j] - dst[i+31:i] := a[i+31:i+16]*b[i+31:i+16] + a[i+15:i]*b[i+15:i] + CASE bc OF + _MM_BROADCAST64_NONE: + CASE conv OF + _MM_UPCONV_PD_NONE: + n := j*64 + dst[i+63:i] := addr[n+63:n] + ESAC + _MM_BROADCAST_1X8: + CASE conv OF + _MM_UPCONV_PD_NONE: + n := j*64 + dst[i+63:i] := addr[63:0] + ESAC + _MM_BROADCAST_4X8: + mod := j%4 + CASE conv OF + _MM_UPCONV_PD_NONE: + n := mod*64 + dst[i+63:i] := addr[n+63:n] + ESAC + ESAC ELSE - dst[i+31:i] := src[i+31:i] + dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - + + +
immintrin.h
 - + Integer - AVX512VL - AVX512BW - Arithmetic - - - - Multiply packed 16-bit integers in "a" and "b", producing intermediate 32-bit integers. Horizontally add adjacent pairs of intermediate 32-bit integers, and pack the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - - + KNCNI + Load + + + + + + Depending on "bc", loads 1, 4, or 8 elements of type and size determined by "conv" from memory address "mt" and converts all elements to 64-bit integer elements, storing the results in "dst". "hint" indicates to the processor whether the data is non-temporal. + addr := MEM[mt] FOR j := 0 to 7 - i := j*32 - IF k[j] - dst[i+31:i] := a[i+31:i+16]*b[i+31:i+16] + a[i+15:i]*b[i+15:i] - ELSE - dst[i+31:i] := 0 - FI -ENDFOR -dst[MAX:256] := 0 - - -
immintrin.h
- - - Integer - AVX512BW - Arithmetic - - - Multiply packed 16-bit integers in "a" and "b", producing intermediate 32-bit integers. Horizontally add adjacent pairs of intermediate 32-bit integers, and pack the results in "dst". - - -FOR j := 0 to 15 - i := j*32 - st[i+31:i] := a[i+31:i+16]*b[i+31:i+16] + a[i+15:i]*b[i+15:i] + i := j*64 + CASE bc OF + _MM_BROADCAST64_NONE: + CASE conv OF + _MM_UPCONV_EPI64_NONE: + n := j*64 + dst[i+63:i] := addr[n+63:n] + ESAC + _MM_BROADCAST_1X8: + CASE conv OF + _MM_UPCONV_EPI64_NONE: + n := j*64 + dst[i+63:i] := addr[63:0] + ESAC + _MM_BROADCAST_4X8: + mod := j%4 + CASE conv OF + _MM_UPCONV_EPI64_NONE: + n := mod*64 + dst[i+63:i] := addr[n+63:n] + ESAC + ESAC ENDFOR dst[MAX:512] := 0 - + + +
immintrin.h
 - + Integer - AVX512BW - Arithmetic - - - - - Multiply packed 16-bit integers in "a" and "b", producing intermediate 32-bit integers. Horizontally add adjacent pairs of intermediate 32-bit integers, and pack the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - - -FOR j := 0 to 15 - i := j*32 + KNCNI + Load + + + + + + + + Depending on "bc", loads 1, 4, or 8 elements of type and size determined by "conv" from memory address "mt" and converts all elements to 64-bit integer elements, storing the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). "hint" indicates to the processor whether the data is non-temporal. + addr := MEM[mt] +FOR j := 0 to 7 + i := j*64 IF k[j] - dst[i+31:i] := a[i+31:i+16]*b[i+31:i+16] + a[i+15:i]*b[i+15:i] + CASE bc OF + _MM_BROADCAST64_NONE: + CASE conv OF + _MM_UPCONV_EPI64_NONE: + n := j*64 + dst[i+63:i] := addr[n+63:n] + ESAC + _MM_BROADCAST_1X8: + CASE conv OF + _MM_UPCONV_EPI64_NONE: + n := j*64 + dst[i+63:i] := addr[63:0] + ESAC + _MM_BROADCAST_4X8: + mod := j%4 + CASE conv OF + _MM_UPCONV_EPI64_NONE: + n := mod*64 + dst[i+63:i] := addr[n+63:n] + ESAC + ESAC ELSE - dst[i+31:i] := src[i+31:i] + dst[i+63:i] := src[i+63:i] FI ENDFOR dst[MAX:512] := 0 - + + +
immintrin.h
 - - Integer - AVX512BW - Arithmetic - - - - Multiply packed 16-bit integers in "a" and "b", producing intermediate 32-bit integers. Horizontally add adjacent pairs of intermediate 32-bit integers, and pack the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - - -FOR j := 0 to 15 - i := j*32 - IF k[j] - dst[i+31:i] := a[i+31:i+16]*b[i+31:i+16] + a[i+15:i]*b[i+15:i] - ELSE - dst[i+31:i] := 0 - FI -ENDFOR + + Floating Point + KNCNI + Swizzle + + + + Performs a swizzle transformation of each of the four groups of packed 4xsingle-precision (32-bit) floating-point elements in "v" using swizzle parameter "s", storing the results in "dst". + CASE s OF +_MM_SWIZ_REG_NONE: + dst[511:0] := v[511:0] +_MM_SWIZ_REG_DCBA: + dst[511:0] := v[511:0] +_MM_SWIZ_REG_CDAB: + FOR j := 0 to 7 + i := j*64 + dst[i+31:i] := v[i+63:i+32] + dst[i+63:i+32] := v[i+31:i] + ENDFOR +_MM_SWIZ_REG_BADC: + FOR j := 0 to 3 + i := j*128 + dst[i+31:i] := v[i+95:i+64] + dst[i+63:i+32] := v[i+127:i+96] + dst[i+95:i+64] := v[i+31:i] + dst[i+127:i+96] := v[i+63:i+32] + ENDFOR +_MM_SWIZ_REG_AAAA: + FOR j := 0 to 3 + i := j*128 + dst[i+31:i] := v[i+31:i] + dst[i+63:i+32] := v[i+31:i] + dst[i+95:i+64] := v[i+31:i] + dst[i+127:i+96] := v[i+31:i] + ENDFOR +_MM_SWIZ_REG_BBBB: + FOR j := 0 to 3 + i := j*128 + dst[i+31:i] := v[i+63:i+32] + dst[i+63:i+32] := v[i+63:i+32] + dst[i+95:i+64] := v[i+63:i+32] + dst[i+127:i+96] := v[i+63:i+32] + ENDFOR +_MM_SWIZ_REG_CCCC: + FOR j := 0 to 3 + i := j*128 + dst[i+31:i] := v[i+95:i+64] + dst[i+63:i+32] := v[i+95:i+64] + dst[i+95:i+64] := v[i+95:i+64] + dst[i+127:i+96] := v[i+95:i+64] + ENDFOR +_MM_SWIZ_REG_DDDD: + FOR j := 0 to 3 + i := j*128 + dst[i+31:i] := v[i+127:i+96] + dst[i+63:i+32] := v[i+127:i+96] + dst[i+95:i+64] := v[i+127:i+96] + dst[i+127:i+96] := v[i+127:i+96] + ENDFOR +_MM_SWIZ_REG_DACB: + FOR j := 0 to 3 + i := j*128 + dst[i+31:i] := v[i+63:i+32] + dst[i+63:i+32] := v[i+95:i+64] + dst[i+95:i+64] := v[i+31:i] + dst[i+127:i+96] := v[i+127:i+96] + ENDFOR +ESAC dst[MAX:512] := 0 -
immintrin.h
 - - Integer - AVX512VL - AVX512BW - Arithmetic - - - - - Multiply packed 16-bit integers in "a" and "b", producing intermediate 32-bit integers. Horizontally add adjacent pairs of intermediate 32-bit integers, and pack the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - - -FOR j := 0 to 3 - i := j*32 - IF k[j] - dst[i+31:i] := a[i+31:i+16]*b[i+31:i+16] + a[i+15:i]*b[i+15:i] - ELSE - dst[i+31:i] := src[i+31:i] - FI -ENDFOR -dst[MAX:128] := 0 + + Floating Point + KNCNI + Swizzle + + + + Performs a swizzle transformation of each of the two groups of packed 4x double-precision (64-bit) floating-point elements in "v" using swizzle parameter "s", storing the results in "dst". + CASE s OF +_MM_SWIZ_REG_NONE: + dst[511:0] := v[511:0] +_MM_SWIZ_REG_DCBA: + dst[511:0] := v[511:0] +_MM_SWIZ_REG_CDAB: + FOR j := 0 to 3 + i := j*64 + dst[i+63:i] := v[i+127:i+64] + dst[i+127:i+64] := v[i+63:i] + ENDFOR +_MM_SWIZ_REG_BADC: + FOR j := 0 to 1 + i := j*256 + dst[i+63:i] := v[i+191:i+128] + dst[i+127:i+64] := v[i+255:i+192] + dst[i+191:i+128] := v[i+63:i] + dst[i+255:i+192] := v[i+127:i+64] + ENDFOR +_MM_SWIZ_REG_AAAA: + FOR j := 0 to 1 + i := j*256 + dst[i+63:i] := v[i+63:i] + dst[i+127:i+64] := v[i+63:i] + dst[i+191:i+128] := v[i+63:i] + dst[i+255:i+192] := v[i+63:i] + ENDFOR +_MM_SWIZ_REG_BBBB: + FOR j := 0 to 1 + i := j*256 + dst[i+63:i] := v[i+127:i+63] + dst[i+127:i+64] := v[i+127:i+63] + dst[i+191:i+128] := v[i+127:i+63] + dst[i+255:i+192] := v[i+127:i+63] + ENDFOR +_MM_SWIZ_REG_CCCC: + FOR j := 0 to 1 + i := j*256 + dst[i+63:i] := v[i+191:i+128] + dst[i+127:i+64] := v[i+191:i+128] + dst[i+191:i+128] := v[i+191:i+128] + dst[i+255:i+192] := v[i+191:i+128] + ENDFOR +_MM_SWIZ_REG_DDDD: + FOR j := 0 to 1 + i := j*256 + dst[i+63:i] := v[i+255:i+192] + dst[i+127:i+64] := v[i+255:i+192] + dst[i+191:i+128] := v[i+255:i+192] + dst[i+255:i+192] := v[i+255:i+192] + ENDFOR +_MM_SWIZ_REG_DACB: + FOR j := 0 to 1 + i := j*256 + dst[i+63:i] := v[i+127:i+64] + dst[i+127:i+64] := v[i+191:i+128] + dst[i+191:i+128] := v[i+63:i] + dst[i+255:i+192] := v[i+255:i+192] + ENDFOR +ESAC +dst[MAX:512] := 0 -
immintrin.h
 - + Integer - AVX512VL - AVX512BW - Arithmetic - - - - Multiply packed 16-bit integers in "a" and "b", producing intermediate 32-bit integers. Horizontally add adjacent pairs of intermediate 32-bit integers, and pack the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - - -FOR j := 0 to 3 - i := j*32 - IF k[j] - dst[i+31:i] := a[i+31:i+16]*b[i+31:i+16] + a[i+15:i]*b[i+15:i] - ELSE - dst[i+31:i] := 0 - FI -ENDFOR -dst[MAX:128] := 0 + KNCNI + Swizzle + + + + Performs a swizzle transformation of each of the four groups of packed 4x 32-bit integer elements in "v" using swizzle parameter "s", storing the results in "dst". + CASE s OF +_MM_SWIZ_REG_NONE: + dst[511:0] := v[511:0] +_MM_SWIZ_REG_DCBA: + dst[511:0] := v[511:0] +_MM_SWIZ_REG_CDAB: + FOR j := 0 to 7 + i := j*64 + dst[i+31:i] := v[i+63:i+32] + dst[i+63:i+32] := v[i+31:i] + ENDFOR +_MM_SWIZ_REG_BADC: + FOR j := 0 to 3 + i := j*128 + dst[i+31:i] := v[i+95:i+64] + dst[i+63:i+32] := v[i+127:i+96] + dst[i+95:i+64] := v[i+31:i] + dst[i+127:i+96] := v[i+63:i+32] + ENDFOR +_MM_SWIZ_REG_AAAA: + FOR j := 0 to 3 + i := j*128 + dst[i+31:i] := v[i+31:i] + dst[i+63:i+32] := v[i+31:i] + dst[i+95:i+64] := v[i+31:i] + dst[i+127:i+96] := v[i+31:i] + ENDFOR +_MM_SWIZ_REG_BBBB: + FOR j := 0 to 3 + i := j*128 + dst[i+31:i] := v[i+63:i+32] + dst[i+63:i+32] := v[i+63:i+32] + dst[i+95:i+64] := v[i+63:i+32] + dst[i+127:i+96] := v[i+63:i+32] + ENDFOR +_MM_SWIZ_REG_CCCC: + FOR j := 0 to 3 + i := j*128 + dst[i+31:i] := v[i+95:i+64] + dst[i+63:i+32] := v[i+95:i+64] + dst[i+95:i+64] := v[i+95:i+64] + dst[i+127:i+96] := v[i+95:i+64] + ENDFOR +_MM_SWIZ_REG_DDDD: + FOR j := 0 to 3 + i := j*128 + dst[i+31:i] := v[i+127:i+96] + dst[i+63:i+32] := v[i+127:i+96] + dst[i+95:i+64] := v[i+127:i+96] + dst[i+127:i+96] := v[i+127:i+96] + ENDFOR +_MM_SWIZ_REG_DACB: + FOR j := 0 to 3 + i := j*128 + dst[i+31:i] := v[i+63:i+32] + dst[i+63:i+32] := v[i+95:i+64] + dst[i+95:i+64] := v[i+31:i] + dst[i+127:i+96] := v[i+127:i+96] + ENDFOR +ESAC +dst[MAX:512] := 0 -
immintrin.h
 - + Integer - AVX512VL - AVX512BW - Arithmetic - - - - - Compare packed 8-bit integers in "a" and "b", and store packed maximum values in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -FOR j := 0 to 31 - i := j*8 - IF k[j] - IF a[i+7:i] > b[i+7:i] - dst[i+7:i] := a[i+7:i] + KNCNI + Swizzle + + + + Performs a swizzle transformation of each of the two groups of packed 4x64-bit integer elements in "v" using swizzle parameter "s", storing the results in "dst". + CASE s OF +_MM_SWIZ_REG_NONE: + dst[511:0] := v[511:0] +_MM_SWIZ_REG_DCBA: + dst[511:0] := v[511:0] +_MM_SWIZ_REG_CDAB: + FOR j := 0 to 3 + i := j*64 + dst[i+63:i] := v[i+127:i+64] + dst[i+127:i+64] := v[i+63:i] + ENDFOR +_MM_SWIZ_REG_BADC: + FOR j := 0 to 1 + i := j*256 + dst[i+63:i] := v[i+191:i+128] + dst[i+127:i+64] := v[i+255:i+192] + dst[i+191:i+128] := v[i+63:i] + dst[i+255:i+192] := v[i+127:i+64] + ENDFOR +_MM_SWIZ_REG_AAAA: + FOR j := 0 to 1 + i := j*256 + dst[i+63:i] := v[i+63:i] + dst[i+127:i+64] := v[i+63:i] + dst[i+191:i+128] := v[i+63:i] + dst[i+255:i+192] := v[i+63:i] + ENDFOR +_MM_SWIZ_REG_BBBB: + FOR j := 0 to 1 + i := j*256 + dst[i+63:i] := v[i+127:i+63] + dst[i+127:i+64] := v[i+127:i+63] + dst[i+191:i+128] := v[i+127:i+63] + dst[i+255:i+192] := v[i+127:i+63] + ENDFOR +_MM_SWIZ_REG_CCCC: + FOR j := 0 to 1 + i := j*256 + dst[i+63:i] := v[i+191:i+128] + dst[i+127:i+64] := v[i+191:i+128] + dst[i+191:i+128] := v[i+191:i+128] + dst[i+255:i+192] := v[i+191:i+128] + ENDFOR +_MM_SWIZ_REG_DDDD: + FOR j := 0 to 1 + i := j*256 + dst[i+63:i] := v[i+255:i+192] + dst[i+127:i+64] := v[i+255:i+192] + dst[i+191:i+128] := v[i+255:i+192] + dst[i+255:i+192] := v[i+255:i+192] + ENDFOR +_MM_SWIZ_REG_DACB: + FOR j := 0 to 1 + i := j*256 + dst[i+63:i] := v[i+127:i+64] + dst[i+127:i+64] := v[i+191:i+128] + dst[i+191:i+128] := v[i+63:i] + dst[i+255:i+192] := v[i+255:i+192] + ENDFOR +ESAC +dst[MAX:512] := 0 + +
immintrin.h
+ + + Floating Point + KNCNI + Swizzle + + + + + + Performs a swizzle transformation of each of the four groups of packed 4x single-precision (32-bit) floating-point elements in "v" using swizzle parameter "s", storing the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + CASE s OF +_MM_SWIZ_REG_NONE: + dst[511:0] := v[511:0] +_MM_SWIZ_REG_DCBA: + dst[511:0] := v[511:0] +_MM_SWIZ_REG_CDAB: + FOR j := 0 to 7 + i := j*64 + IF k[j*2] + dst[i+31:i] := v[i+63:i+32] + ELSE + dst[i+31:i] := src[i+31:i] + FI + IF k[j*2+1] + dst[i+63:i+32] := v[i+31:i] + ELSE + dst[i+63:i+32] := src[i+63:i+32] + FI + ENDFOR +_MM_SWIZ_REG_BADC: + FOR j := 0 to 3 + i := j*128 + IF k[j*4] + dst[i+31:i] := v[i+95:i+64] + ELSE + dst[i+31:i] := src[i+31:i] + FI + IF k[j*4+1] + dst[i+63:i+32] := v[i+127:i+96] + ELSE + dst[i+63:i+32] := src[i+63:i+32] + FI + IF k[j*4+2] + dst[i+95:i+64] := v[i+31:i] + ELSE + dst[i+95:i+64] := src[i+95:i+64] + FI + IF k[j*4+3] + dst[i+127:i+96] := v[i+63:i+32] + ELSE + dst[i+127:i+96] := src[i+127:i+96] + FI + ENDFOR +_MM_SWIZ_REG_AAAA: + FOR j := 0 to 3 + i := j*128 + IF k[j*4] + dst[i+31:i] := v[i+31:i] + ELSE + dst[i+31:i] := src[i+31:i] + FI + IF k[j*4+1] + dst[i+63:i+32] := v[i+31:i] + ELSE + dst[i+63:i+32] := src[i+63:i+32] + FI + IF k[j*4+2] + dst[i+95:i+64] := v[i+31:i] + ELSE + dst[i+95:i+64] := src[i+95:i+64] + FI + IF k[j*4+3] + dst[i+127:i+96] := v[i+31:i] + ELSE + dst[i+127:i+96] := src[i+127:i+96] + FI + ENDFOR +_MM_SWIZ_REG_BBBB: + FOR j := 0 to 3 + i := j*128 + IF k[j*4] + dst[i+31:i] := v[i+63:i+32] + ELSE + dst[i+31:i] := src[i+31:i] + FI + IF k[j*4+1] + dst[i+63:i+32] := v[i+63:i+32] + ELSE + dst[i+63:i+32] := src[i+63:i+32] + FI + IF k[j*4+2] + dst[i+95:i+64] := v[i+63:i+32] + ELSE + dst[i+95:i+64] := src[i+95:i+64] + FI + IF k[j*4+3] + dst[i+127:i+96] := v[i+63:i+32] + ELSE + dst[i+127:i+96] := src[i+127:i+96] + FI + ENDFOR +_MM_SWIZ_REG_CCCC: + FOR j := 0 to 3 + i := j*128 + IF k[j*4] + dst[i+31:i] := v[i+95:i+64] + ELSE + dst[i+31:i] := src[i+31:i] + FI + IF k[j*4+1] + dst[i+63:i+32] := v[i+95:i+64] + ELSE + dst[i+63:i+32] := src[i+63:i+32] + FI + IF k[j*4+2] + dst[i+95:i+64] := v[i+95:i+64] + ELSE + dst[i+95:i+64] := src[i+95:i+64] + FI + IF k[j*4+3] + dst[i+127:i+96] := v[i+95:i+64] + ELSE + dst[i+127:i+96] := src[i+127:i+96] + FI + ENDFOR +_MM_SWIZ_REG_DDDD: + FOR j := 0 to 3 + i := j*128 + IF k[j*4] + dst[i+31:i] := v[i+127:i+96] + ELSE + dst[i+31:i] := src[i+31:i] + FI + IF k[j*4+1] + dst[i+63:i+32] := v[i+127:i+96] + ELSE + dst[i+63:i+32] := src[i+63:i+32] + FI + IF k[j*4+2] + dst[i+95:i+64] := v[i+127:i+96] ELSE - dst[i+7:i] := b[i+7:i] + dst[i+95:i+64] := src[i+95:i+64] FI - ELSE - dst[i+7:i] := src[i+7:i] - FI -ENDFOR -dst[MAX:256] := 0 - - -
immintrin.h
- - - Integer - AVX512VL - AVX512BW - Arithmetic - - - - Compare packed 8-bit integers in "a" and "b", and store packed maximum values in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - -FOR j := 0 to 31 - i := j*8 - IF k[j] - IF a[i+7:i] > b[i+7:i] - dst[i+7:i] := a[i+7:i] + IF k[j*4+3] + dst[i+127:i+96] := v[i+127:i+96] ELSE - dst[i+7:i] := b[i+7:i] + dst[i+127:i+96] := src[i+127:i+96] FI - ELSE - dst[i+7:i] := 0 - FI -ENDFOR -dst[MAX:256] := 0 - - -
immintrin.h
- - - Integer - AVX512BW - Arithmetic - - - - - Compare packed 8-bit integers in "a" and "b", and store packed maximum values in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -FOR j := 0 to 63 - i := j*8 - IF k[j] - IF a[i+7:i] > b[i+7:i] - dst[i+7:i] := a[i+7:i] + ENDFOR +_MM_SWIZ_REG_DACB: + FOR j := 0 to 3 + i := j*128 + IF k[j*4] + dst[i+31:i] := v[i+63:i+32] ELSE - dst[i+7:i] := b[i+7:i] + dst[i+31:i] := src[i+31:i] FI - ELSE - dst[i+7:i] := src[i+7:i] - FI -ENDFOR -dst[MAX:512] := 0 - - -
immintrin.h
- - - Integer - AVX512BW - Arithmetic - - - - Compare packed 8-bit integers in "a" and "b", and store packed maximum values in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - -FOR j := 0 to 63 - i := j*8 - IF k[j] - IF a[i+7:i] > b[i+7:i] - dst[i+7:i] := a[i+7:i] + IF k[j*4+1] + dst[i+63:i+32] := v[i+95:i+64] ELSE - dst[i+7:i] := b[i+7:i] + dst[i+63:i+32] := src[i+63:i+32] FI - ELSE - dst[i+7:i] := 0 - FI -ENDFOR -dst[MAX:512] := 0 - - -
immintrin.h
- - - Integer - AVX512BW - Arithmetic - - - Compare packed 8-bit integers in "a" and "b", and store packed maximum values in "dst". - -FOR j := 0 to 63 - i := j*8 - IF a[i+7:i] > b[i+7:i] - dst[i+7:i] := a[i+7:i] - ELSE - dst[i+7:i] := b[i+7:i] - FI -ENDFOR + IF k[j*4+2] + dst[i+95:i+64] := v[i+31:i] + ELSE + dst[i+95:i+64] := src[i+95:i+64] + FI + IF k[j*4+3] + dst[i+127:i+96] := v[i+127:i+96] + ELSE + dst[i+127:i+96] := src[i+127:i+96] + FI + ENDFOR +ESAC dst[MAX:512] := 0 -
immintrin.h
 - - Integer - AVX512VL - AVX512BW - Arithmetic - - - - - Compare packed 8-bit integers in "a" and "b", and store packed maximum values in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -FOR j := 0 to 15 - i := j*8 - IF k[j] - IF a[i+7:i] > b[i+7:i] - dst[i+7:i] := a[i+7:i] + + Floating Point + KNCNI + Swizzle + + + + + + Performs a swizzle transformation of each of the two groups of packed 4x double-precision (64-bit) floating-point elements in "v" using swizzle parameter "s", storing the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + CASE s OF +_MM_SWIZ_REG_NONE: + dst[511:0] := v[511:0] +_MM_SWIZ_REG_DCBA: + dst[511:0] := v[511:0] +_MM_SWIZ_REG_CDAB: + FOR j := 0 to 3 + i := j*64 + IF k[j*2] + dst[i+63:i] := v[i+127:i+64] ELSE - dst[i+7:i] := b[i+7:i] + dst[i+63:i] := src[i+63:i] FI - ELSE - dst[i+7:i] := src[i+7:i] - FI -ENDFOR -dst[MAX:128] := 0 - - -
immintrin.h
- - - Integer - AVX512VL - AVX512BW - Arithmetic - - - - Compare packed 8-bit integers in "a" and "b", and store packed maximum values in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - -FOR j := 0 to 15 - i := j*8 - IF k[j] - IF a[i+7:i] > b[i+7:i] - dst[i+7:i] := a[i+7:i] + IF k[j*2+1] + dst[i+127:i+64] := v[i+63:i] ELSE - dst[i+7:i] := b[i+7:i] + dst[i+127:i+64] := src[i+127:i+64] FI - ELSE - dst[i+7:i] := 0 - FI -ENDFOR -dst[MAX:128] := 0 - - -
immintrin.h
- - - Integer - AVX512VL - AVX512F - Arithmetic - - - - - Compare packed 32-bit integers in "a" and "b", and store packed maximum values in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -FOR j := 0 to 7 - i := j*32 - IF k[j] - IF a[i+31:i] > b[i+31:i] - dst[i+31:i] := a[i+31:i] + ENDFOR +_MM_SWIZ_REG_BADC: + FOR j := 0 to 1 + i := j*256 + IF k[j*4] + dst[i+63:i] := v[i+191:i+128] ELSE - dst[i+31:i] := b[i+31:i] + dst[i+63:i] := src[i+63:i] FI - ELSE - dst[i+31:i] := src[i+31:i] - FI -ENDFOR -dst[MAX:256] := 0 - - -
immintrin.h
- - - Integer - AVX512VL - AVX512F - Arithmetic - - - - Compare packed 32-bit integers in "a" and "b", and store packed maximum values in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - - -FOR j := 0 to 7 - i := j*32 - IF k[j] - IF a[i+31:i] > b[i+31:i] - dst[i+31:i] := a[i+31:i] + IF k[j*4+1] + dst[i+127:i+64] := v[i+255:i+192] ELSE - dst[i+31:i] := b[i+31:i] + dst[i+127:i+64] := src[i+127:i+64] FI - ELSE - dst[i+31:i] := 0 - FI -ENDFOR -dst[MAX:256] := 0 - - -
immintrin.h
- - - Integer - AVX512VL - AVX512F - Arithmetic - - - - - Compare packed 32-bit integers in "a" and "b", and store packed maximum values in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -FOR j := 0 to 3 - i := j*32 - IF k[j] - IF a[i+31:i] > b[i+31:i] - dst[i+31:i] := a[i+31:i] + IF k[j*4+2] + dst[i+191:i+128] := v[i+63:i] ELSE - dst[i+31:i] := b[i+31:i] + dst[i+191:i+128] := src[i+191:i+128] FI - ELSE - dst[i+31:i] := src[i+31:i] - FI -ENDFOR -dst[MAX:128] := 0 - - -
immintrin.h
- - - Integer - AVX512VL - AVX512F - Arithmetic - - - - Compare packed 32-bit integers in "a" and "b", and store packed maximum values in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - - -FOR j := 0 to 3 - i := j*32 - IF k[j] - IF a[i+31:i] > b[i+31:i] - dst[i+31:i] := a[i+31:i] + IF k[j*4+3] + dst[i+255:i+192] := v[i+127:i+64] ELSE - dst[i+31:i] := b[i+31:i] + dst[i+255:i+192] := src[i+255:i+192] FI - ELSE - dst[i+31:i] := 0 - FI -ENDFOR -dst[MAX:128] := 0 - - -
immintrin.h
- - - Integer - AVX512VL - AVX512F - Arithmetic - - - - - Compare packed 64-bit integers in "a" and "b", and store packed maximum values in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -FOR j := 0 to 3 - i := j*64 - IF k[j] - IF a[i+63:i] > b[i+63:i] - dst[i+63:i] := a[i+63:i] + ENDFOR +_MM_SWIZ_REG_AAAA: + FOR j := 0 to 1 + i := j*256 + IF k[j*4] + dst[i+63:i] := v[i+63:i] ELSE - dst[i+63:i] := b[i+63:i] + dst[i+63:i] := src[i+63:i] FI - ELSE - dst[i+63:i] := src[i+63:i] - FI -ENDFOR -dst[MAX:256] := 0 - - -
immintrin.h
- - - Integer - AVX512VL - AVX512F - Arithmetic - - - - Compare packed 64-bit integers in "a" and "b", and store packed maximum values in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - -FOR j := 0 to 3 - i := j*64 - IF k[j] - IF a[i+63:i] > b[i+63:i] - dst[i+63:i] := a[i+63:i] + IF k[j*4+1] + dst[i+127:i+64] := v[i+63:i] ELSE - dst[i+63:i] := b[i+63:i] + dst[i+127:i+64] := src[i+127:i+64] FI - ELSE - dst[i+63:i] := 0 - FI -ENDFOR -dst[MAX:256] := 0 - - -
immintrin.h
- - - Integer - AVX512VL - AVX512F - Arithmetic - - - Compare packed 64-bit integers in "a" and "b", and store packed maximum values in "dst". - -FOR j := 0 to 3 - i := j*64 - IF a[i+63:i] > b[i+63:i] - dst[i+63:i] := a[i+63:i] - ELSE - dst[i+63:i] := b[i+63:i] - FI -ENDFOR -dst[MAX:256] := 0 - - -
immintrin.h
- - - Integer - AVX512VL - AVX512F - Arithmetic - - - - - Compare packed 64-bit integers in "a" and "b", and store packed maximum values in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -FOR j := 0 to 1 - i := j*64 - IF k[j] - IF a[i+63:i] > b[i+63:i] - dst[i+63:i] := a[i+63:i] + IF k[j*4+2] + dst[i+191:i+128] := v[i+63:i] ELSE - dst[i+63:i] := b[i+63:i] + dst[i+191:i+128] := src[i+191:i+128] FI - ELSE - dst[i+63:i] := src[i+63:i] - FI -ENDFOR -dst[MAX:128] := 0 - - -
immintrin.h
- - - Integer - AVX512VL - AVX512F - Arithmetic - - - - Compare packed 64-bit integers in "a" and "b", and store packed maximum values in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - -FOR j := 0 to 1 - i := j*64 - IF k[j] - IF a[i+63:i] > b[i+63:i] - dst[i+63:i] := a[i+63:i] + IF k[j*4+3] + dst[i+255:i+192] := v[i+63:i] ELSE - dst[i+63:i] := b[i+63:i] + dst[i+255:i+192] := src[i+255:i+192] FI - ELSE - dst[i+63:i] := 0 - FI -ENDFOR -dst[MAX:128] := 0 - - -
immintrin.h
- - - Integer - AVX512VL - AVX512F - Arithmetic - - - Compare packed 64-bit integers in "a" and "b", and store packed maximum values in "dst". - -FOR j := 0 to 1 - i := j*64 - IF a[i+63:i] > b[i+63:i] - dst[i+63:i] := a[i+63:i] - ELSE - dst[i+63:i] := b[i+63:i] - FI -ENDFOR -dst[MAX:128] := 0 - - -
immintrin.h
- - - Integer - AVX512VL - AVX512BW - Arithmetic - - - - - Compare packed 16-bit integers in "a" and "b", and store packed maximum values in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -FOR j := 0 to 15 - i := j*16 - IF k[j] - IF a[i+15:i] > b[i+15:i] - dst[i+15:i] := a[i+15:i] + ENDFOR +_MM_SWIZ_REG_BBBB: + FOR j := 0 to 1 + i := j*256 + IF k[j*4] + dst[i+63:i] := v[i+127:i+63] ELSE - dst[i+15:i] := b[i+15:i] + dst[i+63:i] := src[i+63:i] FI - ELSE - dst[i+15:i] := src[i+15:i] - FI -ENDFOR -dst[MAX:256] := 0 - - -
immintrin.h
- - - Integer - AVX512VL - AVX512BW - Arithmetic - - - - Compare packed 16-bit integers in "a" and "b", and store packed maximum values in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - - -FOR j := 0 to 15 - i := j*16 - IF k[j] - IF a[i+15:i] > b[i+15:i] - dst[i+15:i] := a[i+15:i] + IF k[j*4+1] + dst[i+127:i+64] := v[i+127:i+63] ELSE - dst[i+15:i] := b[i+15:i] + dst[i+127:i+64] := src[i+127:i+64] FI - ELSE - dst[i+15:i] := 0 - FI -ENDFOR -dst[MAX:256] := 0 - - -
immintrin.h
- - - Integer - AVX512BW - Arithmetic - - - - - Compare packed 16-bit integers in "a" and "b", and store packed maximum values in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -FOR j := 0 to 31 - i := j*16 - IF k[j] - IF a[i+15:i] > b[i+15:i] - dst[i+15:i] := a[i+15:i] + IF k[j*4+2] + dst[i+191:i+128] := v[i+127:i+63] ELSE - dst[i+15:i] := b[i+15:i] + dst[i+191:i+128] := src[i+191:i+128] FI - ELSE - dst[i+15:i] := src[i+15:i] - FI -ENDFOR -dst[MAX:512] := 0 - - -
immintrin.h
- - - Integer - AVX512BW - Arithmetic - - - - Compare packed 16-bit integers in "a" and "b", and store packed maximum values in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - - -FOR j := 0 to 31 - i := j*16 - IF k[j] - IF a[i+15:i] > b[i+15:i] - dst[i+15:i] := a[i+15:i] + IF k[j*4+3] + dst[i+255:i+192] := v[i+127:i+63] ELSE - dst[i+15:i] := b[i+15:i] + dst[i+255:i+192] := src[i+255:i+192] FI - ELSE - dst[i+15:i] := 0 - FI -ENDFOR -dst[MAX:512] := 0 - - -
immintrin.h
- - - Integer - AVX512BW - Arithmetic - - - Compare packed 16-bit integers in "a" and "b", and store packed maximum values in "dst". - - -FOR j := 0 to 31 - i := j*16 - IF a[i+15:i] > b[i+15:i] - dst[i+15:i] := a[i+15:i] - ELSE - dst[i+15:i] := b[i+15:i] - FI -ENDFOR -dst[MAX:512] := 0 - - -
immintrin.h
- - - Integer - AVX512VL - AVX512BW - Arithmetic - - - - - Compare packed 16-bit integers in "a" and "b", and store packed maximum values in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -FOR j := 0 to 7 - i := j*16 - IF k[j] - IF a[i+15:i] > b[i+15:i] - dst[i+15:i] := a[i+15:i] + ENDFOR +_MM_SWIZ_REG_CCCC: + FOR j := 0 to 1 + i := j*256 + IF k[j*4] + dst[i+63:i] := v[i+191:i+128] ELSE - dst[i+15:i] := b[i+15:i] + dst[i+63:i] := src[i+63:i] FI - ELSE - dst[i+15:i] := src[i+15:i] - FI -ENDFOR -dst[MAX:128] := 0 - - -
immintrin.h
- - - Integer - AVX512VL - AVX512BW - Arithmetic - - - - Compare packed 16-bit integers in "a" and "b", and store packed maximum values in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - - -FOR j := 0 to 7 - i := j*16 - IF k[j] - IF a[i+15:i] > b[i+15:i] - dst[i+15:i] := a[i+15:i] + IF k[j*4+1] + dst[i+127:i+64] := v[i+191:i+128] + ELSE + dst[i+127:i+64] := src[i+127:i+64] + FI + IF k[j*4+2] + dst[i+191:i+128] := v[i+191:i+128] + ELSE + dst[i+191:i+128] := src[i+191:i+128] + FI + IF k[j*4+3] + dst[i+255:i+192] := v[i+191:i+128] + ELSE + dst[i+255:i+192] := src[i+255:i+192] + FI + ENDFOR +_MM_SWIZ_REG_DDDD: + FOR j := 0 to 1 + i := j*256 + IF k[j*4] + dst[i+63:i] := v[i+255:i+192] + ELSE + dst[i+63:i] := src[i+63:i] + FI + IF k[j*4+1] + dst[i+127:i+64] := v[i+255:i+192] + ELSE + dst[i+127:i+64] := src[i+127:i+64] + FI + IF k[j*4+2] + dst[i+191:i+128] := v[i+255:i+192] + ELSE + dst[i+191:i+128] := src[i+191:i+128] + FI + IF k[j*4+3] + dst[i+255:i+192] := v[i+255:i+192] + ELSE + dst[i+255:i+192] := src[i+255:i+192] + FI + ENDFOR +_MM_SWIZ_REG_DACB: + FOR j := 0 to 1 + i := j*256 + IF k[j*4] + dst[i+63:i] := v[i+127:i+64] + ELSE + dst[i+63:i] := src[i+63:i] + FI + IF k[j*4+1] + dst[i+127:i+64] := v[i+191:i+128] + ELSE + dst[i+127:i+64] := src[i+127:i+64] + FI + IF k[j*4+2] + dst[i+191:i+128] := v[i+63:i] ELSE - dst[i+15:i] := b[i+15:i] + dst[i+191:i+128] := src[i+191:i+128] FI - ELSE - dst[i+15:i] := 0 - FI -ENDFOR -dst[MAX:128] := 0 - - -
immintrin.h
- - - Integer - AVX512VL - AVX512BW - Arithmetic - - - - - Compare packed unsigned 8-bit integers in "a" and "b", and store packed maximum values in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -FOR j := 0 to 31 - i := j*8 - IF k[j] - IF a[i+7:i] > b[i+7:i] - dst[i+7:i] := a[i+7:i] + IF k[j*4+3] + dst[i+255:i+192] := v[i+255:i+192] ELSE - dst[i+7:i] := b[i+7:i] + dst[i+255:i+192] := src[i+255:i+192] FI - ELSE - dst[i+7:i] := src[i+7:i] - FI -ENDFOR -dst[MAX:256] := 0 + ENDFOR +ESAC +dst[MAX:512] := 0 -
immintrin.h
 - + Integer - AVX512VL - AVX512BW - Arithmetic - - - - Compare packed unsigned 8-bit integers in "a" and "b", and store packed maximum values in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - - -FOR j := 0 to 31 - i := j*8 - IF k[j] - IF a[i+7:i] > b[i+7:i] - dst[i+7:i] := a[i+7:i] + KNCNI + Swizzle + + + + + + Performs a swizzle transformation of each of the four groups of packed 4x32-bit integer elements in "v" using swizzle parameter "s", storing the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + CASE s OF +_MM_SWIZ_REG_NONE: + dst[511:0] := v[511:0] +_MM_SWIZ_REG_DCBA: + dst[511:0] := v[511:0] +_MM_SWIZ_REG_CDAB: + FOR j := 0 to 7 + i := j*64 + IF k[j*2] + dst[i+31:i] := v[i+63:i+32] ELSE - dst[i+7:i] := b[i+7:i] + dst[i+31:i] := src[i+31:i] FI - ELSE - dst[i+7:i] := 0 - FI -ENDFOR -dst[MAX:256] := 0 - - -
immintrin.h
- - - Integer - AVX512BW - Arithmetic - - - - - Compare packed unsigned 8-bit integers in "a" and "b", and store packed maximum values in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -FOR j := 0 to 63 - i := j*8 - IF k[j] - IF a[i+7:i] > b[i+7:i] - dst[i+7:i] := a[i+7:i] + IF k[j*2+1] + dst[i+63:i+32] := v[i+31:i] ELSE - dst[i+7:i] := b[i+7:i] + dst[i+63:i+32] := src[i+63:i+32] FI - ELSE - dst[i+7:i] := src[i+7:i] - FI -ENDFOR -dst[MAX:512] := 0 - - -
immintrin.h
- - - Integer - AVX512BW - Arithmetic - - - - Compare packed unsigned 8-bit integers in "a" and "b", and store packed maximum values in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - - -FOR j := 0 to 63 - i := j*8 - IF k[j] - IF a[i+7:i] > b[i+7:i] - dst[i+7:i] := a[i+7:i] + ENDFOR +_MM_SWIZ_REG_BADC: + FOR j := 0 to 3 + i := j*128 + IF k[j*4] + dst[i+31:i] := v[i+95:i+64] ELSE - dst[i+7:i] := b[i+7:i] + dst[i+31:i] := src[i+31:i] FI - ELSE - dst[i+7:i] := 0 - FI -ENDFOR -dst[MAX:512] := 0 - - -
immintrin.h
- - - Integer - AVX512BW - Arithmetic - - - Compare packed unsigned 8-bit integers in "a" and "b", and store packed maximum values in "dst". - - -FOR j := 0 to 63 - i := j*8 - IF a[i+7:i] > b[i+7:i] - dst[i+7:i] := a[i+7:i] - ELSE - dst[i+7:i] := b[i+7:i] - FI -ENDFOR -dst[MAX:512] := 0 - - -
immintrin.h
- - - Integer - AVX512VL - AVX512BW - Arithmetic - - - - - Compare packed unsigned 8-bit integers in "a" and "b", and store packed maximum values in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -FOR j := 0 to 15 - i := j*8 - IF k[j] - IF a[i+7:i] > b[i+7:i] - dst[i+7:i] := a[i+7:i] + IF k[j*4+1] + dst[i+63:i+32] := v[i+127:i+96] ELSE - dst[i+7:i] := b[i+7:i] + dst[i+63:i+32] := src[i+63:i+32] FI - ELSE - dst[i+7:i] := src[i+7:i] - FI -ENDFOR -dst[MAX:128] := 0 - - -
immintrin.h
- - - Integer - AVX512VL - AVX512BW - Arithmetic - - - - Compare packed unsigned 8-bit integers in "a" and "b", and store packed maximum values in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - - -FOR j := 0 to 15 - i := j*8 - IF k[j] - IF a[i+7:i] > b[i+7:i] - dst[i+7:i] := a[i+7:i] + IF k[j*4+2] + dst[i+95:i+64] := v[i+31:i] ELSE - dst[i+7:i] := b[i+7:i] + dst[i+95:i+64] := src[i+95:i+64] FI - ELSE - dst[i+7:i] := 0 - FI -ENDFOR -dst[MAX:128] := 0 - - -
immintrin.h
- - - Integer - AVX512VL - AVX512F - Arithmetic - - - - - Compare packed unsigned 32-bit integers in "a" and "b", and store packed maximum values in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -FOR j := 0 to 7 - i := j*32 - IF k[j] - IF a[i+31:i] > b[i+31:i] - dst[i+31:i] := a[i+31:i] + IF k[j*4+3] + dst[i+127:i+96] := v[i+63:i+32] ELSE - dst[i+31:i] := b[i+31:i] + dst[i+127:i+96] := src[i+127:i+96] FI - ELSE - dst[i+31:i] := src[i+31:i] - FI -ENDFOR -dst[MAX:256] := 0 - - -
immintrin.h
- - - Integer - AVX512VL - AVX512F - Arithmetic - - - - Compare packed unsigned 32-bit integers in "a" and "b", and store packed maximum values in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - - -FOR j := 0 to 7 - i := j*32 - IF k[j] - IF a[i+31:i] > b[i+31:i] - dst[i+31:i] := a[i+31:i] + ENDFOR +_MM_SWIZ_REG_AAAA: + FOR j := 0 to 3 + i := j*128 + IF k[j*4] + dst[i+31:i] := v[i+31:i] ELSE - dst[i+31:i] := b[i+31:i] + dst[i+31:i] := src[i+31:i] FI - ELSE - dst[i+31:i] := 0 - FI -ENDFOR -dst[MAX:256] := 0 - - -
immintrin.h
- - - Integer - AVX512VL - AVX512F - Arithmetic - - - - - Compare packed unsigned 32-bit integers in "a" and "b", and store packed maximum values in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -FOR j := 0 to 3 - i := j*32 - IF k[j] - IF a[i+31:i] > b[i+31:i] - dst[i+31:i] := a[i+31:i] + IF k[j*4+1] + dst[i+63:i+32] := v[i+31:i] ELSE - dst[i+31:i] := b[i+31:i] + dst[i+63:i+32] := src[i+63:i+32] FI - ELSE - dst[i+31:i] := src[i+31:i] - FI -ENDFOR -dst[MAX:128] := 0 - - -
immintrin.h
- - - Integer - AVX512VL - AVX512F - Arithmetic - - - - Compare packed unsigned 32-bit integers in "a" and "b", and store packed maximum values in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - - -FOR j := 0 to 3 - i := j*32 - IF k[j] - IF a[i+31:i] > b[i+31:i] - dst[i+31:i] := a[i+31:i] + IF k[j*4+2] + dst[i+95:i+64] := v[i+31:i] ELSE - dst[i+31:i] := b[i+31:i] + dst[i+95:i+64] := src[i+95:i+64] FI - ELSE - dst[i+31:i] := 0 - FI -ENDFOR -dst[MAX:128] := 0 - - -
immintrin.h
- - - Integer - AVX512VL - AVX512F - Arithmetic - - - - - Compare packed unsigned 64-bit integers in "a" and "b", and store packed maximum values in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -FOR j := 0 to 3 - i := j*64 - IF k[j] - IF a[i+63:i] > b[i+63:i] - dst[i+63:i] := a[i+63:i] + IF k[j*4+3] + dst[i+127:i+96] := v[i+31:i] ELSE - dst[i+63:i] := b[i+63:i] + dst[i+127:i+96] := src[i+127:i+96] FI - ELSE - dst[i+63:i] := src[i+63:i] - FI -ENDFOR -dst[MAX:256] := 0 - - -
immintrin.h
- - - Integer - AVX512VL - AVX512F - Arithmetic - - - - Compare packed unsigned 64-bit integers in "a" and "b", and store packed maximum values in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - - -FOR j := 0 to 3 - i := j*64 - IF k[j] - IF a[i+63:i] > b[i+63:i] - dst[i+63:i] := a[i+63:i] + ENDFOR +_MM_SWIZ_REG_BBBB: + FOR j := 0 to 3 + i := j*128 + IF k[j*4] + dst[i+31:i] := v[i+63:i+32] ELSE - dst[i+63:i] := b[i+63:i] + dst[i+31:i] := src[i+31:i] FI - ELSE - dst[i+63:i] := 0 - FI -ENDFOR -dst[MAX:256] := 0 - - -
immintrin.h
- - - Integer - AVX512VL - AVX512F - Arithmetic - - - Compare packed unsigned 64-bit integers in "a" and "b", and store packed maximum values in "dst". - -FOR j := 0 to 3 - i := j*64 - IF a[i+63:i] > b[i+63:i] - dst[i+63:i] := a[i+63:i] - ELSE - dst[i+63:i] := b[i+63:i] - FI -ENDFOR -dst[MAX:256] := 0 - - -
immintrin.h
- - - Integer - AVX512VL - AVX512F - Arithmetic - - - - - Compare packed unsigned 64-bit integers in "a" and "b", and store packed maximum values in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -FOR j := 0 to 1 - i := j*64 - IF k[j] - IF a[i+63:i] > b[i+63:i] - dst[i+63:i] := a[i+63:i] + IF k[j*4+1] + dst[i+63:i+32] := v[i+63:i+32] ELSE - dst[i+63:i] := b[i+63:i] + dst[i+63:i+32] := src[i+63:i+32] FI - ELSE - dst[i+63:i] := src[i+63:i] - FI -ENDFOR -dst[MAX:128] := 0 - - -
immintrin.h
- - - Integer - AVX512VL - AVX512F - Arithmetic - - - - Compare packed unsigned 64-bit integers in "a" and "b", and store packed maximum values in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - - -FOR j := 0 to 1 - i := j*64 - IF k[j] - IF a[i+63:i] > b[i+63:i] - dst[i+63:i] := a[i+63:i] + IF k[j*4+2] + dst[i+95:i+64] := v[i+63:i+32] ELSE - dst[i+63:i] := b[i+63:i] + dst[i+95:i+64] := src[i+95:i+64] FI - ELSE - dst[i+63:i] := 0 - FI -ENDFOR -dst[MAX:128] := 0 - - -
immintrin.h
- - - Integer - AVX512VL - AVX512F - Arithmetic - - - Compare packed unsigned 64-bit integers in "a" and "b", and store packed maximum values in "dst". - -FOR j := 0 to 1 - i := j*64 - IF a[i+63:i] > b[i+63:i] - dst[i+63:i] := a[i+63:i] - ELSE - dst[i+63:i] := b[i+63:i] - FI -ENDFOR -dst[MAX:128] := 0 - - -
immintrin.h
- - - Integer - AVX512VL - AVX512BW - Arithmetic - - - - - Compare packed unsigned 16-bit integers in "a" and "b", and store packed maximum values in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -FOR j := 0 to 15 - i := j*16 - IF k[j] - IF a[i+15:i] > b[i+15:i] - dst[i+15:i] := a[i+15:i] + IF k[j*4+3] + dst[i+127:i+96] := v[i+63:i+32] ELSE - dst[i+15:i] := b[i+15:i] + dst[i+127:i+96] := src[i+127:i+96] FI - ELSE - dst[i+15:i] := src[i+15:i] - FI -ENDFOR -dst[MAX:256] := 0 - - -
immintrin.h
- - - Integer - AVX512VL - AVX512BW - Arithmetic - - - - Compare packed unsigned 16-bit integers in "a" and "b", and store packed maximum values in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - - -FOR j := 0 to 15 - i := j*16 - IF k[j] - IF a[i+15:i] > b[i+15:i] - dst[i+15:i] := a[i+15:i] + ENDFOR +_MM_SWIZ_REG_CCCC: + FOR j := 0 to 3 + i := j*128 + IF k[j*4] + dst[i+31:i] := v[i+95:i+64] ELSE - dst[i+15:i] := b[i+15:i] + dst[i+31:i] := src[i+31:i] FI - ELSE - dst[i+15:i] := 0 - FI -ENDFOR -dst[MAX:256] := 0 - - -
immintrin.h
- - - Integer - AVX512BW - Arithmetic - - - - - Compare packed unsigned 16-bit integers in "a" and "b", and store packed maximum values in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -FOR j := 0 to 31 - i := j*16 - IF k[j] - IF a[i+15:i] > b[i+15:i] - dst[i+15:i] := a[i+15:i] + IF k[j*4+1] + dst[i+63:i+32] := v[i+95:i+64] ELSE - dst[i+15:i] := b[i+15:i] + dst[i+63:i+32] := src[i+63:i+32] FI - ELSE - dst[i+15:i] := src[i+15:i] - FI -ENDFOR -dst[MAX:512] := 0 - - -
immintrin.h
- - - Integer - AVX512BW - Arithmetic - - - - Compare packed unsigned 16-bit integers in "a" and "b", and store packed maximum values in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - - -FOR j := 0 to 31 - i := j*16 - IF k[j] - IF a[i+15:i] > b[i+15:i] - dst[i+15:i] := a[i+15:i] + IF k[j*4+2] + dst[i+95:i+64] := v[i+95:i+64] ELSE - dst[i+15:i] := b[i+15:i] + dst[i+95:i+64] := src[i+95:i+64] FI - ELSE - dst[i+15:i] := 0 - FI -ENDFOR -dst[MAX:512] := 0 - - -
immintrin.h
- - - Integer - AVX512BW - Arithmetic - - - Compare packed unsigned 16-bit integers in "a" and "b", and store packed maximum values in "dst". - - -FOR j := 0 to 31 - i := j*16 - IF a[i+15:i] > b[i+15:i] - dst[i+15:i] := a[i+15:i] - ELSE - dst[i+15:i] := b[i+15:i] - FI -ENDFOR -dst[MAX:512] := 0 - - -
immintrin.h
- - - Integer - AVX512VL - AVX512BW - Arithmetic - - - - - Compare packed unsigned 16-bit integers in "a" and "b", and store packed maximum values in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -FOR j := 0 to 7 - i := j*16 - IF k[j] - IF a[i+15:i] > b[i+15:i] - dst[i+15:i] := a[i+15:i] + IF k[j*4+3] + dst[i+127:i+96] := v[i+95:i+64] ELSE - dst[i+15:i] := b[i+15:i] + dst[i+127:i+96] := src[i+127:i+96] FI - ELSE - dst[i+15:i] := src[i+15:i] - FI -ENDFOR -dst[MAX:128] := 0 - - -
immintrin.h
- - - Integer - AVX512VL - AVX512BW - Arithmetic - - - - Compare packed unsigned 16-bit integers in "a" and "b", and store packed maximum values in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - - -FOR j := 0 to 7 - i := j*16 - IF k[j] - IF a[i+15:i] > b[i+15:i] - dst[i+15:i] := a[i+15:i] + ENDFOR +_MM_SWIZ_REG_DDDD: + FOR j := 0 to 3 + i := j*128 + IF k[j*4] + dst[i+31:i] := v[i+127:i+96] ELSE - dst[i+15:i] := b[i+15:i] + dst[i+31:i] := src[i+31:i] FI - ELSE - dst[i+15:i] := 0 - FI -ENDFOR -dst[MAX:128] := 0 - - -
immintrin.h
- - - Integer - AVX512VL - AVX512BW - Arithmetic - - - - - Compare packed 8-bit integers in "a" and "b", and store packed minimum values in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -FOR j := 0 to 31 - i := j*8 - IF k[j] - IF a[i+7:i] < b[i+7:i] - dst[i+7:i] := a[i+7:i] + IF k[j*4+1] + dst[i+63:i+32] := v[i+127:i+96] + ELSE + dst[i+63:i+32] := src[i+63:i+32] + FI + IF k[j*4+2] + dst[i+95:i+64] := v[i+127:i+96] + ELSE + dst[i+95:i+64] := src[i+95:i+64] + FI + IF k[j*4+3] + dst[i+127:i+96] := v[i+127:i+96] + ELSE + dst[i+127:i+96] := src[i+127:i+96] + FI + ENDFOR +_MM_SWIZ_REG_DACB: + FOR j := 0 to 3 + i := j*128 + IF k[j*4] + dst[i+31:i] := v[i+63:i+32] + ELSE + dst[i+31:i] := src[i+31:i] + FI + IF k[j*4+1] + dst[i+63:i+32] := v[i+95:i+64] ELSE - dst[i+7:i] := b[i+7:i] + dst[i+63:i+32] := src[i+63:i+32] FI - ELSE - dst[i+7:i] := src[i+7:i] - FI -ENDFOR -dst[MAX:256] := 0 - - -
immintrin.h
- - - Integer - AVX512VL - AVX512BW - Arithmetic - - - - Compare packed 8-bit integers in "a" and "b", and store packed minimum values in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - - -FOR j := 0 to 31 - i := j*8 - IF k[j] - IF a[i+7:i] < b[i+7:i] - dst[i+7:i] := a[i+7:i] + IF k[j*4+2] + dst[i+95:i+64] := v[i+31:i] ELSE - dst[i+7:i] := b[i+7:i] + dst[i+95:i+64] := src[i+95:i+64] FI - ELSE - dst[i+7:i] := 0 - FI -ENDFOR -dst[MAX:256] := 0 - - -
immintrin.h
- - - Integer - AVX512BW - Arithmetic - - - - - Compare packed 8-bit integers in "a" and "b", and store packed minimum values in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -FOR j := 0 to 63 - i := j*8 - IF k[j] - IF a[i+7:i] < b[i+7:i] - dst[i+7:i] := a[i+7:i] + IF k[j*4+3] + dst[i+127:i+96] := v[i+127:i+96] ELSE - dst[i+7:i] := b[i+7:i] + dst[i+127:i+96] := src[i+127:i+96] FI - ELSE - dst[i+7:i] := src[i+7:i] - FI -ENDFOR + ENDFOR +ESAC dst[MAX:512] := 0 -
immintrin.h
 - + Integer - AVX512BW - Arithmetic - - - - Compare packed 8-bit integers in "a" and "b", and store packed minimum values in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - - -FOR j := 0 to 63 - i := j*8 - IF k[j] - IF a[i+7:i] < b[i+7:i] - dst[i+7:i] := a[i+7:i] + KNCNI + Swizzle + + + + + + Performs a swizzle transformation of each of the four groups of packed 4x64-bit integer elements in "v" using swizzle parameter "s", storing the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + CASE s OF +_MM_SWIZ_REG_NONE: + dst[511:0] := v[511:0] +_MM_SWIZ_REG_DCBA: + dst[511:0] := v[511:0] +_MM_SWIZ_REG_CDAB: + FOR j := 0 to 3 + i := j*64 + IF k[j*2] + dst[i+63:i] := v[i+127:i+64] ELSE - dst[i+7:i] := b[i+7:i] + dst[i+63:i] := src[i+63:i] FI - ELSE - dst[i+7:i] := 0 - FI -ENDFOR -dst[MAX:512] := 0 - - -
immintrin.h
- - - Integer - AVX512BW - Arithmetic - - - Compare packed 8-bit integers in "a" and "b", and store packed minimum values in "dst". - - -FOR j := 0 to 63 - i := j*8 - IF a[i+7:i] < b[i+7:i] - dst[i+7:i] := a[i+7:i] - ELSE - dst[i+7:i] := b[i+7:i] - FI -ENDFOR -dst[MAX:512] := 0 - - -
immintrin.h
- - - Integer - AVX512VL - AVX512BW - Arithmetic - - - - - Compare packed 8-bit integers in "a" and "b", and store packed minimum values in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -FOR j := 0 to 15 - i := j*8 - IF k[j] - IF a[i+7:i] < b[i+7:i] - dst[i+7:i] := a[i+7:i] + IF k[j*2+1] + dst[i+127:i+64] := v[i+63:i] ELSE - dst[i+7:i] := b[i+7:i] + dst[i+127:i+64] := src[i+127:i+64] FI - ELSE - dst[i+7:i] := src[i+7:i] - FI -ENDFOR -dst[MAX:128] := 0 - - -
immintrin.h
- - - Integer - AVX512VL - AVX512BW - Arithmetic - - - - Compare packed 8-bit integers in "a" and "b", and store packed minimum values in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - - -FOR j := 0 to 15 - i := j*8 - IF k[j] - IF a[i+7:i] < b[i+7:i] - dst[i+7:i] := a[i+7:i] + ENDFOR +_MM_SWIZ_REG_BADC: + FOR j := 0 to 1 + i := j*256 + IF k[j*4] + dst[i+63:i] := v[i+191:i+128] ELSE - dst[i+7:i] := b[i+7:i] + dst[i+63:i] := src[i+63:i] FI - ELSE - dst[i+7:i] := 0 - FI -ENDFOR -dst[MAX:128] := 0 - - -
immintrin.h
- - - Integer - AVX512VL - AVX512F - Arithmetic - - - - - Compare packed 32-bit integers in "a" and "b", and store packed minimum values in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -FOR j := 0 to 7 - i := j*32 - IF k[j] - IF a[i+31:i] < b[i+31:i] - dst[i+31:i] := a[i+31:i] + IF k[j*4+1] + dst[i+127:i+64] := v[i+255:i+192] ELSE - dst[i+31:i] := b[i+31:i] + dst[i+127:i+64] := src[i+127:i+64] FI - ELSE - dst[i+31:i] := src[i+31:i] - FI -ENDFOR -dst[MAX:256] := 0 - - -
immintrin.h
- - - Integer - AVX512VL - AVX512F - Arithmetic - - - - Compare packed 32-bit integers in "a" and "b", and store packed minimum values in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - - -FOR j := 0 to 7 - i := j*32 - IF k[j] - IF a[i+31:i] < b[i+31:i] - dst[i+31:i] := a[i+31:i] + IF k[j*4+2] + dst[i+191:i+128] := v[i+63:i] ELSE - dst[i+31:i] := b[i+31:i] + dst[i+191:i+128] := src[i+191:i+128] FI - ELSE - dst[i+31:i] := 0 - FI -ENDFOR -dst[MAX:256] := 0 - - -
immintrin.h
- - - Integer - AVX512VL - AVX512F - Arithmetic - - - - - Compare packed 32-bit integers in "a" and "b", and store packed minimum values in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -FOR j := 0 to 3 - i := j*32 - IF k[j] - IF a[i+31:i] < b[i+31:i] - dst[i+31:i] := a[i+31:i] + IF k[j*4+3] + dst[i+255:i+192] := v[i+127:i+64] ELSE - dst[i+31:i] := b[i+31:i] + dst[i+255:i+192] := src[i+255:i+192] FI - ELSE - dst[i+31:i] := src[i+31:i] - FI -ENDFOR -dst[MAX:128] := 0 - - -
immintrin.h
- - - Integer - AVX512VL - AVX512F - Arithmetic - - - - Compare packed 32-bit integers in "a" and "b", and store packed minimum values in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - - -FOR j := 0 to 3 - i := j*32 - IF k[j] - IF a[i+31:i] < b[i+31:i] - dst[i+31:i] := a[i+31:i] + ENDFOR +_MM_SWIZ_REG_AAAA: + FOR j := 0 to 1 + i := j*256 + IF k[j*4] + dst[i+63:i] := v[i+63:i] ELSE - dst[i+31:i] := b[i+31:i] + dst[i+63:i] := src[i+63:i] FI - ELSE - dst[i+31:i] := 0 - FI -ENDFOR -dst[MAX:128] := 0 - - -
immintrin.h
- - - Integer - AVX512VL - AVX512F - Arithmetic - - - - - Compare packed 64-bit integers in "a" and "b", and store packed minimum values in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -FOR j := 0 to 3 - i := j*64 - IF k[j] - IF a[i+63:i] < b[i+63:i] - dst[i+63:i] := a[i+63:i] + IF k[j*4+1] + dst[i+127:i+64] := v[i+63:i] ELSE - dst[i+63:i] := b[i+63:i] + dst[i+127:i+64] := src[i+127:i+64] FI - ELSE - dst[i+63:i] := src[i+63:i] - FI -ENDFOR -dst[MAX:256] := 0 - - -
immintrin.h
- - - Integer - AVX512VL - AVX512F - Arithmetic - - - - Compare packed 64-bit integers in "a" and "b", and store packed minimum values in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - - -FOR j := 0 to 3 - i := j*64 - IF k[j] - IF a[i+63:i] < b[i+63:i] - dst[i+63:i] := a[i+63:i] + IF k[j*4+2] + dst[i+191:i+128] := v[i+63:i] ELSE - dst[i+63:i] := b[i+63:i] + dst[i+191:i+128] := src[i+191:i+128] FI - ELSE - dst[i+63:i] := 0 - FI -ENDFOR -dst[MAX:256] := 0 - - -
immintrin.h
- - - Integer - AVX512VL - AVX512F - Arithmetic - - - Compare packed 64-bit integers in "a" and "b", and store packed minimum values in "dst". - -FOR j := 0 to 3 - i := j*64 - IF a[i+63:i] < b[i+63:i] - dst[i+63:i] := a[i+63:i] - ELSE - dst[i+63:i] := b[i+63:i] - FI -ENDFOR -dst[MAX:256] := 0 - - -
immintrin.h
- - - Integer - AVX512VL - AVX512F - Arithmetic - - - - - Compare packed 64-bit integers in "a" and "b", and store packed minimum values in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -FOR j := 0 to 1 - i := j*64 - IF k[j] - IF a[i+63:i] < b[i+63:i] - dst[i+63:i] := a[i+63:i] + IF k[j*4+3] + dst[i+255:i+192] := v[i+63:i] ELSE - dst[i+63:i] := b[i+63:i] + dst[i+255:i+192] := src[i+255:i+192] FI - ELSE - dst[i+63:i] := src[i+63:i] - FI -ENDFOR -dst[MAX:128] := 0 - - -
immintrin.h
- - - Integer - AVX512VL - AVX512F - Arithmetic - - - - Compare packed 64-bit integers in "a" and "b", and store packed minimum values in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - - -FOR j := 0 to 1 - i := j*64 - IF k[j] - IF a[i+63:i] < b[i+63:i] - dst[i+63:i] := a[i+63:i] + ENDFOR +_MM_SWIZ_REG_BBBB: + FOR j := 0 to 1 + i := j*256 + IF k[j*4] + dst[i+63:i] := v[i+127:i+63] ELSE - dst[i+63:i] := b[i+63:i] + dst[i+63:i] := src[i+63:i] FI - ELSE - dst[i+63:i] := 0 - FI -ENDFOR -dst[MAX:128] := 0 - - -
immintrin.h
- - - Integer - AVX512VL - AVX512F - Arithmetic - - - Compare packed 64-bit integers in "a" and "b", and store packed minimum values in "dst". - -FOR j := 0 to 1 - i := j*64 - IF a[i+63:i] < b[i+63:i] - dst[i+63:i] := a[i+63:i] - ELSE - dst[i+63:i] := b[i+63:i] - FI -ENDFOR -dst[MAX:128] := 0 - - -
immintrin.h
- - - Integer - AVX512VL - AVX512BW - Arithmetic - - - - - Compare packed 16-bit integers in "a" and "b", and store packed minimum values in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -FOR j := 0 to 15 - i := j*16 - IF k[j] - IF a[i+15:i] < b[i+15:i] - dst[i+15:i] := a[i+15:i] + IF k[j*4+1] + dst[i+127:i+64] := v[i+127:i+63] ELSE - dst[i+15:i] := b[i+15:i] + dst[i+127:i+64] := src[i+127:i+64] FI - ELSE - dst[i+15:i] := src[i+15:i] - FI -ENDFOR -dst[MAX:256] := 0 - - -
immintrin.h
- - - Integer - AVX512VL - AVX512BW - Arithmetic - - - - Compare packed 16-bit integers in "a" and "b", and store packed minimum values in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - - -FOR j := 0 to 15 - i := j*16 - IF k[j] - IF a[i+15:i] < b[i+15:i] - dst[i+15:i] := a[i+15:i] + IF k[j*4+2] + dst[i+191:i+128] := v[i+127:i+63] ELSE - dst[i+15:i] := b[i+15:i] + dst[i+191:i+128] := src[i+191:i+128] FI - ELSE - dst[i+15:i] := 0 - FI -ENDFOR -dst[MAX:256] := 0 - - -
immintrin.h
- - - Integer - AVX512BW - Arithmetic - - - - - Compare packed 16-bit integers in "a" and "b", and store packed minimum values in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -FOR j := 0 to 31 - i := j*16 - IF k[j] - IF a[i+15:i] < b[i+15:i] - dst[i+15:i] := a[i+15:i] + IF k[j*4+3] + dst[i+255:i+192] := v[i+127:i+63] ELSE - dst[i+15:i] := b[i+15:i] + dst[i+255:i+192] := src[i+255:i+192] FI - ELSE - dst[i+15:i] := src[i+15:i] - FI -ENDFOR -dst[MAX:512] := 0 - - -
immintrin.h
- - - Integer - AVX512BW - Arithmetic - - - - Compare packed 16-bit integers in "a" and "b", and store packed minimum values in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - - -FOR j := 0 to 31 - i := j*16 - IF k[j] - IF a[i+15:i] < b[i+15:i] - dst[i+15:i] := a[i+15:i] + ENDFOR +_MM_SWIZ_REG_CCCC: + FOR j := 0 to 1 + i := j*256 + IF k[j*4] + dst[i+63:i] := v[i+191:i+128] ELSE - dst[i+15:i] := b[i+15:i] + dst[i+63:i] := src[i+63:i] FI - ELSE - dst[i+15:i] := 0 - FI -ENDFOR -dst[MAX:512] := 0 - - -
immintrin.h
- - - Integer - AVX512BW - Arithmetic - - - Compare packed 16-bit integers in "a" and "b", and store packed minimum values in "dst". - - -FOR j := 0 to 31 - i := j*16 - IF a[i+15:i] < b[i+15:i] - dst[i+15:i] := a[i+15:i] - ELSE - dst[i+15:i] := b[i+15:i] - FI -ENDFOR -dst[MAX:512] := 0 - - -
immintrin.h
- - - Integer - AVX512VL - AVX512BW - Arithmetic - - - - - Compare packed 16-bit integers in "a" and "b", and store packed minimum values in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -FOR j := 0 to 7 - i := j*16 - IF k[j] - IF a[i+15:i] < b[i+15:i] - dst[i+15:i] := a[i+15:i] + IF k[j*4+1] + dst[i+127:i+64] := v[i+191:i+128] ELSE - dst[i+15:i] := b[i+15:i] + dst[i+127:i+64] := src[i+127:i+64] FI - ELSE - dst[i+15:i] := src[i+15:i] - FI -ENDFOR -dst[MAX:128] := 0 - - -
immintrin.h
- - - Integer - AVX512VL - AVX512BW - Arithmetic - - - - Compare packed 16-bit integers in "a" and "b", and store packed minimum values in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - - -FOR j := 0 to 7 - i := j*16 - IF k[j] - IF a[i+15:i] < b[i+15:i] - dst[i+15:i] := a[i+15:i] + IF k[j*4+2] + dst[i+191:i+128] := v[i+191:i+128] ELSE - dst[i+15:i] := b[i+15:i] + dst[i+191:i+128] := src[i+191:i+128] FI - ELSE - dst[i+15:i] := 0 - FI -ENDFOR -dst[MAX:128] := 0 - - -
immintrin.h
- - - Integer - AVX512VL - AVX512BW - Arithmetic - - - - - Compare packed unsigned 8-bit integers in "a" and "b", and store packed minimum values in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -FOR j := 0 to 31 - i := j*8 - IF k[j] - IF a[i+7:i] < b[i+7:i] - dst[i+7:i] := a[i+7:i] + IF k[j*4+3] + dst[i+255:i+192] := v[i+191:i+128] ELSE - dst[i+7:i] := b[i+7:i] + dst[i+255:i+192] := src[i+255:i+192] FI - ELSE - dst[i+7:i] := src[i+7:i] - FI -ENDFOR -dst[MAX:256] := 0 - - -
immintrin.h
- - - Integer - AVX512VL - AVX512BW - Arithmetic - - - - Compare packed unsigned 8-bit integers in "a" and "b", and store packed minimum values in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - - -FOR j := 0 to 31 - i := j*8 - IF k[j] - IF a[i+7:i] < b[i+7:i] - dst[i+7:i] := a[i+7:i] + ENDFOR +_MM_SWIZ_REG_DDDD: + FOR j := 0 to 1 + i := j*256 + IF k[j*4] + dst[i+63:i] := v[i+255:i+192] + ELSE + dst[i+63:i] := src[i+63:i] + FI + IF k[j*4+1] + dst[i+127:i+64] := v[i+255:i+192] + ELSE + dst[i+127:i+64] := src[i+127:i+64] + FI + IF k[j*4+2] + dst[i+191:i+128] := v[i+255:i+192] + ELSE + dst[i+191:i+128] := src[i+191:i+128] + FI + IF k[j*4+3] + dst[i+255:i+192] := v[i+255:i+192] + ELSE + dst[i+255:i+192] := src[i+255:i+192] + FI + ENDFOR +_MM_SWIZ_REG_DACB: + FOR j := 0 to 1 + i := j*256 + IF k[j*4] + dst[i+63:i] := v[i+127:i+64] + ELSE + dst[i+63:i] := src[i+63:i] + FI + IF k[j*4+1] + dst[i+127:i+64] := v[i+191:i+128] ELSE - dst[i+7:i] := b[i+7:i] + dst[i+127:i+64] := src[i+127:i+64] FI - ELSE - dst[i+7:i] := 0 - FI -ENDFOR -dst[MAX:256] := 0 - - -
immintrin.h
- - - Integer - AVX512BW - Arithmetic - - - - - Compare packed unsigned 8-bit integers in "a" and "b", and store packed minimum values in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -FOR j := 0 to 63 - i := j*8 - IF k[j] - IF a[i+7:i] < b[i+7:i] - dst[i+7:i] := a[i+7:i] + IF k[j*4+2] + dst[i+191:i+128] := v[i+63:i] ELSE - dst[i+7:i] := b[i+7:i] + dst[i+191:i+128] := src[i+191:i+128] FI - ELSE - dst[i+7:i] := src[i+7:i] - FI -ENDFOR -dst[MAX:512] := 0 - - -
immintrin.h
- - - Integer - AVX512BW - Arithmetic - - - - Compare packed unsigned 8-bit integers in "a" and "b", and store packed minimum values in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - - -FOR j := 0 to 63 - i := j*8 - IF k[j] - IF a[i+7:i] < b[i+7:i] - dst[i+7:i] := a[i+7:i] + IF k[j*4+3] + dst[i+255:i+192] := v[i+255:i+192] ELSE - dst[i+7:i] := b[i+7:i] + dst[i+255:i+192] := src[i+255:i+192] FI - ELSE - dst[i+7:i] := 0 - FI -ENDFOR -dst[MAX:512] := 0 - - -
immintrin.h
- - - Integer - AVX512BW - Arithmetic - - - Compare packed unsigned 8-bit integers in "a" and "b", and store packed minimum values in "dst". - - -FOR j := 0 to 63 - i := j*8 - IF a[i+7:i] < b[i+7:i] - dst[i+7:i] := a[i+7:i] - ELSE - dst[i+7:i] := b[i+7:i] - FI -ENDFOR + ENDFOR +ESAC dst[MAX:512] := 0 -
immintrin.h
 - - Integer - AVX512VL - AVX512BW - Arithmetic - - - - - Compare packed unsigned 8-bit integers in "a" and "b", and store packed minimum values in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Floating Point + KNCNI + Store + + + + + + Downconverts packed single-precision (32-bit) floating-point elements stored in "v" to a smaller type depending on "conv" and stores them in memory location "mt". "hint" indicates to the processor whether the data is non-temporal. +addr := MEM[mt] FOR j := 0 to 15 - i := j*8 - IF k[j] - IF a[i+7:i] < b[i+7:i] - dst[i+7:i] := a[i+7:i] - ELSE - dst[i+7:i] := b[i+7:i] - FI - ELSE - dst[i+7:i] := src[i+7:i] - FI + i := j*32 + CASE conv OF + _MM_DOWNCONV_PS_NONE: + addr[i+31:i] := v[i+31:i] + _MM_DOWNCONV_PS_FLOAT16: + n := j*16 + addr[n+15:n] := Convert_FP32_To_FP16(v[i+31:i]) + _MM_DOWNCONV_PS_UINT8: + n := j*8 + addr[n+7:n] := Convert_FP32_To_UInt8(v[i+31:i]) + _MM_DOWNCONV_PS_SINT8: + n := j*8 + addr[n+7:n] := Convert_FP32_To_Int8(v[i+31:i]) + _MM_DOWNCONV_PS_UINT16: + n := j*16 + addr[n+15:n] := Convert_FP32_To_UInt16(v[i+31:i]) + _MM_DOWNCONV_PS_SINT16: + n := j*16 + addr[n+15:n] := Convert_FP32_To_Int16(v[i+31:i]) + ESAC ENDFOR -dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - AVX512VL - AVX512BW - Arithmetic - - - - Compare packed unsigned 8-bit integers in "a" and "b", and store packed minimum values in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - - + KNCNI + Store + + + + + + Downconverts packed 32-bit integer elements stored in "v" to a smaller type depending on "conv" and stores them in memory location "mt". "hint" indicates to the processor whether the data is non-temporal. + addr := MEM[mt] FOR j := 0 to 15 - i := j*8 - IF k[j] - IF a[i+7:i] < b[i+7:i] - dst[i+7:i] := a[i+7:i] - ELSE - dst[i+7:i] := b[i+7:i] - FI - ELSE - dst[i+7:i] := 0 - FI -ENDFOR -dst[MAX:128] := 0 - - -
immintrin.h
- - - Integer - AVX512VL - AVX512F - Arithmetic - - - - - Compare packed unsigned 32-bit integers in "a" and "b", and store packed minimum values in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -FOR j := 0 to 7 i := j*32 - IF k[j] - IF a[i+31:i] < b[i+31:i] - dst[i+31:i] := a[i+31:i] - ELSE - dst[i+31:i] := b[i+31:i] - FI - ELSE - dst[i+31:i] := src[i+31:i] - FI + CASE conv OF + _MM_DOWNCONV_EPI32_NONE: + addr[i+31:i] := v[i+31:i] + _MM_DOWNCONV_EPI32_UINT8: + n := j*8 + addr[n+7:n] := Int32ToUInt8(v[i+31:i]) + _MM_DOWNCONV_EPI32_SINT8: + n := j*8 + addr[n+7:n] := Int32ToSInt8(v[i+31:i]) + _MM_DOWNCONV_EPI32_UINT16: + n := j*16 + addr[n+15:n] := Int32ToUInt16(v[i+31:i]) + _MM_DOWNCONV_EPI32_SINT16: + n := j*16 + addr[n+15:n] := Int32ToSInt16(v[i+31:i]) + ESAC ENDFOR -dst[MAX:256] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Arithmetic - - - - Compare packed unsigned 32-bit integers in "a" and "b", and store packed minimum values in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - + + Floating Point + KNCNI + Store + + + + + + Downconverts packed double-precision (64-bit) floating-point elements stored in "v" to a smaller type depending on "conv" and stores them in memory location "mt". "hint" indicates to the processor whether the data is non-temporal. +addr := MEM[mt] FOR j := 0 to 7 - i := j*32 - IF k[j] - IF a[i+31:i] < b[i+31:i] - dst[i+31:i] := a[i+31:i] - ELSE - dst[i+31:i] := b[i+31:i] - FI - ELSE - dst[i+31:i] := 0 - FI + i := j*64 + CASE conv OF + _MM_DOWNCONV_PS_NONE: + addr[i+63:i] := v[i+63:i] + ESAC ENDFOR -dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - AVX512VL - AVX512F - Arithmetic - - - - - Compare packed unsigned 32-bit integers in "a" and "b", and store packed minimum values in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + KNCNI + Store + + + + + + Downconverts packed 64-bit integer elements stored in "v" to a smaller type depending on "conv" and stores them in memory location "mt". "hint" indicates to the processor whether the data is non-temporal. -FOR j := 0 to 3 - i := j*32 - IF k[j] - IF a[i+31:i] < b[i+31:i] - dst[i+31:i] := a[i+31:i] - ELSE - dst[i+31:i] := b[i+31:i] - FI - ELSE - dst[i+31:i] := src[i+31:i] - FI +addr := MEM[mt] +FOR j := 0 to 7 + i := j*64 + CASE conv OF + _MM_DOWNCONV_EPI64_NONE: addr[i+63:i] := v[i+63:i] + ESAC ENDFOR -dst[MAX:128] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Arithmetic - - - - Compare packed unsigned 32-bit integers in "a" and "b", and store packed minimum values in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - + + Floating Point + KNCNI + Store + + + + + + + Downconverts packed single-precision (32-bit) floating-point elements stored in "v" to a smaller type depending on "conv" and stores them in memory location "mt" using writemask "k" (elements are not written to memory when the corresponding mask bit is not set). "hint" indicates to the processor whether the data is non-temporal. -FOR j := 0 to 3 +FOR j := 0 to 15 i := j*32 IF k[j] - IF a[i+31:i] < b[i+31:i] - dst[i+31:i] := a[i+31:i] - ELSE - dst[i+31:i] := b[i+31:i] - FI - ELSE - dst[i+31:i] := 0 - FI + CASE conv OF + _MM_DOWNCONV_PS_NONE: + mt[i+31:i] := v[i+31:i] + _MM_DOWNCONV_PS_FLOAT16: + n := j*16 + mt[n+15:n] := Convert_FP32_To_FP16(v[i+31:i]) + _MM_DOWNCONV_PS_UINT8: + n := j*8 + mt[n+7:n] := Convert_FP32_To_UInt8(v[i+31:i]) + _MM_DOWNCONV_PS_SINT8: + n := j*8 + mt[n+7:n] := Convert_FP32_To_Int8(v[i+31:i]) + _MM_DOWNCONV_PS_UINT16: + n := j*16 + mt[n+15:n] := Convert_FP32_To_UInt16(v[i+31:i]) + _MM_DOWNCONV_PS_SINT16: + n := j*16 + mt[n+15:n] := Convert_FP32_To_Int16(v[i+31:i]) + ESAC + FI ENDFOR -dst[MAX:128] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Arithmetic - - - - - Compare packed unsigned 64-bit integers in "a" and "b", and store packed minimum values in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Floating Point + KNCNI + Store + + + + + + + Downconverts packed double-precision (64-bit) floating-point elements stored in "v" to a smaller type depending on "conv" and stores them in memory location "mt" (elements in "mt" are unaltered when the corresponding mask bit is not set). "hint" indicates to the processor whether the data is non-temporal. -FOR j := 0 to 3 +addr := MEM[mt] +FOR j := 0 to 7 i := j*64 - IF k[j] - IF a[i+63:i] < b[i+63:i] - dst[i+63:i] := a[i+63:i] - ELSE - dst[i+63:i] := b[i+63:i] + CASE conv OF + _MM_DOWNCONV_PD_NONE: + IF k[j] + mt[i+63:i] := v[i+63:i] FI - ELSE - dst[i+63:i] := src[i+63:i] - FI + ESAC ENDFOR -dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - AVX512VL - AVX512F - Arithmetic - - - - Compare packed unsigned 64-bit integers in "a" and "b", and store packed minimum values in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - - -FOR j := 0 to 3 - i := j*64 + KNCNI + Store + + + + + + + Downconverts packed 32-bit integer elements stored in "v" to a smaller type depending on "conv" and stores them in memory location "mt" (elements in "mt" are unaltered when the corresponding mask bit is not set). "hint" indicates to the processor whether the data is non-temporal. + addr := MEM[mt] +FOR j := 0 to 15 + i := j*32 IF k[j] - IF a[i+63:i] < b[i+63:i] - dst[i+63:i] := a[i+63:i] - ELSE - dst[i+63:i] := b[i+63:i] - FI - ELSE - dst[i+63:i] := 0 - FI -ENDFOR -dst[MAX:256] := 0 - - -
immintrin.h
- - - Integer - AVX512VL - AVX512F - Arithmetic - - - Compare packed unsigned 64-bit integers in "a" and "b", and store packed minimum values in "dst". - -FOR j := 0 to 3 - i := j*64 - IF a[i+63:i] < b[i+63:i] - dst[i+63:i] := a[i+63:i] - ELSE - dst[i+63:i] := b[i+63:i] + CASE conv OF + _MM_DOWNCONV_EPI32_NONE: + addr[i+31:i] := v[i+31:i] + _MM_DOWNCONV_EPI32_UINT8: + n := j*8 + addr[n+7:n] := Int32ToUInt8(v[i+31:i]) + _MM_DOWNCONV_EPI32_SINT8: + n := j*8 + addr[n+7:n] := Int32ToSInt8(v[i+31:i]) + _MM_DOWNCONV_EPI32_UINT16: + n := j*16 + addr[n+15:n] := Int32ToUInt16(v[i+31:i]) + _MM_DOWNCONV_EPI32_SINT16: + n := j*16 + addr[n+15:n] := Int32ToSInt16(v[i+31:i]) + ESAC FI ENDFOR -dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - AVX512VL - AVX512F - Arithmetic - - - - - Compare packed unsigned 64-bit integers in "a" and "b", and store packed minimum values in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + KNCNI + Store + + + + + + + Downconverts packed 64-bit integer elements stored in "v" to a smaller type depending on "conv" and stores them in memory location "mt" (elements in "mt" are unaltered when the corresponding mask bit is not set). "hint" indicates to the processor whether the data is non-temporal. -FOR j := 0 to 1 +addr := MEM[mt] +FOR j := 0 to 7 i := j*64 IF k[j] - IF a[i+63:i] < b[i+63:i] - dst[i+63:i] := a[i+63:i] - ELSE - dst[i+63:i] := b[i+63:i] - FI - ELSE - dst[i+63:i] := src[i+63:i] + CASE conv OF + _MM_DOWNCONV_EPI64_NONE: addr[i+63:i] := v[i+63:i] + ESAC FI ENDFOR -dst[MAX:128] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Arithmetic - - - - Compare packed unsigned 64-bit integers in "a" and "b", and store packed minimum values in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - + + Floating Point + KNCNI + Store + + + + Stores packed single-precision (32-bit) floating-point elements from "v" to memory address "mt" with a no-read hint to the processor. -FOR j := 0 to 1 - i := j*64 - IF k[j] - IF a[i+63:i] < b[i+63:i] - dst[i+63:i] := a[i+63:i] - ELSE - dst[i+63:i] := b[i+63:i] - FI - ELSE - dst[i+63:i] := 0 - FI +addr := MEM[mt] +FOR j := 0 to 15 + i := j*32 + addr[i+31:i] := v[i+31:i] ENDFOR -dst[MAX:128] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Arithmetic - - - Compare packed unsigned 64-bit integers in "a" and "b", and store packed minimum values in "dst". + + Floating Point + KNCNI + Store + + + + Stores packed double-precision (64-bit) floating-point elements from "v" to memory address "mt" with a no-read hint to the processor. -FOR j := 0 to 1 +addr := MEM[mt] +FOR j := 0 to 7 i := j*64 - IF a[i+63:i] < b[i+63:i] - dst[i+63:i] := a[i+63:i] - ELSE - dst[i+63:i] := b[i+63:i] - FI -ENDFOR -dst[MAX:128] := 0 - - -
immintrin.h
- - - Integer - AVX512VL - AVX512BW - Arithmetic - - - - - Compare packed unsigned 16-bit integers in "a" and "b", and store packed minimum values in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -FOR j := 0 to 15 - i := j*16 - IF k[j] - IF a[i+15:i] < b[i+15:i] - dst[i+15:i] := a[i+15:i] - ELSE - dst[i+15:i] := b[i+15:i] - FI - ELSE - dst[i+15:i] := src[i+15:i] - FI + addr[i+63:i] := v[i+63:i] ENDFOR -dst[MAX:256] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512BW - Arithmetic - - - - Compare packed unsigned 16-bit integers in "a" and "b", and store packed minimum values in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - + + Floating Point + KNCNI + Store + + + + Stores packed single-precision (32-bit) floating-point elements from "v" to memory address "mt" with a no-read hint and using a weakly-ordered memory consistency model (stores performed with this function are not globally ordered, and subsequent stores from the same thread can be observed before them). +addr := MEM[mt] FOR j := 0 to 15 - i := j*16 - IF k[j] - IF a[i+15:i] < b[i+15:i] - dst[i+15:i] := a[i+15:i] - ELSE - dst[i+15:i] := b[i+15:i] - FI - ELSE - dst[i+15:i] := 0 - FI + i := j*32 + addr[i+31:i] := v[i+31:i] ENDFOR -dst[MAX:256] := 0 - +
immintrin.h
 - - Integer - AVX512BW - Arithmetic - - - - - Compare packed unsigned 16-bit integers in "a" and "b", and store packed minimum values in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Floating Point + KNCNI + Store + + + + Stores packed double-precision (64-bit) floating-point elements from "v" to memory address "mt" with a no-read hint and using a weakly-ordered memory consistency model (stores performed with this function are not globally ordered, and subsequent stores from the same thread can be observed before them). -FOR j := 0 to 31 - i := j*16 - IF k[j] - IF a[i+15:i] < b[i+15:i] - dst[i+15:i] := a[i+15:i] - ELSE - dst[i+15:i] := b[i+15:i] - FI - ELSE - dst[i+15:i] := src[i+15:i] - FI +addr := MEM[mt] +FOR j := 0 to 7 + i := j*64 + addr[i+63:i] := v[i+63:i] ENDFOR -dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX512BW + KNCNI Arithmetic - - - - Compare packed unsigned 16-bit integers in "a" and "b", and store packed minimum values in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - - -FOR j := 0 to 31 - i := j*16 - IF k[j] - IF a[i+15:i] < b[i+15:i] - dst[i+15:i] := a[i+15:i] - ELSE - dst[i+15:i] := b[i+15:i] - FI - ELSE - dst[i+15:i] := 0 - FI + + + + + + Performs element-by-element addition of packed 32-bit integers in "v2" and "v3" and the corresponding bit in "k2", storing the result of the addition in "dst" and the result of the carry in "k2_res". + FOR j := 0 to 15 + i := j*32 + k2_res[j] := Carry(v2[i+31:i] + v3[i+31:i] + k2[j]) + dst[i+31:i] := v2[i+31:i] + v3[i+31:i] + k2[j] ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX512BW + KNCNI Arithmetic - - - Compare packed unsigned 16-bit integers in "a" and "b", and store packed minimum values in "dst". - - -FOR j := 0 to 31 - i := j*16 - IF a[i+15:i] < b[i+15:i] - dst[i+15:i] := a[i+15:i] + + + + + + + Performs element-by-element addition of packed 32-bit integers in "v2" and "v3" and the corresponding bit in "k2", storing the result of the addition in "dst" and the result of the carry in "k2_res" using writemask "k1" (elements are copied from "v2" when the corresponding mask bit is not set). + FOR j := 0 to 15 + i := j*32 + IF k1[j] + k2_res[j] := Carry(v2[i+31:i] + v3[i+31:i] + k2[j]) + dst[i+31:i] := v2[i+31:i] + v3[i+31:i] + k2[j] ELSE - dst[i+15:i] := b[i+15:i] + dst[i+31:i] := v2[i+31:i] FI ENDFOR dst[MAX:512] := 0 - -
immintrin.h
- - - Integer - AVX512VL - AVX512BW - Arithmetic - - - - - Compare packed unsigned 16-bit integers in "a" and "b", and store packed minimum values in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -FOR j := 0 to 7 - i := j*16 - IF k[j] - IF a[i+15:i] < b[i+15:i] - dst[i+15:i] := a[i+15:i] - ELSE - dst[i+15:i] := b[i+15:i] - FI - ELSE - dst[i+15:i] := src[i+15:i] - FI -ENDFOR -dst[MAX:128] := 0 - - -
immintrin.h
- - - Integer - AVX512VL - AVX512BW - Arithmetic - - - - Compare packed unsigned 16-bit integers in "a" and "b", and store packed minimum values in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - - -FOR j := 0 to 7 - i := j*16 - IF k[j] - IF a[i+15:i] < b[i+15:i] - dst[i+15:i] := a[i+15:i] - ELSE - dst[i+15:i] := b[i+15:i] - FI - ELSE - dst[i+15:i] := 0 - FI -ENDFOR -dst[MAX:128] := 0 - - -
immintrin.h
- - - Integer - Mask - AVX512VL - AVX512BW - Miscellaneous - - Set each bit of mask register "k" based on the most significant bit of the corresponding packed 8-bit integer in "a". - -FOR j := 0 to 31 - i := j*8 - IF a[i+7] - k[j] := 1 - ELSE - k[j] := 0 - FI -ENDFOR -k[MAX:32] := 0 - - +
immintrin.h
 - - Integer - Mask - AVX512BW - Miscellaneous - - Set each bit of mask register "k" based on the most significant bit of the corresponding packed 8-bit integer in "a". - -FOR j := 0 to 63 - i := j*8 - IF a[i+7] - k[j] := 1 - ELSE - k[j] := 0 - FI -ENDFOR -k[MAX:64] := 0 - - -
immintrin.h
- - - Integer - Mask - AVX512VL - AVX512BW - Miscellaneous - - Set each bit of mask register "k" based on the most significant bit of the corresponding packed 8-bit integer in "a". - -FOR j := 0 to 15 - i := j*8 - IF a[i+7] - k[j] := 1 - ELSE - k[j] := 0 - FI + + Floating Point + KNCNI + Arithmetic + + + + Performs element-by-element addition between packed double-precision (64-bit) floating-point elements in "v2" and "v3" and negates their sum, storing the results in "dst". + +FOR j := 0 to 7 + i := j*64 + dst[i+63:i] := -(v2[i+63:i] + v3[i+63:i]) ENDFOR -k[MAX:16] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - Mask - AVX512VL - AVX512DQ - Miscellaneous - - Set each bit of mask register "k" based on the most significant bit of the corresponding packed 32-bit integer in "a". + + Floating Point + KNCNI + Arithmetic + + + + + + Performs element-by-element addition between packed double-precision (64-bit) floating-point elements in "v2" and "v3" and negates their sum, storing the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). FOR j := 0 to 7 - i := j*32 - IF a[i+31] - k[j] := 1 + i := j*64 + IF k[j] + dst[i+63:i] := -(v2[i+63:i] + v3[i+63:i]) ELSE - k[j] := 0 + dst[i+63:i] := src[i+63:i] FI ENDFOR -k[MAX:8] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - Mask - AVX512DQ - Miscellaneous - - Set each bit of mask register "k" based on the most significant bit of the corresponding packed 32-bit integer in "a". + + Floating Point + KNCNI + Arithmetic + + + + Performs element-by-element addition between packed single-precision (32-bit) floating-point elements in "v2" and "v3" and negates their sum, storing the results in "dst". FOR j := 0 to 15 i := j*32 - IF a[i+31] - k[j] := 1 - ELSE - k[j] := 0 - FI + dst[i+31:i] := -(v2[i+31:i] + v3[i+31:i]) ENDFOR -k[MAX:16] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - Mask - AVX512VL - AVX512DQ - Miscellaneous - - Set each bit of mask register "k" based on the most significant bit of the corresponding packed 32-bit integer in "a". + + Floating Point + KNCNI + Arithmetic + + + + + + Performs element-by-element addition between packed single-precision (32-bit) floating-point elements in "v2" and "v3" and negates their sum, storing the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 3 +FOR j := 0 to 15 i := j*32 - IF a[i+31] - k[j] := 1 + IF k[j] + dst[i+31:i] := -(v2[i+31:i] + v3[i+31:i]) ELSE - k[j] := 0 + dst[i+31:i] := src[i+31:i] FI ENDFOR -k[MAX:4] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Convert - - Convert packed 32-bit integers in "a" to packed 8-bit integers with truncation, and store the results in "dst". + + Floating Point + KNCNI + Arithmetic + + + + + Performs element by element addition between packed double-precision (64-bit) floating-point elements in "v2" and "v3" and negates the sum, storing the result in "dst". + [round_note] FOR j := 0 to 7 - i := 32*j - k := 8*j - dst[k+7:k] := Truncate_Int32_To_Int8(a[i+31:i]) + i := j*64 + dst[i+63:i] := -(v2[i+63:i] + v3[i+63:i]) ENDFOR -dst[MAX:64] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Convert - - - - Convert packed 32-bit integers in "a" to packed 8-bit integers with truncation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Floating Point + KNCNI + Arithmetic + + + + + + + Performs element by element addition between packed double-precision (64-bit) floating-point elements in "v2" and "v3" and negates the sum, storing the result in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + [round_note] FOR j := 0 to 7 - i := 32*j - l := 8*j + i := j*64 IF k[j] - dst[l+7:l] := Truncate_Int32_To_Int8(a[i+31:i]) + dst[i+63:i] := -(v2[i+63:i] + v3[i+63:i]) ELSE - dst[l+7:l] := src[l+7:l] + dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:64] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Convert - Store - - - - Convert packed 32-bit integers in "a" to packed 8-bit integers with truncation, and store the active results (those with their respective bit set in writemask "k") to unaligned memory at "base_addr". + + Floating Point + KNCNI + Arithmetic + + + + + Performs element by element addition between packed single-precision (32-bit) floating-point elements in "v2" and "v3" and negates the sum, storing the result in "dst". + [round_note] -FOR j := 0 to 7 - i := 32*j - l := 8*j - IF k[j] - MEM[base_addr+l+7:base_addr+l] := Truncate_Int32_To_Int8(a[i+31:i]) - FI +FOR j := 0 to 15 + i := j*32 + dst[i+31:i] := -(v2[i+31:i] + v3[i+31:i]) ENDFOR -dst[MAX:64] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Convert - - - Convert packed 32-bit integers in "a" to packed 8-bit integers with truncation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Floating Point + KNCNI + Arithmetic + + + + + + + Performs element by element addition between packed single-precision (32-bit) floating-point elements in "v2" and "v3" and negates the sum, storing the result in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + [round_note] -FOR j := 0 to 7 - i := 32*j - l := 8*j +FOR j := 0 to 15 + i := j*32 IF k[j] - dst[l+7:l] := Truncate_Int32_To_Int8(a[i+31:i]) + dst[i+31:i] := -(v2[i+31:i] + v3[i+31:i]) ELSE - dst[l+7:l] := 0 + dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:64] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Convert - - Convert packed 32-bit integers in "a" to packed 8-bit integers with truncation, and store the results in "dst". + + Floating Point + KNCNI + Arithmetic + + + + Performs element-by-element subtraction of packed double-precision (64-bit) floating-point elements in "v2" from "v3" storing the results in "dst". -FOR j := 0 to 3 - i := 32*j - k := 8*j - dst[k+7:k] := Truncate_Int32_To_Int8(a[i+31:i]) +FOR j := 0 to 7 + i := j*64 + dst[i+63:i] := v3[i+63:i] - v2[i+63:i] ENDFOR -dst[MAX:32] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Convert - - - - Convert packed 32-bit integers in "a" to packed 8-bit integers with truncation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Floating Point + KNCNI + Arithmetic + + + + + + Performs element-by-element subtraction of packed double-precision (64-bit) floating-point elements in "v2" from "v3" storing the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := 32*j - l := 8*j +FOR j := 0 to 7 + i := j*64 IF k[j] - dst[l+7:l] := Truncate_Int32_To_Int8(a[i+31:i]) + dst[i+63:i] := v3[i+63:i] - v2[i+63:i] ELSE - dst[l+7:l] := src[l+7:l] + dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:32] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Convert - Store - - - - Convert packed 32-bit integers in "a" to packed 8-bit integers with truncation, and store the active results (those with their respective bit set in writemask "k") to unaligned memory at "base_addr". + + Floating Point + KNCNI + Arithmetic + + + + Performs element-by-element subtraction of packed single-precision (32-bit) floating-point elements in "v2" from "v3" storing the results in "dst". -FOR j := 0 to 3 - i := 32*j - l := 8*j - IF k[j] - MEM[base_addr+l+7:base_addr+l] := Truncate_Int32_To_Int8(a[i+31:i]) - FI +FOR j := 0 to 15 + i := j*32 + dst[i+31:i] := v3[i+31:i] - v2[i+31:i] ENDFOR -dst[MAX:32] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Convert - - - Convert packed 32-bit integers in "a" to packed 8-bit integers with truncation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Floating Point + KNCNI + Arithmetic + + + + + + Performs element-by-element subtraction of packed single-precision (32-bit) floating-point elements in "v2" from "v3" storing the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := 32*j - l := 8*j +FOR j := 0 to 15 + i := j*32 IF k[j] - dst[l+7:l] := Truncate_Int32_To_Int8(a[i+31:i]) + dst[i+31:i] := v3[i+31:i] - v2[i+31:i] ELSE - dst[l+7:l] := 0 + dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:32] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Convert - - Convert packed 32-bit integers in "a" to packed 16-bit integers with truncation, and store the results in "dst". + + Floating Point + KNCNI + Arithmetic + + + + + Performs element-by-element subtraction of packed double-precision (64-bit) floating-point elements in "v2" from "v3" storing the results in "dst". + [round_note] FOR j := 0 to 7 - i := 32*j - k := 16*j - dst[k+15:k] := Truncate_Int32_To_Int16(a[i+31:i]) + i := j*64 + dst[i+63:i] := v3[i+63:i] - v2[i+63:i] ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Convert - - - - Convert packed 32-bit integers in "a" to packed 16-bit integers with truncation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Floating Point + KNCNI + Arithmetic + + + + + + + Performs element-by-element subtraction of packed double-precision (64-bit) floating-point elements in "v2" from "v3" storing the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + [round_note] FOR j := 0 to 7 - i := 32*j - l := 16*j + i := j*64 IF k[j] - dst[l+15:l] := Truncate_Int32_To_Int16(a[i+31:i]) + dst[i+63:i] := v3[i+63:i] - v2[i+63:i] ELSE - dst[l+15:l] := src[l+15:l] + dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Convert - Store - - - - Convert packed 32-bit integers in "a" to packed 16-bit integers with truncation, and store the active results (those with their respective bit set in writemask "k") to unaligned memory at "base_addr". + + Floating Point + KNCNI + Arithmetic + + + + + Performs element-by-element subtraction of packed single-precision (32-bit) floating-point elements in "v2" from "v3" storing the results in "dst". + [round_note] -FOR j := 0 to 7 - i := 32*j - l := 16*j - IF k[j] - MEM[base_addr+l+15:base_addr+l] := Truncate_Int32_To_Int16(a[i+31:i]) - FI +FOR j := 0 to 15 + i := j*32 + dst[i+31:i] := v3[i+31:i] - v2[i+31:i] ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Convert - - - Convert packed 32-bit integers in "a" to packed 16-bit integers with truncation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Floating Point + KNCNI + Arithmetic + + + + + + + Performs element-by-element subtraction of packed single-precision (32-bit) floating-point elements in "v2" from "v3" storing the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + [round_note] -FOR j := 0 to 7 - i := 32*j - l := 16*j +FOR j := 0 to 15 + i := j*32 IF k[j] - dst[l+15:l] := Truncate_Int32_To_Int16(a[i+31:i]) + dst[i+31:i] := v3[i+31:i] - v2[i+31:i] ELSE - dst[l+15:l] := 0 + dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX512VL - AVX512F - Convert - - Convert packed 32-bit integers in "a" to packed 16-bit integers with truncation, and store the results in "dst". + KNCNI + Arithmetic + + + + Performs element-by-element subtraction of packed 32-bit integer elements in "v2" from "v3" storing the results in "dst". -FOR j := 0 to 3 - i := 32*j - k := 16*j - dst[k+15:k] := Truncate_Int32_To_Int16(a[i+31:i]) +FOR j := 0 to 15 + i := j*32 + dst[i+31:i] := v3[i+31:i] - v2[i+31:i] ENDFOR -dst[MAX:64] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX512VL - AVX512F - Convert - - - - Convert packed 32-bit integers in "a" to packed 16-bit integers with truncation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + KNCNI + Arithmetic + + + + + + Performs element-by-element subtraction of packed 32-bit integer elements in "v2" from "v3" storing the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set) -FOR j := 0 to 3 - i := 32*j - l := 16*j +FOR j := 0 to 15 + i := j*32 IF k[j] - dst[l+15:l] := Truncate_Int32_To_Int16(a[i+31:i]) + dst[i+31:i] := v3[i+31:i] - v2[i+31:i] ELSE - dst[l+15:l] := src[l+15:l] + dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:64] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX512VL - AVX512F - Convert - Store - - - - Convert packed 32-bit integers in "a" to packed 16-bit integers with truncation, and store the active results (those with their respective bit set in writemask "k") to unaligned memory at "base_addr". - -FOR j := 0 to 3 - i := 32*j - l := 16*j - IF k[j] - MEM[base_addr+l+15:base_addr+l] := Truncate_Int32_To_Int16(a[i+31:i]) - FI + KNCNI + Arithmetic + + + + + Performs element-by-element addition of packed 32-bit integer elements in "v2" and "v3", storing the resultant carry in "k2_res" (carry flag) and the addition results in "dst". + FOR j := 0 to 15 + i := j*32 + dst[i+31:i] := v2[i+31:i] + v3[i+31:i] + k2_res[j] := Carry(v2[i+31:i] + v3[i+31:i]) ENDFOR -dst[MAX:64] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX512VL - AVX512F - Convert - - - Convert packed 32-bit integers in "a" to packed 16-bit integers with truncation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + KNCNI + Arithmetic + + + + + + + Performs element-by-element addition of packed 32-bit integer elements in "v2" and "v3", storing the resultant carry in "k2_res" (carry flag) and the addition results in "dst" using writemask "k" (elements are copied from "v2" and "k_old" when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := 32*j - l := 16*j +FOR j := 0 to 15 + i := j*32 IF k[j] - dst[l+15:l] := Truncate_Int32_To_Int16(a[i+31:i]) + dst[i+31:i] := v2[i+31:i] + v3[i+31:i] ELSE - dst[l+15:l] := 0 + dst[i+31:i] := v2[i+31:i] + k2_res[j] := k_old[j] FI ENDFOR -dst[MAX:64] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX512VL - AVX512BW - Miscellaneous - - Set each packed 8-bit integer in "dst" to all ones or all zeros based on the value of the corresponding bit in "k". + KNCNI + Arithmetic + + + + + Performs an element-by-element addition of packed 32-bit integer elements in "v2" and "v3", storing the results in "dst" and the sign of the sum in "sign" (sign flag). -FOR j := 0 to 31 - i := j*8 - IF k[j] - dst[i+7:i] := 0xFF - ELSE - dst[i+7:i] := 0 - FI +FOR j := 0 to 15 + i := j*32 + dst[i+31:i] := v2[i+31:i] + v3[i+31:i] + sign[j] := v2[i+31:i] & v3[i+31:i] & 0x80000000 ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX512BW - Miscellaneous - - Set each packed 8-bit integer in "dst" to all ones or all zeros based on the value of the corresponding bit in "k". + KNCNI + Arithmetic + + + + + + + Performs an element-by-element addition of packed 32-bit integer elements in "v2" and "v3", storing the results in "dst" and the sign of the sum in "sign" (sign flag). Results are stored using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 63 - i := j*8 +FOR j := 0 to 15 + i := j*32 IF k[j] - dst[i+7:i] := 0xFF + dst[i+31:i] := v2[i+31:i] + v3[i+31:i] + sign[j] := v2[i+31:i] & v3[i+31:i] & 0x80000000 ELSE - dst[i+7:i] := 0 + dst[i+31:i] := src[i+31:i] FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - AVX512BW - AVX512VL - Miscellaneous - - Set each packed 8-bit integer in "dst" to all ones or all zeros based on the value of the corresponding bit in "k". + + Floating Point + KNCNI + Arithmetic + + + + + Performs an element-by-element addition of packed single-precision (32-bit) floating-point elements in "v2" and "v3", storing the results in "dst" and the sign of the sum in "sign" (sign flag). FOR j := 0 to 15 - i := j*8 - IF k[j] - dst[i+7:i] := 0xFF - ELSE - dst[i+7:i] := 0 - FI + i := j*32 + dst[i+31:i] := v2[i+31:i] + v3[i+31:i] + sign[j] := v2[i+31:i] & v3[i+31:i] & 0x80000000 ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512DQ - Miscellaneous - - Set each packed 32-bit integer in "dst" to all ones or all zeros based on the value of the corresponding bit in "k". + + Floating Point + KNCNI + Arithmetic + + + + + + + Performs an element-by-element addition of packed single-precision (32-bit) floating-point elements in "v2" and "v3", storing the results in "dst" and the sign of the sum in "sign" (sign flag). Results are stored using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 7 +FOR j := 0 to 15 i := j*32 IF k[j] - dst[i+31:i] := 0xFFFFFFFF + dst[i+31:i] := v2[i+31:i] + v3[i+31:i] + sign[j] := v2[i+31:i] & v3[i+31:i] & 0x80000000 ELSE - dst[i+31:i] := 0 + dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - AVX512DQ - Miscellaneous - - Set each packed 32-bit integer in "dst" to all ones or all zeros based on the value of the corresponding bit in "k". + + Floating Point + KNCNI + Arithmetic + + + + + + Performs an element-by-element addition of packed single-precision (32-bit) floating-point elements in "v2" and "v3", storing the results in "dst" and the sign of the sum in "sign" (sign flag). + [round_note] FOR j := 0 to 15 i := j*32 - IF k[j] - dst[i+31:i] := 0xFFFFFFFF - ELSE - dst[i+31:i] := 0 - FI + dst[i+31:i] := v2[i+31:i] + v3[i+31:i] + sign[j] := v2[i+31:i] & v3[i+31:i] & 0x80000000 ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512DQ - Miscellaneous - - Set each packed 32-bit integer in "dst" to all ones or all zeros based on the value of the corresponding bit in "k". + + Floating Point + KNCNI + Arithmetic + + + + + + + + Performs an element-by-element addition of packed single-precision (32-bit) floating-point elements in "v2" and "v3", storing the results in "dst" and the sign of the sum in "sign" (sign flag). Results are stored using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + [round_note] -FOR j := 0 to 3 +FOR j := 0 to 15 i := j*32 IF k[j] - dst[i+31:i] := 0xFFFFFFFF + dst[i+31:i] := v2[i+31:i] + v3[i+31:i] + sign[j] := v2[i+31:i] & v3[i+31:i] & 0x80000000 ELSE - dst[i+31:i] := 0 + dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX512VL - AVX512DQ - Miscellaneous - - Set each packed 64-bit integer in "dst" to all ones or all zeros based on the value of the corresponding bit in "k". - -FOR j := 0 to 3 - i := j*64 - IF k[j] - dst[i+63:i] := 0xFFFFFFFFffffffff - ELSE - dst[i+63:i] := 0 - FI + KNCNI + Arithmetic + + + + + Performs element-by-element subtraction of packed 32-bit integer elements in "v3" from "v2", storing the results in "dst" and the nth borrow bit in the nth position of "borrow" (borrow flag). + FOR j := 0 to 15 + i := j*32 + dst[i+31:i] := v2[i+31:i] - v3[i+31:i] + borrow[j] := Borrow(v2[i+31:i] - v3[i+31:i]) ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX512DQ - Miscellaneous - - Set each packed 64-bit integer in "dst" to all ones or all zeros based on the value of the corresponding bit in "k". - -FOR j := 0 to 7 - i := j*64 + KNCNI + Arithmetic + + + + + + + Performs element-by-element subtraction of packed 32-bit integer elements in "v3" from "v2", storing the results in "dst" and the nth borrow bit in the nth position of "borrow" (borrow flag). Results are stored using writemask "k" (elements are copied from "v2" and "k_old" when the corresponding mask bit is not set). + FOR j := 0 to 15 + i := j*32 IF k[j] - dst[i+63:i] := 0xFFFFFFFFffffffff + dst[i+31:i] := v2[i+31:i] - v3[i+31:i] + borrow[j] := Borrow(v2[i+31:i] - v3[i+31:i]) ELSE - dst[i+63:i] := 0 + dst[i+31:i] := v3[i+31:i] + borrow[j] := k_old[j] FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX512VL - AVX512DQ - Miscellaneous - - Set each packed 64-bit integer in "dst" to all ones or all zeros based on the value of the corresponding bit in "k". - -FOR j := 0 to 1 - i := j*64 + KNCNI + Arithmetic + + + + + Performs element-by-element subtraction of packed 32-bit integer elements in "v2" from "v3", storing the results in "dst" and "v2". The borrowed value from the subtraction difference for the nth element is written to the nth bit of "borrow" (borrow flag). + FOR j := 0 to 15 + i := j*32 + dst[i+31:i] := v3[i+31:i] - v2[i+31:i] + borrow[j] := Borrow(v3[i+31:i] - v2[i+31:i]) +ENDFOR +dst[MAX:512] := 0 + + +
immintrin.h
+ + + Integer + KNCNI + Arithmetic + + + + + + + Performs element-by-element subtraction of packed 32-bit integer elements in "v2" from "v3", storing the results in "dst" and "v2". The borrowed value from the subtraction difference for the nth element is written to the nth bit of "borrow" (borrow flag). Results are written using writemask "k" (elements are copied from "k" to "k_old" when the corresponding mask bit is not set). + FOR j := 0 to 15 + i := j*32 IF k[j] - dst[i+63:i] := 0xFFFFFFFFffffffff + diff := v3[i+31:i] - v2[i+31:i] + borrow[j] := Borrow(v3[i+31:i] - v2[i+31:i]) + dst[i+31:i] := diff + v2[i+31:i] := diff ELSE - dst[i+63:i] := 0 + borrow[j] := k_old[j] FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX512VL - AVX512BW - Miscellaneous - - Set each packed 16-bit integer in "dst" to all ones or all zeros based on the value of the corresponding bit in "k". - -FOR j := 0 to 15 - i := j*16 - IF k[j] - dst[i+15:i] := 0xFFFF - ELSE - dst[i+15:i] := 0 - FI + KNCNI + Arithmetic + + + + + + Performs element-by-element three-input subtraction of packed 32-bit integer elements of "v3" as well as the corresponding bit from "k" from "v2". The borrowed value from the subtraction difference for the nth element is written to the nth bit of "borrow" (borrow flag). Results are stored in "dst". + FOR j := 0 to 15 + i := j*32 + dst[i+31:i] := v2[i+31:i] - v3[i+31:i] - k[j] + borrow[j] := Borrow(v2[i+31:i] - v3[i+31:i] - k[j]) ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX512BW - Miscellaneous - - Set each packed 16-bit integer in "dst" to all ones or all zeros based on the value of the corresponding bit in "k". - -FOR j := 0 to 31 - i := j*16 - IF k[j] - dst[i+15:i] := 0xFFFF + KNCNI + Arithmetic + + + + + + + Performs element-by-element three-input subtraction of packed 32-bit integer elements of "v3" as well as the corresponding bit from "k2" from "v2". The borrowed value from the subtraction difference for the nth element is written to the nth bit of "borrow" (borrow flag). Results are stored in "dst" using writemask "k1" (elements are copied from "v2" when the corresponding mask bit is not set). + FOR j := 0 to 15 + i := j*32 + IF k1[j] + dst[i+31:i] := v2[i+31:i] - v3[i+31:i] - k2[j] + borrow[j] := Borrow(v2[i+31:i] - v3[i+31:i] - k2[j]) ELSE - dst[i+15:i] := 0 + dst[i+31:i] := v2[i+31:i] FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX512VL - AVX512BW - Miscellaneous - - Set each packed 16-bit integer in "dst" to all ones or all zeros based on the value of the corresponding bit in "k". - -FOR j := 0 to 7 - i := j*16 - IF k[j] - dst[i+15:i] := 0xFFFF - ELSE - dst[i+15:i] := 0 - FI + KNCNI + Arithmetic + + + + + + Performs element-by-element three-input subtraction of packed 32-bit integer elements of "v2" as well as the corresponding bit from "k" from "v3". The borrowed value from the subtraction difference for the nth element is written to the nth bit of "borrow" (borrow flag). Results are stored in "dst". + FOR j := 0 to 15 + i := j*32 + dst[i+31:i] := v3[i+31:i] - v2[i+31:i] - k[j] + borrow[j] := Borrow(v2[i+31:i] - v3[i+31:i] - k[j]) ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - Mask - AVX512VL - AVX512DQ - Miscellaneous - - Set each bit of mask register "k" based on the most significant bit of the corresponding packed 64-bit integer in "a". - -FOR j := 0 to 3 - i := j*64 - IF a[i+63] - k[j] := 1 + KNCNI + Arithmetic + + + + + + + Performs element-by-element three-input subtraction of packed 32-bit integer elements of "v2" as well as the corresponding bit from "k2" from "v3". The borrowed value from the subtraction difference for the nth element is written to the nth bit of "borrow" (borrow flag). Results are stored in "dst" using writemask "k1" (elements are copied from "v2" when the corresponding mask bit is not set). + FOR j := 0 to 15 + i := j*32 + IF k1[j] + dst[i+31:i] := v3[i+31:i] - v2[i+31:i] - k2[j] + borrow[j] := Borrow(v2[i+31:i] - v3[i+31:i] - k2[j]) ELSE - k[j] := 0 + dst[i+31:i] := v2[i+31:i] FI ENDFOR -k[MAX:4] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - Mask - AVX512DQ - Miscellaneous - - Set each bit of mask register "k" based on the most significant bit of the corresponding packed 64-bit integer in "a". + + Floating Point + KNCNI + Convert + + + + Performs element-by-element conversion of packed double-precision (64-bit) floating-point elements in "v2" to packed single-precision (32-bit) floating-point elements, storing the results in "dst". Results are written to the lower half of "dst", and the upper half locations are set to '0'. + [round_note] FOR j := 0 to 7 i := j*64 - IF a[i+63] - k[j] := 1 - ELSE - k[j] := 0 - FI + k := j*32 + dst[k+31:k] := Convert_FP64_To_FP32(v2[i+63:i]) ENDFOR -k[MAX:8] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - Mask - AVX512VL - AVX512DQ - Miscellaneous - - Set each bit of mask register "k" based on the most significant bit of the corresponding packed 64-bit integer in "a". + + Floating Point + KNCNI + Convert + + + + + + Performs element-by-element conversion of packed double-precision (64-bit) floating-point elements in "v2" to packed single-precision (32-bit) floating-point elements, storing the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). Results are written to the lower half of "dst", and the upper half locations are set to '0'. + [round_note] -FOR j := 0 to 1 +FOR j := 0 to 7 i := j*64 - IF a[i+63] - k[j] := 1 + l := j*32 + IF k[j] + dst[l+31:l] := Convert_FP64_To_FP32(v2[i+63:i]) ELSE - k[j] := 0 + dst[l+31:l] := src[l+31:l] FI ENDFOR -k[MAX:2] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + + Floating Point Integer - AVX512VL - AVX512F + KNCNI Convert - - Convert packed 64-bit integers in "a" to packed 8-bit integers with truncation, and store the results in "dst". + + + + Performs element-by-element conversion of packed double-precision (64-bit) floating-point elements in "v2" to packed 32-bit unsigned integer elements, storing the results in "dst". Results are written to the lower half of "dst", and the upper half locations are set to '0'. + [round_note] -FOR j := 0 to 3 - i := 64*j - k := 8*j - dst[k+7:k] := Truncate_Int64_To_Int8(a[i+63:i]) +FOR j := 0 to 7 + i := j*64 + k := j*32 + dst[k+31:k] := Convert_FP64_To_Int32(v2[i+63:i]) ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + + Floating Point Integer - AVX512VL - AVX512F + KNCNI Convert - - - - Convert packed 64-bit integers in "a" to packed 8-bit integers with truncation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + + + + + Performs element-by-element conversion of packed double-precision (64-bit) floating-point elements in "v2" to packed 32-bit unsigned integer elements, storing the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). Results are written to the lower half of "dst", and the upper half locations are set to '0'. + [round_note] -FOR j := 0 to 3 - i := 64*j - l := 8*j +FOR j := 0 to 7 + i := j*64 + l := j*32 IF k[j] - dst[l+7:l] := Truncate_Int64_To_Int8(a[i+63:i]) + dst[l+31:l] := Convert_FP64_To_Int32(v2[i+63:i]) ELSE - dst[l+7:l] := src[l+7:l] + dst[l+31:l] := src[l+31:l] FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + + Floating Point Integer - AVX512VL - AVX512F + KNCNI Convert - Store - - - - Convert packed 64-bit integers in "a" to packed 8-bit integers with truncation, and store the active results (those with their respective bit set in writemask "k") to unaligned memory at "base_addr". - -FOR j := 0 to 3 - i := 64*j - l := 8*j - IF k[j] - MEM[base_addr+l+7:base_addr+l] := Truncate_Int64_To_Int8(a[i+63:i]) - FI + + + + + Performs element-by-element conversion of packed single-precision (32-bit) floating-point elements in "v2" to packed 32-bit integer elements and performs an optional exponent adjust using "expadj", storing the results in "dst". + [round_note] + FOR j := 0 to 15 + i := j*32 + dst[i+31:i] := v2[i+31:i] + CASE expadj OF + _MM_EXPADJ_NONE: dst[i+31:i] := dst[i+31:i] * (2 << 0) + _MM_EXPADJ_4: dst[i+31:i] := dst[i+31:i] * (2 << 4) + _MM_EXPADJ_5: dst[i+31:i] := dst[i+31:i] * (2 << 5) + _MM_EXPADJ_8: dst[i+31:i] := dst[i+31:i] * (2 << 8) + _MM_EXPADJ_16: dst[i+31:i] := dst[i+31:i] * (2 << 16) + _MM_EXPADJ_24: dst[i+31:i] := dst[i+31:i] * (2 << 24) + _MM_EXPADJ_31: dst[i+31:i] := dst[i+31:i] * (2 << 31) + _MM_EXPADJ_32: dst[i+31:i] := dst[i+31:i] * (2 << 32) + ESAC + dst[i+31:i] := Float32ToInt32(dst[i+31:i]) ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + + Floating Point Integer - AVX512VL - AVX512F + KNCNI Convert - - - Convert packed 64-bit integers in "a" to packed 8-bit integers with truncation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - -FOR j := 0 to 3 - i := 64*j - l := 8*j - IF k[j] - dst[l+7:l] := Truncate_Int64_To_Int8(a[i+63:i]) - ELSE - dst[l+7:l] := 0 - FI + + + + + Performs element-by-element conversion of packed single-precision (32-bit) floating-point elements in "v2" to packed 32-bit unsigned integer elements and performing an optional exponent adjust using "expadj", storing the results in "dst". + [round_note] + FOR j := 0 to 15 + i := j*32 + dst[i+31:i] := v2[i+31:i] + CASE expadj OF + _MM_EXPADJ_NONE: dst[i+31:i] := dst[i+31:i] * (2 << 0) + _MM_EXPADJ_4: dst[i+31:i] := dst[i+31:i] * (2 << 4) + _MM_EXPADJ_5: dst[i+31:i] := dst[i+31:i] * (2 << 5) + _MM_EXPADJ_8: dst[i+31:i] := dst[i+31:i] * (2 << 8) + _MM_EXPADJ_16: dst[i+31:i] := dst[i+31:i] * (2 << 16) + _MM_EXPADJ_24: dst[i+31:i] := dst[i+31:i] * (2 << 24) + _MM_EXPADJ_31: dst[i+31:i] := dst[i+31:i] * (2 << 31) + _MM_EXPADJ_32: dst[i+31:i] := dst[i+31:i] * (2 << 32) + ESAC + dst[i+31:i] := Float32ToUInt32(dst[i+31:i]) ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + + Floating Point Integer - AVX512VL - AVX512F + KNCNI Convert - - Convert packed 64-bit integers in "a" to packed 8-bit integers with truncation, and store the results in "dst". - -FOR j := 0 to 1 - i := 64*j - k := 8*j - dst[k+7:k] := Truncate_Int64_To_Int8(a[i+63:i]) + + + + + Performs element-by-element conversion of packed 32-bit unsigned integer elements in "v2" to packed single-precision (32-bit) floating-point elements and performing an optional exponent adjust using "expadj", storing the results in "dst". + [round_note] + FOR j := 0 to 15 + i := j*32 + dst[i+31:i] := UInt32ToFloat32(v2[i+31:i]) + CASE expadj OF + _MM_EXPADJ_NONE: dst[i+31:i] := dst[i+31:i] * (2 << 0) + _MM_EXPADJ_4: dst[i+31:i] := dst[i+31:i] * (2 << 4) + _MM_EXPADJ_5: dst[i+31:i] := dst[i+31:i] * (2 << 5) + _MM_EXPADJ_8: dst[i+31:i] := dst[i+31:i] * (2 << 8) + _MM_EXPADJ_16: dst[i+31:i] := dst[i+31:i] * (2 << 16) + _MM_EXPADJ_24: dst[i+31:i] := dst[i+31:i] * (2 << 24) + _MM_EXPADJ_31: dst[i+31:i] := dst[i+31:i] * (2 << 31) + _MM_EXPADJ_32: dst[i+31:i] := dst[i+31:i] * (2 << 32) + ESAC ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + + Floating Point Integer - AVX512VL - AVX512F + KNCNI Convert - - - - Convert packed 64-bit integers in "a" to packed 8-bit integers with truncation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -FOR j := 0 to 1 - i := 64*j - l := 8*j + + + + + + + Performs element-by-element conversion of packed 32-bit unsigned integer elements in "v2" to packed single-precision (32-bit) floating-point elements and performing an optional exponent adjust using "expadj", storing the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + [round_note] + FOR j := 0 to 15 + i := j*32 IF k[j] - dst[l+7:l] := Truncate_Int64_To_Int8(a[i+63:i]) + dst[i+31:i] := Int32ToFloat32(v2[i+31:i]) + CASE expadj OF + _MM_EXPADJ_NONE: dst[i+31:i] := dst[i+31:i] * (2 << 0) + _MM_EXPADJ_4: dst[i+31:i] := dst[i+31:i] * (2 << 4) + _MM_EXPADJ_5: dst[i+31:i] := dst[i+31:i] * (2 << 5) + _MM_EXPADJ_8: dst[i+31:i] := dst[i+31:i] * (2 << 8) + _MM_EXPADJ_16: dst[i+31:i] := dst[i+31:i] * (2 << 16) + _MM_EXPADJ_24: dst[i+31:i] := dst[i+31:i] * (2 << 24) + _MM_EXPADJ_31: dst[i+31:i] := dst[i+31:i] * (2 << 31) + _MM_EXPADJ_32: dst[i+31:i] := dst[i+31:i] * (2 << 32) + ESAC ELSE - dst[l+7:l] := src[l+7:l] + dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Convert - Store - - - - Convert packed 64-bit integers in "a" to packed 8-bit integers with truncation, and store the active results (those with their respective bit set in writemask "k") to unaligned memory at "base_addr". - -FOR j := 0 to 1 - i := 64*j - l := 8*j - IF k[j] - MEM[base_addr+l+7:base_addr+l] := Truncate_Int64_To_Int8(a[i+63:i]) - FI + + Floating Point + KNCNI + Elementary Math Functions + + + Approximates the base-2 exponent of the packed single-precision (32-bit) floating-point elements in "v2" with eight bits for sign and magnitude and 24 bits for the fractional part. Results are stored in "dst". + FOR j := 0 to 15 + i := j*32 + dst[i+31:i] := exp223(v2[i+31:i]) ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Convert - - - Convert packed 64-bit integers in "a" to packed 8-bit integers with truncation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - -FOR j := 0 to 1 - i := 64*j - l := 8*j + + Floating Point + KNCNI + Elementary Math Functions + + + + + Approximates the base-2 exponent of the packed single-precision (32-bit) floating-point elements in "v2" with eight bits for sign and magnitude and 24 bits for the fractional part. Results are stored in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + FOR j := 0 to 15 + i := j*32 IF k[j] - dst[l+7:l] := Truncate_Int64_To_Int8(a[i+63:i]) + dst[i+31:i] := exp223(v2[i+31:i]) ELSE - dst[l+7:l] := 0 + dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Convert - - Convert packed 64-bit integers in "a" to packed 32-bit integers with truncation, and store the results in "dst". - -FOR j := 0 to 3 - i := 64*j - k := 32*j - dst[k+31:k] := Truncate_Int64_To_Int32(a[i+63:i]) + + Floating Point + KNCNI + Miscellaneous + + + + + Fixes up NaN's from packed double-precision (64-bit) floating-point elements in "v1" and "v2", storing the results in "dst" and storing the quietized NaN's from "v1" in "v3". + FOR j := 0 to 7 + i := j*64 + dst[i+63:i] := FixupNaNs(v1[i+63:i], v2[i+63:i]) + v3[i+63:i] := QuietizeNaNs(v1[i+63:i]) ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Convert - - - - Convert packed 64-bit integers in "a" to packed 32-bit integers with truncation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -FOR j := 0 to 3 - i := 64*j - l := 32*j + + Floating Point + KNCNI + Miscellaneous + + + + + + Fixes up NaN's from packed double-precision (64-bit) floating-point elements in "v1" and "v2", storing the results in "dst" using writemask "k" (only elements whose corresponding mask bit is set are used in the computation). Quietized NaN's from "v1" are stored in "v3". + FOR j := 0 to 7 + i := j*64 IF k[j] - dst[l+31:l] := Truncate_Int64_To_Int32(a[i+63:i]) - ELSE - dst[l+31:l] := src[l+31:l] + dst[i+63:i] := FixupNaNs(v1[i+63:i], v2[i+63:i]) + v3[i+63:i] := QuietizeNaNs(v1[i+63:i]) FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Convert - Store - - - - Convert packed 64-bit integers in "a" to packed 32-bit integers with truncation, and store the active results (those with their respective bit set in writemask "k") to unaligned memory at "base_addr". - -FOR j := 0 to 3 - i := 64*j - l := 32*j - IF k[j] - MEM[base_addr+l+31:base_addr+l] := Truncate_Int64_To_Int32(a[i+63:i]) - FI + + Floating Point + KNCNI + Miscellaneous + + + + + Fixes up NaN's from packed single-precision (32-bit) floating-point elements in "v1" and "v2", storing the results in "dst" and storing the quietized NaN's from "v1" in "v3". + FOR j := 0 to 15 + i := j*32 + dst[i+31:i] := FixupNaNs(v1[i+31:i], v2[i+31:i]) + v3[i+31:i] := QuietizeNaNs(v1[i+31:i]) ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Convert - - - Convert packed 64-bit integers in "a" to packed 32-bit integers with truncation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - -FOR j := 0 to 3 - i := 64*j - l := 32*j + + Floating Point + KNCNI + Miscellaneous + + + + + + Fixes up NaN's from packed single-precision (32-bit) floating-point elements in "v1" and "v2", storing the results in "dst" using writemask "k" (only elements whose corresponding mask bit is set are used in the computation). Quietized NaN's from "v1" are stored in "v3". + FOR j := 0 to 15 + i := j*32 IF k[j] - dst[l+31:l] := Truncate_Int64_To_Int32(a[i+63:i]) - ELSE - dst[l+31:l] := 0 + dst[i+31:i] := FixupNaNs(v1[i+31:i], v2[i+31:i]) + v3[i+31:i] := QuietizeNaNs(v1[i+31:i]) FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX512VL - AVX512F - Convert - - Convert packed 64-bit integers in "a" to packed 32-bit integers with truncation, and store the results in "dst". - -FOR j := 0 to 1 - i := 64*j - k := 32*j - dst[k+31:k] := Truncate_Int64_To_Int32(a[i+63:i]) + KNCNI + Load + + + + + + Loads the high-64-byte-aligned portion of the byte/word/doubleword stream starting at element-aligned address mt-64, up-converted depending on the value of "conv", and expanded into packed 32-bit integers in "dst". The initial values of "dst" are copied from "src". Only those converted doublewords that occur at or after the first 64-byte-aligned address following (mt-64) are loaded. Elements in the resulting vector that do not map to those doublewords are taken from "src". "hint" indicates to the processor whether the loaded data is non-temporal. + DEFINE UPCONVERT(addr, offset, convertTo) { + CASE conv OF + _MM_UPCONV_EPI32_NONE: + RETURN MEM[addr + 4*offset] + _MM_UPCONV_EPI32_UINT8: + RETURN ZeroExtend32(MEM[addr + offset]) + _MM_UPCONV_EPI32_SINT8: + RETURN SignExtend32(MEM[addr + offset]) + _MM_UPCONV_EPI32_UINT16: + RETURN ZeroExtend32(MEM[addr + 2*offset]) + _MM_UPCONV_EPI32_SINT16: + RETURN SignExtend32(MEM[addr + 2*offset]) + ESAC +} +DEFINE UPCONVERTSIZE(convertTo) { + CASE conv OF + _MM_UPCONV_EPI32_NONE: + RETURN 4 + _MM_UPCONV_EPI32_UINT8: + RETURN 1 + _MM_UPCONV_EPI32_SINT8: + RETURN 1 + _MM_UPCONV_EPI32_UINT16: + RETURN 2 + _MM_UPCONV_EPI32_SINT16: + RETURN 2 + ESAC +} +dst[511:0] := src[511:0] +loadOffset := 0 +foundNext64BytesBoundary := false +upSize := UPCONVERTSIZE(conv) +addr := mt-64 +FOR j := 0 to 15 + IF foundNext64BytesBoundary == false + IF (addr + (loadOffset + 1)*upSize % 64) == 0 + foundNext64BytesBoundary := true + FI + ELSE + i := j*32 + dst[i+31:i] := UPCONVERT(addr, loadOffset, conv) + FI + loadOffset := loadOffset + 1 ENDFOR -dst[MAX:64] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX512VL - AVX512F - Convert - - - - Convert packed 64-bit integers in "a" to packed 32-bit integers with truncation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -FOR j := 0 to 1 - i := 64*j - l := 32*j + KNCNI + Load + + + + + + + Loads the high-64-byte-aligned portion of the byte/word/doubleword stream starting at element-aligned address mt-64, up-converted depending on the value of "conv", and expanded into packed 32-bit integers in "dst". The initial values of "dst" are copied from "src". Only those converted doublewords that occur at or after the first 64-byte-aligned address following (mt-64) are loaded. Elements in the resulting vector that do not map to those doublewords are taken from "src". "hint" indicates to the processor whether the loaded data is non-temporal. Elements are copied to "dst" according to element selector "k" (elements are skipped when the corresponding mask bit is not set). + DEFINE UPCONVERT(addr, offset, convertTo) { + CASE conv OF + _MM_UPCONV_EPI32_NONE: + RETURN MEM[addr + 4*offset] + _MM_UPCONV_EPI32_UINT8: + RETURN ZeroExtend32(MEM[addr + offset]) + _MM_UPCONV_EPI32_SINT8: + RETURN SignExtend32(MEM[addr + offset]) + _MM_UPCONV_EPI32_UINT16: + RETURN ZeroExtend32(MEM[addr + 2*offset]) + _MM_UPCONV_EPI32_SINT16: + RETURN SignExtend32(MEM[addr + 2*offset]) + ESAC +} +DEFINE UPCONVERTSIZE(convertTo) { + CASE conv OF + _MM_UPCONV_EPI32_NONE: + RETURN 4 + _MM_UPCONV_EPI32_UINT8: + RETURN 1 + _MM_UPCONV_EPI32_SINT8: + RETURN 1 + _MM_UPCONV_EPI32_UINT16: + RETURN 2 + _MM_UPCONV_EPI32_SINT16: + RETURN 2 + ESAC +} +dst[511:0] := src[511:0] +loadOffset := 0 +foundNext64BytesBoundary := false +upSize := UPCONVERTSIZE(conv) +addr := mt-64 +FOR j := 0 to 15 IF k[j] - dst[l+31:l] := Truncate_Int64_To_Int32(a[i+63:i]) - ELSE - dst[l+31:l] := src[l+31:l] + IF foundNext64BytesBoundary == false + IF (addr + (loadOffset + 1)*upSize % 64) == 0 + foundNext64BytesBoundary := true + FI + ELSE + i := j*32 + dst[i+31:i] := UPCONVERT(addr, loadOffset, conv) + FI + loadOffset := loadOffset + 1 FI ENDFOR -dst[MAX:64] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX512VL - AVX512F - Convert - Store - - - - Convert packed 64-bit integers in "a" to packed 32-bit integers with truncation, and store the active results (those with their respective bit set in writemask "k") to unaligned memory at "base_addr". + KNCNI + Load + + + + + + Loads the low-64-byte-aligned portion of the byte/word/doubleword stream starting at element-aligned address mt, up-converted depending on the value of "conv", and expanded into packed 32-bit integers in "dst". The initial values of "dst" are copied from "src". Only those converted doublewords that occur before first 64-byte-aligned address following "mt" are loaded. Elements in the resulting vector that do not map to those doublewords are taken from "src". "hint" indicates to the processor whether the loaded data is non-temporal. -FOR j := 0 to 1 - i := 64*j - l := 32*j - IF k[j] - MEM[base_addr+l+31:base_addr+l] := Truncate_Int64_To_Int32(a[i+63:i]) +DEFINE UPCONVERT(addr, offset, convertTo) { + CASE conv OF + _MM_UPCONV_EPI32_NONE: + RETURN MEM[addr + 4*offset] + _MM_UPCONV_EPI32_UINT8: + RETURN ZeroExtend32(MEM[addr + offset]) + _MM_UPCONV_EPI32_SINT8: + RETURN SignExtend32(MEM[addr + offset]) + _MM_UPCONV_EPI32_UINT16: + RETURN ZeroExtend32(MEM[addr + 2*offset]) + _MM_UPCONV_EPI32_SINT16: + RETURN SignExtend32(MEM[addr + 2*offset]) + ESAC +} +DEFINE UPCONVERTSIZE(convertTo) { + CASE conv OF + _MM_UPCONV_EPI32_NONE: + RETURN 4 + _MM_UPCONV_EPI32_UINT8: + RETURN 1 + _MM_UPCONV_EPI32_SINT8: + RETURN 1 + _MM_UPCONV_EPI32_UINT16: + RETURN 2 + _MM_UPCONV_EPI32_SINT16: + RETURN 2 + ESAC +} +dst[511:0] := src[511:0] +loadOffset := 0 +upSize := UPCONVERTSIZE(conv) +addr := mt +FOR j := 0 to 15 + i := j*32 + dst[i+31:i] := UPCONVERT(addr, loadOffset, conv) + loadOffset := loadOffset + 1 + IF (mt + loadOffset * upSize) % 64 == 0 + BREAK FI ENDFOR -dst[MAX:64] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX512VL - AVX512F - Convert - - - Convert packed 64-bit integers in "a" to packed 32-bit integers with truncation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + KNCNI + Load + + + + + + + Loads the low-64-byte-aligned portion of the byte/word/doubleword stream starting at element-aligned address mt, up-converted depending on the value of "conv", and expanded into packed 32-bit integers in "dst". The initial values of "dst" are copied from "src". Only those converted doublewords that occur before first 64-byte-aligned address following "mt" are loaded. Elements in the resulting vector that do not map to those doublewords are taken from "src". "hint" indicates to the processor whether the loaded data is non-temporal. Elements are copied to "dst" according to element selector "k" (elements are skipped when the corresponding mask bit is not set). -FOR j := 0 to 1 - i := 64*j - l := 32*j +DEFINE UPCONVERT(addr, offset, convertTo) { + CASE conv OF + _MM_UPCONV_EPI32_NONE: + RETURN MEM[addr + 4*offset] + _MM_UPCONV_EPI32_UINT8: + RETURN ZeroExtend32(MEM[addr + offset]) + _MM_UPCONV_EPI32_SINT8: + RETURN SignExtend32(MEM[addr + offset]) + _MM_UPCONV_EPI32_UINT16: + RETURN ZeroExtend32(MEM[addr + 2*offset]) + _MM_UPCONV_EPI32_SINT16: + RETURN SignExtend32(MEM[addr + 2*offset]) + ESAC +} +DEFINE UPCONVERTSIZE(convertTo) { + CASE conv OF + _MM_UPCONV_EPI32_NONE: + RETURN 4 + _MM_UPCONV_EPI32_UINT8: + RETURN 1 + _MM_UPCONV_EPI32_SINT8: + RETURN 1 + _MM_UPCONV_EPI32_UINT16: + RETURN 2 + _MM_UPCONV_EPI32_SINT16: + RETURN 2 + ESAC +} +dst[511:0] := src[511:0] +loadOffset := 0 +upSize := UPCONVERTSIZE(conv) +addr := mt +FOR j := 0 to 15 IF k[j] - dst[l+31:l] := Truncate_Int64_To_Int32(a[i+63:i]) - ELSE - dst[l+31:l] := 0 + i := j*32 + dst[i+31:i] := UPCONVERT(addr, loadOffset, conv) + loadOffset := loadOffset + 1 + IF (mt + loadOffset * upSize) % 64 == 0 + BREAK + FI FI ENDFOR -dst[MAX:64] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX512VL - AVX512F - Convert - - Convert packed 64-bit integers in "a" to packed 16-bit integers with truncation, and store the results in "dst". - -FOR j := 0 to 3 - i := 64*j - k := 16*j - dst[k+15:k] := Truncate_Int64_To_Int16(a[i+63:i]) + KNCNI + Load + + + + + + Loads the high-64-byte-aligned portion of the quadword stream starting at element-aligned address mt-64, up-converted depending on the value of "conv", and expanded into packed 64-bit integers in "dst". The initial values of "dst" are copied from "src". Only those converted quadwords that occur at or after the first 64-byte-aligned address following (mt-64) are loaded. Elements in the resulting vector that do not map to those quadwords are taken from "src". "hint" indicates to the processor whether the loaded data is non-temporal. + DEFINE UPCONVERT(addr, offset, convertTo) { + CASE conv OF + _MM_UPCONV_EPI64_NONE: + RETURN MEM[addr + 8*offset] + ESAC +} +DEFINE UPCONVERTSIZE(convertTo) { + CASE conv OF + _MM_UPCONV_EPI64_NONE: + RETURN 8 + ESAC +} +dst[511:0] := src[511:0] +loadOffset := 0 +foundNext64BytesBoundary := false +upSize := UPCONVERTSIZE(conv) +addr := mt-64 +FOR j := 0 to 7 + IF foundNext64BytesBoundary == false + IF (addr + (loadOffset + 1)*upSize) == 0 + foundNext64BytesBoundary := true + FI + ELSE + i := j*64 + dst[i+63:i] := UPCONVERT(addr, loadOffset, conv) + FI + loadOffset := loadOffset + 1 ENDFOR -dst[MAX:64] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX512VL - AVX512F - Convert - - - - Convert packed 64-bit integers in "a" to packed 16-bit integers with truncation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -FOR j := 0 to 3 - i := 64*j - l := 16*j + KNCNI + Load + + + + + + + Loads the high-64-byte-aligned portion of the quadword stream starting at element-aligned address mt-64, up-converted depending on the value of "conv", and expanded into packed 64-bit integers in "dst". The initial values of "dst" are copied from "src". Only those converted quadwords that occur at or after the first 64-byte-aligned address following (mt-64) are loaded. Elements in the resulting vector that do not map to those quadwords are taken from "src". "hint" indicates to the processor whether the loaded data is non-temporal. Elements are copied to "dst" according to element selector "k" (elements are skipped when the corresponding mask bit is not set). + DEFINE UPCONVERT(addr, offset, convertTo) { + CASE conv OF + _MM_UPCONV_EPI64_NONE: + RETURN MEM[addr + 8*offset] + ESAC +} +DEFINE UPCONVERTSIZE(convertTo) { + CASE conv OF + _MM_UPCONV_EPI64_NONE: + RETURN 8 + ESAC +} +dst[511:0] := src[511:0] +loadOffset := 0 +foundNext64BytesBoundary := false +upSize := UPCONVERTSIZE(conv) +addr := mt-64 +FOR j := 0 to 7 IF k[j] - dst[l+15:l] := Truncate_Int64_To_Int16(a[i+63:i]) - ELSE - dst[l+15:l] := src[l+15:l] + IF foundNext64BytesBoundary == false + IF (addr + (loadOffset + 1)*upSize) == 0 + foundNext64BytesBoundary := true + FI + ELSE + i := j*64 + dst[i+63:i] := UPCONVERT(addr, loadOffset, conv) + FI + loadOffset := loadOffset + 1 FI ENDFOR -dst[MAX:64] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX512VL - AVX512F - Convert - Store - - - - Convert packed 64-bit integers in "a" to packed 16-bit integers with truncation, and store the active results (those with their respective bit set in writemask "k") to unaligned memory at "base_addr". + KNCNI + Load + + + + + + Loads the low-64-byte-aligned portion of the quadword stream starting at element-aligned address mt, up-converted depending on the value of "conv", and expanded into packed 64-bit integers in "dst". The initial values of "dst" are copied from "src". Only those converted quad that occur before first 64-byte-aligned address following "mt" are loaded. Elements in the resulting vector that do not map to those quadwords are taken from "src". "hint" indicates to the processor whether the loaded data is non-temporal. -FOR j := 0 to 3 - i := 64*j - l := 16*j - IF k[j] - MEM[base_addr+l+15:base_addr+l] := Truncate_Int64_To_Int16(a[i+63:i]) +DEFINE UPCONVERT(addr, offset, convertTo) { + CASE conv OF + _MM_UPCONV_EPI64_NONE: + RETURN MEM[addr + 8*offset] + ESAC +} +DEFINE UPCONVERTSIZE(convertTo) { + CASE conv OF + _MM_UPCONV_EPI64_NONE: + RETURN 8 + ESAC +} +dst[511:0] := src[511:0] +loadOffset := 0 +upSize := UPCONVERTSIZE(conv) +addr := mt +FOR j := 0 to 7 + i := j*64 + dst[i+63:i] := UPCONVERT(addr, loadOffset, conv) + loadOffset := loadOffset + 1 + IF (addr + loadOffset*upSize % 64) == 0 + BREAK FI ENDFOR -dst[MAX:64] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX512VL - AVX512F - Convert - - - Convert packed 64-bit integers in "a" to packed 16-bit integers with truncation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + KNCNI + Load + + + + + + + Loads the low-64-byte-aligned portion of the quadword stream starting at element-aligned address mt, up-converted depending on the value of "conv", and expanded into packed 64-bit integers in "dst". The initial values of "dst" are copied from "src". Only those converted quadwords that occur before first 64-byte-aligned address following "mt" are loaded. Elements in the resulting vector that do not map to those quadwords are taken from "src". "hint" indicates to the processor whether the loaded data is non-temporal. Elements are copied to "dst" according to element selector "k" (elements are skipped when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := 64*j - l := 16*j +DEFINE UPCONVERT(addr, offset, convertTo) { + CASE conv OF + _MM_UPCONV_EPI64_NONE: + RETURN MEM[addr + 8*offset] + ESAC +} +DEFINE UPCONVERTSIZE(convertTo) { + CASE conv OF + _MM_UPCONV_EPI64_NONE: + RETURN 8 + ESAC +} +dst[511:0] := src[511:0] +loadOffset := 0 +upSize := UPCONVERTSIZE(conv) +addr := mt +FOR j := 0 to 7 IF k[j] - dst[l+15:l] := Truncate_Int64_To_Int16(a[i+63:i]) - ELSE - dst[l+15:l] := 0 + i := j*64 + dst[i+63:i] := UPCONVERT(addr, loadOffset, conv) + loadOffset := loadOffset + 1 + IF (addr + loadOffset*upSize % 64) == 0 + BREAK + FI FI ENDFOR -dst[MAX:64] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Convert - - Convert packed 64-bit integers in "a" to packed 16-bit integers with truncation, and store the results in "dst". - -FOR j := 0 to 1 - i := 64*j - k := 16*j - dst[k+15:k] := Truncate_Int64_To_Int16(a[i+63:i]) + + Floating Point + KNCNI + Load + + + + + + Loads the high-64-byte-aligned portion of the byte/word/doubleword stream starting at element-aligned address mt-64, up-converted depending on the value of "conv", and expanded into packed single-precision (32-bit) floating-point elements in "dst". The initial values of "dst" are copied from "src". Only those converted quadwords that occur at or after the first 64-byte-aligned address following (mt-64) are loaded. Elements in the resulting vector that do not map to those quadwords are taken from "src". "hint" indicates to the processor whether the loaded data is non-temporal. + DEFINE UPCONVERT(addr, offset, convertTo) { + CASE conv OF + _MM_UPCONV_PS_NONE: + RETURN MEM[addr + 4*offset] + _MM_UPCONV_PS_FLOAT16: + RETURN Convert_FP16_To_FP32(MEM[addr + 4*offset]) + _MM_UPCONV_PS_UINT8: + RETURN Convert_UInt8_To_FP32(MEM[addr + offset]) + _MM_UPCONV_PS_SINT8: + RETURN Convert_Int8_To_FP32(MEM[addr + offset]) + _MM_UPCONV_PS_UINT16: + RETURN Convert_UInt16_To_FP32(MEM[addr + 2*offset]) + _MM_UPCONV_PS_SINT16: + RETURN Convert_Int16_To_FP32(MEM[addr + 2*offset]) + ESAC +} +DEFINE UPCONVERTSIZE(convertTo) { + CASE conv OF + _MM_UPCONV_PS_NONE: + RETURN 4 + _MM_UPCONV_PS_FLOAT16: + RETURN 2 + _MM_UPCONV_PS_UINT8: + RETURN 1 + _MM_UPCONV_PS_SINT8: + RETURN 1 + _MM_UPCONV_PS_UINT16: + RETURN 2 + _MM_UPCONV_PS_SINT16: + RETURN 2 + ESAC +} +dst[511:0] := src[511:0] +loadOffset := 0 +foundNext64BytesBoundary := false +upSize := UPCONVERTSIZE(conv) +addr := mt-64 +FOR j := 0 to 15 + IF foundNext64BytesBoundary == false + IF (addr + (loadOffset + 1)*upSize % 64) == 0 + foundNext64BytesBoundary := true + FI + ELSE + i := j*32 + dst[i+31:i] := UPCONVERT(addr, loadOffset, conv) + FI + loadOffset := loadOffset + 1 ENDFOR -dst[MAX:32] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Convert - - - - Convert packed 64-bit integers in "a" to packed 16-bit integers with truncation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -FOR j := 0 to 1 - i := 64*j - l := 16*j + + Floating Point + KNCNI + Load + + + + + + + Loads the high-64-byte-aligned portion of the byte/word/doubleword stream starting at element-aligned address mt-64, up-converted depending on the value of "conv", and expanded into packed single-precision (32-bit) floating-point elements in "dst". The initial values of "dst" are copied from "src". Only those converted quadwords that occur at or after the first 64-byte-aligned address following (mt-64) are loaded. Elements in the resulting vector that do not map to those quadwords are taken from "src". "hint" indicates to the processor whether the loaded data is non-temporal. Elements are copied to "dst" according to element selector "k" (elements are skipped when the corresponding mask bit is not set). + DEFINE UPCONVERT(addr, offset, convertTo) { + CASE conv OF + _MM_UPCONV_PS_NONE: + RETURN MEM[addr + 4*offset] + _MM_UPCONV_PS_FLOAT16: + RETURN Convert_FP16_To_FP32(MEM[addr + 4*offset]) + _MM_UPCONV_PS_UINT8: + RETURN Convert_UInt8_To_FP32(MEM[addr + offset]) + _MM_UPCONV_PS_SINT8: + RETURN Convert_Int8_To_FP32(MEM[addr + offset]) + _MM_UPCONV_PS_UINT16: + RETURN Convert_UInt16_To_FP32(MEM[addr + 2*offset]) + _MM_UPCONV_PS_SINT16: + RETURN Convert_Int16_To_FP32(MEM[addr + 2*offset]) + ESAC +} +DEFINE UPCONVERT(addr, offset, convertTo) { + CASE conv OF + _MM_UPCONV_PS_NONE: + RETURN MEM[addr + 4*offset] + _MM_UPCONV_PS_FLOAT16: + RETURN Convert_FP16_To_FP32(MEM[addr + 4*offset]) + _MM_UPCONV_PS_UINT8: + RETURN Convert_UInt8_To_FP32(MEM[addr + offset]) + _MM_UPCONV_PS_SINT8: + RETURN Convert_Int8_To_FP32(MEM[addr + offset]) + _MM_UPCONV_PS_UINT16: + RETURN Convert_UInt16_To_FP32(MEM[addr + 2*offset]) + _MM_UPCONV_PS_SINT16: + RETURN Convert_Int16_To_FP32(MEM[addr + 2*offset]) + ESAC +} +dst[511:0] := src[511:0] +loadOffset := 0 +foundNext64BytesBoundary := false +upSize := UPCONVERTSIZE(conv) +addr := mt-64 +FOR j := 0 to 15 IF k[j] - dst[l+15:l] := Truncate_Int64_To_Int16(a[i+63:i]) - ELSE - dst[l+15:l] := src[l+15:l] + IF foundNext64BytesBoundary == false + IF (addr + (loadOffset + 1)*upSize % 64) == 0 + foundNext64BytesBoundary := true + FI + ELSE + i := j*32 + dst[i+31:i] := UPCONVERT(addr, loadOffset, conv) + FI + loadOffset := loadOffset + 1 FI ENDFOR -dst[MAX:32] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Convert - Store - - - - Convert packed 64-bit integers in "a" to packed 16-bit integers with truncation, and store the active results (those with their respective bit set in writemask "k") to unaligned memory at "base_addr". - -FOR j := 0 to 1 - i := 64*j - l := 16*j - IF k[j] - MEM[base_addr+l+15:base_addr+l] := Truncate_Int64_To_Int16(a[i+63:i]) + + Floating Point + KNCNI + Load + + + + + + Loads the low-64-byte-aligned portion of the byte/word/doubleword stream starting at element-aligned address mt, up-converted depending on the value of "conv", and expanded into packed single-precision (32-bit) floating-point elements in "dst". The initial values of "dst" are copied from "src". Only those converted doublewords that occur before first 64-byte-aligned address following "mt" are loaded. Elements in the resulting vector that do not map to those doublewords are taken from "src". "hint" indicates to the processor whether the loaded data is non-temporal. + DEFINE UPCONVERT(addr, offset, convertTo) { + CASE conv OF + _MM_UPCONV_PS_NONE: + RETURN MEM[addr + 4*offset] + _MM_UPCONV_PS_FLOAT16: + RETURN Convert_FP16_To_FP32(MEM[addr + 4*offset]) + _MM_UPCONV_PS_UINT8: + RETURN Convert_UInt8_To_FP32(MEM[addr + offset]) + _MM_UPCONV_PS_SINT8: + RETURN Convert_Int8_To_FP32(MEM[addr + offset]) + _MM_UPCONV_PS_UINT16: + RETURN Convert_UInt16_To_FP32(MEM[addr + 2*offset]) + _MM_UPCONV_PS_SINT16: + RETURN Convert_Int16_To_FP32(MEM[addr + 2*offset]) + ESAC +} +DEFINE UPCONVERT(addr, offset, convertTo) { + CASE conv OF + _MM_UPCONV_PS_NONE: + RETURN MEM[addr + 4*offset] + _MM_UPCONV_PS_FLOAT16: + RETURN Convert_FP16_To_FP32(MEM[addr + 4*offset]) + _MM_UPCONV_PS_UINT8: + RETURN Convert_UInt8_To_FP32(MEM[addr + offset]) + _MM_UPCONV_PS_SINT8: + RETURN Convert_Int8_To_FP32(MEM[addr + offset]) + _MM_UPCONV_PS_UINT16: + RETURN Convert_UInt16_To_FP32(MEM[addr + 2*offset]) + _MM_UPCONV_PS_SINT16: + RETURN Convert_Int16_To_FP32(MEM[addr + 2*offset]) + ESAC +} +dst[511:0] := src[511:0] +loadOffset := 0 +upSize := UPCONVERTSIZE(conv) +addr := MEM[mt] +FOR j := 0 to 15 + i := j*32 + dst[i+31:i] := UPCONVERT(addr, loadOffset, conv) + loadOffset := loadOffset + 1 + IF (mt + loadOffset * upSize) % 64 == 0 + BREAK FI ENDFOR -dst[MAX:32] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Convert - - - Convert packed 64-bit integers in "a" to packed 16-bit integers with truncation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - -FOR j := 0 to 1 - i := 64*j - l := 16*j + + Floating Point + KNCNI + Load + + + + + + + Loads the low-64-byte-aligned portion of the byte/word/doubleword stream starting at element-aligned address mt, up-converted depending on the value of "conv", and expanded into packed single-precision (32-bit) floating-point elements in "dst". The initial values of "dst" are copied from "src". Only those converted doublewords that occur before first 64-byte-aligned address following "mt" are loaded. Elements in the resulting vector that do not map to those doublewords are taken from "src". "hint" indicates to the processor whether the loaded data is non-temporal. Elements are copied to "dst" according to element selector "k" (elements are skipped when the corresponding mask bit is not set). + DEFINE UPCONVERT(addr, offset, convertTo) { + CASE conv OF + _MM_UPCONV_PS_NONE: + RETURN MEM[addr + 4*offset] + _MM_UPCONV_PS_FLOAT16: + RETURN Convert_FP16_To_FP32(MEM[addr + 4*offset]) + _MM_UPCONV_PS_UINT8: + RETURN Convert_UInt8_To_FP32(MEM[addr + offset]) + _MM_UPCONV_PS_SINT8: + RETURN Convert_Int8_To_FP32(MEM[addr + offset]) + _MM_UPCONV_PS_UINT16: + RETURN Convert_UInt16_To_FP32(MEM[addr + 2*offset]) + _MM_UPCONV_PS_SINT16: + RETURN Convert_Int16_To_FP32(MEM[addr + 2*offset]) + ESAC +} +DEFINE UPCONVERT(addr, offset, convertTo) { + CASE conv OF + _MM_UPCONV_PS_NONE: + RETURN MEM[addr + 4*offset] + _MM_UPCONV_PS_FLOAT16: + RETURN Convert_FP16_To_FP32(MEM[addr + 4*offset]) + _MM_UPCONV_PS_UINT8: + RETURN Convert_UInt8_To_FP32(MEM[addr + offset]) + _MM_UPCONV_PS_SINT8: + RETURN Convert_Int8_To_FP32(MEM[addr + offset]) + _MM_UPCONV_PS_UINT16: + RETURN Convert_UInt16_To_FP32(MEM[addr + 2*offset]) + _MM_UPCONV_PS_SINT16: + RETURN Convert_Int16_To_FP32(MEM[addr + 2*offset]) + ESAC +} +dst[511:0] := src[511:0] +loadOffset := 0 +upSize := UPCONVERTSIZE(conv) +addr := MEM[mt] +FOR j := 0 to 15 IF k[j] - dst[l+15:l] := Truncate_Int64_To_Int16(a[i+63:i]) - ELSE - dst[l+15:l] := 0 + i := j*32 + dst[i+31:i] := UPCONVERT(addr, loadOffset, conv) + loadOffset := loadOffset + 1 + IF (mt + loadOffset * upSize) % 64 == 0 + BREAK + FI FI ENDFOR -dst[MAX:32] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Convert - - Convert packed 32-bit integers in "a" to packed 8-bit integers with signed saturation, and store the results in "dst". - + + Floating Point + KNCNI + Load + + + + + + Loads the high-64-byte-aligned portion of the quadword stream starting at element-aligned address mt-64, up-converted depending on the value of "conv", and expanded into packed double-precision (64-bit) floating-point values in "dst". The initial values of "dst" are copied from "src". Only those converted quadwords that occur at or after the first 64-byte-aligned address following (mt-64) are loaded. Elements in the resulting vector that do not map to those quadwords are taken from "src". "hint" indicates to the processor whether the loaded data is non-temporal. + DEFINE UPCONVERT(addr, offset, convertTo) { + CASE conv OF + _MM_UPCONV_PD_NONE: + RETURN MEM[addr + 8*offset] + ESAC +} +DEFINE UPCONVERTSIZE(convertTo) { + CASE conv OF + _MM_UPCONV_PD_NONE: + RETURN 8 + ESAC +} +dst[511:0] := src[511:0] +loadOffset := 0 +foundNext64BytesBoundary := false +upSize := UPCONVERTSIZE(conv) +addr := mt-64 FOR j := 0 to 7 - i := 32*j - k := 8*j - dst[k+7:k] := Saturate_Int32_To_Int8(a[i+31:i]) + IF foundNext64BytesBoundary == false + IF (addr + (loadOffset + 1)*upSize) % 64 == 0 + foundNext64BytesBoundary := true + FI + ELSE + i := j*64 + dst[i+63:i] := UPCONVERT(addr, loadOffset, conv) + FI + loadOffset := loadOffset + 1 ENDFOR -dst[MAX:64] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Convert - - - - Convert packed 32-bit integers in "a" to packed 8-bit integers with signed saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - + + Floating Point + KNCNI + Load + + + + + + + Loads the high-64-byte-aligned portion of the quadword stream starting at element-aligned address mt-64, up-converted depending on the value of "conv", and expanded into packed double-precision (64-bit) floating-point values in "dst". The initial values of "dst" are copied from "src". Only those converted quadwords that occur at or after the first 64-byte-aligned address following (mt-64) are loaded. Elements in the resulting vector that do not map to those quadwords are taken from "src". "hint" indicates to the processor whether the loaded data is non-temporal. Elements are copied to "dst" according to element selector "k" (elements are skipped when the corresponding mask bit is not set). + DEFINE UPCONVERT(addr, offset, convertTo) { + CASE conv OF + _MM_UPCONV_PD_NONE: + RETURN MEM[addr + 8*offset] + ESAC +} +DEFINE UPCONVERTSIZE(convertTo) { + CASE conv OF + _MM_UPCONV_PD_NONE: + RETURN 8 + ESAC +} +dst[511:0] := src[511:0] +loadOffset := 0 +foundNext64BytesBoundary := false +upSize := UPCONVERTSIZE(conv) +addr := mt-64 FOR j := 0 to 7 - i := 32*j - l := 8*j IF k[j] - dst[l+7:l] := Saturate_Int32_To_Int8(a[i+31:i]) - ELSE - dst[l+7:l] := src[l+7:l] + IF foundNext64BytesBoundary == false + IF (addr + (loadOffset + 1)*upSize) % 64 == 0 + foundNext64BytesBoundary := true + FI + ELSE + i := j*64 + dst[i+63:i] := UPCONVERT(addr, loadOffset, conv) + FI + loadOffset := loadOffset + 1 FI ENDFOR -dst[MAX:64] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Convert - Store - - - - Convert packed 32-bit integers in "a" to packed 8-bit integers with signed saturation, and store the active results (those with their respective bit set in writemask "k") to unaligned memory at "base_addr". + + Floating Point + KNCNI + Load + + + + + + Loads the low-64-byte-aligned portion of the quadword stream starting at element-aligned address mt, up-converted depending on the value of "conv", and expanded into packed double-precision (64-bit) floating-point elements in "dst". The initial values of "dst" are copied from "src". Only those converted quad that occur before first 64-byte-aligned address following "mt" are loaded. Elements in the resulting vector that do not map to those quadwords are taken from "src". "hint" indicates to the processor whether the loaded data is non-temporal. +DEFINE UPCONVERT(addr, offset, convertTo) { + CASE conv OF + _MM_UPCONV_PD_NONE: + RETURN MEM[addr + 8*offset] + ESAC +} +DEFINE UPCONVERTSIZE(convertTo) { + CASE conv OF + _MM_UPCONV_PD_NONE: + RETURN 8 + ESAC +} +dst[511:0] := src[511:0] +loadOffset := 0 +upSize := UPCONVERTSIZE(conv) +addr := mt FOR j := 0 to 7 - i := 32*j - l := 8*j - IF k[j] - MEM[base_addr+l+7:base_addr+l] := Saturate_Int32_To_Int8(a[i+31:i]) + i := j*64 + dst[i+63:i] := UPCONVERT(addr, loadOffset, conv) + loadOffset := loadOffset + 1 + IF (mt + loadOffset * upSize) % 64 == 0 + BREAK FI ENDFOR -dst[MAX:64] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Convert - - - Convert packed 32-bit integers in "a" to packed 8-bit integers with signed saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Floating Point + KNCNI + Load + + + + + + + Loads the low-64-byte-aligned portion of the quadword stream starting at element-aligned address mt, up-converted depending on the value of "conv", and expanded into packed double-precision (64-bit) floating-point elements in "dst". The initial values of "dst" are copied from "src". Only those converted quad that occur before first 64-byte-aligned address following "mt" are loaded. Elements in the resulting vector that do not map to those quadwords are taken from "src". "hint" indicates to the processor whether the loaded data is non-temporal. Elements are copied to "dst" according to element selector "k" (elemenst are skipped when the corresponding mask bit is not set). +DEFINE UPCONVERT(addr, offset, convertTo) { + CASE conv OF + _MM_UPCONV_PD_NONE: + RETURN MEM[addr + 8*offset] + ESAC +} +DEFINE UPCONVERTSIZE(convertTo) { + CASE conv OF + _MM_UPCONV_PD_NONE: + RETURN 8 + ESAC +} +dst[511:0] := src[511:0] +loadOffset := 0 +upSize := UPCONVERTSIZE(conv) +addr := mt FOR j := 0 to 7 - i := 32*j - l := 8*j IF k[j] - dst[l+7:l] := Saturate_Int32_To_Int8(a[i+31:i]) - ELSE - dst[l+7:l] := 0 + i := j*64 + dst[i+63:i] := UPCONVERT(addr, loadOffset, conv) + loadOffset := loadOffset + 1 + IF (mt + loadOffset * upSize) % 64 == 0 + BREAK + FI FI ENDFOR -dst[MAX:64] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX512VL - AVX512F - Convert - - Convert packed 32-bit integers in "a" to packed 8-bit integers with signed saturation, and store the results in "dst". - -FOR j := 0 to 3 - i := 32*j - k := 8*j - dst[k+7:k] := Saturate_Int32_To_Int8(a[i+31:i]) + KNCNI + Store + + + + + + Down-converts and stores packed 32-bit integer elements of "v1" into a byte/word/doubleword stream according to "conv" at a logically mapped starting address (mt-64), storing the high-64-byte elements of that stream (those elemetns of the stream that map at or after the first 64-byte-aligned address following (m5-64)). "hint" indicates to the processor whether the data is non-temporal. + DEFINE DOWNCONVERT(element, convertTo) { + CASE convertTo OF + _MM_DOWNCONV_EPI32_NONE: + RETURN element[31:0] + _MM_DOWNCONV_EPI32_UINT8: + RETURN Truncate8(element[31:0]) + _MM_DOWNCONV_EPI32_SINT8: + RETURN Saturate8(element[31:0]) + _MM_DOWNCONV_EPI32_UINT16: + RETURN Truncate16(element[31:0]) + _MM_DOWNCONV_EPI32_SINT16: + RETURN Saturate16(element[31:0]) + ESAC +} +DEFINE DOWNCONVERTSIZE(convertTo) { + CASE convertTo OF + _MM_DOWNCONV_EPI32_NONE: + RETURN 4 + _MM_DOWNCONV_EPI32_UINT8: + RETURN 1 + _MM_DOWNCONV_EPI32_SINT8: + RETURN 1 + _MM_DOWNCONV_EPI32_UINT16: + RETURN 2 + _MM_DOWNCONV_EPI32_SINT16: + RETURN 2 + ESAC +} +storeOffset := 0 +foundNext64BytesBoundary := false +downSize := DOWNCONVERTSIZE(conv) +addr := mt-64 +FOR j := 0 to 15 + IF foundNext64BytesBoundary == false + IF ((addr + (storeOffset + 1)*downSize) % 64) == 0 + foundNext64BytesBoundary := true + FI + ELSE + i := j*32 + tmp := DOWNCONVERT(v1[i+31:i], conv) + storeAddr := addr + storeOffset * downSize + CASE downSize OF + 4: MEM[storeAddr] := tmp[31:0] + 2: MEM[storeAddr] := tmp[15:0] + 1: MEM[storeAddr] := tmp[7:0] + ESAC + FI + storeOffset := storeOffset + 1 ENDFOR -dst[MAX:32] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX512VL - AVX512F - Convert - - - - Convert packed 32-bit integers in "a" to packed 8-bit integers with signed saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -FOR j := 0 to 3 - i := 32*j - l := 8*j + KNCNI + Store + + + + + + + Down-converts and stores packed 32-bit integer elements of "v1" into a byte/word/doubleword stream according to "conv" at a logically mapped starting address (mt-64), storing the high-64-byte elements of that stream (those elemetns of the stream that map at or after the first 64-byte-aligned address following (m5-64)). "hint" indicates to the processor whether the data is non-temporal. Elements are stored to memory according to element selector "k" (elements are skipped when the corresonding mask bit is not set). + DEFINE DOWNCONVERT(element, convertTo) { + CASE convertTo OF + _MM_DOWNCONV_EPI32_NONE: + RETURN element[31:0] + _MM_DOWNCONV_EPI32_UINT8: + RETURN Truncate8(element[31:0]) + _MM_DOWNCONV_EPI32_SINT8: + RETURN Saturate8(element[31:0]) + _MM_DOWNCONV_EPI32_UINT16: + RETURN Truncate16(element[31:0]) + _MM_DOWNCONV_EPI32_SINT16: + RETURN Saturate16(element[31:0]) + ESAC +} +DEFINE DOWNCONVERTSIZE(convertTo) { + CASE convertTo OF + _MM_DOWNCONV_EPI32_NONE: + RETURN 4 + _MM_DOWNCONV_EPI32_UINT8: + RETURN 1 + _MM_DOWNCONV_EPI32_SINT8: + RETURN 1 + _MM_DOWNCONV_EPI32_UINT16: + RETURN 2 + _MM_DOWNCONV_EPI32_SINT16: + RETURN 2 + ESAC +} +storeOffset := 0 +foundNext64BytesBoundary := false +downSize := DOWNCONVERTSIZE(conv) +addr := mt-64 +FOR j := 0 to 15 IF k[j] - dst[l+7:l] := Saturate_Int32_To_Int8(a[i+31:i]) - ELSE - dst[l+7:l] := src[l+7:l] + IF foundNext64BytesBoundary == false + IF ((addr + (storeOffset + 1)*downSize) % 64) == 0 + foundNext64BytesBoundary := true + FI + ELSE + i := j*32 + tmp := DOWNCONVERT(v1[i+31:i], conv) + storeAddr := addr + storeOffset * downSize + CASE downSize OF + 4: MEM[storeAddr] := tmp[31:0] + 2: MEM[storeAddr] := tmp[15:0] + 1: MEM[storeAddr] := tmp[7:0] + ESAC + FI + storeOffset := storeOffset + 1 FI ENDFOR -dst[MAX:32] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX512VL - AVX512F - Convert + KNCNI Store - - - - Convert packed 32-bit integers in "a" to packed 8-bit integers with signed saturation, and store the active results (those with their respective bit set in writemask "k") to unaligned memory at "base_addr". + + + + + + Down-converts and stores packed 32-bit integer elements of "v1" into a byte/word/doubleword stream according to "conv" at a logically mapped starting address "mt", storing the low-64-byte elements of that stream (those elements of the stream that map before the first 64-byte-aligned address follwing "mt"). "hint" indicates to the processor whether the data is non-temporal. -FOR j := 0 to 3 - i := 32*j - l := 8*j - IF k[j] - MEM[base_addr+l+7:base_addr+l] := Saturate_Int32_To_Int8(a[i+31:i]) +DEFINE DOWNCONVERT(element, convertTo) { + CASE convertTo OF + _MM_DOWNCONV_EPI32_NONE: + RETURN element[31:0] + _MM_DOWNCONV_EPI32_UINT8: + RETURN Truncate8(element[31:0]) + _MM_DOWNCONV_EPI32_SINT8: + RETURN Saturate8(element[31:0]) + _MM_DOWNCONV_EPI32_UINT16: + RETURN Truncate16(element[31:0]) + _MM_DOWNCONV_EPI32_SINT16: + RETURN Saturate16(element[31:0]) + ESAC +} +DEFINE DOWNCONVERTSIZE(convertTo) { + CASE convertTo OF + _MM_DOWNCONV_EPI32_NONE: + RETURN 4 + _MM_DOWNCONV_EPI32_UINT8: + RETURN 1 + _MM_DOWNCONV_EPI32_SINT8: + RETURN 1 + _MM_DOWNCONV_EPI32_UINT16: + RETURN 2 + _MM_DOWNCONV_EPI32_SINT16: + RETURN 2 + ESAC +} +storeOffset := 0 +downSize := DOWNCONVERTSIZE(conv) +addr := mt +FOR j := 0 to 15 + i := j*32 + tmp := DOWNCONVERT(v1[i+31:i], conv) + storeAddr := addr + storeOffset * downSize + CASE downSize OF + 4: MEM[storeAddr] := tmp[31:0] + 2: MEM[storeAddr] := tmp[15:0] + 1: MEM[storeAddr] := tmp[7:0] + ESAC + storeOffset := storeOffset + 1 + IF ((addr + storeOffset * downSize) % 64) == 0 + BREAK FI ENDFOR -dst[MAX:32] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX512VL - AVX512F - Convert - - - Convert packed 32-bit integers in "a" to packed 8-bit integers with signed saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + KNCNI + Store + + + + + + + Down-converts and stores packed 32-bit integer elements of "v1" into a byte/word/doubleword stream according to "conv" at a logically mapped starting address "mt", storing the low-64-byte elements of that stream (those elements of the stream that map before the first 64-byte-aligned address follwing "mt"). "hint" indicates to the processor whether the data is non-temporal. Elements are written to memory according to element selector "k" (elements are skipped when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := 32*j - l := 8*j +DEFINE DOWNCONVERT(element, convertTo) { + CASE convertTo OF + _MM_DOWNCONV_EPI32_NONE: + RETURN element[31:0] + _MM_DOWNCONV_EPI32_UINT8: + RETURN Truncate8(element[31:0]) + _MM_DOWNCONV_EPI32_SINT8: + RETURN Saturate8(element[31:0]) + _MM_DOWNCONV_EPI32_UINT16: + RETURN Truncate16(element[31:0]) + _MM_DOWNCONV_EPI32_SINT16: + RETURN Saturate16(element[31:0]) + ESAC +} +DEFINE DOWNCONVERTSIZE(convertTo) { + CASE convertTo OF + _MM_DOWNCONV_EPI32_NONE: + RETURN 4 + _MM_DOWNCONV_EPI32_UINT8: + RETURN 1 + _MM_DOWNCONV_EPI32_SINT8: + RETURN 1 + _MM_DOWNCONV_EPI32_UINT16: + RETURN 2 + _MM_DOWNCONV_EPI32_SINT16: + RETURN 2 + ESAC +} +storeOffset := 0 +downSize := DOWNCONVERTSIZE(conv) +addr := mt +FOR j := 0 to 15 IF k[j] - dst[l+7:l] := Saturate_Int32_To_Int8(a[i+31:i]) - ELSE - dst[l+7:l] := 0 + i := j*32 + tmp := DOWNCONVERT(v1[i+31:i], conv) + storeAddr := addr + storeOffset * downSize + CASE downSize OF + 4: MEM[storeAddr] := tmp[31:0] + 2: MEM[storeAddr] := tmp[15:0] + 1: MEM[storeAddr] := tmp[7:0] + ESAC + storeOffset := storeOffset + 1 + IF ((addr + storeOffset * downSize) % 64) == 0 + BREAK + FI FI ENDFOR -dst[MAX:32] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX512VL - AVX512F - Convert - - Convert packed 32-bit integers in "a" to packed 16-bit integers with signed saturation, and store the results in "dst". - + KNCNI + Store + + + + + + Down-converts and stores packed 64-bit integer elements of "v1" into a quadword stream according to "conv" at a logically mapped starting address (mt-64), storing the high-64-byte elements of that stream (those elemetns of the stream that map at or after the first 64-byte-aligned address following (m5-64)). "hint" indicates to the processor whether the data is non-temporal. + DEFINE DOWNCONVERT(element, convertTo) { + CASE convertTo OF + _MM_UPCONV_EPI64_NONE: + RETURN element[63:0] + ESAC +} +DEFINE DOWNCONVERTSIZE(convertTo) { + CASE convertTo OF + _MM_UPCONV_EPI64_NONE: + RETURN 8 + ESAC +} +storeOffset := 0 +foundNext64BytesBoundary := false +downSize := DOWNCONVERTSIZE(conv) +addr := mt-64 FOR j := 0 to 7 - i := 32*j - k := 16*j - dst[k+15:k] := Saturate_Int32_To_Int16(a[i+31:i]) + IF foundNext64BytesBoundary == false + IF ((addr + (storeOffset + 1)*downSize) % 64) == 0 + foundNext64BytesBoundary := true + FI + ELSE + i := j*64 + tmp := DOWNCONVERT(v1[i+63:i], conv) + storeAddr := addr + storeOffset * downSize + CASE downSize OF + 8: MEM[storeAddr] := tmp[63:0] + ESAC + FI + storeOffset := storeOffset + 1 ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX512VL - AVX512F - Convert - - - - Convert packed 32-bit integers in "a" to packed 16-bit integers with signed saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - + KNCNI + Store + + + + + + + Down-converts and stores packed 64-bit integer elements of "v1" into a quadword stream according to "conv" at a logically mapped starting address (mt-64), storing the high-64-byte elements of that stream (those elemetns of the stream that map at or after the first 64-byte-aligned address following (mt-64)). "hint" indicates to the processor whether the data is non-temporal. Elements are stored to memory according to element selector "k" (elements are skipped when the corresonding mask bit is not set). + DEFINE DOWNCONVERT(element, convertTo) { + CASE convertTo OF + _MM_UPCONV_EPI64_NONE: + RETURN element[63:0] + ESAC +} +DEFINE DOWNCONVERTSIZE(convertTo) { + CASE convertTo OF + _MM_UPCONV_EPI64_NONE: + RETURN 8 + ESAC +} +storeOffset := 0 +foundNext64BytesBoundary := false +downSize := DOWNCONVERTSIZE(conv) +addr := mt-64 FOR j := 0 to 7 - i := 32*j - l := 16*j IF k[j] - dst[l+15:l] := Saturate_Int32_To_Int16(a[i+31:i]) - ELSE - dst[l+15:l] := src[l+15:l] + IF foundNext64BytesBoundary == false + IF ((addr + (storeOffset + 1)*downSize) % 64) == 0 + foundNext64BytesBoundary := true + FI + ELSE + i := j*64 + tmp := DOWNCONVERT(v1[i+63:i], conv) + storeAddr := addr + storeOffset * downSize + CASE downSize OF + 8: MEM[storeAddr] := tmp[63:0] + ESAC + FI + storeOffset := storeOffset + 1 FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX512VL - AVX512F - Convert + KNCNI Store - - - - Convert packed 32-bit integers in "a" to packed 16-bit integers with signed saturation, and store the active results (those with their respective bit set in writemask "k") to unaligned memory at "base_addr". + + + + + + Down-converts and stores packed 64-bit integer elements of "v1" into a quadword stream according to "conv" at a logically mapped starting address "mt", storing the low-64-byte elements of that stream (those elements of the stream that map before the first 64-byte-aligned address follwing "mt"). "hint" indicates to the processor whether the data is non-temporal. +DEFINE DOWNCONVERT(element, convertTo) { + CASE convertTo OF + _MM_UPCONV_EPI64_NONE: + RETURN element[63:0] + ESAC +} +DEFINE DOWNCONVERTSIZE(convertTo) { + CASE convertTo OF + _MM_UPCONV_EPI64_NONE: + RETURN 8 + ESAC +} +storeOffset := 0 +downSize := DOWNCONVERTSIZE(conv) +addr := mt FOR j := 0 to 7 - i := 32*j - l := 16*j - IF k[j] - MEM[base_addr+l+15:base_addr+l] := Saturate_Int32_To_Int16(a[i+31:i]) + i := j*63 + tmp := DOWNCONVERT(v1[i+63:i], conv) + storeAddr := addr + storeOffset * downSize + CASE downSize OF + 8: MEM[storeAddr] := tmp[63:0] + ESAC + storeOffset := storeOffset + 1 + IF ((addr + storeOffset * downSize) % 64) == 0 + BREAK FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX512VL - AVX512F - Convert - - - Convert packed 32-bit integers in "a" to packed 16-bit integers with signed saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + KNCNI + Store + + + + + + + Down-converts and stores packed 64-bit integer elements of "v1" into a quadword stream according to "conv" at a logically mapped starting address "mt", storing the low-64-byte elements of that stream (those elements of the stream that map before the first 64-byte-aligned address follwing "mt"). "hint" indicates to the processor whether the data is non-temporal. Elements are stored to memory according to element selector "k" (elements are skipped whent he corresponding mask bit is not set). +DEFINE DOWNCONVERT(element, convertTo) { + CASE convertTo OF + _MM_UPCONV_EPI64_NONE: + RETURN element[63:0] + ESAC +} +DEFINE DOWNCONVERTSIZE(convertTo) { + CASE convertTo OF + _MM_UPCONV_EPI64_NONE: + RETURN 8 + ESAC +} +storeOffset := 0 +downSize := DOWNCONVERTSIZE(conv) +addr := mt FOR j := 0 to 7 - i := 32*j - l := 16*j IF k[j] - dst[l+15:l] := Saturate_Int32_To_Int16(a[i+31:i]) - ELSE - dst[l+15:l] := 0 + i := j*63 + tmp := DOWNCONVERT(v1[i+63:i], conv) + storeAddr := addr + storeOffset * downSize + CASE downSize OF + 8: MEM[storeAddr] := tmp[63:0] + ESAC + storeOffset := storeOffset + 1 + IF ((addr + storeOffset * downSize) % 64) == 0 + BREAK + FI FI ENDFOR -dst[MAX:128] := 0 - - -
immintrin.h
- - - Integer - AVX512VL - AVX512F - Convert - - Convert packed 32-bit integers in "a" to packed 16-bit integers with signed saturation, and store the results in "dst". - -FOR j := 0 to 3 - i := 32*j - k := 16*j - dst[k+15:k] := Saturate_Int32_To_Int16(a[i+31:i]) -ENDFOR -dst[MAX:64] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Convert - - - - Convert packed 32-bit integers in "a" to packed 16-bit integers with signed saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -FOR j := 0 to 3 - i := 32*j - l := 16*j - IF k[j] - dst[l+15:l] := Saturate_Int32_To_Int16(a[i+31:i]) + + Floating Point + KNCNI + Store + + + + + + Down-converts and stores packed single-precision (32-bit) floating-point elements of "v1" into a byte/word/doubleword stream according to "conv" at a logically mapped starting address (mt-64), storing the high-64-byte elements of that stream (those elemetns of the stream that map at or after the first 64-byte-aligned address following (m5-64)). "hint" indicates to the processor whether the data is non-temporal. + DEFINE DOWNCONVERT(element, convertTo) { + CASE convertTo OF + _MM_UPCONV_PS_NONE: + RETURN element[31:0] + _MM_UPCONV_PS_FLOAT16: + RETURN Convert_FP32_To_FP16(element[31:0]) + _MM_UPCONV_PS_UINT8: + RETURN Truncate8(element[31:0]) + _MM_UPCONV_PS_SINT8: + RETURN Saturate8(element[31:0]) + _MM_UPCONV_PS_UINT16: + RETURN Truncate16(element[31:0]) + _MM_UPCONV_PS_SINT16: + RETURN Saturate16(element[31:0]) + ESAC +} +DEFINE DOWNCONVERTSIZE(convertTo) { + CASE convertTo OF + _MM_UPCONV_PS_NONE: + RETURN 4 + _MM_UPCONV_PS_FLOAT16: + RETURN 2 + _MM_UPCONV_PS_UINT8: + RETURN 1 + _MM_UPCONV_PS_SINT8: + RETURN 1 + _MM_UPCONV_PS_UINT16: + RETURN 2 + _MM_UPCONV_PS_SINT16: + RETURN 2 + ESAC +} +storeOffset := 0 +foundNext64BytesBoundary := false +downSize := DOWNCONVERTSIZE(conv) +addr := mt-64 +FOR j := 0 to 15 + IF foundNext64BytesBoundary == false + IF ((addr + (storeOffset + 1)*downSize) % 64) == 0 + foundNext64BytesBoundary := true + FI ELSE - dst[l+15:l] := src[l+15:l] + i := j*32 + tmp := DOWNCONVERT(v1[i+31:i], conv) + storeAddr := addr + storeOffset * downSize + CASE downSize OF + 4: MEM[storeAddr] := tmp[31:0] + 2: MEM[storeAddr] := tmp[15:0] + 1: MEM[storeAddr] := tmp[7:0] + ESAC FI + storeOffset := storeOffset + 1 ENDFOR -dst[MAX:64] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Convert + + Floating Point + KNCNI Store - - - - Convert packed 32-bit integers in "a" to packed 16-bit integers with signed saturation, and store the active results (those with their respective bit set in writemask "k") to unaligned memory at "base_addr". - -FOR j := 0 to 3 - i := 32*j - l := 16*j + + + + + + + Down-converts and stores packed single-precision (32-bit) floating-point elements of "v1" into a byte/word/doubleword stream according to "conv" at a logically mapped starting address (mt-64), storing the high-64-byte elements of that stream (those elemetns of the stream that map at or after the first 64-byte-aligned address following (m5-64)). "hint" indicates to the processor whether the data is non-temporal. Elements are stored to memory according to element selector "k" (elements are skipped when the corresponding mask bit is not set). + DEFINE DOWNCONVERT(element, convertTo) { + CASE convertTo OF + _MM_UPCONV_PS_NONE: + RETURN element[31:0] + _MM_UPCONV_PS_FLOAT16: + RETURN Convert_FP32_To_FP16(element[31:0]) + _MM_UPCONV_PS_UINT8: + RETURN Truncate8(element[31:0]) + _MM_UPCONV_PS_SINT8: + RETURN Saturate8(element[31:0]) + _MM_UPCONV_PS_UINT16: + RETURN Truncate16(element[31:0]) + _MM_UPCONV_PS_SINT16: + RETURN Saturate16(element[31:0]) + ESAC +} +DEFINE DOWNCONVERTSIZE(convertTo) { + CASE convertTo OF + _MM_UPCONV_PS_NONE: + RETURN 4 + _MM_UPCONV_PS_FLOAT16: + RETURN 2 + _MM_UPCONV_PS_UINT8: + RETURN 1 + _MM_UPCONV_PS_SINT8: + RETURN 1 + _MM_UPCONV_PS_UINT16: + RETURN 2 + _MM_UPCONV_PS_SINT16: + RETURN 2 + ESAC +} +storeOffset := 0 +foundNext64BytesBoundary := false +downSize := DOWNCONVERTSIZE(conv) +addr := mt-64 +FOR j := 0 to 15 IF k[j] - MEM[base_addr+l+15:base_addr+l] := Saturate_Int32_To_Int16(a[i+31:i]) + IF foundNext64BytesBoundary == false + IF ((addr + (storeOffset + 1)*downSize) % 64) == 0 + foundNext64BytesBoundary := true + FI + ELSE + i := j*32 + tmp := DOWNCONVERT(v1[i+31:i], conv) + storeAddr := addr + storeOffset * downSize + CASE downSize OF + 4: MEM[storeAddr] := tmp[31:0] + 2: MEM[storeAddr] := tmp[15:0] + 1: MEM[storeAddr] := tmp[7:0] + ESAC + FI + storeOffset := storeOffset + 1 FI ENDFOR -dst[MAX:64] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Convert - - - Convert packed 32-bit integers in "a" to packed 16-bit integers with signed saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Floating Point + KNCNI + Store + + + + + + Down-converts and stores packed single-precision (32-bit) floating-point elements of "v1" into a byte/word/doubleword stream according to "conv" at a logically mapped starting address "mt", storing the low-64-byte elements of that stream (those elements of the stream that map before the first 64-byte-aligned address follwing "mt"). "hint" indicates to the processor whether the data is non-temporal. -FOR j := 0 to 3 - i := 32*j - l := 16*j - IF k[j] - dst[l+15:l] := Saturate_Int32_To_Int16(a[i+31:i]) - ELSE - dst[l+15:l] := 0 +DEFINE DOWNCONVERT(element, convertTo) { + CASE convertTo OF + _MM_UPCONV_PS_NONE: + RETURN element[31:0] + _MM_UPCONV_PS_FLOAT16: + RETURN Convert_FP32_To_FP16(element[31:0]) + _MM_UPCONV_PS_UINT8: + RETURN Truncate8(element[31:0]) + _MM_UPCONV_PS_SINT8: + RETURN Saturate8(element[31:0]) + _MM_UPCONV_PS_UINT16: + RETURN Truncate16(element[31:0]) + _MM_UPCONV_PS_SINT16: + RETURN Saturate16(element[31:0]) + ESAC +} +DEFINE DOWNCONVERTSIZE(convertTo) { + CASE convertTo OF + _MM_UPCONV_PS_NONE: + RETURN 4 + _MM_UPCONV_PS_FLOAT16: + RETURN 2 + _MM_UPCONV_PS_UINT8: + RETURN 1 + _MM_UPCONV_PS_SINT8: + RETURN 1 + _MM_UPCONV_PS_UINT16: + RETURN 2 + _MM_UPCONV_PS_SINT16: + RETURN 2 + ESAC +} +storeOffset := 0 +downSize := DOWNCONVERTSIZE(conv) +addr := mt +FOR j := 0 to 15 + i := j*32 + tmp := DOWNCONVERT(v1[i+31:i], conv) + storeAddr := addr + storeOffset * downSize + CASE downSize OF + 4: MEM[storeAddr] := tmp[31:0] + 2: MEM[storeAddr] := tmp[15:0] + 1: MEM[storeAddr] := tmp[7:0] + ESAC + storeOffset := storeOffset + 1 + IF ((addr + storeOffset * downSize) % 64) == 0 + BREAK FI ENDFOR -dst[MAX:64] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Convert - - Convert packed 64-bit integers in "a" to packed 8-bit integers with signed saturation, and store the results in "dst". + + Floating Point + KNCNI + Store + + + + + + + Down-converts and stores packed single-precision (32-bit) floating-point elements of "v1" into a byte/word/doubleword stream according to "conv" at a logically mapped starting address "mt", storing the low-64-byte elements of that stream (those elements of the stream that map before the first 64-byte-aligned address follwing "mt"). "hint" indicates to the processor whether the data is non-temporal. Elements are stored to memory according to element selector "k" (elements are skipped when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := 64*j - k := 8*j - dst[k+7:k] := Saturate_Int64_To_Int8(a[i+63:i]) +DEFINE DOWNCONVERT(element, convertTo) { + CASE convertTo OF + _MM_UPCONV_PS_NONE: + RETURN element[31:0] + _MM_UPCONV_PS_FLOAT16: + RETURN Convert_FP32_To_FP16(element[31:0]) + _MM_UPCONV_PS_UINT8: + RETURN Truncate8(element[31:0]) + _MM_UPCONV_PS_SINT8: + RETURN Saturate8(element[31:0]) + _MM_UPCONV_PS_UINT16: + RETURN Truncate16(element[31:0]) + _MM_UPCONV_PS_SINT16: + RETURN Saturate16(element[31:0]) + ESAC +} +DEFINE DOWNCONVERTSIZE(convertTo) { + CASE convertTo OF + _MM_UPCONV_PS_NONE: + RETURN 4 + _MM_UPCONV_PS_FLOAT16: + RETURN 2 + _MM_UPCONV_PS_UINT8: + RETURN 1 + _MM_UPCONV_PS_SINT8: + RETURN 1 + _MM_UPCONV_PS_UINT16: + RETURN 2 + _MM_UPCONV_PS_SINT16: + RETURN 2 + ESAC +} +storeOffset := 0 +downSize := DOWNCONVERTSIZE(conv) +addr := mt +FOR j := 0 to 15 + IF k[j] + i := j*32 + tmp := DOWNCONVERT(v1[i+31:i], conv) + storeAddr := addr + storeOffset * downSize + CASE downSize OF + 4: MEM[storeAddr] := tmp[31:0] + 2: MEM[storeAddr] := tmp[15:0] + 1: MEM[storeAddr] := tmp[7:0] + ESAC + storeOffset := storeOffset + 1 + IF ((addr + storeOffset * downSize) % 64) == 0 + BREAK + FI + FI ENDFOR -dst[MAX:32] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Convert - - - - Convert packed 64-bit integers in "a" to packed 8-bit integers with signed saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -FOR j := 0 to 3 - i := 64*j - l := 8*j - IF k[j] - dst[l+7:l] := Saturate_Int64_To_Int8(a[i+63:i]) + + Floating Point + KNCNI + Store + + + + + + Down-converts and stores packed double-precision (64-bit) floating-point elements of "v1" into a quadword stream according to "conv" at a logically mapped starting address (mt-64), storing the high-64-byte elements of that stream (those elemetns of the stream that map at or after the first 64-byte-aligned address following (m5-64)). "hint" indicates to the processor whether the data is non-temporal. + DEFINE DOWNCONVERT(element, convertTo) { + CASE convertTo OF + _MM_UPCONV_PD_NONE: + RETURN element[63:0] + ESAC +} +DEFINE DOWNCONVERTSIZE(convertTo) { + CASE convertTo OF + _MM_UPCONV_PD_NONE: + RETURN 8 + ESAC +} +storeOffset := 0 +foundNext64BytesBoundary := false +downSize := DOWNCONVERTSIZE(conv) +addr := mt-64 +FOR j := 0 to 7 + IF foundNext64BytesBoundary == false + IF ((addr + (storeOffset + 1)*downSize) % 64) == 0 + foundNext64BytesBoundary := true + FI ELSE - dst[l+7:l] := src[l+7:l] + i := j*64 + tmp := DOWNCONVERT(v1[i+63:i], conv) + storeAddr := addr + storeOffset * downSize + CASE downSize OF + 8: MEM[storeAddr] := tmp[63:0] + ESAC FI + storeOffset := storeOffset + 1 ENDFOR -dst[MAX:32] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Convert + + Floating Point + KNCNI Store - - - - Convert packed 64-bit integers in "a" to packed 8-bit integers with signed saturation, and store the active results (those with their respective bit set in writemask "k") to unaligned memory at "base_addr". - -FOR j := 0 to 3 - i := 64*j - l := 8*j + + + + + + + Down-converts and stores packed double-precision (64-bit) floating-point elements of "v1" into a quadword stream according to "conv" at a logically mapped starting address (mt-64), storing the high-64-byte elements of that stream (those elemetns of the stream that map at or after the first 64-byte-aligned address following (m5-64)). "hint" indicates to the processor whether the data is non-temporal. Elements are stored to memory according to element selector "k" (elements are skipped when the corresponding mask bit is not set). + DEFINE DOWNCONVERT(element, convertTo) { + CASE convertTo OF + _MM_UPCONV_PD_NONE: + RETURN element[63:0] + ESAC +} +DEFINE DOWNCONVERTSIZE(convertTo) { + CASE convertTo OF + _MM_UPCONV_PD_NONE: + RETURN 8 + ESAC +} +storeOffset := 0 +foundNext64BytesBoundary := false +downSize := DOWNCONVERTSIZE(conv) +addr := mt-64 +FOR j := 0 to 7 IF k[j] - MEM[base_addr+l+7:base_addr+l] := Saturate_Int64_To_Int8(a[i+63:i]) + IF foundNext64BytesBoundary == false + IF ((addr + (storeOffset + 1)*downSize) % 64) == 0 + foundNext64BytesBoundary := true + FI + ELSE + i := j*64 + tmp := DOWNCONVERT(v1[i+63:i], conv) + storeAddr := addr + storeOffset * downSize + CASE downSize OF + 8: MEM[storeAddr] := tmp[63:0] + ESAC + FI + storeOffset := storeOffset + 1 FI ENDFOR -dst[MAX:32] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Convert - - - Convert packed 64-bit integers in "a" to packed 8-bit integers with signed saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Floating Point + KNCNI + Store + + + + + + Down-converts and stores packed double-precision (64-bit) floating-point elements of "v1" into a quadword stream according to "conv" at a logically mapped starting address "mt", storing the low-64-byte elements of that stream (those elements of the stream that map before the first 64-byte-aligned address follwing "mt"). "hint" indicates to the processor whether the data is non-temporal. -FOR j := 0 to 3 - i := 64*j - l := 8*j - IF k[j] - dst[l+7:l] := Saturate_Int64_To_Int8(a[i+63:i]) - ELSE - dst[l+7:l] := 0 +DEFINE DOWNCONVERT(element, convertTo) { + CASE convertTo OF + _MM_UPCONV_PD_NONE: + RETURN element[63:0] + ESAC +} +DEFINE DOWNCONVERTSIZE(convertTo) { + CASE convertTo OF + _MM_UPCONV_PD_NONE: + RETURN 8 + ESAC +} +storeOffset := 0 +downSize := DOWNCONVERTSIZE(conv) +addr := mt +FOR j := 0 to 7 + i := j*63 + tmp := DOWNCONVERT(v1[i+63:i], conv) + storeAddr := addr + storeOffset * downSize + CASE downSize OF + 8: MEM[storeAddr] := tmp[63:0] + ESAC + storeOffset := storeOffset + 1 + IF ((addr + storeOffset * downSize) % 64) == 0 + BREAK FI ENDFOR -dst[MAX:32] := 0 - - -
immintrin.h
- - - Integer - AVX512VL - AVX512F - Convert - - Convert packed 64-bit integers in "a" to packed 8-bit integers with signed saturation, and store the results in "dst". - -FOR j := 0 to 1 - i := 64*j - k := 8*j - dst[k+7:k] := Saturate_Int64_To_Int8(a[i+63:i]) -ENDFOR -dst[MAX:16] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Convert - - - - Convert packed 64-bit integers in "a" to packed 8-bit integers with signed saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Floating Point + KNCNI + Store + + + + + + + Down-converts and stores packed double-precision (64-bit) floating-point elements of "v1" into a quadword stream according to "conv" at a logically mapped starting address "mt", storing the low-64-byte elements of that stream (those elements of the stream that map before the first 64-byte-aligned address follwing "mt"). "hint" indicates to the processor whether the data is non-temporal. Elements are stored to memory according to element selector "k" (elements are skipped when the corresponding mask bit is not set). -FOR j := 0 to 1 - i := 64*j - l := 8*j +DEFINE DOWNCONVERT(element, convertTo) { + CASE convertTo OF + _MM_UPCONV_PD_NONE: + RETURN element[63:0] + ESAC +} +DEFINE DOWNCONVERTSIZE(convertTo) { + CASE convertTo OF + _MM_UPCONV_PD_NONE: + RETURN 8 + ESAC +} +storeOffset := 0 +downSize := DOWNCONVERTSIZE(conv) +addr := mt +FOR j := 0 to 7 IF k[j] - dst[l+7:l] := Saturate_Int64_To_Int8(a[i+63:i]) - ELSE - dst[l+7:l] := src[l+7:l] + i := j*63 + tmp := DOWNCONVERT(v1[i+63:i], conv) + storeAddr := addr + storeOffset * downSize + CASE downSize OF + 8: MEM[storeAddr] := tmp[63:0] + ESAC + storeOffset := storeOffset + 1 + IF ((addr + storeOffset * downSize) % 64) == 0 + BREAK + FI FI ENDFOR -dst[MAX:16] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX512VL - AVX512F - Convert + KNCNI Store - - - - Convert packed 64-bit integers in "a" to packed 8-bit integers with signed saturation, and store the active results (those with their respective bit set in writemask "k") to unaligned memory at "base_addr". + + + + + + Stores 8 packed 64-bit integer elements located in "a" and stores them in memory locations starting at location "base_addr" at packed 32-bit integer indices stored in "vindex" scaled by "scale". -FOR j := 0 to 1 - i := 64*j - l := 8*j - IF k[j] - MEM[base_addr+l+7:base_addr+l] := Saturate_Int64_To_Int8(a[i+63:i]) - FI +FOR j := 0 to 7 + i := j*64 + m := j*32 + addr := base_addr + SignExtend64(vindex[m+31:m]) * ZeroExtend64(scale) * 8 + MEM[addr+63:addr] := a[i+63:i] ENDFOR -dst[MAX:16] := 0 - +
immintrin.h
 - + Integer - AVX512VL - AVX512F - Convert - - - Convert packed 64-bit integers in "a" to packed 8-bit integers with signed saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + KNCNI + Store + + + + + + + Stores 8 packed 64-bit integer elements located in "a" and stores them in memory locations starting at location "base_addr" at packed 32-bit integer indices stored in "vindex" scaled by "scale" using writemask "k" (elements whose corresponding mask bit is not set are not written to memory). -FOR j := 0 to 1 - i := 64*j - l := 8*j +FOR j := 0 to 7 + i := j*64 + m := j*32 IF k[j] - dst[l+7:l] := Saturate_Int64_To_Int8(a[i+63:i]) - ELSE - dst[l+7:l] := 0 + addr := base_addr + SignExtend64(vindex[m+31:m]) * ZeroExtend64(scale) * 8 + MEM[addr+63:addr] := a[i+63:i] FI ENDFOR -dst[MAX:16] := 0 - +
immintrin.h
 - + Integer - AVX512VL - AVX512F - Convert - - Convert packed 64-bit integers in "a" to packed 32-bit integers with signed saturation, and store the results in "dst". - -FOR j := 0 to 3 - i := 64*j - k := 32*j - dst[k+31:k] := Saturate_Int64_To_Int32(a[i+63:i]) + KNCNI + Load + + + + Loads the high-64-byte-aligned portion of the byte/word/doubleword stream starting at element-aligned address mt-64 and expands them into packed 32-bit integers in "dst". The initial values of "dst" are copied from "src". Only those converted doublewords that occur at or after the first 64-byte-aligned address following (mt-64) are loaded. Elements in the resulting vector that do not map to those doublewords are taken from "src". + dst[511:0] := src[511:0] +loadOffset := 0 +foundNext64BytesBoundary := false +addr := mt-64 +FOR j := 0 to 15 + IF foundNext64BytesBoundary == false + IF (addr + (loadOffset + 1)*4 % 64) == 0 + foundNext64BytesBoundary := true + FI + ELSE + i := j*32 + tmp := MEM[addr + loadOffset*4] + dst[i+31:i] := tmp[i+31:i] + FI + loadOffset := loadOffset + 1 ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX512VL - AVX512F - Convert - - - - Convert packed 64-bit integers in "a" to packed 32-bit integers with signed saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -FOR j := 0 to 3 - i := 64*j - l := 32*j + KNCNI + Load + + + + + Loads the high-64-byte-aligned portion of the byte/word/doubleword stream starting at element-aligned address mt-64 and expands them into packed 32-bit integers in "dst". The initial values of "dst" are copied from "src". Only those converted doublewords that occur at or after the first 64-byte-aligned address following (mt-64) are loaded. Elements in the resulting vector that do not map to those doublewords are taken from "src". Elements are loaded from memory according to element selector "k" (elements are skipped when the corresponding mask bit is not set). + dst[511:0] := src[511:0] +loadOffset := 0 +foundNext64BytesBoundary := false +addr := mt-64 +FOR j := 0 to 15 IF k[j] - dst[l+31:l] := Saturate_Int64_To_Int32(a[i+63:i]) - ELSE - dst[l+31:l] := src[l+31:l] + IF foundNext64BytesBoundary == false + IF (addr + (loadOffset + 1)*4 % 64) == 0 + foundNext64BytesBoundary := true + FI + ELSE + i := j*32 + tmp := MEM[addr + loadOffset*4] + dst[i+31:i] := tmp[i+31:i] + FI + loadOffset := loadOffset + 1 FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX512VL - AVX512F - Convert - Store - - - - Convert packed 64-bit integers in "a" to packed 32-bit integers with signed saturation, and store the active results (those with their respective bit set in writemask "k") to unaligned memory at "base_addr". + KNCNI + Load + + + + Loads the low-64-byte-aligned portion of the byte/word/doubleword stream starting at element-aligned address mt and expanded into packed 32-bit integers in "dst". The initial values of "dst" are copied from "src". Only those converted doublewords that occur before first 64-byte-aligned address following "mt" are loaded. Elements in the resulting vector that do not map to those doublewords are taken from "src". -FOR j := 0 to 3 - i := 64*j - l := 32*j - IF k[j] - MEM[base_addr+l+31:base_addr+l] := Saturate_Int64_To_Int32(a[i+63:i]) +dst[511:0] := src[511:0] +loadOffset := 0 +addr := mt +FOR j := 0 to 15 + i := j*32 + tmp := MEM[addr + loadOffset*4] + dst[i+31:i] := tmp[i+31:i] + loadOffset := loadOffset + 1 + IF (mt + loadOffset * 4) % 64 == 0 + BREAK FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX512VL - AVX512F - Convert - - - Convert packed 64-bit integers in "a" to packed 32-bit integers with signed saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + KNCNI + Load + + + + + Loads the low-64-byte-aligned portion of the byte/word/doubleword stream starting at element-aligned address mt and expands them into packed 32-bit integers in "dst". The initial values of "dst" are copied from "src". Only those converted doublewords that occur before first 64-byte-aligned address following "mt" are loaded. Elements in the resulting vector that do not map to those doublewords are taken from "src". Elements are loaded from memory according to element selector "k" (elements are skipped when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := 64*j - l := 32*j +dst[511:0] := src[511:0] +loadOffset := 0 +addr := mt +FOR j := 0 to 15 + i := j*32 IF k[j] - dst[l+31:l] := Saturate_Int64_To_Int32(a[i+63:i]) - ELSE - dst[l+31:l] := 0 + tmp := MEM[addr + loadOffset*4] + dst[i+31:i] := tmp[i+31:i] + loadOffset := loadOffset + 1 + IF (mt + loadOffset * 4) % 64 == 0 + BREAK + FI FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX512VL - AVX512F - Convert - - Convert packed 64-bit integers in "a" to packed 32-bit integers with signed saturation, and store the results in "dst". - -FOR j := 0 to 1 - i := 64*j - k := 32*j - dst[k+31:k] := Saturate_Int64_To_Int32(a[i+63:i]) + KNCNI + Load + + + + Loads the high-64-byte-aligned portion of the quadword stream starting at element-aligned address mt-64 and expands them into packed 64-bit integers in "dst". The initial values of "dst" are copied from "src". Only those converted quadwords that occur at or after the first 64-byte-aligned address following (mt-64) are loaded. Elements in the resulting vector that do not map to those quadwords are taken from "src". + dst[511:0] := src[511:0] +loadOffset := 0 +foundNext64BytesBoundary := false +addr := mt-64 +FOR j := 0 to 7 + IF foundNext64BytesBoundary == false + IF (addr + (loadOffset + 1)*8) == 0 + foundNext64BytesBoundary := true + FI + ELSE + i := j*64 + tmp := MEM[addr + loadOffset*8] + dst[i+63:i] := tmp[i+63:i] + FI + loadOffset := loadOffset + 1 ENDFOR -dst[MAX:64] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX512VL - AVX512F - Convert - - - - Convert packed 64-bit integers in "a" to packed 32-bit integers with signed saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -FOR j := 0 to 1 - i := 64*j - l := 32*j + KNCNI + Load + + + + + Loads the high-64-byte-aligned portion of the quadword stream starting at element-aligned address mt-64 and expands them into packed 64-bit integers in "dst". The initial values of "dst" are copied from "src". Only those converted quadwords that occur at or after the first 64-byte-aligned address following (mt-64) are loaded. Elements in the resulting vector that do not map to those quadwords are taken from "src". Elements are loaded from memory according to element selector "k" (elements are skipped when the corresponding mask bit is not set). + dst[511:0] := src[511:0] +loadOffset := 0 +foundNext64BytesBoundary := false +addr := mt-64 +FOR j := 0 to 7 IF k[j] - dst[l+31:l] := Saturate_Int64_To_Int32(a[i+63:i]) - ELSE - dst[l+31:l] := src[l+31:l] + IF foundNext64BytesBoundary == false + IF (addr + (loadOffset + 1)*8) == 0 + foundNext64BytesBoundary := true + FI + ELSE + i := j*64 + tmp := MEM[addr + loadOffset*8] + dst[i+63:i] := tmp[i+63:i] + FI + loadOffset := loadOffset + 1 FI ENDFOR -dst[MAX:64] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX512VL - AVX512F - Convert - Store - - - - Convert packed 64-bit integers in "a" to packed 32-bit integers with signed saturation, and store the active results (those with their respective bit set in writemask "k") to unaligned memory at "base_addr". + KNCNI + Load + + + + Loads the low-64-byte-aligned portion of the quadword stream starting at element-aligned address mt and expands them into packed 64-bit integers in "dst". The initial values of "dst" are copied from "src". Only those converted quad that occur before first 64-byte-aligned address following "mt" are loaded. Elements in the resulting vector that do not map to those quadwords are taken from "src". -FOR j := 0 to 1 - i := 64*j - l := 32*j - IF k[j] - MEM[base_addr+l+31:base_addr+l] := Saturate_Int64_To_Int32(a[i+63:i]) +dst[511:0] := src[511:0] +loadOffset := 0 +addr := mt +FOR j := 0 to 7 + i := j*64 + tmp := MEM[addr + loadOffset*8] + dst[i+63:i] := tmp[i+63:i] + loadOffset := loadOffset + 1 + IF (addr + loadOffset*8 % 64) == 0 + BREAK FI ENDFOR -dst[MAX:64] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX512VL - AVX512F - Convert - - - Convert packed 64-bit integers in "a" to packed 32-bit integers with signed saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + KNCNI + Load + + + + + Loads the low-64-byte-aligned portion of the quadword stream starting at element-aligned address mt and expands them into packed 64-bit integers in "dst". The initial values of "dst" are copied from "src". Only those converted quad that occur before first 64-byte-aligned address following "mt" are loaded. Elements in the resulting vector that do not map to those quadwords are taken from "src". Elements are loaded from memory according to element selector "k" (elements are skipped when the corresponding mask bit is not set). -FOR j := 0 to 1 - i := 64*j - l := 32*j +dst[511:0] := src[511:0] +loadOffset := 0 +addr := mt +FOR j := 0 to 7 + i := j*64 IF k[j] - dst[l+31:l] := Saturate_Int64_To_Int32(a[i+63:i]) - ELSE - dst[l+31:l] := 0 + tmp := MEM[addr + loadOffset*8] + dst[i+63:i] := tmp[i+63:i] + loadOffset := loadOffset + 1 + IF (addr + loadOffset*8 % 64) == 0 + BREAK + FI FI ENDFOR -dst[MAX:64] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Convert - - Convert packed 64-bit integers in "a" to packed 16-bit integers with signed saturation, and store the results in "dst". - -FOR j := 0 to 3 - i := 64*j - k := 16*j - dst[k+15:k] := Saturate_Int64_To_Int16(a[i+63:i]) + + Floating Point + KNCNI + Load + + + + Loads the high-64-byte-aligned portion of the byte/word/doubleword stream starting at element-aligned address mt-64 and expands them into packed single-precision (32-bit) floating-point elements in "dst". The initial values of "dst" are copied from "src". Only those converted quadwords that occur at or after the first 64-byte-aligned address following (mt-64) are loaded. Elements in the resulting vector that do not map to those quadwords are taken from "src". + dst[511:0] := src[511:0] +loadOffset := 0 +foundNext64BytesBoundary := false +addr := mt-64 +FOR j := 0 to 15 + IF foundNext64BytesBoundary == false + IF (addr + (loadOffset + 1)*4 % 64) == 0 + foundNext64BytesBoundary := true + FI + ELSE + i := j*32 + tmp := MEM[addr + loadOffset*4] + dst[i+31:i] := tmp[i+31:i] + FI + loadOffset := loadOffset + 1 ENDFOR -dst[MAX:64] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Convert - - - - Convert packed 64-bit integers in "a" to packed 16-bit integers with signed saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -FOR j := 0 to 3 - i := 64*j - l := 16*j + + Floating Point + KNCNI + Load + + + + + Loads the high-64-byte-aligned portion of the doubleword stream starting at element-aligned address mt-64 and expands them into packed single-precision (32-bit) floating-point elements in "dst". The initial values of "dst" are copied from "src". Only those converted quadwords that occur at or after the first 64-byte-aligned address following (mt-64) are loaded. Elements in the resulting vector that do not map to those quadwords are taken from "src". Elements are loaded from memory according to element selector "k" (elements are skipped when the corresponding mask bit is not set). + dst[511:0] := src[511:0] +loadOffset := 0 +foundNext64BytesBoundary := false +addr := mt-64 +FOR j := 0 to 15 IF k[j] - dst[l+15:l] := Saturate_Int64_To_Int16(a[i+63:i]) - ELSE - dst[l+15:l] := src[l+15:l] + IF foundNext64BytesBoundary == false + IF (addr + (loadOffset + 1)*4 % 64) == 0 + foundNext64BytesBoundary := true + FI + ELSE + i := j*32 + tmp := MEM[addr + loadOffset*4] + dst[i+31:i] := tmp[i+31:i] + FI + loadOffset := loadOffset + 1 FI ENDFOR -dst[MAX:64] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Convert - Store - - - - Convert packed 64-bit integers in "a" to packed 16-bit integers with signed saturation, and store the active results (those with their respective bit set in writemask "k") to unaligned memory at "base_addr". + + Floating Point + KNCNI + Load + + + + Loads the low-64-byte-aligned portion of the doubleword stream starting at element-aligned address mt and expanded into packed single-precision (32-bit) floating-point elements in "dst". The initial values of "dst" are copied from "src". Only those converted doublewords that occur before first 64-byte-aligned address following "mt" are loaded. Elements in the resulting vector that do not map to those doublewords are taken from "src". -FOR j := 0 to 3 - i := 64*j - l := 16*j - IF k[j] - MEM[base_addr+l+15:base_addr+l] := Saturate_Int64_To_Int16(a[i+63:i]) +dst[511:0] := src[511:0] +loadOffset := 0 +addr := mt +FOR j := 0 to 15 + i := j*32 + tmp := MEM[addr + loadOffset*4] + dst[i+31:i] := tmp[i+31:i] + loadOffset := loadOffset + 1 + IF (mt + loadOffset * 4) % 64 == 0 + BREAK FI ENDFOR -dst[MAX:64] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Convert - - - Convert packed 64-bit integers in "a" to packed 16-bit integers with signed saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Floating Point + KNCNI + Load + + + + + Loads the low-64-byte-aligned portion of the doubleword stream starting at element-aligned address mt and expanded into packed single-precision (32-bit) floating-point elements in "dst". The initial values of "dst" are copied from "src". Only those converted doublewords that occur before first 64-byte-aligned address following "mt" are loaded. Elements in the resulting vector that do not map to those doublewords are taken from "src". Elements are loaded from memory according to element selector "k" (elements are skipped when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := 64*j - l := 16*j +dst[511:0] := src[511:0] +loadOffset := 0 +addr := mt +FOR j := 0 to 15 + i := j*32 IF k[j] - dst[l+15:l] := Saturate_Int64_To_Int16(a[i+63:i]) - ELSE - dst[l+15:l] := 0 + tmp := MEM[addr + loadOffset*4] + dst[i+31:i] := tmp[i+31:i] + loadOffset := loadOffset + 1 + IF (mt + loadOffset * 4) % 64 == 0 + BREAK + FI FI ENDFOR -dst[MAX:64] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Convert - - Convert packed 64-bit integers in "a" to packed 16-bit integers with signed saturation, and store the results in "dst". - -FOR j := 0 to 1 - i := 64*j - k := 16*j - dst[k+15:k] := Saturate_Int64_To_Int16(a[i+63:i]) + + Floating Point + KNCNI + Load + + + + Loads the high-64-byte-aligned portion of the quadword stream starting at element-aligned address mt-64 and expands them into packed double-precision (64-bit) floating-point values in "dst". The initial values of "dst" are copied from "src". Only those converted quadwords that occur at or after the first 64-byte-aligned address following (mt-64) are loaded. Elements in the resulting vector that do not map to those quadwords are taken from "src". + dst[511:0] := src[511:0] +loadOffset := 0 +foundNext64BytesBoundary := false +addr := mt-64 +FOR j := 0 to 7 + IF foundNext64BytesBoundary == false + IF (addr + (loadOffset + 1)*8) % 64 == 0 + foundNext64BytesBoundary := true + FI + ELSE + i := j*64 + tmp := MEM[addr + loadOffset*8] + dst[i+63:i] := tmp[i+63:i] + FI + loadOffset := loadOffset + 1 ENDFOR -dst[MAX:32] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Convert - - - - Convert packed 64-bit integers in "a" to packed 16-bit integers with signed saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -FOR j := 0 to 1 - i := 64*j - l := 16*j + + Floating Point + KNCNI + Load + + + + + Loads the high-64-byte-aligned portion of the quadword stream starting at element-aligned address mt-64 and expands them into packed double-precision (64-bit) floating-point values in "dst". The initial values of "dst" are copied from "src". Only those converted quadwords that occur at or after the first 64-byte-aligned address following (mt-64) are loaded. Elements in the resulting vector that do not map to those quadwords are taken from "src". Elements are loaded from memory according to element selector "k" (elements are skipped when the corresponding mask bit is not set). + dst[511:0] := src[511:0] +loadOffset := 0 +foundNext64BytesBoundary := false +addr := mt-64 +FOR j := 0 to 7 IF k[j] - dst[l+15:l] := Saturate_Int64_To_Int16(a[i+63:i]) - ELSE - dst[l+15:l] := src[l+15:l] + IF foundNext64BytesBoundary == false + IF (addr + (loadOffset + 1)*8) % 64 == 0 + foundNext64BytesBoundary := true + FI + ELSE + i := j*64 + tmp := MEM[addr + loadOffset*8] + dst[i+63:i] := tmp[i+63:i] + FI + loadOffset := loadOffset + 1 FI ENDFOR -dst[MAX:32] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Convert - Store - - - - Convert packed 64-bit integers in "a" to packed 16-bit integers with signed saturation, and store the active results (those with their respective bit set in writemask "k") to unaligned memory at "base_addr". + + Floating Point + KNCNI + Load + + + + Loads the low-64-byte-aligned portion of the quadword stream starting at element-aligned address mt and expands them into packed double-precision (64-bit) floating-point elements in "dst". The initial values of "dst" are copied from "src". Only those converted quad that occur before first 64-byte-aligned address following "mt" are loaded. Elements in the resulting vector that do not map to those quadwords are taken from "src". -FOR j := 0 to 1 - i := 64*j - l := 16*j - IF k[j] - MEM[base_addr+l+15:base_addr+l] := Saturate_Int64_To_Int16(a[i+63:i]) +dst[511:0] := src[511:0] +loadOffset := 0 +addr := mt +FOR j := 0 to 7 + i := j*64 + tmp := MEM[addr + loadOffset*8] + dst[i+63:i] := tmp[i+63:i] + loadOffset := loadOffset + 1 + IF ((addr + 8*loadOffset) % 64) == 0 + BREAK FI ENDFOR -dst[MAX:32] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Convert - - - Convert packed 64-bit integers in "a" to packed 16-bit integers with signed saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Floating Point + KNCNI + Load + + + + + Loads the low-64-byte-aligned portion of the quadword stream starting at element-aligned address mt and expands them into packed double-precision (64-bit) floating-point values in "dst". The initial values of "dst" are copied from "src". Only those converted quad that occur before first 64-byte-aligned address following "mt" are loaded. Elements in the resulting vector that do not map to those quadwords are taken from "src". Elements are loaded from memory according to element selector "k" (elements are skipped when the corresponding mask bit is not set). -FOR j := 0 to 1 - i := 64*j - l := 16*j +dst[511:0] := src[511:0] +loadOffset := 0 +addr := mt +FOR j := 0 to 7 + i := j*64 IF k[j] - dst[l+15:l] := Saturate_Int64_To_Int16(a[i+63:i]) - ELSE - dst[l+15:l] := 0 + tmp := MEM[addr + loadOffset*8] + dst[i+63:i] := tmp[i+63:i] + loadOffset := loadOffset + 1 + IF ((addr + 8*loadOffset) % 64) == 0 + BREAK + FI FI ENDFOR -dst[MAX:32] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX512VL - AVX512BW - Convert - - Convert packed 16-bit integers in "a" to packed 8-bit integers with signed saturation, and store the results in "dst". + KNCNI + Store + + + + Stores packed 32-bit integer elements of "v1" into a doubleword stream at a logically mapped starting address (mt-64), storing the high-64-byte elements of that stream (those elements of the stream that map at or after the first 64-byte-aligned address following (m5-64)). +storeOffset := 0 +foundNext64BytesBoundary := 0 +addr := mt-64 FOR j := 0 to 15 - i := 16*j - l := 8*j - dst[l+7:l] := Saturate_Int16_To_Int8(a[i+15:i]) + IF foundNext64BytesBoundary == 0 + IF ((addr + (storeOffset + 1)*4) % 64) == 0 + foundNext64BytesBoundary := 1 + FI + ELSE + i := j*32 + MEM[addr + storeOffset*4] := v1[i+31:i] + FI + storeOffset := storeOffset + 1 ENDFOR -dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - AVX512VL - AVX512BW - Convert - - - - Convert packed 16-bit integers in "a" to packed 8-bit integers with signed saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + KNCNI + Store + + + + + Stores packed 32-bit integer elements of "v1" into a doubleword stream at a logically mapped starting address (mt-64), storing the high-64-byte elements of that stream (those elements of the stream that map at or after the first 64-byte-aligned address following (m5-64)). Elements are loaded from memory according to element selector "k" (elements are skipped when the corresponding mask bit is not set). +storeOffset := 0 +foundNext64BytesBoundary := 0 +addr := mt-64 FOR j := 0 to 15 - i := 16*j - l := 8*j IF k[j] - dst[l+7:l] := Saturate_Int16_To_Int8(a[i+15:i]) - ELSE - dst[l+7:l] := src[l+7:l] + IF foundNext64BytesBoundary == 0 + IF ((addr + (storeOffset + 1)*4) % 64) == 0 + foundNext64BytesBoundary := 1 + FI + ELSE + i := j*32 + MEM[addr + storeOffset*4] := v1[i+31:i] + FI + storeOffset := storeOffset + 1 FI ENDFOR -dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - AVX512VL - AVX512BW - Convert + KNCNI Store - - - - Convert packed 16-bit integers in "a" to packed 8-bit integers with signed saturation, and store the active results (those with their respective bit set in writemask "k") to unaligned memory at "base_addr". + + + + Stores packed 32-bit integer elements of "v1" into a doubleword stream at a logically mapped starting address "mt", storing the low-64-byte elements of that stream (those elements of the stream that map before the first 64-byte-aligned address follwing "mt"). +storeOffset := 0 +addr := mt FOR j := 0 to 15 - i := 16*j - l := 8*j - IF k[j] - MEM[base_addr+l+7:base_addr+l] := Saturate_Int16_To_Int8(a[i+15:i]) + i := j*32 + MEM[addr + storeOffset*4] := v1[i+31:i] + storeOffset := storeOffset + 1 + IF ((addr + storeOffset*4) % 64) == 0 + BREAK FI ENDFOR -dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - AVX512VL - AVX512BW - Convert - - - Convert packed 16-bit integers in "a" to packed 8-bit integers with signed saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + KNCNI + Store + + + + + Stores packed 32-bit integer elements of "v1" into a doubleword stream at a logically mapped starting address "mt", storing the low-64-byte elements of that stream (those elements of the stream that map before the first 64-byte-aligned address follwing "mt"). Elements are loaded from memory according to element selector "k" (elements are skipped when the corresponding mask bit is not set). +storeOffset := 0 +addr := mt FOR j := 0 to 15 - i := 16*j - l := 8*j IF k[j] - dst[l+7:l] := Saturate_Int16_To_Int8(a[i+15:i]) - ELSE - dst[l+7:l] := 0 + i := j*32 + MEM[addr + storeOffset*4] := v1[i+31:i] + storeOffset := storeOffset + 1 + IF ((addr + storeOffset*4) % 64) == 0 + BREAK + FI FI ENDFOR -dst[MAX:128] := 0 - +
immintrin.h
 - + Integer - AVX512BW - Convert - - Convert packed 16-bit integers in "a" to packed 8-bit integers with signed saturation, and store the results in "dst". + KNCNI + Store + + + + Stores packed 64-bit integer elements of "v1" into a quadword stream at a logically mapped starting address (mt-64), storing the high-64-byte elements of that stream (those elemetns of the stream that map at or after the first 64-byte-aligned address following (m5-64)). -FOR j := 0 to 31 - i := 16*j - l := 8*j - dst[l+7:l] := Saturate_Int16_To_Int8(a[i+15:i]) +storeOffset := 0 +foundNext64BytesBoundary := 0 +addr := mt-64 +FOR j := 0 to 7 + IF foundNext64BytesBoundary == 0 + IF ((addr + (storeOffset + 1)*8) % 64) == 0 + foundNext64BytesBoundary := 1 + FI + ELSE + i := j*64 + MEM[addr + storeOffset*8] := v1[i+63:i] + FI + storeOffset := storeOffset + 1 ENDFOR -dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - AVX512BW - Convert - - - - Convert packed 16-bit integers in "a" to packed 8-bit integers with signed saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + KNCNI + Store + + + + + Stores packed 64-bit integer elements of "v1" into a quadword stream at a logically mapped starting address (mt-64), storing the high-64-byte elements of that stream (those elemetns of the stream that map at or after the first 64-byte-aligned address following (m5-64)). Elements are loaded from memory according to element selector "k" (elements are skipped when the corresponding mask bit is not set). -FOR j := 0 to 31 - i := 16*j - l := 8*j +storeOffset := 0 +foundNext64BytesBoundary := 0 +addr := mt-64 +FOR j := 0 to 7 IF k[j] - dst[l+7:l] := Saturate_Int16_To_Int8(a[i+15:i]) - ELSE - dst[l+7:l] := src[l+7:l] + IF foundNext64BytesBoundary == 0 + IF ((addr + (storeOffset + 1)*8) % 64) == 0 + foundNext64BytesBoundary := 1 + FI + ELSE + i := j*64 + MEM[addr + storeOffset*8] := v1[i+63:i] + FI + storeOffset := storeOffset + 1 FI ENDFOR -dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - AVX512BW - Convert + KNCNI Store - - - - Convert packed 16-bit integers in "a" to packed 8-bit integers with signed saturation, and store the active results (those with their respective bit set in writemask "k") to unaligned memory at "base_addr". + + + + Stores packed 64-bit integer elements of "v1" into a quadword stream at a logically mapped starting address "mt", storing the low-64-byte elements of that stream (those elements of the stream that map before the first 64-byte-aligned address follwing "mt"). -FOR j := 0 to 31 - i := 16*j - l := 8*j - IF k[j] - MEM[base_addr+l+7:base_addr+l] := Saturate_Int16_To_Int8(a[i+15:i]) +storeOffset := 0 +addr := mt +FOR j := 0 to 7 + i := j*64 + MEM[addr + storeOffset*8] := v1[i+63:i] + storeOffset := storeOffset + 1 + IF ((addr + storeOffset*8) % 64) == 0 + BREAK FI ENDFOR -dst[MAX:256] := 0 - +
immintrin.h
 - + Integer - AVX512BW - Convert - - - Convert packed 16-bit integers in "a" to packed 8-bit integers with signed saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + KNCNI + Store + + + + + Stores packed 64-bit integer elements of "v1" into a quadword stream at a logically mapped starting address "mt", storing the low-64-byte elements of that stream (those elements of the stream that map before the first 64-byte-aligned address follwing "mt"). Elements are loaded from memory according to element selector "k" (elements are skipped when the corresponding mask bit is not set). -FOR j := 0 to 31 - i := 16*j - l := 8*j +storeOffset := 0 +addr := mt +FOR j := 0 to 7 IF k[j] - dst[l+7:l] := Saturate_Int16_To_Int8(a[i+15:i]) - ELSE - dst[l+7:l] := 0 + i := j*64 + MEM[addr + storeOffset*8] := v1[i+63:i] + storeOffset := storeOffset + 1 + IF ((addr + storeOffset*8) % 64) == 0 + BREAK + FI FI ENDFOR -dst[MAX:256] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512BW - Convert - - Convert packed 16-bit integers in "a" to packed 8-bit integers with signed saturation, and store the results in "dst". + + Floating Point + KNCNI + Store + + + + Stores packed single-precision (32-bit) floating-point elements of "v1" into a doubleword stream at a logically mapped starting address (mt-64), storing the high-64-byte elements of that stream (those elemetns of the stream that map at or after the first 64-byte-aligned address following (m5-64)). -FOR j := 0 to 7 - i := 16*j - l := 8*j - dst[l+7:l] := Saturate_Int16_To_Int8(a[i+15:i]) +storeOffset := 0 +foundNext64BytesBoundary := 0 +addr := mt-64 +FOR j := 0 to 15 + IF foundNext64BytesBoundary == 0 + IF ((addr + (storeOffset + 1)*4) % 64) == 0 + foundNext64BytesBoundary := 1 + FI + ELSE + i := j*32 + MEM[addr + storeOffset*4] := v1[i+31:i] + FI + storeOffset := storeOffset + 1 ENDFOR -dst[MAX:64] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512BW - Convert - - - - Convert packed 16-bit integers in "a" to packed 8-bit integers with signed saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Floating Point + KNCNI + Store + + + + + Stores packed single-precision (32-bit) floating-point elements of "v1" into a doubleword stream at a logically mapped starting address (mt-64), storing the high-64-byte elements of that stream (those elemetns of the stream that map at or after the first 64-byte-aligned address following (m5-64)). Elements are loaded from memory according to element selector "k" (elements are skipped when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := 16*j - l := 8*j +storeOffset := 0 +foundNext64BytesBoundary := 0 +addr := mt-64 +FOR j := 0 to 15 IF k[j] - dst[l+7:l] := Saturate_Int16_To_Int8(a[i+15:i]) - ELSE - dst[l+7:l] := src[l+7:l] + IF foundNext64BytesBoundary == 0 + IF ((addr + (storeOffset + 1)*4) % 64) == 0 + foundNext64BytesBoundary := 1 + FI + ELSE + i := j*32 + MEM[addr + storeOffset*4] := v1[i+31:i] + FI + storeOffset := storeOffset + 1 FI ENDFOR -dst[MAX:64] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512BW - Convert + + Floating Point + KNCNI Store - - - - Convert packed 16-bit integers in "a" to packed 8-bit integers with signed saturation, and store the active results (those with their respective bit set in writemask "k") to unaligned memory at "base_addr". + + + + Stores packed single-precision (32-bit) floating-point elements of "v1" into a doubleword stream at a logically mapped starting address "mt", storing the low-64-byte elements of that stream (those elements of the stream that map before the first 64-byte-aligned address follwing "mt"). -FOR j := 0 to 7 - i := 16*j - l := 8*j - IF k[j] - MEM[base_addr+l+7:base_addr+l] := Saturate_Int16_To_Int8(a[i+15:i]) +storeOffset := 0 +addr := mt +FOR j := 0 to 15 + i := j*32 + MEM[addr + storeOffset*4] := v1[i+31:i] + storeOffset := storeOffset + 1 + IF ((addr + storeOffset*4) % 64) == 0 + BREAK FI ENDFOR -dst[MAX:64] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512BW - Convert - - - Convert packed 16-bit integers in "a" to packed 8-bit integers with signed saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Floating Point + KNCNI + Store + + + + + Stores packed single-precision (32-bit) floating-point elements of "v1" into a doubleword stream at a logically mapped starting address "mt", storing the low-64-byte elements of that stream (those elements of the stream that map before the first 64-byte-aligned address follwing "mt"). Elements are loaded from memory according to element selector "k" (elements are skipped when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := 16*j - l := 8*j +storeOffset := 0 +addr := mt +FOR j := 0 to 15 IF k[j] - dst[l+7:l] := Saturate_Int16_To_Int8(a[i+15:i]) - ELSE - dst[l+7:l] := 0 + i := j*32 + MEM[addr + storeOffset*4] := v1[i+31:i] + storeOffset := storeOffset + 1 + IF ((addr + storeOffset*4) % 64) == 0 + BREAK + FI FI ENDFOR -dst[MAX:64] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Convert - - - - Sign extend packed 8-bit integers in the low 8 bytes of "a" to packed 32-bit integers, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Floating Point + KNCNI + Store + + + + Stores packed double-precision (64-bit) floating-point elements of "v1" into a quadword stream at a logically mapped starting address (mt-64), storing the high-64-byte elements of that stream (those elemetns of the stream that map at or after the first 64-byte-aligned address following (m5-64)). +storeOffset := 0 +foundNext64BytesBoundary := 0 +addr := mt-64 FOR j := 0 to 7 - i := 32*j - l := 8*j - IF k[j] - dst[i+31:i] := SignExtend(a[l+7:l]) + IF foundNext64BytesBoundary == 0 + IF ((addr + (storeOffset + 1)*8) % 64) == 0 + foundNext64BytesBoundary := 1 + FI ELSE - dst[i+31:i] := src[i+31:i] + i := j*64 + MEM[addr + storeOffset*4] := v1[i+63:i] FI + storeOffset := storeOffset + 1 ENDFOR -dst[MAX:256] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Convert - - - Sign extend packed 8-bit integers in the low 8 bytes of "a" to packed 32-bit integers, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Floating Point + KNCNI + Store + + + + + Stores packed double-precision (64-bit) floating-point elements of "v1" into a quadword stream at a logically mapped starting address (mt-64), storing the high-64-byte elements of that stream (those elemetns of the stream that map at or after the first 64-byte-aligned address following (m5-64)). Elements are loaded from memory according to element selector "k" (elements are skipped when the corresponding mask bit is not set). +storeOffset := 0 +foundNext64BytesBoundary := 0 +addr := mt-64 FOR j := 0 to 7 - i := 32*j - l := 8*j IF k[j] - dst[i+31:i] := SignExtend(a[l+7:l]) - ELSE - dst[i+31:i] := 0 + IF foundNext64BytesBoundary == 0 + IF ((addr + (storeOffset + 1)*8) % 64) == 0 + foundNext64BytesBoundary := 1 + FI + ELSE + i := j*64 + MEM[addr + storeOffset*4] := v1[i+63:i] + FI + storeOffset := storeOffset + 1 FI ENDFOR -dst[MAX:256] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Convert - - - - Sign extend packed 8-bit integers in the low 4 bytes of "a" to packed 32-bit integers, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Floating Point + KNCNI + Store + + + + Stores packed double-precision (64-bit) floating-point elements of "v1" into a quadword stream at a logically mapped starting address "mt", storing the low-64-byte elements of that stream (those elements of the stream that map before the first 64-byte-aligned address follwing "mt"). -FOR j := 0 to 3 - i := 32*j - l := 8*j - IF k[j] - dst[i+31:i] := SignExtend(a[l+7:l]) - ELSE - dst[i+31:i] := src[i+31:i] +storeOffset := 0 +addr := mt +FOR j := 0 to 7 + i := j*64 + MEM[addr + storeOffset*8] := v1[i+63:i] + storeOffset := storeOffset + 1 + IF ((addr + storeOffset*8) % 64) == 0 + BREAK FI ENDFOR -dst[MAX:128] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Convert - - - Sign extend packed 8-bit integers in the low 4 bytes of "a" to packed 32-bit integers, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Floating Point + KNCNI + Store + + + + + Stores packed double-precision (64-bit) floating-point elements of "v1" into a quadword stream at a logically mapped starting address "mt", storing the low-64-byte elements of that stream (those elements of the stream that map before the first 64-byte-aligned address follwing "mt"). Elements are loaded from memory according to element selector "k" (elements are skipped when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := 32*j - l := 8*j +storeOffset := 0 +addr := mt +FOR j := 0 to 7 IF k[j] - dst[i+31:i] := SignExtend(a[l+7:l]) - ELSE - dst[i+31:i] := 0 + i := j*64 + MEM[addr + storeOffset*8] := v1[i+63:i] + storeOffset := storeOffset + 1 + IF ((addr + storeOffset*8) % 64) == 0 + BREAK + FI FI ENDFOR -dst[MAX:128] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Convert - - - - Sign extend packed 8-bit integers in the low 4 bytes of "a" to packed 64-bit integers, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + KNCNI + Bit Manipulation + + + Counts the number of set bits in 32-bit unsigned integer "r1", returning the results in "dst". + dst[31:0] := PopCount(r1[31:0]) + + +
immintrin.h
+ + + KNCNI + Bit Manipulation + + + Counts the number of set bits in 64-bit unsigned integer "r1", returning the results in "dst". + dst[63:0] := PopCount(r1[63:0]) + + +
immintrin.h
+ + + Mask + KNCNI + Mask + + + + Inserts the low byte of mask "k2" into the high byte of "dst", and copies the low byte of "k1" to the low byte of "dst". -FOR j := 0 to 3 - i := 64*j - l := 8*j - IF k[j] - dst[i+63:i] := SignExtend(a[l+7:l]) - ELSE - dst[i+63:i] := src[i+63:i] - FI -ENDFOR -dst[MAX:256] := 0 +dst[7:0] := k1[7:0] +dst[15:8] := k2[7:0] - +
immintrin.h
 - + + Floating Point Integer - AVX512VL - AVX512F + KNCNI Convert - - - Sign extend packed 8-bit integers in the low 4 bytes of "a" to packed 64-bit integers, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + + + Performs an element-by-element conversion of elements in packed double-precision (64-bit) floating-point vector "v2" to 32-bit integer elements, storing them in the lower half of "dst". The elements in the upper half of "dst" are set to 0. + [round_note] -FOR j := 0 to 3 - i := 64*j - l := 8*j - IF k[j] - dst[i+63:i] := SignExtend(a[l+7:l]) - ELSE - dst[i+63:i] := 0 - FI +FOR j := 0 to 7 + i := j*64 + k := j*32 + dst[k+31:k] := Convert_FP64_To_Int32(v2[i+63:i]) ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + + Floating Point Integer - AVX512VL - AVX512F + KNCNI Convert - - - - Sign extend packed 8-bit integers in the low 2 bytes of "a" to packed 64-bit integers, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + + + + + Performs an element-by-element conversion of elements in packed double-precision (64-bit) floating-point vector "v2" to 32-bit integer elements, storing them in the lower half of "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). The elements in the upper half of "dst" are set to 0. + [round_note] -FOR j := 0 to 1 - i := 64*j - l := 8*j +FOR j := 0 to 7 + i := j*64 + l := j*32 IF k[j] - dst[i+63:i] := SignExtend(a[l+7:l]) + dst[l+31:l] := Convert_FP64_To_Int32(v2[i+63:i]) ELSE - dst[i+63:i] := src[i+63:i] + dst[l+31:l] := src[l+31:l] FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + + Floating Point Integer - AVX512VL - AVX512F + KNCNI Convert - - - Sign extend packed 8-bit integers in the low 2 bytes of "a" to packed 64-bit integers, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - -FOR j := 0 to 1 - i := 64*j - l := 8*j - IF k[j] - dst[i+63:i] := SignExtend(a[l+7:l]) - ELSE - dst[i+63:i] := 0 - FI + + + + + Performs element-by-element conversion of packed 32-bit integer elements in "v2" to packed single-precision (32-bit) floating-point elements and performing an optional exponent adjust using "expadj", storing the results in "dst". + [round_note] + FOR j := 0 to 15 + i := j*32 + dst[i+31:i] := Int32ToFloat32(v2[i+31:i]) + CASE expadj OF + _MM_EXPADJ_NONE: dst[i+31:i] := dst[i+31:i] * (2 << 0) + _MM_EXPADJ_4: dst[i+31:i] := dst[i+31:i] * (2 << 4) + _MM_EXPADJ_5: dst[i+31:i] := dst[i+31:i] * (2 << 5) + _MM_EXPADJ_8: dst[i+31:i] := dst[i+31:i] * (2 << 8) + _MM_EXPADJ_16: dst[i+31:i] := dst[i+31:i] * (2 << 16) + _MM_EXPADJ_24: dst[i+31:i] := dst[i+31:i] * (2 << 24) + _MM_EXPADJ_31: dst[i+31:i] := dst[i+31:i] * (2 << 31) + _MM_EXPADJ_32: dst[i+31:i] := dst[i+31:i] * (2 << 32) + ESAC ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512BW - Convert - - - - Sign extend packed 8-bit integers in "a" to packed 16-bit integers, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Floating Point + KNCNI + Elementary Math Functions + + + Compute the base-2 logarithm of packed single-precision (32-bit) floating-point elements in "a" with absolute error of 2^(-23) and store the results in "dst". FOR j := 0 to 15 - i := j*8 - l := j*16 - IF k[j] - dst[l+15:l] := SignExtend(a[i+7:i]) - ELSE - dst[l+15:l] := src[l+15:l] - FI + i := j*32 + dst[i+31:i] := LOG(a[i+31:i]) / LOG(2.0) ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512BW - Convert - - - Sign extend packed 8-bit integers in "a" to packed 16-bit integers, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - + + Floating Point + KNCNI + Elementary Math Functions + + + + + Compute the base-2 logarithm of packed single-precision (32-bit) floating-point elements in "a" with absolute error of 2^(-23) and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). FOR j := 0 to 15 - i := j*8 - l := j*16 + i := j*32 IF k[j] - dst[l+15:l] := SignExtend(a[i+7:i]) + dst[i+31:i] := LOG(a[i+31:i]) / LOG(2.0) ELSE - dst[l+15:l] := 0 + dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX512BW - Convert - - Sign extend packed 8-bit integers in "a" to packed 16-bit integers, and store the results in "dst". + KNCNI + Arithmetic + + + + + Multiply packed 32-bit integer elements in "a" and "b", add the intermediate result to packed elements in "c" and store the results in "dst". -FOR j := 0 to 31 - i := j*8 - l := j*16 - dst[l+15:l] := SignExtend(a[i+7:i]) +FOR j := 0 to 15 + i := j*32 + dst[i+31:i] := (a[i+31:i] * b[i+31:i]) + c[i+31:i] ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX512BW - Convert - - - - Sign extend packed 8-bit integers in "a" to packed 16-bit integers, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + KNCNI + Arithmetic + + + + + + Multiply packed 32-bit integer elements in "a" and "b", add the intermediate result to packed elements in "c" and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). -FOR j := 0 to 31 - i := j*8 - l := j*16 +FOR j := 0 to 15 + i := j*32 IF k[j] - dst[l+15:l] := SignExtend(a[i+7:i]) + dst[i+31:i] := (a[i+31:i] * b[i+31:i]) + c[i+31:i] ELSE - dst[l+15:l] := src[l+15:l] + dst[i+31:i] := a[i+31:i] FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX512BW - Convert - - - Sign extend packed 8-bit integers in "a" to packed 16-bit integers, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - + KNCNI + Arithmetic + + + + + + Multiply packed 32-bit integer elements in "a" and "b", add the intermediate result to packed elements in "c" and store the results in "dst" using writemask "k" (elements are copied from "c" when the corresponding mask bit is not set). -FOR j := 0 to 31 - i := j*8 - l := j*16 +FOR j := 0 to 15 + i := j*32 IF k[j] - dst[l+15:l] := SignExtend(a[i+7:i]) + dst[i+31:i] := (a[i+31:i] * b[i+31:i]) + c[i+31:i] ELSE - dst[l+15:l] := 0 + dst[i+31:i] := c[i+31:i] FI ENDFOR dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX512VL - AVX512BW - Convert - - - - Sign extend packed 8-bit integers in "a" to packed 16-bit integers, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + KNCNI + Arithmetic + + + + Multiply packed 32-bit integer elements in each 4-element set of "a" and by element 1 of the corresponding 4-element set from "b", add the intermediate result to element 0 of the corresponding 4-element set from "b", and store the results in "dst". -FOR j := 0 to 7 - i := j*8 - l := j*16 - IF k[j] - dst[l+15:l] := SignExtend(a[i+7:i]) - ELSE - dst[l+15:l] := src[l+15:l] - FI +FOR j := 0 to 15 + i := j*32 + base := (j & ~0x3) * 32 + scale[31:0] := b[base+63:base+32] + bias[31:0] := b[base+31:base] + dst[i+31:i] := (a[i+31:i] * scale[31:0]) + bias[31:0] ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX512VL - AVX512BW - Convert - - - Sign extend packed 8-bit integers in "a" to packed 16-bit integers, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - + KNCNI + Arithmetic + + + + + + Multiply packed 32-bit integer elements in each 4-element set of "a" and by element 1 of the corresponding 4-element set from "b", add the intermediate result to element 0 of the corresponding 4-element set from "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := j*8 - l := j*16 +FOR j := 0 to 15 + i := j*32 IF k[j] - dst[l+15:l] := SignExtend(a[i+7:i]) + base := (j & ~0x3) * 32 + scale[31:0] := b[base+63:base+32] + bias[31:0] := b[base+31:base] + dst[i+31:i] := (a[i+31:i] * scale[31:0]) + bias[31:0] ELSE - dst[l+15:l] := 0 + dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Convert - - - - Sign extend packed 32-bit integers in "a" to packed 64-bit integers, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Floating Point + KNCNI + Arithmetic + + + + + Multiply packed single-precision (32-bit) floating-point elements in each 4-element set of "a" and by element 1 of the corresponding 4-element set from "b", add the intermediate result to element 0 of the corresponding 4-element set from "b", and store the results in "dst". + [round_note] -FOR j := 0 to 3 - i := 64*j - l := 32*j - IF k[j] - dst[i+63:i] := SignExtend(a[l+31:l]) - ELSE - dst[i+63:i] := src[i+63:i] - FI +FOR j := 0 to 15 + i := j*32 + base := (j & ~0x3) * 32 + scale[31:0] := b[base+63:base+32] + bias[31:0] := b[base+31:base] + dst[i+31:i] := (a[i+31:i] * scale[31:0]) + bias[31:0] ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Convert - - - Sign extend packed 32-bit integers in "a" to packed 64-bit integers, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Floating Point + KNCNI + Arithmetic + + + + + + + Multiply packed single-precision (32-bit) floating-point elements in each 4-element set of "a" and by element 1 of the corresponding 4-element set from "b", add the intermediate result to element 0 of the corresponding 4-element set from "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + [round_note] -FOR j := 0 to 3 - i := 64*j - l := 32*j +FOR j := 0 to 15 + i := j*32 IF k[j] - dst[i+63:i] := SignExtend(a[l+31:l]) + base := (j & ~0x3) * 32 + scale[31:0] := b[base+63:base+32] + bias[31:0] := b[base+31:base] + dst[i+31:i] := (a[i+31:i] * scale[31:0]) + bias[31:0] ELSE - dst[i+63:i] := 0 + dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Convert - - - - Sign extend packed 32-bit integers in "a" to packed 64-bit integers, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -FOR j := 0 to 1 - i := 64*j - l := 32*j - IF k[j] - dst[i+63:i] := SignExtend(a[l+31:l]) - ELSE - dst[i+63:i] := src[i+63:i] - FI + + Floating Point + KNCNI + Special Math Functions + + + + Determines the maximum of the absolute elements of each pair of corresponding elements of packed single-precision (32-bit) floating-point elements in "a" and "b", storing the results in "dst". + FOR j := 0 to 15 + i := j*32 + dst[i+31:i] := FpMax(ABS(a[i+31:i]), ABS(b[i+31:i])) ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Convert - - - Sign extend packed 32-bit integers in "a" to packed 64-bit integers, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - -FOR j := 0 to 1 - i := 64*j - l := 32*j + + Floating Point + KNCNI + Special Math Functions + + + + + + Determines the maximum of the absolute elements of each pair of corresponding elements of packed single-precision (32-bit) floating-point elements in "a" and "b", storing the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + FOR j := 0 to 15 + i := j*32 IF k[j] - dst[i+63:i] := SignExtend(a[l+31:l]) + dst[i+31:i] := FpMax(ABS(a[i+31:i]), ABS(b[i+31:i])) ELSE - dst[i+63:i] := 0 + dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Convert - - - - Sign extend packed 16-bit integers in "a" to packed 32-bit integers, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -FOR j := 0 to 7 + + Floating Point + KNCNI + Special Math Functions + + + + Determines the maximum of each pair of corresponding elements in packed single-precision (32-bit) floating-point elements in "a" and "b", storing the results in "dst". + FOR j := 0 to 15 + i := j*32 + dst[i+31:i] := FpMax(a[i+31:i], b[i+31:i]) +ENDFOR +dst[MAX:512] := 0 + + +
immintrin.h
+ + + Floating Point + KNCNI + Special Math Functions + + + + + + Determines the maximum of each pair of corresponding elements of packed single-precision (32-bit) floating-point elements in "a" and "b", storing the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + FOR j := 0 to 15 i := j*32 - l := j*16 IF k[j] - dst[i+31:i] := SignExtend(a[l+15:l]) + dst[i+31:i] := FpMax(a[i+31:i], b[i+31:i]) ELSE dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Convert - - - Sign extend packed 16-bit integers in "a" to packed 32-bit integers, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - -FOR j := 0 to 7 - i := 32*j - l := 16*j - IF k[j] - dst[i+31:i] := SignExtend(a[l+15:l]) - ELSE - dst[i+31:i] := 0 - FI + + Floating Point + KNCNI + Special Math Functions + + + + Determines the maximum of the absolute elements of each pair of corresponding elements of packed single-precision (32-bit) floating-point elements in "a" and "b", storing the results in "dst". + FOR j := 0 to 15 + i := j*32 + dst[i+31:i] := FpMax(ABS(a[i+31:i]), ABS(b[i+31:i])) ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Convert - - - - Sign extend packed 16-bit integers in "a" to packed 32-bit integers, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -FOR j := 0 to 3 + + Floating Point + KNCNI + Special Math Functions + + + + + + Determines the maximum of the absolute elements of each pair of corresponding elements of packed single-precision (32-bit) floating-point elements in "a" and "b", storing the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + FOR j := 0 to 15 i := j*32 - l := j*16 IF k[j] - dst[i+31:i] := SignExtend(a[l+15:l]) + dst[i+31:i] := FpMax(ABS(a[i+31:i]), ABS(b[i+31:i])) ELSE dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Convert - - - Sign extend packed 16-bit integers in "a" to packed 32-bit integers, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - -FOR j := 0 to 3 - i := 32*j - l := 16*j - IF k[j] - dst[i+31:i] := SignExtend(a[l+15:l]) - ELSE - dst[i+31:i] := 0 - FI + + Floating Point + KNCNI + Special Math Functions + + + + Determines the maximum of each pair of corresponding elements in packed double-precision (64-bit) floating-point elements in "a" and "b", storing the results in "dst". + FOR j := 0 to 7 + i := j*64 + dst[i+63:i] := FpMax(a[i+63:i], b[i+63:i]) ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Convert - - - - Sign extend packed 16-bit integers in the low 8 bytes of "a" to packed 64-bit integers, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -FOR j := 0 to 3 - i := 64*j - l := 16*j + + Floating Point + KNCNI + Special Math Functions + + + + + + Determines the maximum of each pair of corresponding elements of packed double-precision (64-bit) floating-point elements in "a" and "b", storing the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + FOR j := 0 to 7 + i := j*64 IF k[j] - dst[i+63:i] := SignExtend(a[l+15:l]) + dst[i+63:i] := FpMax(a[i+63:i], b[i+63:i]) ELSE dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Convert - - - Sign extend packed 16-bit integers in the low 8 bytes of "a" to packed 64-bit integers, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - -FOR j := 0 to 3 - i := 64*j - l := 16*j - IF k[j] - dst[i+63:i] := SignExtend(a[l+15:l]) - ELSE - dst[i+63:i] := 0 - FI + + Floating Point + KNCNI + Special Math Functions + + + + Determines the minimum of each pair of corresponding elements in packed single-precision (32-bit) floating-point elements in "a" and "b", storing the results in "dst". + FOR j := 0 to 15 + i := j*32 + dst[i+31:i] := FpMin(a[i+31:i], b[i+31:i]) ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Convert - - - - Sign extend packed 16-bit integers in the low 4 bytes of "a" to packed 64-bit integers, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -FOR j := 0 to 1 - i := 64*j - l := 16*j + + Floating Point + KNCNI + Special Math Functions + + + + + + Determines the maximum of each pair of corresponding elements of packed single-precision (32-bit) floating-point elements in "a" and "b", storing the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + FOR j := 0 to 15 + i := j*32 IF k[j] - dst[i+63:i] := SignExtend(a[l+15:l]) + dst[i+31:i] := FpMin(a[i+31:i], b[i+31:i]) ELSE - dst[i+63:i] := src[i+63:i] + dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Convert - - - Sign extend packed 16-bit integers in the low 4 bytes of "a" to packed 64-bit integers, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - -FOR j := 0 to 1 - i := 64*j - l := 16*j + + Floating Point + KNCNI + Special Math Functions + + + + Determines the minimum of each pair of corresponding elements in packed double-precision (64-bit) floating-point elements in "a" and "b", storing the results in "dst". + FOR j := 0 to 7 + i := j*64 + dst[i+63:i] := FpMin(a[i+63:i], b[i+63:i]) +ENDFOR +dst[MAX:512] := 0 + + +
immintrin.h
+ + + Floating Point + KNCNI + Special Math Functions + + + + + + Determines the maximum of each pair of corresponding elements of packed double-precision (64-bit) floating-point elements in "a" and "b", storing the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + FOR j := 0 to 7 + i := j*64 IF k[j] - dst[i+63:i] := SignExtend(a[l+15:l]) + dst[i+63:i] := FpMin(a[i+63:i], b[i+63:i]) ELSE - dst[i+63:i] := 0 + dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX512VL - AVX512F - Convert - - Convert packed unsigned 32-bit integers in "a" to packed unsigned 8-bit integers with unsigned saturation, and store the results in "dst". + KNCNI + Arithmetic + + + + Performs element-by-element multiplication between packed 32-bit integer elements in "a" and "b" and stores the high 32 bits of each result into "dst". -FOR j := 0 to 7 - i := 32*j - k := 8*j - dst[k+7:k] := Saturate_UnsignedInt32_To_Int8(a[i+31:i]) +FOR j := 0 to 15 + i := j*32 + dst[i+31:i] := (a[i+31:i] * b[i+31:i]) >> 32 ENDFOR -dst[MAX:64] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX512VL - AVX512F - Convert - - - - Convert packed unsigned 32-bit integers in "a" to packed unsigned 8-bit integers with unsigned saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + KNCNI + Arithmetic + + + + + + Performs element-by-element multiplication between packed 32-bit integer elements in "a" and "b" and stores the high 32 bits of each result into "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := 32*j - l := 8*j +FOR j := 0 to 15 + i := j*32 IF k[j] - dst[l+7:l] := Saturate_UnsignedInt32_To_Int8(a[i+31:i]) + dst[i+31:i] := (a[i+31:i] * b[i+31:i]) >> 32 ELSE - dst[l+7:l] := src[l+7:l] + dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:64] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX512VL - AVX512F - Convert - Store - - - - Convert packed unsigned 32-bit integers in "a" to packed unsigned 8-bit integers with unsigned saturation, and store the active results (those with their respective bit set in writemask "k") to unaligned memory at "base_addr". + KNCNI + Arithmetic + + + + Performs element-by-element multiplication between packed unsigned 32-bit integer elements in "a" and "b" and stores the high 32 bits of each result into "dst". -FOR j := 0 to 7 - i := 32*j - l := 8*j - IF k[j] - MEM[base_addr+l+7:base_addr+l] := Saturate_UnsignedInt32_To_Int8(a[i+31:i]) - FI +FOR j := 0 to 15 + i := j*32 + dst[i+31:i] := (a[i+31:i] * b[i+31:i]) >> 32 ENDFOR -dst[MAX:64] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX512VL - AVX512F - Convert - - - Convert packed unsigned 32-bit integers in "a" to packed unsigned 8-bit integers with unsigned saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + KNCNI + Arithmetic + + + + + + Performs element-by-element multiplication between packed unsigned 32-bit integer elements in "a" and "b" and stores the high 32 bits of each result into "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 7 - i := 32*j - l := 8*j +FOR j := 0 to 15 + i := j*32 IF k[j] - dst[l+7:l] := Saturate_UnsignedInt32_To_Int8(a[i+31:i]) + dst[i+31:i] := (a[i+31:i] * b[i+31:i]) >> 32 ELSE - dst[l+7:l] := 0 + dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:64] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX512VL - AVX512F - Convert - - Convert packed unsigned 32-bit integers in "a" to packed unsigned 8-bit integers with unsigned saturation, and store the results in "dst". + KNCNI + Swizzle + + + + Permutes 128-bit blocks of the packed 32-bit integer vector "a" using constant "imm8". The results are stored in "dst". +DEFINE SELECT4(src, control) { + CASE control[1:0] OF + 0: tmp[127:0] := src[127:0] + 1: tmp[127:0] := src[255:128] + 2: tmp[127:0] := src[383:256] + 3: tmp[127:0] := src[511:384] + ESAC + RETURN tmp[127:0] +} FOR j := 0 to 3 - i := 32*j - k := 8*j - dst[k+7:k] := Saturate_UnsignedInt32_To_Int8(a[i+31:i]) + i := j*128 + n := j*2 + dst[i+127:i] := SELECT4(a[511:0], imm8[n+1:n]) ENDFOR -dst[MAX:32] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX512VL - AVX512F - Convert - - - - Convert packed unsigned 32-bit integers in "a" to packed unsigned 8-bit integers with unsigned saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + KNCNI + Swizzle + + + + + + Permutes 128-bit blocks of the packed 32-bit integer vector "a" using constant "imm8". The results are stored in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). +DEFINE SELECT4(src, control) { + CASE control[1:0] OF + 0: tmp[127:0] := src[127:0] + 1: tmp[127:0] := src[255:128] + 2: tmp[127:0] := src[383:256] + 3: tmp[127:0] := src[511:384] + ESAC + RETURN tmp[127:0] +} +tmp[511:0] := 0 FOR j := 0 to 3 - i := 32*j - l := 8*j + i := j*128 + n := j*2 + tmp[i+127:i] := SELECT4(a[511:0], imm8[n+1:n]) +ENDFOR +FOR j := 0 to 15 IF k[j] - dst[l+7:l] := Saturate_UnsignedInt32_To_Int8(a[i+31:i]) + dst[i+31:i] := tmp[i+31:i] ELSE - dst[l+7:l] := src[l+7:l] + dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:32] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Convert - Store - - - - Convert packed unsigned 32-bit integers in "a" to packed unsigned 8-bit integers with unsigned saturation, and store the active results (those with their respective bit set in writemask "k") to unaligned memory at "base_addr". - -FOR j := 0 to 3 - i := 32*j - l := 8*j - IF k[j] - MEM[base_addr+l+7:base_addr+l] := Saturate_UnsignedInt32_To_Int8(a[i+31:i]) - FI + + Floating Point + KNCNI + Elementary Math Functions + + + Approximates the reciprocals of packed single-precision (32-bit) floating-point elements in "a" to 23 bits of precision, storing the results in "dst". + FOR j := 0 to 15 + i := j*32 + dst[i+31:i] := (1.0 / a[i+31:i]) ENDFOR -dst[MAX:32] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Convert - - - Convert packed unsigned 32-bit integers in "a" to packed unsigned 8-bit integers with unsigned saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - -FOR j := 0 to 3 - i := 32*j - l := 8*j + + Floating Point + KNCNI + Elementary Math Functions + + + + + Approximates the reciprocals of packed single-precision (32-bit) floating-point elements in "a" to 23 bits of precision, storing the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + FOR j := 0 to 15 + i := j*32 IF k[j] - dst[l+7:l] := Saturate_UnsignedInt32_To_Int8(a[i+31:i]) + dst[i+31:i] := (1.0 / a[i+31:i]) ELSE - dst[l+7:l] := 0 + dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:32] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512F + + Floating Point + KNCNI Convert - - Convert packed unsigned 32-bit integers in "a" to packed unsigned 16-bit integers with unsigned saturation, and store the results in "dst". - -FOR j := 0 to 7 - i := 32*j - k := 16*j - dst[k+15:k] := Saturate_UnsignedInt32_To_Int16(a[i+31:i]) + + + + + Round the packed single-precision (32-bit) floating-point elements in "a" to the nearest integer value using "expadj" and in the direction of "rounding", and store the results as packed single-precision floating-point elements in "dst". + [round_note] + FOR j := 0 to 15 + i := j*32 + dst[i+31:i] := ROUND(a[i+31:i]) + CASE expadj OF + _MM_EXPADJ_NONE: dst[i+31:i] := dst[i+31:i] * (2 << 0) + _MM_EXPADJ_4: dst[i+31:i] := dst[i+31:i] * (2 << 4) + _MM_EXPADJ_5: dst[i+31:i] := dst[i+31:i] * (2 << 5) + _MM_EXPADJ_8: dst[i+31:i] := dst[i+31:i] * (2 << 8) + _MM_EXPADJ_16: dst[i+31:i] := dst[i+31:i] * (2 << 16) + _MM_EXPADJ_24: dst[i+31:i] := dst[i+31:i] * (2 << 24) + _MM_EXPADJ_31: dst[i+31:i] := dst[i+31:i] * (2 << 31) + _MM_EXPADJ_32: dst[i+31:i] := dst[i+31:i] * (2 << 32) + ESAC ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512F + + Floating Point + KNCNI Convert - - - - Convert packed unsigned 32-bit integers in "a" to packed unsigned 16-bit integers with unsigned saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -FOR j := 0 to 7 - i := 32*j - l := 16*j + + + + + + + Round the packed single-precision (32-bit) floating-point elements in "a" to the nearest integer value using "expadj" and in the direction of "rounding", and store the results as packed single-precision floating-point elements in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + [round_note] + FOR j := 0 to 15 + i := j*32 IF k[j] - dst[l+15:l] := Saturate_UnsignedInt32_To_Int16(a[i+31:i]) + dst[i+31:i] := ROUND(a[i+31:i]) + CASE expadj OF + _MM_EXPADJ_NONE: dst[i+31:i] := dst[i+31:i] * (2 << 0) + _MM_EXPADJ_4: dst[i+31:i] := dst[i+31:i] * (2 << 4) + _MM_EXPADJ_5: dst[i+31:i] := dst[i+31:i] * (2 << 5) + _MM_EXPADJ_8: dst[i+31:i] := dst[i+31:i] * (2 << 8) + _MM_EXPADJ_16: dst[i+31:i] := dst[i+31:i] * (2 << 16) + _MM_EXPADJ_24: dst[i+31:i] := dst[i+31:i] * (2 << 24) + _MM_EXPADJ_31: dst[i+31:i] := dst[i+31:i] * (2 << 31) + _MM_EXPADJ_32: dst[i+31:i] := dst[i+31:i] * (2 << 32) + ESAC ELSE - dst[l+15:l] := src[l+15:l] + dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Convert - Store - - - - Convert packed unsigned 32-bit integers in "a" to packed unsigned 16-bit integers with unsigned saturation, and store the active results (those with their respective bit set in writemask "k") to unaligned memory at "base_addr". - -FOR j := 0 to 7 - i := 32*j - l := 16*j - IF k[j] - MEM[base_addr+l+15:base_addr+l] := Saturate_UnsignedInt32_To_Int16(a[i+31:i]) - FI + + Floating Point + KNCNI + Miscellaneous + + + + + Performs element-by-element rounding of packed single-precision (32-bit) floating-point elements in "a" using "expadj" and in the direction of "rounding" and stores results in "dst". + [round_note] + FOR j := 0 to 15 + i := j*32 + dst[i+31:i] := ROUND(a[i+31:i]) + CASE expadj OF + _MM_EXPADJ_NONE: dst[i+31:i] := dst[i+31:i] * (2 << 0) + _MM_EXPADJ_4: dst[i+31:i] := dst[i+31:i] * (2 << 4) + _MM_EXPADJ_5: dst[i+31:i] := dst[i+31:i] * (2 << 5) + _MM_EXPADJ_8: dst[i+31:i] := dst[i+31:i] * (2 << 8) + _MM_EXPADJ_16: dst[i+31:i] := dst[i+31:i] * (2 << 16) + _MM_EXPADJ_24: dst[i+31:i] := dst[i+31:i] * (2 << 24) + _MM_EXPADJ_31: dst[i+31:i] := dst[i+31:i] * (2 << 31) + _MM_EXPADJ_32: dst[i+31:i] := dst[i+31:i] * (2 << 32) + ESAC ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Convert - - - Convert packed unsigned 32-bit integers in "a" to packed unsigned 16-bit integers with unsigned saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - -FOR j := 0 to 7 - i := 32*j - l := 16*j + + Floating Point + KNCNI + Miscellaneous + + + + + + + Performs element-by-element rounding of packed single-precision (32-bit) floating-point elements in "a" using "expadj" and in the direction of "rounding" and stores results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + [round_note] + FOR j := 0 to 15 + i := j*32 IF k[j] - dst[l+15:l] := Saturate_UnsignedInt32_To_Int16(a[i+31:i]) + dst[i+31:i] := ROUND(a[i+31:i]) + CASE expadj OF + _MM_EXPADJ_NONE: dst[i+31:i] := dst[i+31:i] * (2 << 0) + _MM_EXPADJ_4: dst[i+31:i] := dst[i+31:i] * (2 << 4) + _MM_EXPADJ_5: dst[i+31:i] := dst[i+31:i] * (2 << 5) + _MM_EXPADJ_8: dst[i+31:i] := dst[i+31:i] * (2 << 8) + _MM_EXPADJ_16: dst[i+31:i] := dst[i+31:i] * (2 << 16) + _MM_EXPADJ_24: dst[i+31:i] := dst[i+31:i] * (2 << 24) + _MM_EXPADJ_31: dst[i+31:i] := dst[i+31:i] * (2 << 31) + _MM_EXPADJ_32: dst[i+31:i] := dst[i+31:i] * (2 << 32) + ESAC ELSE - dst[l+15:l] := 0 + dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Convert - - Convert packed unsigned 32-bit integers in "a" to packed unsigned 16-bit integers with unsigned saturation, and store the results in "dst". - -FOR j := 0 to 3 - i := 32*j - k := 16*j - dst[k+15:k] := Saturate_UnsignedInt32_To_Int16(a[i+31:i]) + + Floating Point + KNCNI + Miscellaneous + + + + + Performs element-by-element rounding of packed double-precision (64-bit) floating-point elements in "a" using "expadj" and in the direction of "rounding" and stores results in "dst". + [round_note] + FOR j := 0 to 7 + i := j*64 + dst[i+63:i] := ROUND(a[i+63:i]) + CASE expadj OF + _MM_EXPADJ_NONE: dst[i+31:i] := dst[i+31:i] * (2 << 0) + _MM_EXPADJ_4: dst[i+31:i] := dst[i+31:i] * (2 << 4) + _MM_EXPADJ_5: dst[i+31:i] := dst[i+31:i] * (2 << 5) + _MM_EXPADJ_8: dst[i+31:i] := dst[i+31:i] * (2 << 8) + _MM_EXPADJ_16: dst[i+31:i] := dst[i+31:i] * (2 << 16) + _MM_EXPADJ_24: dst[i+31:i] := dst[i+31:i] * (2 << 24) + _MM_EXPADJ_31: dst[i+31:i] := dst[i+31:i] * (2 << 31) + _MM_EXPADJ_32: dst[i+31:i] := dst[i+31:i] * (2 << 32) + ESAC ENDFOR -dst[MAX:64] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Convert - - - - Convert packed unsigned 32-bit integers in "a" to packed unsigned 16-bit integers with unsigned saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -FOR j := 0 to 3 - i := 32*j - l := 16*j + + Floating Point + KNCNI + Miscellaneous + + + + + + + Performs element-by-element rounding of packed double-precision (64-bit) floating-point elements in "a" using "expadj" and in the direction of "rounding" and stores results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + [round_note] + FOR j := 0 to 7 + i := j*64 IF k[j] - dst[l+15:l] := Saturate_UnsignedInt32_To_Int16(a[i+31:i]) + dst[i+63:i] := ROUND(a[i+63:i]) + CASE expadj OF + _MM_EXPADJ_NONE: dst[i+31:i] := dst[i+31:i] * (2 << 0) + _MM_EXPADJ_4: dst[i+31:i] := dst[i+31:i] * (2 << 4) + _MM_EXPADJ_5: dst[i+31:i] := dst[i+31:i] * (2 << 5) + _MM_EXPADJ_8: dst[i+31:i] := dst[i+31:i] * (2 << 8) + _MM_EXPADJ_16: dst[i+31:i] := dst[i+31:i] * (2 << 16) + _MM_EXPADJ_24: dst[i+31:i] := dst[i+31:i] * (2 << 24) + _MM_EXPADJ_31: dst[i+31:i] := dst[i+31:i] * (2 << 31) + _MM_EXPADJ_32: dst[i+31:i] := dst[i+31:i] * (2 << 32) + ESAC ELSE - dst[l+15:l] := src[l+15:l] + dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:64] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Convert - Store - - - - Convert packed unsigned 32-bit integers in "a" to packed unsigned 16-bit integers with unsigned saturation, and store the active results (those with their respective bit set in writemask "k") to unaligned memory at "base_addr". - -FOR j := 0 to 3 - i := 32*j - l := 16*j - IF k[j] - MEM[base_addr+l+15:base_addr+l] := Saturate_UnsignedInt32_To_Int16(a[i+31:i]) - FI + + Floating Point + KNCNI + Elementary Math Functions + + + Calculates the reciprocal square root of packed single-precision (32-bit) floating-point elements in "a" to 23 bits of accuracy and stores the result in "dst". + FOR j := 0 to 15 + i := j*32 + dst[i+31:i] := Sqrt(1.0 / a[i+31:i]) ENDFOR -dst[MAX:64] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Convert - - - Convert packed unsigned 32-bit integers in "a" to packed unsigned 16-bit integers with unsigned saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - -FOR j := 0 to 3 - i := 32*j - l := 16*j + + Floating Point + KNCNI + Elementary Math Functions + + + + + Calculates the reciprocal square root of packed single-precision (32-bit) floating-point elements in "a" to 23 bits of accuracy and stores the result in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + FOR j := 0 to 15 + i := j*32 IF k[j] - dst[l+15:l] := Saturate_UnsignedInt32_To_Int16(a[i+31:i]) + dst[i+31:i] := Sqrt(1.0 / a[i+31:i]) ELSE - dst[l+15:l] := 0 + dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:64] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Convert - - Convert packed unsigned 64-bit integers in "a" to packed unsigned 8-bit integers with unsigned saturation, and store the results in "dst". + + Floating Point + KNCNI + Arithmetic + + + + Scales each single-precision (32-bit) floating-point element in "a" by multiplying it by 2**exponent, where the exponent is the corresponding 32-bit integer element in "b", storing results in "dst". -FOR j := 0 to 3 - i := 64*j - k := 8*j - dst[k+7:k] := Saturate_UnsignedInt64_To_Int8(a[i+63:i]) +FOR j := 0 to 15 + i := j*32 + dst[i+31:i] := a[i+31:i] * POW(2.0, FP32(b[i+31:i])) ENDFOR -dst[MAX:32] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Convert - - - - Convert packed unsigned 64-bit integers in "a" to packed unsigned 8-bit integers with unsigned saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Floating Point + KNCNI + Arithmetic + + + + + + Scales each single-precision (32-bit) floating-point element in "a" by multiplying it by 2**exponent, where the exponent is the corresponding 32-bit integer element in "b", storing results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 3 - i := 64*j - l := 8*j +FOR j := 0 to 15 + i := j*32 IF k[j] - dst[l+7:l] := Saturate_UnsignedInt64_To_Int8(a[i+63:i]) + dst[i+31:i] := a[i+31:i] * POW(2.0, FP32(b[i+31:i])) ELSE - dst[l+7:l] := src[l+7:l] + dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:32] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Convert - Store - - - - Convert packed unsigned 64-bit integers in "a" to packed unsigned 8-bit integers with unsigned saturation, and store the active results (those with their respective bit set in writemask "k") to unaligned memory at "base_addr". + + Floating Point + KNCNI + Arithmetic + + + + + Scales each single-precision (32-bit) floating-point element in "a" by multiplying it by 2**exponent, where the exponent is the corresponding 32-bit integer element in "b", storing results in "dst". Intermediate elements are rounded using "rounding". + [round_note] -FOR j := 0 to 3 - i := 64*j - l := 8*j - IF k[j] - MEM[base_addr+l+7:base_addr+l] := Saturate_UnsignedInt64_To_Int8(a[i+63:i]) - FI +FOR j := 0 to 15 + i := j*32 + dst[i+31:i] := a[i+31:i] * POW(2.0,FP32(b[i+31:i])) ENDFOR -dst[MAX:32] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Convert - - - Convert packed unsigned 64-bit integers in "a" to packed unsigned 8-bit integers with unsigned saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Floating Point + KNCNI + Arithmetic + + + + + + + Scales each single-precision (32-bit) floating-point element in "a" by multiplying it by 2**exp, where the exp is the corresponding 32-bit integer element in "b", storing results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). Results are rounded using constant "rounding". + [round_note] -FOR j := 0 to 3 - i := 64*j - l := 8*j +FOR j := 0 to 15 + i := j*32 IF k[j] - dst[l+7:l] := Saturate_UnsignedInt64_To_Int8(a[i+63:i]) + dst[i+31:i] := a[i+31:i] * POW(2.0, FP32(b[i+31:i])) ELSE - dst[l+7:l] := 0 + dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:32] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Convert - - Convert packed unsigned 64-bit integers in "a" to packed unsigned 8-bit integers with unsigned saturation, and store the results in "dst". - -FOR j := 0 to 1 - i := 64*j - k := 8*j - dst[k+7:k] := Saturate_UnsignedInt64_To_Int8(a[i+63:i]) + + Floating Point + KNCNI + Special Math Functions + + + Determines the minimum element of the packed single-precision (32-bit) floating-point elements stored in "a" and stores the result in "dst". + min := a[31:0] +FOR j := 1 to 15 + i := j*32 + dst := FpMin(min, a[i+31:i]) ENDFOR -dst[MAX:16] := 0 +dst := min -
immintrin.h
 - - Integer - AVX512VL - AVX512F - Convert - - - - Convert packed unsigned 64-bit integers in "a" to packed unsigned 8-bit integers with unsigned saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -FOR j := 0 to 1 - i := 64*j - l := 8*j + + Floating Point + KNCNI + Special Math Functions + + + + Determines the minimum element of the packed single-precision (32-bit) floating-point elements stored in "a" and stores the result in "dst" using writemask "k" (elements are ignored when the corresponding mask bit is not set). + min := a[31:0] +FOR j := 1 to 15 + i := j*32 IF k[j] - dst[l+7:l] := Saturate_UnsignedInt64_To_Int8(a[i+63:i]) + CONTINUE ELSE - dst[l+7:l] := src[l+7:l] + dst := FpMin(min, a[i+31:i]) FI ENDFOR -dst[MAX:16] := 0 +dst := min -
immintrin.h
 - - Integer - AVX512VL - AVX512F - Convert - Store - - - - Convert packed unsigned 64-bit integers in "a" to packed unsigned 8-bit integers with unsigned saturation, and store the active results (those with their respective bit set in writemask "k") to unaligned memory at "base_addr". - -FOR j := 0 to 1 - i := 64*j - l := 8*j - IF k[j] - MEM[base_addr+l+7:base_addr+l] := Saturate_UnsignedInt64_To_Int8(a[i+63:i]) - FI + + Floating Point + KNCNI + Special Math Functions + + + Determines the minimum element of the packed double-precision (64-bit) floating-point elements stored in "a" and stores the result in "dst". + min := a[63:0] +FOR j := 1 to 7 + i := j*64 + dst := FpMin(min, a[i+63:i]) ENDFOR -dst[MAX:16] := 0 +dst := min -
immintrin.h
 - - Integer - AVX512VL - AVX512F - Convert - - - Convert packed unsigned 64-bit integers in "a" to packed unsigned 8-bit integers with unsigned saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - -FOR j := 0 to 1 - i := 64*j - l := 8*j + + Floating Point + KNCNI + Special Math Functions + + + + Determines the minimum element of the packed double-precision (64-bit) floating-point elements stored in "a" and stores the result in "dst". Bitmask "k" is used to exclude certain elements (elements are ignored when the corresponding mask bit is not set). + min := a[63:0] +FOR j := 1 to 7 + i := j*64 IF k[j] - dst[l+7:l] := Saturate_UnsignedInt64_To_Int8(a[i+63:i]) + CONTINUE ELSE - dst[l+7:l] := 0 + dst := FpMin(min, a[i+63:i]) FI ENDFOR -dst[MAX:16] := 0 +dst := min -
immintrin.h
 - - Integer - AVX512VL - AVX512F - Convert - - Convert packed unsigned 64-bit integers in "a" to packed unsigned 32-bit integers with unsigned saturation, and store the results in "dst". - -FOR j := 0 to 3 - i := 64*j - k := 32*j - dst[k+31:k] := Saturate_UnsignedInt64_To_Int32(a[i+63:i]) + + Floating Point + KNCNI + Special Math Functions + + + Determines the maximum element of the packed single-precision (32-bit) floating-point elements stored in "a" and stores the result in "dst". + max := a[31:0] +FOR j := 1 to 15 + i := j*32 + dst := FpMax(max, a[i+31:i]) ENDFOR -dst[MAX:128] := 0 +dst := max -
immintrin.h
 - - Integer - AVX512VL - AVX512F - Convert - - - - Convert packed unsigned 64-bit integers in "a" to packed unsigned 32-bit integers with unsigned saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -FOR j := 0 to 3 - i := 64*j - l := 32*j + + Floating Point + KNCNI + Special Math Functions + + + + Determines the maximum element of the packed single-precision (32-bit) floating-point elements stored in "a" and stores the result in "dst". Bitmask "k" is used to exclude certain elements (elements are ignored when the corresponding mask bit is not set). + max := a[31:0] +FOR j := 1 to 15 + i := j*32 IF k[j] - dst[l+31:l] := Saturate_UnsignedInt64_To_Int32(a[i+63:i]) + CONTINUE ELSE - dst[l+31:l] := src[l+31:l] + dst := FpMax(max, a[i+31:i]) FI ENDFOR -dst[MAX:128] := 0 +dst := max -
immintrin.h
 - - Integer - AVX512VL - AVX512F - Convert - Store - - - - Convert packed unsigned 64-bit integers in "a" to packed unsigned 32-bit integers with unsigned saturation, and store the active results (those with their respective bit set in writemask "k") to unaligned memory at "base_addr". - -FOR j := 0 to 3 - i := 64*j - l := 32*j - IF k[j] - MEM[base_addr+l+31:base_addr+l] := Saturate_UnsignedInt64_To_Int32(a[i+63:i]) - FI + + Floating Point + KNCNI + Special Math Functions + + + Determines the maximum element of the packed double-precision (64-bit) floating-point elements stored in "a" and stores the result in "dst". + max := a[63:0] +FOR j := 1 to 7 + i := j*64 + dst := FpMax(max, a[i+63:i]) ENDFOR -dst[MAX:128] := 0 +dst := max -
immintrin.h
 - - Integer - AVX512VL - AVX512F - Convert - - - Convert packed unsigned 64-bit integers in "a" to packed unsigned 32-bit integers with unsigned saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - -FOR j := 0 to 3 - i := 64*j - l := 32*j + + Floating Point + KNCNI + Special Math Functions + + + + Determines the maximum element of the packed double-precision (64-bit) floating-point elements stored in "a" and stores the result in "dst". Bitmask "k" is used to exclude certain elements (elements are ignored when the corresponding mask bit is not set). + max := a[63:0] +FOR j := 1 to 7 + i := j*64 IF k[j] - dst[l+31:l] := Saturate_UnsignedInt64_To_Int32(a[i+63:i]) + CONTINUE ELSE - dst[l+31:l] := 0 + dst := FpMax(max, a[i+63:i]) FI ENDFOR -dst[MAX:128] := 0 +dst := max -
immintrin.h
 - - Integer - AVX512VL - AVX512F - Convert - - Convert packed unsigned 64-bit integers in "a" to packed unsigned 32-bit integers with unsigned saturation, and store the results in "dst". + + KNCNI + Bit Manipulation + + + + Count the number of trailing zero bits in unsigned 32-bit integer "x" starting at bit "a", and return that count in "dst". -FOR j := 0 to 1 - i := 64*j - k := 32*j - dst[k+31:k] := Saturate_UnsignedInt64_To_Int32(a[i+63:i]) -ENDFOR -dst[MAX:64] := 0 +tmp := a +IF tmp < 0 + tmp := 0 +FI +dst := 0 +IF tmp > 31 + dst := 32 +ELSE + DO WHILE ((tmp < 32) AND x[tmp] == 0) + tmp := tmp + 1 + dst := dst + 1 + OD +FI - +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Convert - - - - Convert packed unsigned 64-bit integers in "a" to packed unsigned 32-bit integers with unsigned saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + KNCNI + Bit Manipulation + + + + Count the number of trailing zero bits in unsigned 64-bit integer "x" starting at bit "a", and return that count in "dst". -FOR j := 0 to 1 - i := 64*j - l := 32*j - IF k[j] - dst[l+31:l] := Saturate_UnsignedInt64_To_Int32(a[i+63:i]) - ELSE - dst[l+31:l] := src[l+31:l] - FI -ENDFOR -dst[MAX:64] := 0 +tmp := a +IF tmp < 0 + tmp := 0 +FI +dst := 0 +IF tmp > 63 + dst := 64 +ELSE + DO WHILE ((tmp < 64) AND x[tmp] == 0) + tmp := tmp + 1 + dst := dst + 1 + OD +FI - +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Convert - Store - - - - Convert packed unsigned 64-bit integers in "a" to packed unsigned 32-bit integers with unsigned saturation, and store the active results (those with their respective bit set in writemask "k") to unaligned memory at "base_addr". - -FOR j := 0 to 1 - i := 64*j - l := 32*j - IF k[j] - MEM[base_addr+l+31:base_addr+l] := Saturate_UnsignedInt64_To_Int32(a[i+63:i]) - FI -ENDFOR -dst[MAX:64] := 0 + + KNCNI + General Support + + + Stalls a thread without blocking other threads for 32-bit unsigned integer "r1" clock cycles. + BlockThread(r1) - +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Convert - - - Convert packed unsigned 64-bit integers in "a" to packed unsigned 32-bit integers with unsigned saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - -FOR j := 0 to 1 - i := 64*j - l := 32*j - IF k[j] - dst[l+31:l] := Saturate_UnsignedInt64_To_Int32(a[i+63:i]) - ELSE - dst[l+31:l] := 0 - FI -ENDFOR -dst[MAX:64] := 0 + + KNCNI + General Support + + + Stalls a thread without blocking other threads for 64-bit unsigned integer "r1" clock cycles. + BlockThread(r1) - +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Convert - - Convert packed unsigned 64-bit integers in "a" to packed unsigned 16-bit integers with unsigned saturation, and store the results in "dst". - -FOR j := 0 to 3 - i := 64*j - k := 16*j - dst[k+15:k] := Saturate_UnsignedInt64_To_Int16(a[i+63:i]) -ENDFOR -dst[MAX:64] := 0 + + KNCNI + General Support + + + Set performance monitoring filtering mask to 32-bit unsigned integer "r1". + SetPerfMonMask(r1[31:0]) - +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Convert - - - - Convert packed unsigned 64-bit integers in "a" to packed unsigned 16-bit integers with unsigned saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -FOR j := 0 to 3 - i := 64*j - l := 16*j - IF k[j] - dst[l+15:l] := Saturate_UnsignedInt64_To_Int16(a[i+63:i]) - ELSE - dst[l+15:l] := src[l+15:l] - FI -ENDFOR -dst[MAX:64] := 0 + + KNCNI + General Support + + + Set performance monitoring filtering mask to 64-bit unsigned integer "r1". + SetPerfMonMask(r1[63:0]) - +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Convert - Store - - - - Convert packed unsigned 64-bit integers in "a" to packed unsigned 16-bit integers with unsigned saturation, and store the active results (those with their respective bit set in writemask "k") to unaligned memory at "base_addr". - -FOR j := 0 to 3 - i := 64*j - l := 16*j - IF k[j] - MEM[base_addr+l+15:base_addr+l] := Saturate_UnsignedInt64_To_Int16(a[i+63:i]) - FI -ENDFOR -dst[MAX:64] := 0 + + KNCNI + General Support + + + + Evicts the cache line containing the address "ptr" from cache level "level" (can be either 0 or 1). + CacheLineEvict(ptr, level) - + +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Convert - - - Convert packed unsigned 64-bit integers in "a" to packed unsigned 16-bit integers with unsigned saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - -FOR j := 0 to 3 - i := 64*j - l := 16*j - IF k[j] - dst[l+15:l] := Saturate_UnsignedInt64_To_Int16(a[i+63:i]) - ELSE - dst[l+15:l] := 0 - FI -ENDFOR -dst[MAX:64] := 0 + + Mask + KNCNI + Mask + + + + Performs a bitwise AND operation between NOT of "k2" and "k1", storing the result in "dst". + dst[15:0] := NOT(k2[15:0]) & k1[15:0] - +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Convert - - Convert packed unsigned 64-bit integers in "a" to packed unsigned 16-bit integers with unsigned saturation, and store the results in "dst". + + Mask + KNCNI + Mask + + + + Moves high byte from "k2" to low byte of "k1", and moves low byte of "k2" to high byte of "k1". -FOR j := 0 to 1 - i := 64*j - k := 16*j - dst[k+15:k] := Saturate_UnsignedInt64_To_Int16(a[i+63:i]) -ENDFOR -dst[MAX:32] := 0 +tmp[7:0] := k2[15:8] +k2[15:8] := k1[7:0] +k1[7:0] := tmp[7:0] +tmp[7:0] := k2[7:0] +k2[7:0] := k1[15:8] +k1[15:8] := tmp[7:0] - +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Convert - - - - Convert packed unsigned 64-bit integers in "a" to packed unsigned 16-bit integers with unsigned saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -FOR j := 0 to 1 - i := 64*j - l := 16*j - IF k[j] - dst[l+15:l] := Saturate_UnsignedInt64_To_Int16(a[i+63:i]) - ELSE - dst[l+15:l] := src[l+15:l] - FI -ENDFOR -dst[MAX:32] := 0 + + Mask + KNCNI + Mask + + + + Performs bitwise OR between "k1" and "k2", storing the result in "dst". ZF flag is set if "dst" is 0. + dst[15:0] := k1[15:0] | k2[15:0] +IF dst == 0 + SetZF() +FI - +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Convert - Store - - - - Convert packed unsigned 64-bit integers in "a" to packed unsigned 16-bit integers with unsigned saturation, and store the active results (those with their respective bit set in writemask "k") to unaligned memory at "base_addr". + + Mask + KNCNI + Mask + + + + Performs bitwise OR between "k1" and "k2", storing the result in "dst". CF flag is set if "dst" consists of all 1's. + dst[15:0] := k1[15:0] | k2[15:0] +IF PopCount(dst[15:0]) == 16 + SetCF() +FI + + +
immintrin.h
+ + + KNCNI + Mask + + + Converts bit mask "k1" into an integer value, storing the results in "dst". -FOR j := 0 to 1 - i := 64*j - l := 16*j - IF k[j] - MEM[base_addr+l+15:base_addr+l] := Saturate_UnsignedInt64_To_Int16(a[i+63:i]) - FI -ENDFOR -dst[MAX:32] := 0 +dst := ZeroExtend32(k1) - +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Convert - - - Convert packed unsigned 64-bit integers in "a" to packed unsigned 16-bit integers with unsigned saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + KNCNI + Mask + + + Converts integer "mask" into bitmask, storing the result in "dst". -FOR j := 0 to 1 - i := 64*j - l := 16*j - IF k[j] - dst[l+15:l] := Saturate_UnsignedInt64_To_Int16(a[i+63:i]) - ELSE - dst[l+15:l] := 0 - FI -ENDFOR -dst[MAX:32] := 0 +dst := mask[15:0] - +
immintrin.h
 - - Integer - AVX512VL - AVX512BW - Convert - - Convert packed unsigned 16-bit integers in "a" to packed unsigned 8-bit integers with unsigned saturation, and store the results in "dst". + + Mask + KNCNI + Mask + + + + Packs masks "k1" and "k2" into the high 32 bits of "dst". The rest of "dst" is set to 0. -FOR j := 0 to 15 - i := 16*j - l := 8*j - dst[l+7:l] := Saturate_UnsignedInt16_To_Int8(a[i+15:i]) -ENDFOR -dst[MAX:128] := 0 +dst[63:48] := k1[15:0] +dst[47:32] := k2[15:0] +dst[31:0] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512BW - Convert - - - - Convert packed unsigned 16-bit integers in "a" to packed unsigned 8-bit integers with unsigned saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Mask + KNCNI + Mask + + + + Packs masks "k1" and "k2" into the low 32 bits of "dst". The rest of "dst" is set to 0. -FOR j := 0 to 15 - i := 16*j - l := 8*j - IF k[j] - dst[l+7:l] := Saturate_UnsignedInt16_To_Int8(a[i+15:i]) - ELSE - dst[l+7:l] := src[l+7:l] - FI -ENDFOR -dst[MAX:128] := 0 +dst[31:16] := k1[15:0] +dst[15:0] := k2[15:0] +dst[63:32] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512BW - Convert - Store - - - - Convert packed unsigned 16-bit integers in "a" to packed unsigned 8-bit integers with unsigned saturation, and store the active results (those with their respective bit set in writemask "k") to unaligned memory at "base_addr". + + Mask + KNCNI + Mask + + + + Extracts 16-bit value "b" from 64-bit integer "a", storing the result in "dst". -FOR j := 0 to 15 - i := 16*j - l := 8*j - IF k[j] - MEM[base_addr+l+7:base_addr+l] := Saturate_UnsignedInt16_To_Int8(a[i+15:i]) - FI +CASE b[1:0] OF +0: dst[15:0] := a[63:48] +1: dst[15:0] := a[47:32] +2: dst[15:0] := a[31:16] +3: dst[15:0] := a[15:0] +ESAC +dst[MAX:15] := 0 + + +
immintrin.h
+ + + Floating Point + KNCNI + Arithmetic + + + + Multiply packed single-precision (32-bit) floating-point elements in each 4-element set of "a" and by element 1 of the corresponding 4-element set from "b", add the intermediate result to element 0 of the corresponding 4-element set from "b", and store the results in "dst". + +FOR j := 0 to 15 + i := j*32 + base := (j & ~0x3) * 32 + scale[31:0] := b[base+63:base+32] + bias[31:0] := b[base+31:base] + dst[i+31:i] := (a[i+31:i] * scale[31:0]) + bias[31:0] ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512BW - Convert - - - Convert packed unsigned 16-bit integers in "a" to packed unsigned 8-bit integers with unsigned saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Floating Point + KNCNI + Arithmetic + + + + + + Multiply packed single-precision (32-bit) floating-point elements in each 4-element set of "a" and by element 1 of the corresponding 4-element set from "b", add the intermediate result to element 0 of the corresponding 4-element set from "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). FOR j := 0 to 15 - i := 16*j - l := 8*j + i := j*32 IF k[j] - dst[l+7:l] := Saturate_UnsignedInt16_To_Int8(a[i+15:i]) + base := (j & ~0x3) * 32 + scale[31:0] := b[base+63:base+32] + bias[31:0] := b[base+31:base] + dst[i+31:i] := (a[i+31:i] * scale[31:0]) + bias[31:0] ELSE - dst[l+7:l] := 0 + dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX512BW - Convert - - Convert packed unsigned 16-bit integers in "a" to packed unsigned 8-bit integers with unsigned saturation, and store the results in "dst". + KNCNI + Load + + + + + + + Up-converts 8 single-precision (32-bit) memory locations starting at location "base_addr" at packed 64-bit integer indices stored in "vindex" scaled by "scale" using "conv" to 32-bit integer elements and stores them in "dst". "hint" indicates to the processor whether the data is non-temporal. -FOR j := 0 to 31 - i := 16*j - l := 8*j - dst[l+7:l] := Saturate_UnsignedInt16_To_Int8(a[i+15:i]) +FOR j := 0 to 7 + i := j*32 + m := j*64 + addr := base_addr + vindex[m+63:m] * ZeroExtend64(scale) * 8 + + CASE conv OF + _MM_UPCONV_EPI32_NONE: dst[i+31:i] := MEM[addr+31:addr] + _MM_UPCONV_EPI32_UINT8: dst[i+31:i] := ZeroExtend32(MEM[addr+7:addr]) + _MM_UPCONV_EPI32_SINT8: dst[i+31:i] := SignExtend32(MEM[addr+7:addr]) + _MM_UPCONV_EPI32_UINT16: dst[i+31:i] := ZeroExtend32(MEM[addr+15:addr]) + _MM_UPCONV_EPI32_SINT16: dst[i+31:i] := SignExtend32(MEM[addr+15:addr]) + ESAC ENDFOR dst[MAX:256] := 0 -
immintrin.h
 - + Integer - AVX512BW - Convert - - - - Convert packed unsigned 16-bit integers in "a" to packed unsigned 8-bit integers with unsigned saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + KNCNI + Load + + + + + + + + + Up-converts 8 single-precision (32-bit) memory locations starting at location "base_addr" at packed 64-bit integer indices stored in "vindex" scaled by "scale" using "conv" to 32-bit integer elements and stores them in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). "hint" indicates to the processor whether the data is non-temporal. -FOR j := 0 to 31 - i := 16*j - l := 8*j +FOR j := 0 to 7 + i := j*32 + m := j*64 + addr := base_addr + vindex[m+63:m] * ZeroExtend64(scale) * 8 + IF k[j] - dst[l+7:l] := Saturate_UnsignedInt16_To_Int8(a[i+15:i]) + CASE conv OF + _MM_UPCONV_EPI32_NONE: dst[i+31:i] := MEM[addr+31:addr] + _MM_UPCONV_EPI32_UINT8: dst[i+31:i] := ZeroExtend32(MEM[addr+7:addr]) + _MM_UPCONV_EPI32_SINT8: dst[i+31:i] := SignExtend32(MEM[addr+7:addr]) + _MM_UPCONV_EPI32_UINT16: dst[i+31:i] := ZeroExtend32(MEM[addr+15:addr]) + _MM_UPCONV_EPI32_SINT16: dst[i+31:i] := SignExtend32(MEM[addr+15:addr]) + ESAC ELSE - dst[l+7:l] := src[l+7:l] + dst[i+31:i] := src[i+31:i] FI ENDFOR dst[MAX:256] := 0 -
immintrin.h
 - + Integer - AVX512BW - Convert - Store - - - - Convert packed unsigned 16-bit integers in "a" to packed unsigned 8-bit integers with unsigned saturation, and store the active results (those with their respective bit set in writemask "k") to unaligned memory at "base_addr". + KNCNI + Load + + + + + + + Up-converts 8 double-precision (64-bit) memory locations starting at location "base_addr" at packed 64-bit integer indices stored in "vindex" scaled by "scale" using "conv" to 64-bit integer elements and stores them in "dst". "hint" indicates to the processor whether the load is non-temporal. -FOR j := 0 to 31 - i := 16*j - l := 8*j - IF k[j] - MEM[base_addr+l+7:base_addr+l] := Saturate_UnsignedInt16_To_Int8(a[i+15:i]) - FI +FOR j := 0 to 7 + i := j*32 + m := j*64 + addr := base_addr + vindex[m+63:m] * ZeroExtend64(scale) * 8 + + CASE conv OF + _MM_UPCONV_EPI64_NONE: dst[i+63:i] := MEM[addr+63:addr] + ESAC ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 -
immintrin.h
 - + Integer - AVX512BW - Convert - - - Convert packed unsigned 16-bit integers in "a" to packed unsigned 8-bit integers with unsigned saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + KNCNI + Load + + + + + + + + + Up-converts 8 double-precision (64-bit) memory locations starting at location "base_addr" at packed 64-bit integer indices stored in "vindex" scaled by "scale" using "conv" to 64-bit integer elements and stores them in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). "hint" indicates to the processor whether the load is non-temporal. -FOR j := 0 to 31 - i := 16*j - l := 8*j +FOR j := 0 to 7 + i := j*32 + m := j*64 + addr := base_addr + vindex[m+63:m] * ZeroExtend64(scale) * 8 + IF k[j] - dst[l+7:l] := Saturate_UnsignedInt16_To_Int8(a[i+15:i]) + CASE conv OF + _MM_UPCONV_EPI64_NONE: dst[i+63:i] := MEM[addr+63:addr] + ESAC ELSE - dst[l+7:l] := 0 + dst[i+63:i] := src[i+63:i] FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 -
immintrin.h
 - - Integer - AVX512VL - AVX512BW - Convert - - Convert packed unsigned 16-bit integers in "a" to packed unsigned 8-bit integers with unsigned saturation, and store the results in "dst". + + Floating Point + KNCNI + Load + + + + + + + Up-converts 8 memory locations starting at location "base_addr" at packed 64-bit integer indices stored in "vindex" scaled by "scale" using "conv" to single-precision (32-bit) floating-point elements and stores them in the lower half of "dst". "hint" indicates to the processor whether the load is non-temporal. FOR j := 0 to 7 - i := 16*j - l := 8*j - dst[l+7:l] := Saturate_UnsignedInt16_To_Int8(a[i+15:i]) + i := j*32 + m := j*64 + addr := base_addr + vindex[m+63:m] * ZeroExtend64(scale) * 8 + + CASE conv OF + _MM_UPCONV_PS_NONE: dst[i+31:i] := MEM[addr+31:addr] + _MM_UPCONV_PS_FLOAT16: dst[i+31:i] := Convert_FP16_To_FP32(MEM[addr+15:addr]) + _MM_UPCONV_PS_UINT8: dst[i+31:i] := Convert_UInt8_To_FP32(MEM[addr+7:addr]) + _MM_UPCONV_PS_SINT8: dst[i+31:i] := Convert_Int8_To_FP32(MEM[addr+7:addr]) + _MM_UPCONV_PS_UINT16: dst[i+31:i] := Convert_UInt16_To_FP32(MEM[addr+15:addr]) + _MM_UPCONV_PS_SINT16: dst[i+31:i] := Convert_Int16_To_FP32(MEM[addr+15:addr]) + ESAC ENDFOR -dst[MAX:64] := 0 +dst[MAX:256] := 0 -
immintrin.h
 - - Integer - AVX512VL - AVX512BW - Convert - - - - Convert packed unsigned 16-bit integers in "a" to packed unsigned 8-bit integers with unsigned saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Floating Point + KNCNI + Load + + + + + + + + + Up-converts 8 memory locations starting at location "base_addr" at packed 64-bit integer indices stored in "vindex" scaled by "scale" using "conv" to single-precision (32-bit) floating-point elements and stores them in the lower half of "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). "hint" indicates to the processor whether the load is non-temporal. FOR j := 0 to 7 - i := 16*j - l := 8*j + i := j*32 + m := j*64 + addr := base_addr + vindex[m+63:m] * ZeroExtend64(scale) * 8 + IF k[j] - dst[l+7:l] := Saturate_UnsignedInt16_To_Int8(a[i+15:i]) + CASE conv OF + _MM_UPCONV_PS_NONE: dst[i+31:i] := MEM[addr+31:addr] + _MM_UPCONV_PS_FLOAT16: dst[i+31:i] := Convert_FP16_To_FP32(MEM[addr+15:addr]) + _MM_UPCONV_PS_UINT8: dst[i+31:i] := Convert_UInt8_To_FP32(MEM[addr+7:addr]) + _MM_UPCONV_PS_SINT8: dst[i+31:i] := Convert_Int8_To_FP32(MEM[addr+7:addr]) + _MM_UPCONV_PS_UINT16: dst[i+31:i] := Convert_UInt16_To_FP32(MEM[addr+15:addr]) + _MM_UPCONV_PS_SINT16: dst[i+31:i] := Convert_Int16_To_FP32(MEM[addr+15:addr]) + ESAC ELSE - dst[l+7:l] := src[l+7:l] + dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:64] := 0 +dst[MAX:256] := 0 -
immintrin.h
 - - Integer - AVX512VL - AVX512BW - Convert - Store - - - - Convert packed unsigned 16-bit integers in "a" to packed unsigned 8-bit integers with unsigned saturation, and store the active results (those with their respective bit set in writemask "k") to unaligned memory at "base_addr". + + Floating Point + KNCNI + Load + + + + + + + Up-converts 8 double-precision (64-bit) floating-point elements stored in memory starting at location "base_addr" at packed 64-bit integer indices stored in "vindex" scaled by "scale" using "conv" to 64-bit floating-point elements and stores them in "dst". "hint" indicates to the processor whether the data is non-temporal. FOR j := 0 to 7 - i := 16*j - l := 8*j - IF k[j] - MEM[base_addr+l+7:base_addr+l] := Saturate_UnsignedInt16_To_Int8(a[i+15:i]) - FI + i := j*64 + m := j*64 + addr := base_addr + vindex[m+63:m] * ZeroExtend64(scale) * 8 + + CASE conv OF + _MM_UPCONV_PD_NONE: dst[i+63:i] := MEM[addr+63:addr] + ESAC ENDFOR -dst[MAX:64] := 0 +dst[MAX:512] := 0 -
immintrin.h
 - - Integer - AVX512VL - AVX512BW - Convert - - - Convert packed unsigned 16-bit integers in "a" to packed unsigned 8-bit integers with unsigned saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Floating Point + KNCNI + Load + + + + + + + + + Up-converts 8 double-precision (64-bit) floating-point elements stored in memory starting at location "base_addr" at packed 64-bit integer indices stored in "vindex" scaled by "scale" using "conv" to 64-bit floating-point elements and stores them in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). "hint" indicates to the processor whether the data is non-temporal. FOR j := 0 to 7 - i := 16*j - l := 8*j + i := j*64 + m := j*64 + addr := base_addr + vindex[m+63:m] * ZeroExtend64(scale) * 8 + IF k[j] - dst[l+7:l] := Saturate_UnsignedInt16_To_Int8(a[i+15:i]) - ELSE - dst[l+7:l] := 0 - FI -ENDFOR -dst[MAX:64] := 0 - - -
immintrin.h
- - - Integer - Mask - AVX512VL - AVX512BW - Miscellaneous - - Set each bit of mask register "k" based on the most significant bit of the corresponding packed 16-bit integer in "a". - -FOR j := 0 to 15 - i := j*16 - IF a[i+15] - k[j] := 1 + CASE conv OF + _MM_UPCONV_PD_NONE: dst[i+63:i] := MEM[addr+63:addr] + ESAC ELSE - k[j] := 0 + dst[i+63:i] := src[i+63:i] FI ENDFOR -k[MAX:16] := 0 +dst[MAX:512] := 0 -
immintrin.h
 - - Integer - Mask - AVX512BW - Miscellaneous - - Set each bit of mask register "k" based on the most significant bit of the corresponding packed 16-bit integer in "a". + + Floating Point + KNCNI + Store + + + + + + + + Down-converts 8 packed single-precision (32-bit) floating-point elements in "a" using "conv" and stores them in memory locations starting at location "base_addr" at packed 64-bit integer indices stored in "vindex" scaled by "scale". "hint" indicates to the processor whether the data is non-temporal. -FOR j := 0 to 31 - i := j*16 - IF a[i+15] - k[j] := 1 - ELSE - k[j] := 0 - FI +FOR j := 0 to 7 + i := j*32 + m := j*64 + addr := base_addr + vindex[m+63:m] * ZeroExtend64(scale) * 8 + + CASE conv OF + _MM_DOWNCONV_PS_NONE: MEM[addr+31:addr] := a[i+31:i] + _MM_DOWNCONV_PS_FLOAT16: MEM[addr+15:addr] := Convert_FP32_To_FP16(a[i+31:i]) + _MM_DOWNCONV_PS_UINT8: MEM[addr+ 7:addr] := Convert_FP32_To_UInt8(a[i+31:i]) + _MM_DOWNCONV_PS_SINT8: MEM[addr+ 7:addr] := Convert_FP32_To_Int8(a[i+31:i]) + _MM_DOWNCONV_PS_UINT16: MEM[addr+15:addr] := Convert_FP32_To_UInt16(a[i+31:i]) + _MM_DOWNCONV_PS_SINT16: MEM[addr+15:addr] := Convert_FP32_To_Int16(a[i+31:i]) + ESAC ENDFOR -k[MAX:32] := 0 -
immintrin.h
 - - Integer - Mask - AVX512VL - AVX512BW - Miscellaneous - - Set each bit of mask register "k" based on the most significant bit of the corresponding packed 16-bit integer in "a". + + Floating Point + KNCNI + Store + + + + + + + + + Down-converts 8 packed single-precision (32-bit) floating-point elements in "a" using "conv" and stores them in memory locations starting at location "base_addr" at packed 64-bit integer indices stored in "vindex" scaled by "scale". Elements are only written when the corresponding mask bit is set in "k"; otherwise, elements are unchanged in memory. "hint" indicates to the processor whether the data is non-temporal. FOR j := 0 to 7 - i := j*16 - IF a[i+15] - k[j] := 1 - ELSE - k[j] := 0 + i := j*32 + m := j*64 + addr := base_addr + vindex[m+63:m] * ZeroExtend64(scale) * 8 + + IF k[j] + CASE conv OF + _MM_DOWNCONV_PS_NONE: MEM[addr+31:addr] := a[i+31:i] + _MM_DOWNCONV_PS_FLOAT16: MEM[addr+15:addr] := Convert_FP32_To_FP16(a[i+31:i]) + _MM_DOWNCONV_PS_UINT8: MEM[addr+ 7:addr] := Convert_FP32_To_UInt8(a[i+31:i]) + _MM_DOWNCONV_PS_SINT8: MEM[addr+ 7:addr] := Convert_FP32_To_Int8(a[i+31:i]) + _MM_DOWNCONV_PS_UINT16: MEM[addr+15:addr] := Convert_FP32_To_UInt16(a[i+31:i]) + _MM_DOWNCONV_PS_SINT16: MEM[addr+15:addr] := Convert_FP32_To_Int16(a[i+31:i]) + ESAC FI ENDFOR -k[MAX:8] := 0 -
immintrin.h
 - - Integer - AVX512VL - AVX512BW - Convert - - Convert packed 16-bit integers in "a" to packed 8-bit integers with truncation, and store the results in "dst". + + Floating Point + KNCNI + Store + + + + + + + + Down-converts 8 packed double-precision (64-bit) floating-point elements in "a" using "conv" and stores them in memory locations starting at location "base_addr" at packed 64-bit integer indices stored in "vindex" scaled by "scale". "hint" indicates to the processor whether the data is non-temporal. -FOR j := 0 to 15 - i := 16*j - l := 8*j - dst[l+7:l] := Truncate_Int16_To_Int8(a[i+15:i]) +FOR j := 0 to 7 + i := j*64 + m := j*64 + addr := base_addr + vindex[m+63:m] * ZeroExtend64(scale) * 8 + + CASE conv OF + _MM_DOWNCONV_EPI64_NONE: MEM[addr+63:addr] := a[i+63:i] + ESAC ENDFOR -dst[MAX:128] := 0 -
immintrin.h
 - - Integer - AVX512VL - AVX512BW - Convert - - - - Convert packed 16-bit integers in "a" to packed 8-bit integers with truncation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Floating Point + KNCNI + Store + + + + + + + + + Down-converts 8 packed double-precision (64-bit) floating-point elements in "a" using "conv" and stores them in memory locations starting at location "base_addr" at packed 64-bit integer indices stored in "vindex" scaled by "scale". Elements are written to memory using writemask "k" (elements are not stored to memory when the corresponding mask bit is not set; the memory location is left unchagned). "hint" indicates to the processor whether the data is non-temporal. -FOR j := 0 to 15 - i := 16*j - l := 8*j +FOR j := 0 to 7 + i := j*64 + m := j*64 + addr := base_addr + vindex[m+63:m] * ZeroExtend64(scale) * 8 + IF k[j] - dst[l+7:l] := Truncate_Int16_To_Int8(a[i+15:i]) - ELSE - dst[l+7:l] := src[l+7:l] + CASE conv OF + _MM_DOWNCONV_EPI64_NONE: MEM[addr+63:addr] := a[i+63:i] + ESAC FI ENDFOR -dst[MAX:128] := 0 -
immintrin.h
 - + Integer - AVX512VL - AVX512BW - Convert + KNCNI Store - - - - Convert packed 16-bit integers in "a" to packed 8-bit integers with truncation, and store the active results (those with their respective bit set in writemask "k") to unaligned memory at "base_addr". + + + + + + + + Down-converts the low 8 packed 32-bit integer elements in "a" using "conv" and stores them in memory locations starting at location "base_addr" at packed 64-bit integer indices stored in "vindex" scaled by "scale". "hint" indicates to the processor whether the data is non-temporal. -FOR j := 0 to 15 - i := 16*j - l := 8*j - IF k[j] - MEM[base_addr+l+7:base_addr+l] := Truncate_Int16_To_Int8(a[i+15:i]) - FI +FOR j := 0 to 7 + i := j*32 + m := j*64 + addr := base_addr + vindex[m+63:m] * ZeroExtend64(scale) * 8 + + CASE conv OF + _MM_DOWNCONV_EPI32_NONE: MEM[addr+31:addr] := a[i+31:i] + _MM_DOWNCONV_EPI32_UINT8: MEM[addr+ 7:addr] := Truncate8(a[i+31:i]) + _MM_DOWNCONV_EPI32_SINT8: MEM[addr+ 7:addr] := Saturate8(a[i+31:i]) + _MM_DOWNCONV_EPI32_UINT16: MEM[addr+15:addr] := Truncate16(a[i+31:i]) + _MM_DOWNCONV_EPI32_SINT16: MEM[addr+15:addr] := Saturate16(a[i+31:i]) + ESAC ENDFOR -dst[MAX:128] := 0 -
immintrin.h
 - + Integer - AVX512VL - AVX512BW - Convert - - - Convert packed 16-bit integers in "a" to packed 8-bit integers with truncation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + KNCNI + Store + + + + + + + + + Down-converts the low 8 packed 32-bit integer elements in "a" using "conv" and stores them in memory locations starting at location "base_addr" at packed 64-bit integer indices stored in "vindex" scaled by "scale". Elements are written to memory using writemask "k" (elements are only written when the corresponding mask bit is set; otherwise, the memory location is left unchanged). "hint" indicates to the processor whether the data is non-temporal. -FOR j := 0 to 15 - i := 16*j - l := 8*j +FOR j := 0 to 7 + i := j*32 + m := j*64 + addr := base_addr + vindex[m+63:m] * ZeroExtend64(scale) * 8 + IF k[j] - dst[l+7:l] := Truncate_Int16_To_Int8(a[i+15:i]) - ELSE - dst[l+7:l] := 0 + CASE conv OF + _MM_DOWNCONV_EPI32_NONE: MEM[addr+31:addr] := a[i+31:i] + _MM_DOWNCONV_EPI32_UINT8: MEM[addr+ 7:addr] := Truncate8(a[i+31:i]) + _MM_DOWNCONV_EPI32_SINT8: MEM[addr+ 7:addr] := Saturate8(a[i+31:i]) + _MM_DOWNCONV_EPI32_UINT16: MEM[addr+15:addr] := Truncate16(a[i+31:i]) + _MM_DOWNCONV_EPI32_SINT16: MEM[addr+15:addr] := Saturate16(a[i+31:i]) + ESAC FI ENDFOR -dst[MAX:128] := 0 -
immintrin.h
 - + Integer - AVX512BW - Convert - - Convert packed 16-bit integers in "a" to packed 8-bit integers with truncation, and store the results in "dst". + KNCNI + Store + + + + + + + + Down-converts 8 packed 64-bit integer elements in "a" using "conv" and stores them in memory locations starting at location "base_addr" at packed 64-bit integer indices stored in "vindex" scaled by "scale". "hint" indicates to the processor whether the load is non-temporal. -FOR j := 0 to 31 - i := 16*j - l := 8*j - dst[l+7:l] := Truncate_Int16_To_Int8(a[i+15:i]) +FOR j := 0 to 7 + i := j*64 + m := j*64 + addr := base_addr + vindex[m+63:m] * ZeroExtend64(scale) * 8 + + CASE conv OF + _MM_DOWNCONV_EPI64_NONE: MEM[addr+63:addr] := a[i+63:i] + ESAC ENDFOR -dst[MAX:256] := 0 -
immintrin.h
 - + Integer - AVX512BW - Convert - - - - Convert packed 16-bit integers in "a" to packed 8-bit integers with truncation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + KNCNI + Store + + + + + + + + + Down-converts 8 packed 64-bit integer elements in "a" using "conv" and stores them in memory locations starting at location "base_addr" at packed 64-bit integer indices stored in "vindex" scaled by "scale". Only those elements whose corresponding mask bit is set in writemask "k" are written to memory. -FOR j := 0 to 31 - i := 16*j - l := 8*j +FOR j := 0 to 7 + i := j*64 + m := j*64 + addr := base_addr + vindex[m+63:m] * ZeroExtend64(scale) * 8 + IF k[j] - dst[l+7:l] := Truncate_Int16_To_Int8(a[i+15:i]) - ELSE - dst[l+7:l] := src[l+7:l] + CASE conv OF + _MM_DOWNCONV_EPI64_NONE: MEM[addr+63:addr] := a[i+63:i] + ESAC FI ENDFOR -dst[MAX:256] := 0 -
immintrin.h
 - - Integer - AVX512BW - Convert + + Floating Point + KNCNI Store - - - - Convert packed 16-bit integers in "a" to packed 8-bit integers with truncation, and store the active results (those with their respective bit set in writemask "k") to unaligned memory at "base_addr". + + + + Permutes 128-bit blocks of the packed single-precision (32-bit) floating-point elements in "a" using constant "imm8". The results are stored in "dst". -FOR j := 0 to 31 - i := 16*j - l := 8*j - IF k[j] - MEM[base_addr+l+7:base_addr+l] := Truncate_Int16_To_Int8(a[i+15:i]) - FI +DEFINE SELECT4(src, control) { + CASE control[1:0] OF + 0: tmp[127:0] := src[127:0] + 1: tmp[127:0] := src[255:128] + 2: tmp[127:0] := src[383:256] + 3: tmp[127:0] := src[511:384] + ESAC + RETURN tmp[127:0] +} +FOR j := 0 to 3 + i := j*128 + n := j*2 + dst[i+127:i] := SELECT4(a[511:0], imm8[n+1:n]) ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - Integer - AVX512BW - Convert - - - Convert packed 16-bit integers in "a" to packed 8-bit integers with truncation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Floating Point + KNCNI + Swizzle + + + + + + Permutes 128-bit blocks of the packed single-precision (32-bit) floating-point elements in "a" using constant "imm8". The results are stored in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). -FOR j := 0 to 31 - i := 16*j - l := 8*j +DEFINE SELECT4(src, control) { + CASE control[1:0] OF + 0: tmp[127:0] := src[127:0] + 1: tmp[127:0] := src[255:128] + 2: tmp[127:0] := src[383:256] + 3: tmp[127:0] := src[511:384] + ESAC + RETURN tmp[127:0] +} +tmp[511:0] := 0 +FOR j := 0 to 3 + i := j*128 + n := j*2 + tmp[i+127:i] := SELECT4(a[511:0], imm8[n+1:n]) +ENDFOR +FOR j := 0 to 15 IF k[j] - dst[l+7:l] := Truncate_Int16_To_Int8(a[i+15:i]) + dst[i+31:i] := tmp[i+31:i] ELSE - dst[l+7:l] := 0 + dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:256] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX512VL - AVX512BW - Convert - - Convert packed 16-bit integers in "a" to packed 8-bit integers with truncation, and store the results in "dst". + KNCNI + Load + + + + + Loads 8 32-bit integer memory locations starting at location "base_addr" at packed 64-bit integer indices stored in "vindex" scaled by "scale" to "dst". FOR j := 0 to 7 - i := 16*j - l := 8*j - dst[l+7:l] := Truncate_Int16_To_Int8(a[i+15:i]) + i := j*32 + m := j*64 + addr := base_addr + vindex[m+63:m] * ZeroExtend64(scale) * 8 + dst[i+31:i] := MEM[addr+31:addr] ENDFOR -dst[MAX:64] := 0 +dst[MAX:256] := 0 -
immintrin.h
 - + Integer - AVX512VL - AVX512BW - Convert - - - - Convert packed 16-bit integers in "a" to packed 8-bit integers with truncation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + KNCNI + Load + + + + + + + Loads 8 32-bit integer memory locations starting at location "base_addr" at packed 64-bit integer indices stored in "vindex" scaled by "scale" to "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). FOR j := 0 to 7 - i := 16*j - l := 8*j + i := j*32 + m := j*64 IF k[j] - dst[l+7:l] := Truncate_Int16_To_Int8(a[i+15:i]) + addr := base_addr + vindex[m+63:m] * ZeroExtend64(scale) * 8 + dst[i+31:i] := MEM[addr+31:addr] ELSE - dst[l+7:l] := src[l+7:l] - FI -ENDFOR -dst[MAX:64] := 0 - - -
immintrin.h
- - - Integer - AVX512VL - AVX512BW - Convert - Store - - - - Convert packed 16-bit integers in "a" to packed 8-bit integers with truncation, and store the active results (those with their respective bit set in writemask "k") to unaligned memory at "base_addr". - -FOR j := 0 to 7 - i := 16*j - l := 8*j - IF k[j] - MEM[base_addr+l+7:base_addr+l] := Truncate_Int16_To_Int8(a[i+15:i]) + dst[i+31:i] := src[i+31:i] FI ENDFOR -dst[MAX:64] := 0 +dst[MAX:256] := 0 -
immintrin.h
 - - Integer - AVX512VL - AVX512BW - Convert - - - Convert packed 16-bit integers in "a" to packed 8-bit integers with truncation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Floating Point + KNCNI + Load + + + + + Loads 8 single-precision (32-bit) floating-point memory locations starting at location "base_addr" at packed 64-bit integer indices stored in "vindex" scaled by "scale" to "dst". FOR j := 0 to 7 - i := 16*j - l := 8*j - IF k[j] - dst[l+7:l] := Truncate_Int16_To_Int8(a[i+15:i]) - ELSE - dst[l+7:l] := 0 - FI + i := j*32 + m := j*64 + addr := base_addr + vindex[m+63:m] * ZeroExtend64(scale) * 8 + dst[i+31:i] := MEM[addr+31:addr] ENDFOR -dst[MAX:64] := 0 +dst[MAX:256] := 0 -
immintrin.h
 - - Integer - AVX512VL - AVX512F - Convert - - - - Zero extend packed unsigned 8-bit integers in the low 8 bytes of "a" to packed 32-bit integers, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Floating Point + KNCNI + Load + + + + + + + Loads 8 single-precision (32-bit) floating-point memory locations starting at location "base_addr" at packed 64-bit integer indices stored in "vindex" scaled by "scale" to "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). FOR j := 0 to 7 - i := 32*j - l := 8*j + i := j*32 + m := j*64 IF k[j] - dst[i+31:i] := ZeroExtend(a[l+7:l]) + addr := base_addr + vindex[m+63:m] * ZeroExtend64(scale) * 8 + dst[i+31:i] := MEM[addr+31:addr] ELSE dst[i+31:i] := src[i+31:i] FI ENDFOR dst[MAX:256] := 0 -
immintrin.h
 - - Integer - AVX512VL - AVX512F - Convert - - - Zero extend packed unsigned 8-bit integers in the low 8 bytes of "a" to packed 32-bit integers, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Floating Point + KNCNI + Store + + + + + + Stores 8 packed single-precision (32-bit) floating-point elements in "a" in memory locations starting at location "base_addr" at packed 64-bit integer indices stored in "vindex" scaled by "scale". FOR j := 0 to 7 - i := 32*j - l := 8*j - IF k[j] - dst[i+31:i] := ZeroExtend(a[l+7:l]) - ELSE - dst[i+31:i] := 0 - FI + i := j*32 + m := j*64 + addr := base_addr + vindex[m+63:m] * ZeroExtend64(scale) * 8 + MEM[addr+31:addr] := a[i+31:i] ENDFOR -dst[MAX:256] := 0 -
immintrin.h
 - - Integer - AVX512VL - AVX512F - Convert - - - - Zero extend packed unsigned 8-bit integers in the low 4 bytes of "a" to packed 32-bit integers, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Floating Point + KNCNI + Store + + + + + + + Stores 8 packed single-precision (32-bit) floating-point elements in "a" in memory locations starting at location "base_addr" at packed 64-bit integer indices stored in "vindex" scaled by "scale" using writemask "k" (elements are only written to memory when the corresponding mask bit is set). -FOR j := 0 to 3 - i := 32*j - l := 8*j +FOR j := 0 to 7 + i := j*32 + m := j*64 IF k[j] - dst[i+31:i] := ZeroExtend(a[l+7:l]) - ELSE - dst[i+31:i] := src[i+31:i] - FI + addr := base_addr + vindex[m+63:m] * ZeroExtend64(scale) * 8 + MEM[addr+31:addr] := a[i+31:i] + FI ENDFOR -dst[MAX:128] := 0 -
immintrin.h
 - + Integer - AVX512VL - AVX512F - Convert - - - Zero extend packed unsigned 8-bit integers in th elow 4 bytes of "a" to packed 32-bit integers, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + KNCNI + Store + + + + + + Stores 8 packed 32-bit integer elements in "a" in memory locations starting at location "base_addr" at packed 64-bit integer indices stored in "vindex" scaled by "scale". -FOR j := 0 to 3 - i := 32*j - l := 8*j - IF k[j] - dst[i+31:i] := ZeroExtend(a[l+7:l]) - ELSE - dst[i+31:i] := 0 - FI +FOR j := 0 to 7 + i := j*32 + m := j*64 + addr := base_addr + vindex[m+63:m] * ZeroExtend64(scale) * 8 + MEM[addr+31:addr] := a[i+31:i] ENDFOR -dst[MAX:128] := 0 -
immintrin.h
 - + Integer - AVX512VL - AVX512F - Convert - - - - Zero extend packed unsigned 8-bit integers in the low 4 bytes of "a" to packed 64-bit integers, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + KNCNI + Store + + + + + + + Stores 8 packed 32-bit integer elements in "a" in memory locations starting at location "base_addr" at packed 64-bit integer indices stored in "vindex" scaled by "scale" using writemask "k" (elements are only written to memory when the corresponding mask bit is set). -FOR j := 0 to 3 - i := 64*j - l := 8*j +FOR j := 0 to 7 + i := j*32 + m := j*64 IF k[j] - dst[i+63:i] := ZeroExtend(a[l+7:l]) - ELSE - dst[i+63:i] := src[i+63:i] - FI + addr := base_addr + vindex[m+63:m] * ZeroExtend64(scale) * 8 + MEM[addr+31:addr] := a[i+31:i] + FI ENDFOR -dst[MAX:256] := 0 -
immintrin.h
 - - Integer - AVX512VL - AVX512F - Convert - - - Zero extend packed unsigned 8-bit integers in the low 4 bytes of "a" to packed 64-bit integers, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Mask + KNCNI + Mask + + + + Move the high element from "k1" to the low element of "k1", and insert the low element of "k2" into the high element of "k1". -FOR j := 0 to 3 - i := 64*j - l := 8*j - IF k[j] - dst[i+63:i] := ZeroExtend(a[l+7:l]) - ELSE - dst[i+63:i] := 0 - FI -ENDFOR -dst[MAX:256] := 0 +tmp[7:0] := k1[15:8] +k1[15:8] := k2[7:0] +k1[7:0] := tmp[7:0] - +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Convert - - - - Zero extend packed unsigned 8-bit integers in the low 2 bytes of "a" to packed 64-bit integers, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Mask + KNCNI + Mask + + + + Insert the low element of "k2" into the high element of "k1". -FOR j := 0 to 1 - i := 64*j - l := 8*j - IF k[j] - dst[i+63:i] := ZeroExtend(a[l+7:l]) - ELSE - dst[i+63:i] := src[i+63:i] - FI -ENDFOR -dst[MAX:128] := 0 +k1[15:8] := k2[7:0] - +
immintrin.h
 - + Integer - AVX512VL - AVX512F - Convert - - - Zero extend packed unsigned 8-bit integers in the low 2 bytes of "a" to packed 64-bit integers, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + LZCNT + Bit Manipulation + + + Count the number of leading zero bits in unsigned 32-bit integer "a", and return that count in "dst". -FOR j := 0 to 1 - i := 64*j - l := 8*j - IF k[j] - dst[i+63:i] := ZeroExtend(a[l+7:l]) - ELSE - dst[i+63:i] := 0 - FI -ENDFOR -dst[MAX:128] := 0 +tmp := 31 +dst := 0 +DO WHILE (tmp >= 0 AND a[tmp] == 0) + tmp := tmp - 1 + dst := dst + 1 +OD - +
immintrin.h
 - + Integer - AVX512VL - AVX512BW - Convert - - - - Zero extend packed unsigned 8-bit integers in "a" to packed 16-bit integers, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + LZCNT + Bit Manipulation + + + Count the number of leading zero bits in unsigned 64-bit integer "a", and return that count in "dst". -FOR j := 0 to 15 - i := j*8 - l := j*16 - IF k[j] - dst[l+15:l] := ZeroExtend(a[i+7:i]) - ELSE - dst[l+15:l] := src[l+15:l] - FI -ENDFOR -dst[MAX:256] := 0 +tmp := 63 +dst := 0 +DO WHILE (tmp >= 0 AND a[tmp] == 0) + tmp := tmp - 1 + dst := dst + 1 +OD - +
immintrin.h
 - + Integer - AVX512VL - AVX512BW + MMX Convert - - - Zero extend packed unsigned 8-bit integers in "a" to packed 16-bit integers, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - + + + Copy 64-bit integer "a" to "dst". -FOR j := 0 to 15 - i := j*8 - l := j*16 - IF k[j] - dst[l+15:l] := ZeroExtend(a[i+7:i]) - ELSE - dst[l+15:l] := 0 - FI -ENDFOR -dst[MAX:256] := 0 +dst[63:0] := a[63:0] - -
immintrin.h
+ +
mmintrin.h
 - + Integer - AVX512BW + MMX Convert - - Zero extend packed unsigned 8-bit integers in "a" to packed 16-bit integers, and store the results in "dst". + + + Copy 64-bit integer "a" to "dst". -FOR j := 0 to 31 - i := j*8 - l := j*16 - dst[l+15:l] := ZeroExtend(a[i+7:i]) -ENDFOR -dst[MAX:512] := 0 +dst[63:0] := a[63:0] - -
immintrin.h
+ +
mmintrin.h
+ + + MMX + General Support + + + Empty the MMX state, which marks the x87 FPU registers as available for use by x87 instructions. This instruction must be used at the end of all MMX technology procedures. + +
mmintrin.h
 - + Integer - AVX512BW + MMX Convert - - - - Zero extend packed unsigned 8-bit integers in "a" to packed 16-bit integers, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + + Copy 32-bit integer "a" to the lower elements of "dst", and zero the upper element of "dst". -FOR j := 0 to 31 - i := j*8 - l := j*16 - IF k[j] - dst[l+15:l] := ZeroExtend(a[i+7:i]) - ELSE - dst[l+15:l] := src[l+15:l] - FI -ENDFOR -dst[MAX:512] := 0 +dst[31:0] := a[31:0] +dst[63:32] := 0 - -
immintrin.h
+ +
mmintrin.h
 - + Integer - AVX512BW + MMX Convert - - - Zero extend packed unsigned 8-bit integers in "a" to packed 16-bit integers, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - + + + Copy the lower 32-bit integer in "a" to "dst". -FOR j := 0 to 31 - i := j*8 - l := j*16 - IF k[j] - dst[l+15:l] := ZeroExtend(a[i+7:i]) - ELSE - dst[l+15:l] := 0 - FI -ENDFOR -dst[MAX:512] := 0 +dst[31:0] := a[31:0] - -
immintrin.h
+ +
mmintrin.h
 - + Integer - AVX512VL - AVX512BW - Convert - - - - Zero extend packed unsigned 8-bit integers in "a" to packed 16-bit integers, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + MMX + Miscellaneous + + + + Convert packed signed 16-bit integers from "a" and "b" to packed 8-bit integers using signed saturation, and store the results in "dst". + +dst[7:0] := Saturate8(a[15:0]) +dst[15:8] := Saturate8(a[31:16]) +dst[23:16] := Saturate8(a[47:32]) +dst[31:24] := Saturate8(a[63:48]) +dst[39:32] := Saturate8(b[15:0]) +dst[47:40] := Saturate8(b[31:16]) +dst[55:48] := Saturate8(b[47:32]) +dst[63:56] := Saturate8(b[63:48]) + + +
mmintrin.h
+ + + Integer + MMX + Miscellaneous + + + + Convert packed signed 32-bit integers from "a" and "b" to packed 16-bit integers using signed saturation, and store the results in "dst". -FOR j := 0 to 7 - i := j*8 - l := j*16 - IF k[j] - dst[l+15:l] := ZeroExtend(a[i+7:i]) - ELSE - dst[l+15:l] := src[l+15:l] - FI -ENDFOR -dst[MAX:128] := 0 +dst[15:0] := Saturate16(a[31:0]) +dst[31:16] := Saturate16(a[63:32]) +dst[47:32] := Saturate16(b[31:0]) +dst[63:48] := Saturate16(b[63:32]) - -
immintrin.h
+ +
mmintrin.h
 - + Integer - AVX512VL - AVX512BW - Convert - - - Zero extend packed unsigned 8-bit integers in "a" to packed 16-bit integers, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - + MMX + Miscellaneous + + + + Convert packed signed 16-bit integers from "a" and "b" to packed 8-bit integers using unsigned saturation, and store the results in "dst". + +dst[7:0] := SaturateU8(a[15:0]) +dst[15:8] := SaturateU8(a[31:16]) +dst[23:16] := SaturateU8(a[47:32]) +dst[31:24] := SaturateU8(a[63:48]) +dst[39:32] := SaturateU8(b[15:0]) +dst[47:40] := SaturateU8(b[31:16]) +dst[55:48] := SaturateU8(b[47:32]) +dst[63:56] := SaturateU8(b[63:48]) + + +
mmintrin.h
+ + + Integer + MMX + Swizzle + + + + Unpack and interleave 8-bit integers from the high half of "a" and "b", and store the results in "dst". -FOR j := 0 to 7 - i := j*8 - l := j*16 - IF k[j] - dst[l+15:l] := ZeroExtend(a[i+7:i]) - ELSE - dst[l+15:l] := 0 - FI -ENDFOR -dst[MAX:128] := 0 +DEFINE INTERLEAVE_HIGH_BYTES(src1[63:0], src2[63:0]) { + dst[7:0] := src1[39:32] + dst[15:8] := src2[39:32] + dst[23:16] := src1[47:40] + dst[31:24] := src2[47:40] + dst[39:32] := src1[55:48] + dst[47:40] := src2[55:48] + dst[55:48] := src1[63:56] + dst[63:56] := src2[63:56] + RETURN dst[63:0] +} +dst[63:0] := INTERLEAVE_HIGH_BYTES(a[63:0], b[63:0]) - -
immintrin.h
+ +
mmintrin.h
 - + Integer - AVX512VL - AVX512F - Convert - - - - Zero extend packed unsigned 32-bit integers in "a" to packed 64-bit integers, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + MMX + Swizzle + + + + Unpack and interleave 16-bit integers from the high half of "a" and "b", and store the results in "dst". -FOR j := 0 to 3 - i := 64*j - l := 32*j - IF k[j] - dst[i+63:i] := ZeroExtend(a[l+31:l]) - ELSE - dst[i+63:i] := src[i+63:i] - FI -ENDFOR -dst[MAX:256] := 0 +DEFINE INTERLEAVE_HIGH_WORDS(src1[63:0], src2[63:0]) { + dst[15:0] := src1[47:32] + dst[31:16] := src2[47:32] + dst[47:32] := src1[63:48] + dst[63:48] := src2[63:48] + RETURN dst[63:0] +} +dst[63:0] := INTERLEAVE_HIGH_WORDS(a[63:0], b[63:0]) - -
immintrin.h
+ +
mmintrin.h
 - + Integer - AVX512VL - AVX512F - Convert - - - Zero extend packed unsigned 32-bit integers in "a" to packed 64-bit integers, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + MMX + Swizzle + + + + Unpack and interleave 32-bit integers from the high half of "a" and "b", and store the results in "dst". -FOR j := 0 to 3 - i := 64*j - l := 32*j - IF k[j] - dst[i+63:i] := ZeroExtend(a[l+31:l]) - ELSE - dst[i+63:i] := 0 - FI -ENDFOR -dst[MAX:256] := 0 +dst[31:0] := a[63:32] +dst[63:32] := b[63:32] - -
immintrin.h
+ +
mmintrin.h
 - + Integer - AVX512VL - AVX512F - Convert - - - - Zero extend packed unsigned 32-bit integers in "a" to packed 64-bit integers, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + MMX + Swizzle + + + + Unpack and interleave 8-bit integers from the low half of "a" and "b", and store the results in "dst". -FOR j := 0 to 1 - i := 64*j - l := 32*j - IF k[j] - dst[i+63:i] := ZeroExtend(a[l+31:l]) - ELSE - dst[i+63:i] := src[i+63:i] - FI -ENDFOR -dst[MAX:128] := 0 +DEFINE INTERLEAVE_BYTES(src1[63:0], src2[63:0]) { + dst[7:0] := src1[7:0] + dst[15:8] := src2[7:0] + dst[23:16] := src1[15:8] + dst[31:24] := src2[15:8] + dst[39:32] := src1[23:16] + dst[47:40] := src2[23:16] + dst[55:48] := src1[31:24] + dst[63:56] := src2[31:24] + RETURN dst[63:0] +} +dst[63:0] := INTERLEAVE_BYTES(a[63:0], b[63:0]) - -
immintrin.h
+ +
mmintrin.h
 - + Integer - AVX512VL - AVX512F - Convert - - - Zero extend packed unsigned 32-bit integers in "a" to packed 64-bit integers, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + MMX + Swizzle + + + + Unpack and interleave 16-bit integers from the low half of "a" and "b", and store the results in "dst". -FOR j := 0 to 1 - i := 64*j - l := 32*j - IF k[j] - dst[i+63:i] := ZeroExtend(a[l+31:l]) - ELSE - dst[i+63:i] := 0 - FI -ENDFOR -dst[MAX:128] := 0 +DEFINE INTERLEAVE_WORDS(src1[63:0], src2[63:0]) { + dst[15:0] := src1[15:0] + dst[31:16] := src2[15:0] + dst[47:32] := src1[31:16] + dst[63:48] := src2[31:16] + RETURN dst[63:0] +} +dst[63:0] := INTERLEAVE_WORDS(a[63:0], b[63:0]) - -
immintrin.h
+ +
mmintrin.h
 - + Integer - AVX512VL - AVX512F - Convert - - - - Zero extend packed unsigned 16-bit integers in "a" to packed 32-bit integers, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + MMX + Swizzle + + + + Unpack and interleave 32-bit integers from the low half of "a" and "b", and store the results in "dst". -FOR j := 0 to 7 - i := 32*j - l := 16*j - IF k[j] - dst[i+31:i] := ZeroExtend(a[l+15:l]) - ELSE - dst[i+31:i] := src[i+31:i] - FI -ENDFOR -dst[MAX:256] := 0 +dst[31:0] := a[31:0] +dst[63:32] := b[31:0] - -
immintrin.h
+ +
mmintrin.h
 - + Integer - AVX512VL - AVX512F - Convert - - - Zero extend packed unsigned 16-bit integers in "a" to packed 32-bit integers, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + MMX + Arithmetic + + + + Add packed 8-bit integers in "a" and "b", and store the results in "dst". FOR j := 0 to 7 - i := 32*j - l := 16*j - IF k[j] - dst[i+31:i] := ZeroExtend(a[l+15:l]) - ELSE - dst[i+31:i] := 0 - FI + i := j*8 + dst[i+7:i] := a[i+7:i] + b[i+7:i] ENDFOR -dst[MAX:256] := 0 - -
immintrin.h
+ +
mmintrin.h
 - + Integer - AVX512VL - AVX512F - Convert - - - - Zero extend packed unsigned 16-bit integers in "a" to packed 32-bit integers, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + MMX + Arithmetic + + + + Add packed 16-bit integers in "a" and "b", and store the results in "dst". FOR j := 0 to 3 - i := 32*j - l := 16*j - IF k[j] - dst[i+31:i] := ZeroExtend(a[l+15:l]) - ELSE - dst[i+31:i] := src[i+31:i] - FI + i := j*16 + dst[i+15:i] := a[i+15:i] + b[i+15:i] ENDFOR -dst[MAX:128] := 0 - -
immintrin.h
+ +
mmintrin.h
 - + Integer - AVX512VL - AVX512F - Convert - - - Zero extend packed unsigned 16-bit integers in "a" to packed 32-bit integers, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + MMX + Arithmetic + + + + Add packed 32-bit integers in "a" and "b", and store the results in "dst". -FOR j := 0 to 3 - i := 32*j - l := 16*j - IF k[j] - dst[i+31:i] := ZeroExtend(a[l+15:l]) - ELSE - dst[i+31:i] := 0 - FI +FOR j := 0 to 1 + i := j*32 + dst[i+31:i] := a[i+31:i] + b[i+31:i] ENDFOR -dst[MAX:128] := 0 - -
immintrin.h
+ +
mmintrin.h
 - + Integer - AVX512VL - AVX512F - Convert - - - - Zero extend packed unsigned 16-bit integers in the low 8 bytes of "a" to packed 64-bit integers, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + MMX + Arithmetic + + + + Add packed signed 8-bit integers in "a" and "b" using saturation, and store the results in "dst". -FOR j := 0 to 3 - i := 64*j - l := 16*j - IF k[j] - dst[i+63:i] := ZeroExtend(a[l+15:l]) - ELSE - dst[i+63:i] := src[i+63:i] - FI +FOR j := 0 to 7 + i := j*8 + dst[i+7:i] := Saturate8( a[i+7:i] + b[i+7:i] ) ENDFOR -dst[MAX:256] := 0 - -
immintrin.h
+ +
mmintrin.h
 - + Integer - AVX512VL - AVX512F - Convert - - - Zero extend packed unsigned 16-bit integers in the low 8 bytes of "a" to packed 64-bit integers, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + MMX + Arithmetic + + + + Add packed signed 16-bit integers in "a" and "b" using saturation, and store the results in "dst". FOR j := 0 to 3 - i := 64*j - l := 16*j - IF k[j] - dst[i+63:i] := ZeroExtend(a[l+15:l]) - ELSE - dst[i+63:i] := 0 - FI + i := j*16 + dst[i+15:i] := Saturate16( a[i+15:i] + b[i+15:i] ) ENDFOR -dst[MAX:256] := 0 - -
immintrin.h
+ +
mmintrin.h
 - + Integer - AVX512VL - AVX512F - Convert - - - - Zero extend packed unsigned 16-bit integers in the low 4 bytes of "a" to packed 64-bit integers, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + MMX + Arithmetic + + + + Add packed unsigned 8-bit integers in "a" and "b" using saturation, and store the results in "dst". -FOR j := 0 to 1 - i := 64*j - l := 16*j - IF k[j] - dst[i+63:i] := ZeroExtend(a[l+15:l]) - ELSE - dst[i+63:i] := src[i+63:i] - FI +FOR j := 0 to 7 + i := j*8 + dst[i+7:i] := SaturateU8( a[i+7:i] + b[i+7:i] ) ENDFOR -dst[MAX:128] := 0 - -
immintrin.h
+ +
mmintrin.h
 - + Integer - AVX512VL - AVX512F - Convert - - - Zero extend packed unsigned 16-bit integers in the low 4 bytes of "a" to packed 64-bit integers, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + MMX + Arithmetic + + + + Add packed unsigned 16-bit integers in "a" and "b" using saturation, and store the results in "dst". -FOR j := 0 to 1 - i := 64*j - l := 16*j - IF k[j] - dst[i+63:i] := ZeroExtend(a[l+15:l]) - ELSE - dst[i+63:i] := 0 - FI +FOR j := 0 to 3 + i := j*16 + dst[i+15:i] := SaturateU16( a[i+15:i] + b[i+15:i] ) ENDFOR -dst[MAX:128] := 0 - -
immintrin.h
+ +
mmintrin.h
 - + + Floating Point Integer - AVX512VL - AVX512F + MMX Arithmetic - - - - - Multiply the low 32-bit integers from each packed 64-bit element in "a" and "b", and store the signed 64-bit results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + + + Subtract packed 8-bit integers in "b" from packed 8-bit integers in "a", and store the results in "dst". -FOR j := 0 to 3 - i := j*64 - IF k[j] - dst[i+63:i] := a[i+31:i] * b[i+31:i] - ELSE - dst[i+63:i] := src[i+63:i] - FI +FOR j := 0 to 7 + i := j*8 + dst[i+7:i] := a[i+7:i] - b[i+7:i] ENDFOR -dst[MAX:256] := 0 - -
immintrin.h
+ +
mmintrin.h
 - + + Floating Point Integer - AVX512VL - AVX512F + MMX Arithmetic - - - - Multiply the low 32-bit integers from each packed 64-bit element in "a" and "b", and store the signed 64-bit results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + + + Subtract packed 16-bit integers in "b" from packed 16-bit integers in "a", and store the results in "dst". FOR j := 0 to 3 - i := j*64 - IF k[j] - dst[i+63:i] := a[i+31:i] * b[i+31:i] - ELSE - dst[i+63:i] := 0 - FI + i := j*16 + dst[i+15:i] := a[i+15:i] - b[i+15:i] ENDFOR -dst[MAX:256] := 0 - -
immintrin.h
+ +
mmintrin.h
 - + + Floating Point Integer - AVX512VL - AVX512F + MMX Arithmetic - - - - - Multiply the low 32-bit integers from each packed 64-bit element in "a" and "b", and store the signed 64-bit results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + + + Subtract packed 32-bit integers in "b" from packed 32-bit integers in "a", and store the results in "dst". FOR j := 0 to 1 - i := j*64 - IF k[j] - dst[i+63:i] := a[i+31:i] * b[i+31:i] - ELSE - dst[i+63:i] := src[i+63:i] - FI + i := j*32 + dst[i+31:i] := a[i+31:i] - b[i+31:i] ENDFOR -dst[MAX:128] := 0 - -
immintrin.h
+ +
mmintrin.h
 - + + Floating Point Integer - AVX512VL - AVX512F + MMX Arithmetic - - - - Multiply the low 32-bit integers from each packed 64-bit element in "a" and "b", and store the signed 64-bit results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + + + Subtract packed signed 8-bit integers in "b" from packed 8-bit integers in "a" using saturation, and store the results in "dst". -FOR j := 0 to 1 - i := j*64 - IF k[j] - dst[i+63:i] := a[i+31:i] * b[i+31:i] - ELSE - dst[i+63:i] := 0 - FI +FOR j := 0 to 7 + i := j*8 + dst[i+7:i] := Saturate8(a[i+7:i] - b[i+7:i]) ENDFOR -dst[MAX:128] := 0 - -
immintrin.h
+ +
mmintrin.h
 - + + Floating Point Integer - AVX512VL - AVX512BW + MMX Arithmetic - - - - - Multiply packed 16-bit integers in "a" and "b", producing intermediate signed 32-bit integers. Truncate each intermediate integer to the 18 most significant bits, round by adding 1, and store bits [16:1] to "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + + + Subtract packed signed 16-bit integers in "b" from packed 16-bit integers in "a" using saturation, and store the results in "dst". -FOR j := 0 to 15 +FOR j := 0 to 3 i := j*16 - IF k[j] - tmp[31:0] := ((a[i+15:i] * b[i+15:i]) >> 14) + 1 - dst[i+15:i] := tmp[16:1] - ELSE - dst[i+15:i] := src[i+15:i] - FI + dst[i+15:i] := Saturate16(a[i+15:i] - b[i+15:i]) ENDFOR -dst[MAX:256] := 0 - -
immintrin.h
+ +
mmintrin.h
 - + + Floating Point Integer - AVX512VL - AVX512BW + MMX Arithmetic - - - - Multiply packed 16-bit integers in "a" and "b", producing intermediate signed 32-bit integers. Truncate each intermediate integer to the 18 most significant bits, round by adding 1, and store bits [16:1] to "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + + + Subtract packed unsigned 8-bit integers in "b" from packed unsigned 8-bit integers in "a" using saturation, and store the results in "dst". -FOR j := 0 to 15 - i := j*16 - IF k[j] - tmp[31:0] := ((a[i+15:i] * b[i+15:i]) >> 14) + 1 - dst[i+15:i] := tmp[16:1] - ELSE - dst[i+15:i] := 9 - FI +FOR j := 0 to 7 + i := j*8 + dst[i+7:i] := SaturateU8(a[i+7:i] - b[i+7:i]) ENDFOR -dst[MAX:256] := 0 - -
immintrin.h
+ +
mmintrin.h
 - + + Floating Point Integer - AVX512BW + MMX Arithmetic - - - - - Multiply packed 16-bit integers in "a" and "b", producing intermediate signed 32-bit integers. Truncate each intermediate integer to the 18 most significant bits, round by adding 1, and store bits [16:1] to "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + + + Subtract packed unsigned 16-bit integers in "b" from packed unsigned 16-bit integers in "a" using saturation, and store the results in "dst". -FOR j := 0 to 31 +FOR j := 0 to 3 i := j*16 - IF k[j] - tmp[31:0] := ((a[i+15:i] * b[i+15:i]) >> 14) + 1 - dst[i+15:i] := tmp[16:1] - ELSE - dst[i+15:i] := src[i+15:i] - FI + dst[i+15:i] := SaturateU16(a[i+15:i] - b[i+15:i]) ENDFOR -dst[MAX:512] := 0 - -
immintrin.h
+ +
mmintrin.h
 - + Integer - AVX512BW + MMX Arithmetic - - - - Multiply packed 16-bit integers in "a" and "b", producing intermediate signed 32-bit integers. Truncate each intermediate integer to the 18 most significant bits, round by adding 1, and store bits [16:1] to "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + + + Multiply packed signed 16-bit integers in "a" and "b", producing intermediate signed 32-bit integers. Horizontally add adjacent pairs of intermediate 32-bit integers, and pack the results in "dst". -FOR j := 0 to 31 - i := j*16 - IF k[j] - tmp[31:0] := ((a[i+15:i] * b[i+15:i]) >> 14) + 1 - dst[i+15:i] := tmp[16:1] - ELSE - dst[i+15:i] := 9 - FI +FOR j := 0 to 1 + i := j*32 + dst[i+31:i] := SignExtend32(a[i+31:i+16]*b[i+31:i+16]) + SignExtend32(a[i+15:i]*b[i+15:i]) ENDFOR -dst[MAX:512] := 0 - -
immintrin.h
+ +
mmintrin.h
 - + Integer - AVX512BW + MMX Arithmetic - - - Multiply packed 16-bit integers in "a" and "b", producing intermediate signed 32-bit integers. Truncate each intermediate integer to the 18 most significant bits, round by adding 1, and store bits [16:1] to "dst". + + + + Multiply the packed signed 16-bit integers in "a" and "b", producing intermediate 32-bit integers, and store the high 16 bits of the intermediate integers in "dst". -FOR j := 0 to 31 +FOR j := 0 to 3 i := j*16 - tmp[31:0] := ((a[i+15:i] * b[i+15:i]) >> 14) + 1 - dst[i+15:i] := tmp[16:1] + tmp[31:0] := SignExtend32(a[i+15:i]) * SignExtend32(b[i+15:i]) + dst[i+15:i] := tmp[31:16] ENDFOR -dst[MAX:512] := 0 - -
immintrin.h
+ +
mmintrin.h
 - + Integer - AVX512VL - AVX512BW + MMX Arithmetic - - - - - Multiply packed 16-bit integers in "a" and "b", producing intermediate signed 32-bit integers. Truncate each intermediate integer to the 18 most significant bits, round by adding 1, and store bits [16:1] to "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + + + Multiply the packed 16-bit integers in "a" and "b", producing intermediate 32-bit integers, and store the low 16 bits of the intermediate integers in "dst". -FOR j := 0 to 7 +FOR j := 0 to 3 i := j*16 - IF k[j] - tmp[31:0] := ((a[i+15:i] * b[i+15:i]) >> 14) + 1 - dst[i+15:i] := tmp[16:1] - ELSE - dst[i+15:i] := src[i+15:i] - FI + tmp[31:0] := a[i+15:i] * b[i+15:i] + dst[i+15:i] := tmp[15:0] ENDFOR -dst[MAX:128] := 0 - -
immintrin.h
+ +
mmintrin.h
 - + + Floating Point Integer - AVX512VL - AVX512BW - Arithmetic - - - - Multiply packed 16-bit integers in "a" and "b", producing intermediate signed 32-bit integers. Truncate each intermediate integer to the 18 most significant bits, round by adding 1, and store bits [16:1] to "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + MMX + Shift + + + + Shift packed 16-bit integers in "a" left by "count" while shifting in zeros, and store the results in "dst". -FOR j := 0 to 7 +FOR j := 0 to 3 i := j*16 - IF k[j] - tmp[31:0] := ((a[i+15:i] * b[i+15:i]) >> 14) + 1 - dst[i+15:i] := tmp[16:1] + IF count[63:0] > 15 + dst[i+15:i] := 0 ELSE - dst[i+15:i] := 9 + dst[i+15:i] := ZeroExtend16(a[i+15:i] << count[63:0]) FI ENDFOR -dst[MAX:128] := 0 - -
immintrin.h
+ +
mmintrin.h
 - + + Floating Point Integer - AVX512VL - AVX512BW - Arithmetic - - - - - Multiply the packed unsigned 16-bit integers in "a" and "b", producing intermediate 32-bit integers, and store the high 16 bits of the intermediate integers in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + MMX + Shift + + + + Shift packed 16-bit integers in "a" left by "imm8" while shifting in zeros, and store the results in "dst". -FOR j := 0 to 15 +FOR j := 0 to 3 i := j*16 - IF k[j] - tmp[31:0] := a[i+15:i] * b[i+15:i] - dst[i+15:i] := tmp[31:16] + IF imm8[7:0] > 15 + dst[i+15:i] := 0 ELSE - dst[i+15:i] := src[i+15:i] + dst[i+15:i] := ZeroExtend16(a[i+15:i] << imm8[7:0]) FI ENDFOR -dst[MAX:256] := 0 - -
immintrin.h
+ +
mmintrin.h
 - + + Floating Point Integer - AVX512VL - AVX512BW - Arithmetic - - - - Multiply the packed unsigned 16-bit integers in "a" and "b", producing intermediate 32-bit integers, and store the high 16 bits of the intermediate integers in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + MMX + Shift + + + + Shift packed 32-bit integers in "a" left by "count" while shifting in zeros, and store the results in "dst". -FOR j := 0 to 15 - i := j*16 - IF k[j] - tmp[31:0] := a[i+15:i] * b[i+15:i] - dst[i+15:i] := tmp[31:16] +FOR j := 0 to 1 + i := j*32 + IF count[63:0] > 31 + dst[i+31:i] := 0 ELSE - dst[i+15:i] := o + dst[i+31:i] := ZeroExtend32(a[i+31:i] << count[63:0]) FI ENDFOR -dst[MAX:256] := 0 - -
immintrin.h
+ +
mmintrin.h
 - + + Floating Point Integer - AVX512BW - Arithmetic - - - - - Multiply the packed unsigned 16-bit integers in "a" and "b", producing intermediate 32-bit integers, and store the high 16 bits of the intermediate integers in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + MMX + Shift + + + + Shift packed 32-bit integers in "a" left by "imm8" while shifting in zeros, and store the results in "dst". -FOR j := 0 to 31 - i := j*16 - IF k[j] - tmp[31:0] := a[i+15:i] * b[i+15:i] - dst[i+15:i] := tmp[31:16] +FOR j := 0 to 1 + i := j*32 + IF imm8[7:0] > 31 + dst[i+31:i] := 0 ELSE - dst[i+15:i] := src[i+15:i] + dst[i+31:i] := ZeroExtend32(a[i+31:i] << imm8[7:0]) FI ENDFOR -dst[MAX:512] := 0 - -
immintrin.h
+ +
mmintrin.h
 - + + Floating Point Integer - AVX512BW - Arithmetic - - - - Multiply the packed unsigned 16-bit integers in "a" and "b", producing intermediate 32-bit integers, and store the high 16 bits of the intermediate integers in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + MMX + Shift + + + + Shift 64-bit integer "a" left by "count" while shifting in zeros, and store the result in "dst". -FOR j := 0 to 31 +IF count[63:0] > 63 + dst[63:0] := 0 +ELSE + dst[63:0] := ZeroExtend64(a[63:0] << count[63:0]) +FI + + +
mmintrin.h
+ + + Floating Point + Integer + MMX + Shift + + + + Shift 64-bit integer "a" left by "imm8" while shifting in zeros, and store the result in "dst". + +IF imm8[7:0] > 63 + dst[63:0] := 0 +ELSE + dst[63:0] := ZeroExtend64(a[63:0] << imm8[7:0]) +FI + + +
mmintrin.h
+ + + Floating Point + Integer + MMX + Shift + + + + Shift packed 16-bit integers in "a" right by "count" while shifting in sign bits, and store the results in "dst". + +FOR j := 0 to 3 i := j*16 - IF k[j] - tmp[31:0] := a[i+15:i] * b[i+15:i] - dst[i+15:i] := tmp[31:16] + IF count[63:0] > 15 + dst[i+15:i] := (a[i+15] ? 0xFFFF : 0x0) ELSE - dst[i+15:i] := o + dst[i+15:i] := SignExtend16(a[i+15:i] >> count[63:0]) FI ENDFOR -dst[MAX:512] := 0 - -
immintrin.h
+ +
mmintrin.h
 - + + Floating Point Integer - AVX512BW - Arithmetic - - - Multiply the packed unsigned 16-bit integers in "a" and "b", producing intermediate 32-bit integers, and store the high 16 bits of the intermediate integers in "dst". + MMX + Shift + + + + Shift packed 16-bit integers in "a" right by "imm8" while shifting in sign bits, and store the results in "dst". -FOR j := 0 to 31 +FOR j := 0 to 3 i := j*16 - tmp[31:0] := a[i+15:i] * b[i+15:i] - dst[i+15:i] := tmp[31:16] + IF imm8[7:0] > 15 + dst[i+15:i] := (a[i+15] ? 0xFFFF : 0x0) + ELSE + dst[i+15:i] := SignExtend16(a[i+15:i] >> imm8[7:0]) + FI ENDFOR -dst[MAX:512] := 0 - -
immintrin.h
+ +
mmintrin.h
 - + + Floating Point Integer - AVX512VL - AVX512BW - Arithmetic - - - - - Multiply the packed unsigned 16-bit integers in "a" and "b", producing intermediate 32-bit integers, and store the high 16 bits of the intermediate integers in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + MMX + Shift + + + + Shift packed 32-bit integers in "a" right by "count" while shifting in sign bits, and store the results in "dst". -FOR j := 0 to 7 - i := j*16 - IF k[j] - tmp[31:0] := a[i+15:i] * b[i+15:i] - dst[i+15:i] := tmp[31:16] +FOR j := 0 to 1 + i := j*32 + IF count[63:0] > 31 + dst[i+31:i] := (a[i+31] ? 0xFFFFFFFF : 0x0) ELSE - dst[i+15:i] := src[i+15:i] + dst[i+31:i] := SignExtend32(a[i+31:i] >> count[63:0]) FI ENDFOR -dst[MAX:128] := 0 - -
immintrin.h
+ +
mmintrin.h
 - + + Floating Point Integer - AVX512VL - AVX512BW - Arithmetic - - - - Multiply the packed unsigned 16-bit integers in "a" and "b", producing intermediate 32-bit integers, and store the high 16 bits of the intermediate integers in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + MMX + Shift + + + + Shift packed 32-bit integers in "a" right by "imm8" while shifting in sign bits, and store the results in "dst". -FOR j := 0 to 7 - i := j*16 - IF k[j] - tmp[31:0] := a[i+15:i] * b[i+15:i] - dst[i+15:i] := tmp[31:16] +FOR j := 0 to 1 + i := j*32 + IF imm8[7:0] > 31 + dst[i+31:i] := (a[i+31] ? 0xFFFFFFFF : 0x0) ELSE - dst[i+15:i] := o + dst[i+31:i] := SignExtend32(a[i+31:i] >> imm8[7:0]) FI ENDFOR -dst[MAX:128] := 0 - -
immintrin.h
+ +
mmintrin.h
 - + + Floating Point Integer - AVX512VL - AVX512BW - Arithmetic - - - - - Multiply the packed 16-bit integers in "a" and "b", producing intermediate 32-bit integers, and store the high 16 bits of the intermediate integers in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + MMX + Shift + + + + Shift packed 16-bit integers in "a" right by "count" while shifting in zeros, and store the results in "dst". -FOR j := 0 to 15 +FOR j := 0 to 3 i := j*16 - IF k[j] - tmp[31:0] := a[i+15:i] * b[i+15:i] - dst[i+15:i] := tmp[31:16] + IF count[63:0] > 15 + dst[i+15:i] := 0 ELSE - dst[i+15:i] := src[i+15:i] + dst[i+15:i] := ZeroExtend16(a[i+15:i] >> count[63:0]) FI ENDFOR -dst[MAX:256] := 0 - -
immintrin.h
+ +
mmintrin.h
 - + + Floating Point Integer - AVX512VL - AVX512BW - Arithmetic - - - - Multiply the packed 16-bit integers in "a" and "b", producing intermediate 32-bit integers, and store the high 16 bits of the intermediate integers in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + MMX + Shift + + + + Shift packed 16-bit integers in "a" right by "imm8" while shifting in zeros, and store the results in "dst". -FOR j := 0 to 15 +FOR j := 0 to 3 i := j*16 - IF k[j] - tmp[31:0] := a[i+15:i] * b[i+15:i] - dst[i+15:i] := tmp[31:16] - ELSE + IF imm8[7:0] > 15 dst[i+15:i] := 0 + ELSE + dst[i+15:i] := ZeroExtend16(a[i+15:i] >> imm8[7:0]) FI ENDFOR -dst[MAX:256] := 0 - -
immintrin.h
+ +
mmintrin.h
 - + + Floating Point Integer - AVX512BW - Arithmetic - - - - - Multiply the packed 16-bit integers in "a" and "b", producing intermediate 32-bit integers, and store the high 16 bits of the intermediate integers in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + MMX + Shift + + + + Shift packed 32-bit integers in "a" right by "count" while shifting in zeros, and store the results in "dst". -FOR j := 0 to 31 - i := j*16 - IF k[j] - tmp[31:0] := a[i+15:i] * b[i+15:i] - dst[i+15:i] := tmp[31:16] +FOR j := 0 to 1 + i := j*32 + IF count[63:0] > 31 + dst[i+31:i] := 0 ELSE - dst[i+15:i] := src[i+15:i] + dst[i+31:i] := ZeroExtend32(a[i+31:i] >> count[63:0]) FI ENDFOR -dst[MAX:512] := 0 - -
immintrin.h
+ +
mmintrin.h
 - + + Floating Point Integer - AVX512BW - Arithmetic - - - - Multiply the packed 16-bit integers in "a" and "b", producing intermediate 32-bit integers, and store the high 16 bits of the intermediate integers in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + MMX + Shift + + + + Shift packed 32-bit integers in "a" right by "imm8" while shifting in zeros, and store the results in "dst". -FOR j := 0 to 31 - i := j*16 - IF k[j] - tmp[31:0] := a[i+15:i] * b[i+15:i] - dst[i+15:i] := tmp[31:16] +FOR j := 0 to 1 + i := j*32 + IF imm8[7:0] > 31 + dst[i+31:i] := 0 ELSE - dst[i+15:i] := 0 + dst[i+31:i] := ZeroExtend32(a[i+31:i] >> imm8[7:0]) FI ENDFOR -dst[MAX:512] := 0 - -
immintrin.h
+ +
mmintrin.h
 - + + Floating Point Integer - AVX512BW - Arithmetic - - - Multiply the packed 16-bit integers in "a" and "b", producing intermediate 32-bit integers, and store the high 16 bits of the intermediate integers in "dst". + MMX + Shift + + + + Shift 64-bit integer "a" right by "count" while shifting in zeros, and store the result in "dst". -FOR j := 0 to 31 - i := j*16 - tmp[31:0] := a[i+15:i] * b[i+15:i] - dst[i+15:i] := tmp[31:16] -ENDFOR -dst[MAX:512] := 0 +IF count[63:0] > 63 + dst[63:0] := 0 +ELSE + dst[63:0] := ZeroExtend64(a[63:0] >> count[63:0]) +FI - -
immintrin.h
+ +
mmintrin.h
 - + + Floating Point Integer - AVX512VL - AVX512BW - Arithmetic - - - - - Multiply the packed 16-bit integers in "a" and "b", producing intermediate 32-bit integers, and store the high 16 bits of the intermediate integers in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + MMX + Shift + + + + Shift 64-bit integer "a" right by "imm8" while shifting in zeros, and store the result in "dst". + +IF imm8[7:0] > 63 + dst[63:0] := 0 +ELSE + dst[63:0] := ZeroExtend64(a[63:0] >> imm8[7:0]) +FI + + +
mmintrin.h
+ + + Integer + MMX + Logical + + + + Compute the bitwise AND of 64 bits (representing integer data) in "a" and "b", and store the result in "dst". + +dst[63:0] := (a[63:0] AND b[63:0]) + + +
mmintrin.h
+ + + Integer + MMX + Logical + + + + Compute the bitwise NOT of 64 bits (representing integer data) in "a" and then AND with "b", and store the result in "dst". + +dst[63:0] := ((NOT a[63:0]) AND b[63:0]) + + +
mmintrin.h
+ + + Integer + MMX + Logical + + + + Compute the bitwise OR of 64 bits (representing integer data) in "a" and "b", and store the result in "dst". + +dst[63:0] := (a[63:0] OR b[63:0]) + + +
mmintrin.h
+ + + Integer + MMX + Logical + + + + Compute the bitwise XOR of 64 bits (representing integer data) in "a" and "b", and store the result in "dst". + +dst[63:0] := (a[63:0] XOR b[63:0]) + + +
mmintrin.h
+ + + Integer + MMX + Compare + + + + Compare packed 8-bit integers in "a" and "b" for equality, and store the results in "dst". FOR j := 0 to 7 - i := j*16 - IF k[j] - tmp[31:0] := a[i+15:i] * b[i+15:i] - dst[i+15:i] := tmp[31:16] - ELSE - dst[i+15:i] := src[i+15:i] - FI + i := j*8 + dst[i+7:i] := ( a[i+7:i] == b[i+7:i] ) ? 0xFF : 0 ENDFOR -dst[MAX:128] := 0 - -
immintrin.h
+ +
mmintrin.h
 - + Integer - AVX512VL - AVX512BW - Arithmetic - - - - Multiply the packed 16-bit integers in "a" and "b", producing intermediate 32-bit integers, and store the high 16 bits of the intermediate integers in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + MMX + Compare + + + + Compare packed 16-bit integers in "a" and "b" for equality, and store the results in "dst". -FOR j := 0 to 7 +FOR j := 0 to 3 i := j*16 - IF k[j] - tmp[31:0] := a[i+15:i] * b[i+15:i] - dst[i+15:i] := tmp[31:16] - ELSE - dst[i+15:i] := 0 - FI + dst[i+15:i] := ( a[i+15:i] == b[i+15:i] ) ? 0xFFFF : 0 ENDFOR -dst[MAX:128] := 0 - -
immintrin.h
+ +
mmintrin.h
 - + Integer - AVX512VL - AVX512F - Arithmetic - - - - - Multiply the packed 32-bit integers in "a" and "b", producing intermediate 64-bit integers, and store the low 32 bits of the intermediate integers in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + MMX + Compare + + + + Compare packed 32-bit integers in "a" and "b" for equality, and store the results in "dst". -FOR j := 0 to 7 +FOR j := 0 to 1 i := j*32 - IF k[j] - tmp[63:0] := a[i+31:i] * b[i+31:i] - dst[i+31:i] := tmp[31:0] - ELSE - dst[i+31:i] := src[i+31:i] - FI + dst[i+31:i] := ( a[i+31:i] == b[i+31:i] ) ? 0xFFFFFFFF : 0 ENDFOR -dst[MAX:256] := 0 - -
immintrin.h
+ +
mmintrin.h
 - + Integer - AVX512VL - AVX512F - Arithmetic - - - - Multiply the packed 32-bit integers in "a" and "b", producing intermediate 64-bit integers, and store the low 32 bits of the intermediate integers in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + MMX + Compare + + + + Compare packed 8-bit integers in "a" and "b" for greater-than, and store the results in "dst". FOR j := 0 to 7 - i := j*32 - IF k[j] - tmp[63:0] := a[i+31:i] * b[i+31:i] - dst[i+31:i] := tmp[31:0] - ELSE - dst[i+31:i] := 0 - FI + i := j*8 + dst[i+7:i] := ( a[i+7:i] > b[i+7:i] ) ? 0xFF : 0 ENDFOR -dst[MAX:256] := 0 - -
immintrin.h
+ +
mmintrin.h
 - + Integer - AVX512F - Arithmetic - - - - Multiply the packed 32-bit integers in "a" and "b", producing intermediate 64-bit integers, and store the low 32 bits of the intermediate integers in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + MMX + Compare + + + + Compare packed 16-bit integers in "a" and "b" for greater-than, and store the results in "dst". -FOR j := 0 to 15 - i := j*32 - IF k[j] - tmp[63:0] := a[i+31:i] * b[i+31:i] - dst[i+31:i] := tmp[31:0] - ELSE - dst[i+31:i] := 0 - FI +FOR j := 0 to 3 + i := j*16 + dst[i+15:i] := ( a[i+15:i] > b[i+15:i] ) ? 0xFFFF : 0 ENDFOR -dst[MAX:512] := 0 - -
immintrin.h
+ +
mmintrin.h
 - + Integer - AVX512VL - AVX512F - Arithmetic - - - - - Multiply the packed 32-bit integers in "a" and "b", producing intermediate 64-bit integers, and store the low 32 bits of the intermediate integers in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + MMX + Compare + + + + Compare packed 32-bit integers in "a" and "b" for greater-than, and store the results in "dst". -FOR j := 0 to 3 +FOR j := 0 to 1 i := j*32 - IF k[j] - tmp[63:0] := a[i+31:i] * b[i+31:i] - dst[i+31:i] := tmp[31:0] - ELSE - dst[i+31:i] := src[i+31:i] - FI + dst[i+31:i] := ( a[i+31:i] > b[i+31:i] ) ? 0xFFFFFFFF : 0 ENDFOR -dst[MAX:128] := 0 - -
immintrin.h
+ +
mmintrin.h
+ + + MMX + General Support + + + Empty the MMX state, which marks the x87 FPU registers as available for use by x87 instructions. This instruction must be used at the end of all MMX technology procedures. + +
mmintrin.h
 - + Integer - AVX512VL - AVX512F + MMX Arithmetic - - - - Multiply the packed 32-bit integers in "a" and "b", producing intermediate 64-bit integers, and store the low 32 bits of the intermediate integers in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + + + Add packed 8-bit integers in "a" and "b", and store the results in "dst". -FOR j := 0 to 3 - i := j*32 - IF k[j] - tmp[63:0] := a[i+31:i] * b[i+31:i] - dst[i+31:i] := tmp[31:0] - ELSE - dst[i+31:i] := 0 - FI +FOR j := 0 to 7 + i := j*8 + dst[i+7:i] := a[i+7:i] + b[i+7:i] ENDFOR -dst[MAX:128] := 0 - -
immintrin.h
+ +
mmintrin.h
 - + Integer - AVX512VL - AVX512DQ + MMX Arithmetic - - - - - Multiply the packed 64-bit integers in "a" and "b", producing intermediate 128-bit integers, and store the low 64 bits of the intermediate integers in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + + + Add packed 16-bit integers in "a" and "b", and store the results in "dst". FOR j := 0 to 3 - i := j*64 - IF k[j] - tmp[127:0] := a[i+63:i] * b[i+63:i] - dst[i+63:i] := tmp[63:0] - ELSE - dst[i+63:i] := src[i+63:i] - FI + i := j*16 + dst[i+15:i] := a[i+15:i] + b[i+15:i] +ENDFOR + + +
mmintrin.h
+ + + Integer + MMX + Arithmetic + + + + Add packed 32-bit integers in "a" and "b", and store the results in "dst". + +FOR j := 0 to 1 + i := j*32 + dst[i+31:i] := a[i+31:i] + b[i+31:i] ENDFOR -dst[MAX:256] := 0 - -
immintrin.h
+ +
mmintrin.h
 - + Integer - AVX512VL - AVX512DQ + MMX Arithmetic - - - - Multiply the packed 64-bit integers in "a" and "b", producing intermediate 128-bit integers, and store the low 64 bits of the intermediate integers in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + + + Add packed signed 8-bit integers in "a" and "b" using saturation, and store the results in "dst". -FOR j := 0 to 3 - i := j*64 - IF k[j] - tmp[127:0] := a[i+63:i] * b[i+63:i] - dst[i+63:i] := tmp[63:0] - ELSE - dst[i+63:i] := 0 - FI +FOR j := 0 to 7 + i := j*8 + dst[i+7:i] := Saturate8( a[i+7:i] + b[i+7:i] ) ENDFOR -dst[MAX:256] := 0 - -
immintrin.h
+ +
mmintrin.h
 - + Integer - AVX512VL - AVX512DQ + MMX Arithmetic - - - Multiply the packed 64-bit integers in "a" and "b", producing intermediate 128-bit integers, and store the low 64 bits of the intermediate integers in "dst". + + + + Add packed signed 16-bit integers in "a" and "b" using saturation, and store the results in "dst". FOR j := 0 to 3 - i := j*64 - tmp[127:0] := a[i+63:i] * b[i+63:i] - dst[i+63:i] := tmp[63:0] + i := j*16 + dst[i+15:i] := Saturate16( a[i+15:i] + b[i+15:i] ) ENDFOR -dst[MAX:256] := 0 - -
immintrin.h
+ +
mmintrin.h
 - + Integer - AVX512DQ + MMX Arithmetic - - - - - Multiply the packed 64-bit integers in "a" and "b", producing intermediate 128-bit integers, and store the low 64 bits of the intermediate integers in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + + + Add packed unsigned 8-bit integers in "a" and "b" using saturation, and store the results in "dst". FOR j := 0 to 7 - i := j*64 - IF k[j] - tmp[127:0] := a[i+63:i] * b[i+63:i] - dst[i+63:i] := tmp[63:0] - ELSE - dst[i+63:i] := src[i+63:i] - FI + i := j*8 + dst[i+7:i] := SaturateU8( a[i+7:i] + b[i+7:i] ) ENDFOR -dst[MAX:512] := 0 - -
immintrin.h
+ +
mmintrin.h
 - + Integer - AVX512DQ + MMX Arithmetic - - - - Multiply the packed 64-bit integers in "a" and "b", producing intermediate 128-bit integers, and store the low 64 bits of the intermediate integers in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + + + Add packed unsigned 16-bit integers in "a" and "b" using saturation, and store the results in "dst". -FOR j := 0 to 7 - i := j*64 - IF k[j] - tmp[127:0] := a[i+63:i] * b[i+63:i] - dst[i+63:i] := tmp[63:0] - ELSE - dst[i+63:i] := 0 - FI +FOR j := 0 to 3 + i := j*16 + dst[i+15:i] := SaturateU16( a[i+15:i] + b[i+15:i] ) ENDFOR -dst[MAX:512] := 0 - -
immintrin.h
+ +
mmintrin.h
 - + Integer - AVX512DQ + MMX Arithmetic - - - Multiply the packed 64-bit integers in "a" and "b", producing intermediate 128-bit integers, and store the low 64 bits of the intermediate integers in "dst". + + + + Subtract packed 8-bit integers in "b" from packed 8-bit integers in "a", and store the results in "dst". FOR j := 0 to 7 - i := j*64 - tmp[127:0] := a[i+63:i] * b[i+63:i] - dst[i+63:i] := tmp[63:0] + i := j*8 + dst[i+7:i] := a[i+7:i] - b[i+7:i] ENDFOR -dst[MAX:512] := 0 - -
immintrin.h
+ +
mmintrin.h
 - + Integer - AVX512VL - AVX512DQ + MMX Arithmetic - - - - - Multiply the packed 64-bit integers in "a" and "b", producing intermediate 128-bit integers, and store the low 64 bits of the intermediate integers in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + + + Subtract packed 16-bit integers in "b" from packed 16-bit integers in "a", and store the results in "dst". -FOR j := 0 to 1 - i := j*64 - IF k[j] - tmp[127:0] := a[i+63:i] * b[i+63:i] - dst[i+63:i] := tmp[63:0] - ELSE - dst[i+63:i] := src[i+63:i] - FI +FOR j := 0 to 3 + i := j*16 + dst[i+15:i] := a[i+15:i] - b[i+15:i] ENDFOR -dst[MAX:128] := 0 - -
immintrin.h
+ +
mmintrin.h
 - + Integer - AVX512VL - AVX512DQ + MMX Arithmetic - - - - Multiply the packed 64-bit integers in "a" and "b", producing intermediate 128-bit integers, and store the low 64 bits of the intermediate integers in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + + + Subtract packed 32-bit integers in "b" from packed 32-bit integers in "a", and store the results in "dst". FOR j := 0 to 1 - i := j*64 - IF k[j] - tmp[127:0] := a[i+63:i] * b[i+63:i] - dst[i+63:i] := tmp[63:0] - ELSE - dst[i+63:i] := 0 - FI + i := j*32 + dst[i+31:i] := a[i+31:i] - b[i+31:i] ENDFOR -dst[MAX:128] := 0 - -
immintrin.h
+ +
mmintrin.h
 - + Integer - AVX512VL - AVX512DQ + MMX Arithmetic - - - Multiply the packed 64-bit integers in "a" and "b", producing intermediate 128-bit integers, and store the low 64 bits of the intermediate integers in "dst". + + + + Subtract packed signed 8-bit integers in "b" from packed 8-bit integers in "a" using saturation, and store the results in "dst". -FOR j := 0 to 1 - i := j*64 - tmp[127:0] := a[i+63:i] * b[i+63:i] - dst[i+63:i] := tmp[63:0] +FOR j := 0 to 7 + i := j*8 + dst[i+7:i] := Saturate8(a[i+7:i] - b[i+7:i]) ENDFOR -dst[MAX:128] := 0 - -
immintrin.h
+ +
mmintrin.h
 - + Integer - AVX512VL - AVX512BW + MMX Arithmetic - - - - - Multiply the packed 16-bit integers in "a" and "b", producing intermediate 32-bit integers, and store the low 16 bits of the intermediate integers in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + + + Subtract packed signed 16-bit integers in "b" from packed 16-bit integers in "a" using saturation, and store the results in "dst". -FOR j := 0 to 15 +FOR j := 0 to 3 i := j*16 - IF k[j] - tmp[31:0] := a[i+15:i] * b[i+15:i] - dst[i+15:i] := tmp[15:0] - ELSE - dst[i+15:i] := src[i+15:i] - FI + dst[i+15:i] := Saturate16(a[i+15:i] - b[i+15:i]) ENDFOR -dst[MAX:256] := 0 - -
immintrin.h
+ +
mmintrin.h
 - + Integer - AVX512VL - AVX512BW + MMX Arithmetic - - - - Multiply the packed 16-bit integers in "a" and "b", producing intermediate 32-bit integers, and store the low 16 bits of the intermediate integers in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + + + Subtract packed unsigned 8-bit integers in "b" from packed unsigned 8-bit integers in "a" using saturation, and store the results in "dst". -FOR j := 0 to 15 - i := j*16 - IF k[j] - tmp[31:0] := a[i+15:i] * b[i+15:i] - dst[i+15:i] := tmp[15:0] - ELSE - dst[i+15:i] := 0 - FI +FOR j := 0 to 7 + i := j*8 + dst[i+7:i] := SaturateU8(a[i+7:i] - b[i+7:i]) ENDFOR -dst[MAX:256] := 0 - -
immintrin.h
+ +
mmintrin.h
 - + Integer - AVX512BW + MMX Arithmetic - - - - - Multiply the packed 16-bit integers in "a" and "b", producing intermediate 32-bit integers, and store the low 16 bits of the intermediate integers in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + + + Subtract packed unsigned 16-bit integers in "b" from packed unsigned 16-bit integers in "a" using saturation, and store the results in "dst". -FOR j := 0 to 31 +FOR j := 0 to 3 i := j*16 - IF k[j] - tmp[31:0] := a[i+15:i] * b[i+15:i] - dst[i+15:i] := tmp[15:0] - ELSE - dst[i+15:i] := src[i+15:i] - FI + dst[i+15:i] := SaturateU16(a[i+15:i] - b[i+15:i]) ENDFOR -dst[MAX:512] := 0 - -
immintrin.h
+ +
mmintrin.h
 - + Integer - AVX512BW + MMX Arithmetic - - - - Multiply the packed 16-bit integers in "a" and "b", producing intermediate 32-bit integers, and store the low 16 bits of the intermediate integers in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + + + Multiply packed signed 16-bit integers in "a" and "b", producing intermediate signed 32-bit integers. Horizontally add adjacent pairs of intermediate 32-bit integers, and pack the results in "dst". -FOR j := 0 to 31 - i := j*16 - IF k[j] - tmp[31:0] := a[i+15:i] * b[i+15:i] - dst[i+15:i] := tmp[15:0] - ELSE - dst[i+15:i] := 0 - FI +FOR j := 0 to 1 + i := j*32 + dst[i+31:i] := SignExtend32(a[i+31:i+16]*b[i+31:i+16]) + SignExtend32(a[i+15:i]*b[i+15:i]) ENDFOR -dst[MAX:512] := 0 - -
immintrin.h
+ +
mmintrin.h
 - + Integer - AVX512BW + MMX Arithmetic - - - Multiply the packed 16-bit integers in "a" and "b", producing intermediate 32-bit integers, and store the low 16 bits of the intermediate integers in "dst". + + + + Multiply the packed signed 16-bit integers in "a" and "b", producing intermediate 32-bit integers, and store the high 16 bits of the intermediate integers in "dst". -FOR j := 0 to 31 +FOR j := 0 to 3 i := j*16 - tmp[31:0] := a[i+15:i] * b[i+15:i] - dst[i+15:i] := tmp[15:0] + tmp[31:0] := SignExtend32(a[i+15:i]) * SignExtend32(b[i+15:i]) + dst[i+15:i] := tmp[31:16] ENDFOR -dst[MAX:512] := 0 - -
immintrin.h
+ +
mmintrin.h
 - + Integer - AVX512VL - AVX512BW + MMX Arithmetic - - - - - Multiply the packed 16-bit integers in "a" and "b", producing intermediate 32-bit integers, and store the low 16 bits of the intermediate integers in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + + + Multiply the packed 16-bit integers in "a" and "b", producing intermediate 32-bit integers, and store the low 16 bits of the intermediate integers in "dst". -FOR j := 0 to 7 +FOR j := 0 to 3 i := j*16 - IF k[j] - tmp[31:0] := a[i+15:i] * b[i+15:i] - dst[i+15:i] := tmp[15:0] - ELSE - dst[i+15:i] := src[i+15:i] - FI + tmp[31:0] := a[i+15:i] * b[i+15:i] + dst[i+15:i] := tmp[15:0] ENDFOR -dst[MAX:128] := 0 - -
immintrin.h
+ +
mmintrin.h
 - + Integer - AVX512VL - AVX512BW - Arithmetic - - - - Multiply the packed 16-bit integers in "a" and "b", producing intermediate 32-bit integers, and store the low 16 bits of the intermediate integers in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + MMX + Shift + + + + Shift packed 16-bit integers in "a" left by "count" while shifting in zeros, and store the results in "dst". -FOR j := 0 to 7 +FOR j := 0 to 3 i := j*16 - IF k[j] - tmp[31:0] := a[i+15:i] * b[i+15:i] - dst[i+15:i] := tmp[15:0] - ELSE + IF count[63:0] > 15 dst[i+15:i] := 0 + ELSE + dst[i+15:i] := ZeroExtend16(a[i+15:i] << count[63:0]) FI ENDFOR -dst[MAX:128] := 0 - -
immintrin.h
+ +
mmintrin.h
 - + Integer - AVX512VL - AVX512F - Arithmetic - - - - - Multiply the low unsigned 32-bit integers from each packed 64-bit element in "a" and "b", and store the unsigned 64-bit results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + MMX + Shift + + + + Shift packed 16-bit integers in "a" left by "imm8" while shifting in zeros, and store the results in "dst". FOR j := 0 to 3 - i := j*64 - IF k[j] - dst[i+63:i] := a[i+31:i] * b[i+31:i] + i := j*16 + IF imm8[7:0] > 15 + dst[i+15:i] := 0 ELSE - dst[i+63:i] := src[i+63:i] + dst[i+15:i] := ZeroExtend16(a[i+15:i] << imm8[7:0]) FI ENDFOR -dst[MAX:256] := 0 - -
immintrin.h
+ +
mmintrin.h
 - + Integer - AVX512VL - AVX512F - Arithmetic - - - - Multiply the low unsigned 32-bit integers from each packed 64-bit element in "a" and "b", and store the unsigned 64-bit results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + MMX + Shift + + + + Shift packed 32-bit integers in "a" left by "count" while shifting in zeros, and store the results in "dst". -FOR j := 0 to 3 - i := j*64 - IF k[j] - dst[i+63:i] := a[i+31:i] * b[i+31:i] +FOR j := 0 to 1 + i := j*32 + IF count[63:0] > 31 + dst[i+31:i] := 0 ELSE - dst[i+63:i] := 0 + dst[i+31:i] := ZeroExtend32(a[i+31:i] << count[63:0]) FI ENDFOR -dst[MAX:256] := 0 - -
immintrin.h
+ +
mmintrin.h
 - + Integer - AVX512VL - AVX512F - Arithmetic - - - - - Multiply the low unsigned 32-bit integers from each packed 64-bit element in "a" and "b", and store the unsigned 64-bit results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + MMX + Shift + + + + Shift packed 32-bit integers in "a" left by "imm8" while shifting in zeros, and store the results in "dst". FOR j := 0 to 1 - i := j*64 - IF k[j] - dst[i+63:i] := a[i+31:i] * b[i+31:i] + i := j*32 + IF imm8[7:0] > 31 + dst[i+31:i] := 0 ELSE - dst[i+63:i] := src[i+63:i] + dst[i+31:i] := ZeroExtend32(a[i+31:i] << imm8[7:0]) FI ENDFOR -dst[MAX:128] := 0 - -
immintrin.h
+ +
mmintrin.h
 - + Integer - AVX512VL - AVX512F - Arithmetic - - - - Multiply the low unsigned 32-bit integers from each packed 64-bit element in "a" and "b", and store the unsigned 64-bit results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + MMX + Shift + + + + Shift 64-bit integer "a" left by "count" while shifting in zeros, and store the result in "dst". -FOR j := 0 to 1 - i := j*64 - IF k[j] - dst[i+63:i] := a[i+31:i] * b[i+31:i] - ELSE - dst[i+63:i] := 0 - FI -ENDFOR -dst[MAX:128] := 0 +IF count[63:0] > 63 + dst[63:0] := 0 +ELSE + dst[63:0] := ZeroExtend64(a[63:0] << count[63:0]) +FI - -
immintrin.h
+ +
mmintrin.h
 - + Integer - AVX512VL - AVX512F - Logical - - - - - Compute the bitwise OR of packed 32-bit integers in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + MMX + Shift + + + + Shift 64-bit integer "a" left by "imm8" while shifting in zeros, and store the result in "dst". -FOR j := 0 to 7 - i := j*32 - IF k[j] - dst[i+31:i] := a[i+31:i] OR b[i+31:i] +IF imm8[7:0] > 63 + dst[63:0] := 0 +ELSE + dst[63:0] := ZeroExtend64(a[63:0] << imm8[7:0]) +FI + + +
mmintrin.h
+ + + Integer + MMX + Shift + + + + Shift packed 16-bit integers in "a" right by "count" while shifting in sign bits, and store the results in "dst". + +FOR j := 0 to 3 + i := j*16 + IF count[63:0] > 15 + dst[i+15:i] := (a[i+15] ? 0xFFFF : 0x0) ELSE - dst[i+31:i] := src[i+31:i] + dst[i+15:i] := SignExtend16(a[i+15:i] >> count[63:0]) FI ENDFOR -dst[MAX:256] := 0 - -
immintrin.h
+ +
mmintrin.h
 - + Integer - AVX512VL - AVX512F - Logical - - - - Compute the bitwise OR of packed 32-bit integers in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - + MMX + Shift + + + + Shift packed 16-bit integers in "a" right by "imm8" while shifting in sign bits, and store the results in "dst". -FOR j := 0 to 7 - i := j*32 - IF k[j] - dst[i+31:i] := a[i+31:i] OR b[i+31:i] +FOR j := 0 to 3 + i := j*16 + IF imm8[7:0] > 15 + dst[i+15:i] := (a[i+15] ? 0xFFFF : 0x0) ELSE - dst[i+31:i] := 0 + dst[i+15:i] := SignExtend16(a[i+15:i] >> imm8[7:0]) FI ENDFOR -dst[MAX:256] := 0 - -
immintrin.h
+ +
mmintrin.h
 - + Integer - AVX512VL - AVX512F - Logical - - - - - Compute the bitwise OR of packed 32-bit integers in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + MMX + Shift + + + + Shift packed 32-bit integers in "a" right by "count" while shifting in sign bits, and store the results in "dst". -FOR j := 0 to 3 +FOR j := 0 to 1 i := j*32 - IF k[j] - dst[i+31:i] := a[i+31:i] OR b[i+31:i] + IF count[63:0] > 31 + dst[i+31:i] := (a[i+31] ? 0xFFFFFFFF : 0x0) ELSE - dst[i+31:i] := src[i+31:i] + dst[i+31:i] := SignExtend32(a[i+31:i] >> count[63:0]) FI ENDFOR -dst[MAX:128] := 0 - -
immintrin.h
+ +
mmintrin.h
 - + Integer - AVX512VL - AVX512F - Logical - - - - Compute the bitwise OR of packed 32-bit integers in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - + MMX + Shift + + + + Shift packed 32-bit integers in "a" right by "imm8" while shifting in sign bits, and store the results in "dst". -FOR j := 0 to 3 +FOR j := 0 to 1 i := j*32 - IF k[j] - dst[i+31:i] := a[i+31:i] OR b[i+31:i] + IF imm8[7:0] > 31 + dst[i+31:i] := (a[i+31] ? 0xFFFFFFFF : 0x0) ELSE - dst[i+31:i] := 0 + dst[i+31:i] := SignExtend32(a[i+31:i] >> imm8[7:0]) FI ENDFOR -dst[MAX:128] := 0 - -
immintrin.h
+ +
mmintrin.h
 - + Integer - AVX512VL - AVX512F - Logical - - - - - Compute the bitwise OR of packed 64-bit integers in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + MMX + Shift + + + + Shift packed 16-bit integers in "a" right by "count" while shifting in zeros, and store the results in "dst". FOR j := 0 to 3 - i := j*64 - IF k[j] - dst[i+63:i] := a[i+63:i] OR b[i+63:i] + i := j*16 + IF count[63:0] > 15 + dst[i+15:i] := 0 ELSE - dst[i+63:i] := src[i+63:i] + dst[i+15:i] := ZeroExtend16(a[i+15:i] >> count[63:0]) FI ENDFOR -dst[MAX:256] := 0 - -
immintrin.h
+ +
mmintrin.h
 - + Integer - AVX512VL - AVX512F - Logical - - - - Compute the bitwise OR of packed 64-bit integers in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - + MMX + Shift + + + + Shift packed 16-bit integers in "a" right by "imm8" while shifting in zeros, and store the results in "dst". FOR j := 0 to 3 - i := j*64 - IF k[j] - dst[i+63:i] := a[i+63:i] OR b[i+63:i] + i := j*16 + IF imm8[7:0] > 15 + dst[i+15:i] := 0 ELSE - dst[i+63:i] := 0 + dst[i+15:i] := ZeroExtend16(a[i+15:i] >> imm8[7:0]) FI ENDFOR -dst[MAX:256] := 0 - -
immintrin.h
+ +
mmintrin.h
 - + Integer - AVX512VL - AVX512F - Logical - - - - - Compute the bitwise OR of packed 64-bit integers in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + MMX + Shift + + + + Shift packed 32-bit integers in "a" right by "count" while shifting in zeros, and store the results in "dst". FOR j := 0 to 1 - i := j*64 - IF k[j] - dst[i+63:i] := a[i+63:i] OR b[i+63:i] + i := j*32 + IF count[63:0] > 31 + dst[i+31:i] := 0 ELSE - dst[i+63:i] := src[i+63:i] + dst[i+31:i] := ZeroExtend32(a[i+31:i] >> count[63:0]) FI ENDFOR -dst[MAX:128] := 0 - -
immintrin.h
+ +
mmintrin.h
 - + Integer - AVX512VL - AVX512F - Logical - - - - Compute the bitwise OR of packed 64-bit integers in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - + MMX + Shift + + + + Shift packed 32-bit integers in "a" right by "imm8" while shifting in zeros, and store the results in "dst". FOR j := 0 to 1 - i := j*64 - IF k[j] - dst[i+63:i] := a[i+63:i] OR b[i+63:i] + i := j*32 + IF imm8[7:0] > 31 + dst[i+31:i] := 0 ELSE - dst[i+63:i] := 0 + dst[i+31:i] := ZeroExtend32(a[i+31:i] >> imm8[7:0]) FI ENDFOR -dst[MAX:128] := 0 - -
immintrin.h
+ +
mmintrin.h
 - + Integer - AVX512VL - AVX512F + MMX Shift - - - - - Rotate the bits in each packed 32-bit integer in "a" to the left by the number of bits specified in "imm8", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -LEFT_ROTATE_DWORDS(src, count_src){ - count := count_src modulo 32 - RETURN (src << count) OR (src >> (32 - count)) -} + + + + Shift 64-bit integer "a" right by "count" while shifting in zeros, and store the result in "dst". + +IF count[63:0] > 63 + dst[63:0] := 0 +ELSE + dst[63:0] := ZeroExtend64(a[63:0] >> count[63:0]) +FI + + +
mmintrin.h
+ + + Integer + MMX + Shift + + + + Shift 64-bit integer "a" right by "imm8" while shifting in zeros, and store the result in "dst". + +IF imm8[7:0] > 63 + dst[63:0] := 0 +ELSE + dst[63:0] := ZeroExtend64(a[63:0] >> imm8[7:0]) +FI + + +
mmintrin.h
+ + + Integer + MMX + Logical + + + + Compute the bitwise AND of 64 bits (representing integer data) in "a" and "b", and store the result in "dst". + +dst[63:0] := (a[63:0] AND b[63:0]) + + +
mmintrin.h
+ + + Integer + MMX + Logical + + + + Compute the bitwise NOT of 64 bits (representing integer data) in "a" and then AND with "b", and store the result in "dst". + +dst[63:0] := ((NOT a[63:0]) AND b[63:0]) + + +
mmintrin.h
+ + + Integer + MMX + Logical + + + + Compute the bitwise OR of 64 bits (representing integer data) in "a" and "b", and store the result in "dst". + +dst[63:0] := (a[63:0] OR b[63:0]) + + +
mmintrin.h
+ + + Integer + MMX + Logical + + + + Compute the bitwise XOR of 64 bits (representing integer data) in "a" and "b", and store the result in "dst". + +dst[63:0] := (a[63:0] XOR b[63:0]) + + +
mmintrin.h
+ + + Integer + MMX + Compare + + + + Compare packed 8-bit integers in "a" and "b" for equality, and store the results in "dst". + FOR j := 0 to 7 - i := j*32 - IF k[j] - dst[i+31:i] := LEFT_ROTATE_DWORDS(a[i+31:i], imm8[7:0]) - ELSE - dst[i+31:i] := src[i+31:i] - FI + i := j*8 + dst[i+7:i] := ( a[i+7:i] == b[i+7:i] ) ? 0xFF : 0 ENDFOR -dst[MAX:256] := 0 - -
immintrin.h
+ +
mmintrin.h
+ + + Integer + MMX + Compare + + + + Compare packed 16-bit integers in "a" and "b" for equality, and store the results in "dst". + +FOR j := 0 to 3 + i := j*16 + dst[i+15:i] := ( a[i+15:i] == b[i+15:i] ) ? 0xFFFF : 0 +ENDFOR + + +
mmintrin.h
 - + Integer - AVX512VL - AVX512F - Shift - - - - Rotate the bits in each packed 32-bit integer in "a" to the left by the number of bits specified in "imm8", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + MMX + Compare + + + + Compare packed 32-bit integers in "a" and "b" for equality, and store the results in "dst". -LEFT_ROTATE_DWORDS(src, count_src){ - count := count_src modulo 32 - RETURN (src << count) OR (src >> (32 - count)) -} -FOR j := 0 to 7 +FOR j := 0 to 1 i := j*32 - IF k[j] - dst[i+31:i] := LEFT_ROTATE_DWORDS(a[i+31:i], imm8[7:0]) - ELSE - dst[i+31:i] := 0 - FI + dst[i+31:i] := ( a[i+31:i] == b[i+31:i] ) ? 0xFFFFFFFF : 0 ENDFOR -dst[MAX:256] := 0 - -
immintrin.h
+ +
mmintrin.h
 - + Integer - AVX512VL - AVX512F - Shift - - - Rotate the bits in each packed 32-bit integer in "a" to the left by the number of bits specified in "imm8", and store the results in "dst". + MMX + Compare + + + + Compare packed signed 8-bit integers in "a" and "b" for greater-than, and store the results in "dst". -LEFT_ROTATE_DWORDS(src, count_src){ - count := count_src modulo 32 - RETURN (src << count) OR (src >> (32 - count)) -} FOR j := 0 to 7 - i := j*32 - dst[i+31:i] := LEFT_ROTATE_DWORDS(a[i+31:i], imm8[7:0]) + i := j*8 + dst[i+7:i] := ( a[i+7:i] > b[i+7:i] ) ? 0xFF : 0 ENDFOR -dst[MAX:256] := 0 - -
immintrin.h
+ +
mmintrin.h
 - + Integer - AVX512VL - AVX512F - Shift - - - - - Rotate the bits in each packed 32-bit integer in "a" to the left by the number of bits specified in "imm8", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -LEFT_ROTATE_DWORDS(src, count_src){ - count := count_src modulo 32 - RETURN (src << count) OR (src >> (32 - count)) -} + MMX + Compare + + + + Compare packed signed 16-bit integers in "a" and "b" for greater-than, and store the results in "dst". + FOR j := 0 to 3 - i := j*32 - IF k[j] - dst[i+31:i] := LEFT_ROTATE_DWORDS(a[i+31:i], imm8[7:0]) - ELSE - dst[i+31:i] := src[i+31:i] - FI + i := j*16 + dst[i+15:i] := ( a[i+15:i] > b[i+15:i] ) ? 0xFFFF : 0 ENDFOR -dst[MAX:128] := 0 - -
immintrin.h
+ +
mmintrin.h
 - + Integer - AVX512VL - AVX512F - Shift - - - - Rotate the bits in each packed 32-bit integer in "a" to the left by the number of bits specified in "imm8", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + MMX + Compare + + + + Compare packed signed 32-bit integers in "a" and "b" for greater-than, and store the results in "dst". -LEFT_ROTATE_DWORDS(src, count_src){ - count := count_src modulo 32 - RETURN (src << count) OR (src >> (32 - count)) -} -FOR j := 0 to 3 +FOR j := 0 to 1 i := j*32 - IF k[j] - dst[i+31:i] := LEFT_ROTATE_DWORDS(a[i+31:i], imm8[7:0]) - ELSE - dst[i+31:i] := 0 - FI + dst[i+31:i] := ( a[i+31:i] > b[i+31:i] ) ? 0xFFFFFFFF : 0 ENDFOR -dst[MAX:128] := 0 - -
immintrin.h
+ +
mmintrin.h
 - + Integer - AVX512VL - AVX512F - Shift - - - Rotate the bits in each packed 32-bit integer in "a" to the left by the number of bits specified in "imm8", and store the results in "dst". + MMX + Convert + + + Copy 32-bit integer "a" to the lower elements of "dst", and zero the upper element of "dst". -LEFT_ROTATE_DWORDS(src, count_src){ - count := count_src modulo 32 - RETURN (src << count) OR (src >> (32 - count)) -} -FOR j := 0 to 3 - i := j*32 - dst[i+31:i] := LEFT_ROTATE_DWORDS(a[i+31:i], imm8[7:0]) -ENDFOR -dst[MAX:128] := 0 +dst[31:0] := a[31:0] +dst[63:32] := 0 - -
immintrin.h
+ +
mmintrin.h
 - + Integer - AVX512VL - AVX512F - Shift - - - - - Rotate the bits in each packed 64-bit integer in "a" to the left by the number of bits specified in "imm8", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -LEFT_ROTATE_QWORDS(src, count_src){ - count := count_src modulo 64 - RETURN (src << count) OR (src >> (64 - count)) -} -FOR j := 0 to 3 - i := j*64 - IF k[j] - dst[i+63:i] := LEFT_ROTATE_QWORDS(a[i+63:i], imm8[7:0]) - ELSE - dst[i+63:i] := src[i+63:i] - FI -ENDFOR -dst[MAX:256] := 0 + MMX + Convert + + + Copy the lower 32-bit integer in "a" to "dst". + +dst[31:0] := a[31:0] - -
immintrin.h
+ +
mmintrin.h
 - + Integer - AVX512VL - AVX512F - Shift - - - - Rotate the bits in each packed 64-bit integer in "a" to the left by the number of bits specified in "imm8", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + MMX + Convert + + + Copy 64-bit integer "a" to "dst". -LEFT_ROTATE_QWORDS(src, count_src){ - count := count_src modulo 64 - RETURN (src << count) OR (src >> (64 - count)) -} -FOR j := 0 to 3 - i := j*64 - IF k[j] - dst[i+63:i] := LEFT_ROTATE_QWORDS(a[i+63:i], imm8[7:0]) - ELSE - dst[i+63:i] := 0 - FI -ENDFOR -dst[MAX:256] := 0 +dst[63:0] := a[63:0] - -
immintrin.h
+ +
mmintrin.h
 - + Integer - AVX512VL - AVX512F - Shift - - - Rotate the bits in each packed 64-bit integer in "a" to the left by the number of bits specified in "imm8", and store the results in "dst". + MMX + Convert + + + Copy 64-bit integer "a" to "dst". -LEFT_ROTATE_QWORDS(src, count_src){ - count := count_src modulo 64 - RETURN (src << count) OR (src >> (64 - count)) -} -FOR j := 0 to 3 - i := j*64 - dst[i+63:i] := LEFT_ROTATE_QWORDS(a[i+63:i], imm8[7:0]) -ENDFOR -dst[MAX:256] := 0 +dst[63:0] := a[63:0] - -
immintrin.h
+ +
mmintrin.h
 - + Integer - AVX512VL - AVX512F - Shift - - - - - Rotate the bits in each packed 64-bit integer in "a" to the left by the number of bits specified in "imm8", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -LEFT_ROTATE_QWORDS(src, count_src){ - count := count_src modulo 64 - RETURN (src << count) OR (src >> (64 - count)) -} -FOR j := 0 to 1 - i := j*64 - IF k[j] - dst[i+63:i] := LEFT_ROTATE_QWORDS(a[i+63:i], imm8[7:0]) - ELSE - dst[i+63:i] := src[i+63:i] - FI -ENDFOR -dst[MAX:128] := 0 + MMX + Set + + + Return vector of type __m64 with all elements set to zero. + +dst[MAX:0] := 0 - -
immintrin.h
+ +
mmintrin.h
 - + Integer - AVX512VL - AVX512F - Shift - - - - Rotate the bits in each packed 64-bit integer in "a" to the left by the number of bits specified in "imm8", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + MMX + Set + + + + Set packed 32-bit integers in "dst" with the supplied values. -LEFT_ROTATE_QWORDS(src, count_src){ - count := count_src modulo 64 - RETURN (src << count) OR (src >> (64 - count)) -} -FOR j := 0 to 1 - i := j*64 - IF k[j] - dst[i+63:i] := LEFT_ROTATE_QWORDS(a[i+63:i], imm8[7:0]) - ELSE - dst[i+63:i] := 0 - FI -ENDFOR -dst[MAX:128] := 0 +dst[31:0] := e0 +dst[63:32] := e1 - -
immintrin.h
+
mmintrin.h
 - + Integer - AVX512VL - AVX512F - Shift - - - Rotate the bits in each packed 64-bit integer in "a" to the left by the number of bits specified in "imm8", and store the results in "dst". + MMX + Set + + + + + + Set packed 16-bit integers in "dst" with the supplied values. -LEFT_ROTATE_QWORDS(src, count_src){ - count := count_src modulo 64 - RETURN (src << count) OR (src >> (64 - count)) -} -FOR j := 0 to 1 - i := j*64 - dst[i+63:i] := LEFT_ROTATE_QWORDS(a[i+63:i], imm8[7:0]) -ENDFOR -dst[MAX:128] := 0 +dst[15:0] := e0 +dst[31:16] := e1 +dst[47:32] := e2 +dst[63:48] := e3 - -
immintrin.h
+
mmintrin.h
 - + Integer - AVX512VL - AVX512F - Shift - - - - - Rotate the bits in each packed 32-bit integer in "a" to the left by the number of bits specified in the corresponding element of "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -LEFT_ROTATE_DWORDS(src, count_src){ - count := count_src modulo 32 - RETURN (src << count) OR (src >> (32 - count)) -} -FOR j := 0 to 7 + MMX + Set + + + + + + + + + + Set packed 8-bit integers in "dst" with the supplied values. + +dst[7:0] := e0 +dst[15:8] := e1 +dst[23:16] := e2 +dst[31:24] := e3 +dst[39:32] := e4 +dst[47:40] := e5 +dst[55:48] := e6 +dst[63:56] := e7 + +
mmintrin.h
+ + + Integer + MMX + Set + + + Broadcast 32-bit integer "a" to all elements of "dst". + +FOR j := 0 to 1 i := j*32 - IF k[j] - dst[i+31:i] := LEFT_ROTATE_DWORDS(a[i+31:i], b[i+31:i]) - ELSE - dst[i+31:i] := src[i+31:i] - FI + dst[i+31:i] := a[31:0] ENDFOR -dst[MAX:256] := 0 - -
immintrin.h
+
mmintrin.h
 - + Integer - AVX512VL - AVX512F - Shift - - - - Rotate the bits in each packed 32-bit integer in "a" to the left by the number of bits specified in the corresponding element of "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + MMX + Set + + + Broadcast 16-bit integer "a" to all all elements of "dst". -LEFT_ROTATE_DWORDS(src, count_src){ - count := count_src modulo 32 - RETURN (src << count) OR (src >> (32 - count)) -} -FOR j := 0 to 7 - i := j*32 - IF k[j] - dst[i+31:i] := LEFT_ROTATE_DWORDS(a[i+31:i], b[i+31:i]) - ELSE - dst[i+31:i] := 0 - FI +FOR j := 0 to 3 + i := j*16 + dst[i+15:i] := a[15:0] ENDFOR -dst[MAX:256] := 0 - -
immintrin.h
+
mmintrin.h
 - + Integer - AVX512VL - AVX512F - Shift - - - Rotate the bits in each packed 32-bit integer in "a" to the left by the number of bits specified in the corresponding element of "b", and store the results in "dst". + MMX + Set + + + Broadcast 8-bit integer "a" to all elements of "dst". -LEFT_ROTATE_DWORDS(src, count_src){ - count := count_src modulo 32 - RETURN (src << count) OR (src >> (32 - count)) -} FOR j := 0 to 7 - i := j*32 - dst[i+31:i] := LEFT_ROTATE_DWORDS(a[i+31:i], b[i+31:i]) + i := j*8 + dst[i+7:i] := a[7:0] ENDFOR -dst[MAX:256] := 0 - -
immintrin.h
+
mmintrin.h
 - + Integer - AVX512VL - AVX512F - Shift - - - - - Rotate the bits in each packed 32-bit integer in "a" to the left by the number of bits specified in the corresponding element of "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -LEFT_ROTATE_DWORDS(src, count_src){ - count := count_src modulo 32 - RETURN (src << count) OR (src >> (32 - count)) -} -FOR j := 0 to 3 - i := j*32 - IF k[j] - dst[i+31:i] := LEFT_ROTATE_DWORDS(a[i+31:i], b[i+31:i]) - ELSE - dst[i+31:i] := src[i+31:i] - FI -ENDFOR -dst[MAX:128] := 0 + MMX + Set + + + + Set packed 32-bit integers in "dst" with the supplied values in reverse order. + +dst[31:0] := e1 +dst[63:32] := e0 - -
immintrin.h
+
mmintrin.h
 - + Integer - AVX512VL - AVX512F - Shift - - - - Rotate the bits in each packed 32-bit integer in "a" to the left by the number of bits specified in the corresponding element of "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + MMX + Set + + + + + + Set packed 16-bit integers in "dst" with the supplied values in reverse order. -LEFT_ROTATE_DWORDS(src, count_src){ - count := count_src modulo 32 - RETURN (src << count) OR (src >> (32 - count)) -} -FOR j := 0 to 3 - i := j*32 - IF k[j] - dst[i+31:i] := LEFT_ROTATE_DWORDS(a[i+31:i], b[i+31:i]) - ELSE - dst[i+31:i] := 0 - FI -ENDFOR -dst[MAX:128] := 0 +dst[15:0] := e3 +dst[31:16] := e2 +dst[47:32] := e1 +dst[63:48] := e0 - -
immintrin.h
+
mmintrin.h
 - + Integer - AVX512VL - AVX512F - Shift - - - Rotate the bits in each packed 32-bit integer in "a" to the left by the number of bits specified in the corresponding element of "b", and store the results in "dst". + MMX + Set + + + + + + + + + + Set packed 8-bit integers in "dst" with the supplied values in reverse order. -LEFT_ROTATE_DWORDS(src, count_src){ - count := count_src modulo 32 - RETURN (src << count) OR (src >> (32 - count)) -} -FOR j := 0 to 3 - i := j*32 - dst[i+31:i] := LEFT_ROTATE_DWORDS(a[i+31:i], b[i+31:i]) -ENDFOR -dst[MAX:128] := 0 +dst[7:0] := e7 +dst[15:8] := e6 +dst[23:16] := e5 +dst[31:24] := e4 +dst[39:32] := e3 +dst[47:40] := e2 +dst[55:48] := e1 +dst[63:56] := e0 - -
immintrin.h
+
mmintrin.h
 - + Integer - AVX512VL - AVX512F - Shift - - - - - Rotate the bits in each packed 64-bit integer in "a" to the left by the number of bits specified in the corresponding element of "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -LEFT_ROTATE_QWORDS(src, count_src){ - count := count_src modulo 64 - RETURN (src << count) OR (src >> (64 - count)) -} -FOR j := 0 to 3 - i := j*64 - IF k[j] - dst[i+63:i] := LEFT_ROTATE_QWORDS(a[i+63:i], b[i+63:i]) - ELSE - dst[i+63:i] := src[i+63:i] - FI -ENDFOR -dst[MAX:256] := 0 + MMX + Miscellaneous + + + + Convert packed signed 16-bit integers from "a" and "b" to packed 8-bit integers using signed saturation, and store the results in "dst". + +dst[7:0] := Saturate8(a[15:0]) +dst[15:8] := Saturate8(a[31:16]) +dst[23:16] := Saturate8(a[47:32]) +dst[31:24] := Saturate8(a[63:48]) +dst[39:32] := Saturate8(b[15:0]) +dst[47:40] := Saturate8(b[31:16]) +dst[55:48] := Saturate8(b[47:32]) +dst[63:56] := Saturate8(b[63:48]) + + +
mmintrin.h
+ + + Integer + MMX + Miscellaneous + + + + Convert packed signed 32-bit integers from "a" and "b" to packed 16-bit integers using signed saturation, and store the results in "dst". + +dst[15:0] := Saturate16(a[31:0]) +dst[31:16] := Saturate16(a[63:32]) +dst[47:32] := Saturate16(b[31:0]) +dst[63:48] := Saturate16(b[63:32]) - -
immintrin.h
+ +
mmintrin.h
 - + Integer - AVX512VL - AVX512F - Shift - - - - Rotate the bits in each packed 64-bit integer in "a" to the left by the number of bits specified in the corresponding element of "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + MMX + Miscellaneous + + + + Convert packed signed 16-bit integers from "a" and "b" to packed 8-bit integers using unsigned saturation, and store the results in "dst". + +dst[7:0] := SaturateU8(a[15:0]) +dst[15:8] := SaturateU8(a[31:16]) +dst[23:16] := SaturateU8(a[47:32]) +dst[31:24] := SaturateU8(a[63:48]) +dst[39:32] := SaturateU8(b[15:0]) +dst[47:40] := SaturateU8(b[31:16]) +dst[55:48] := SaturateU8(b[47:32]) +dst[63:56] := SaturateU8(b[63:48]) + + +
mmintrin.h
+ + + Integer + MMX + Swizzle + + + + Unpack and interleave 8-bit integers from the high half of "a" and "b", and store the results in "dst". -LEFT_ROTATE_QWORDS(src, count_src){ - count := count_src modulo 64 - RETURN (src << count) OR (src >> (64 - count)) +DEFINE INTERLEAVE_HIGH_BYTES(src1[63:0], src2[63:0]) { + dst[7:0] := src1[39:32] + dst[15:8] := src2[39:32] + dst[23:16] := src1[47:40] + dst[31:24] := src2[47:40] + dst[39:32] := src1[55:48] + dst[47:40] := src2[55:48] + dst[55:48] := src1[63:56] + dst[63:56] := src2[63:56] + RETURN dst[63:0] } -FOR j := 0 to 3 - i := j*64 - IF k[j] - dst[i+63:i] := LEFT_ROTATE_QWORDS(a[i+63:i], b[i+63:i]) - ELSE - dst[i+63:i] := 0 - FI -ENDFOR -dst[MAX:256] := 0 +dst[63:0] := INTERLEAVE_HIGH_BYTES(a[63:0], b[63:0]) - -
immintrin.h
+ +
mmintrin.h
 - + Integer - AVX512VL - AVX512F - Shift - - - Rotate the bits in each packed 64-bit integer in "a" to the left by the number of bits specified in the corresponding element of "b", and store the results in "dst". + MMX + Swizzle + + + + Unpack and interleave 16-bit integers from the high half of "a" and "b", and store the results in "dst". -LEFT_ROTATE_QWORDS(src, count_src){ - count := count_src modulo 64 - RETURN (src << count) OR (src >> (64 - count)) +DEFINE INTERLEAVE_HIGH_WORDS(src1[63:0], src2[63:0]) { + dst[15:0] := src1[47:32] + dst[31:16] := src2[47:32] + dst[47:32] := src1[63:48] + dst[63:48] := src2[63:48] + RETURN dst[63:0] } -FOR j := 0 to 3 - i := j*64 - dst[i+63:i] := LEFT_ROTATE_QWORDS(a[i+63:i], b[i+63:i]) -ENDFOR -dst[MAX:256] := 0 +dst[63:0] := INTERLEAVE_HIGH_WORDS(a[63:0], b[63:0]) - -
immintrin.h
+ +
mmintrin.h
 - + Integer - AVX512VL - AVX512F - Shift - - - - - Rotate the bits in each packed 64-bit integer in "a" to the left by the number of bits specified in the corresponding element of "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -LEFT_ROTATE_QWORDS(src, count_src){ - count := count_src modulo 64 - RETURN (src << count) OR (src >> (64 - count)) -} -FOR j := 0 to 1 - i := j*64 - IF k[j] - dst[i+63:i] := LEFT_ROTATE_QWORDS(a[i+63:i], b[i+63:i]) - ELSE - dst[i+63:i] := src[i+63:i] - FI -ENDFOR -dst[MAX:128] := 0 + MMX + Swizzle + + + + Unpack and interleave 32-bit integers from the high half of "a" and "b", and store the results in "dst". + +dst[31:0] := a[63:32] +dst[63:32] := b[63:32] - -
immintrin.h
+ +
mmintrin.h
 - + Integer - AVX512VL - AVX512F - Shift - - - - Rotate the bits in each packed 64-bit integer in "a" to the left by the number of bits specified in the corresponding element of "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + MMX + Swizzle + + + + Unpack and interleave 8-bit integers from the low half of "a" and "b", and store the results in "dst". -LEFT_ROTATE_QWORDS(src, count_src){ - count := count_src modulo 64 - RETURN (src << count) OR (src >> (64 - count)) +DEFINE INTERLEAVE_BYTES(src1[63:0], src2[63:0]) { + dst[7:0] := src1[7:0] + dst[15:8] := src2[7:0] + dst[23:16] := src1[15:8] + dst[31:24] := src2[15:8] + dst[39:32] := src1[23:16] + dst[47:40] := src2[23:16] + dst[55:48] := src1[31:24] + dst[63:56] := src2[31:24] + RETURN dst[63:0] } -FOR j := 0 to 1 - i := j*64 - IF k[j] - dst[i+63:i] := LEFT_ROTATE_QWORDS(a[i+63:i], b[i+63:i]) - ELSE - dst[i+63:i] := 0 - FI -ENDFOR -dst[MAX:128] := 0 +dst[63:0] := INTERLEAVE_BYTES(a[63:0], b[63:0]) - -
immintrin.h
+ +
mmintrin.h
 - + Integer - AVX512VL - AVX512F - Shift - - - Rotate the bits in each packed 64-bit integer in "a" to the left by the number of bits specified in the corresponding element of "b", and store the results in "dst". + MMX + Swizzle + + + + Unpack and interleave 16-bit integers from the low half of "a" and "b", and store the results in "dst". -LEFT_ROTATE_QWORDS(src, count_src){ - count := count_src modulo 64 - RETURN (src << count) OR (src >> (64 - count)) +DEFINE INTERLEAVE_WORDS(src1[63:0], src2[63:0]) { + dst[15:0] := src1[15:0] + dst[31:16] := src2[15:0] + dst[47:32] := src1[31:16] + dst[63:48] := src2[31:16] + RETURN dst[63:0] } -FOR j := 0 to 1 - i := j*64 - dst[i+63:i] := LEFT_ROTATE_QWORDS(a[i+63:i], b[i+63:i]) -ENDFOR -dst[MAX:128] := 0 +dst[63:0] := INTERLEAVE_WORDS(a[63:0], b[63:0]) - -
immintrin.h
+ +
mmintrin.h
 - + Integer - AVX512VL - AVX512F - Shift - - - - - Rotate the bits in each packed 32-bit integer in "a" to the right by the number of bits specified in "imm8", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -RIGHT_ROTATE_DWORDS(src, count_src){ - count := count_src modulo 32 - RETURN (src >>count) OR (src << (32 - count)) -} -FOR j := 0 to 7 - i := j*32 - IF k[j] - dst[i+31:i] := RIGHT_ROTATE_DWORDS(a[i+31:i], imm8[7:0]) - ELSE - dst[i+31:i] := src[i+31:i] - FI + MMX + Swizzle + + + + Unpack and interleave 32-bit integers from the low half of "a" and "b", and store the results in "dst". + +dst[31:0] := a[31:0] +dst[63:32] := b[31:0] + + +
mmintrin.h
+ + + MONITOR + General Support + + + + + Arm address monitoring hardware using the address specified in "p". A store to an address within the specified address range triggers the monitoring hardware. Specify optional extensions in "extensions", and optional hints in "hints". + +
pmmintrin.h
+ + + MONITOR + General Support + + + + Hint to the processor that it can enter an implementation-dependent-optimized state while waiting for an event or store operation to the address range specified by MONITOR. + +
pmmintrin.h
+ + + MOVBE + Load + + + Load 16 bits from memory, perform a byte swap operation, and store the result in "dst". + +FOR j := 0 to 1 + i := j*8 + dst[i+7:i] := MEM[ptr+15-i:ptr+8-i] ENDFOR -dst[MAX:256] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Shift - - - - Rotate the bits in each packed 32-bit integer in "a" to the right by the number of bits specified in "imm8", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + MOVBE + Load + + + Load 32 bits from memory, perform a byte swap operation, and store the result in "dst". -RIGHT_ROTATE_DWORDS(src, count_src){ - count := count_src modulo 32 - RETURN (src >>count) OR (src << (32 - count)) -} -FOR j := 0 to 7 - i := j*32 - IF k[j] - dst[i+31:i] := RIGHT_ROTATE_DWORDS(a[i+31:i], imm8[7:0]) - ELSE - dst[i+31:i] := 0 - FI +FOR j := 0 to 3 + i := j*8 + dst[i+7:i] := MEM[ptr+31-i:ptr+24-i] ENDFOR -dst[MAX:256] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Shift - - - Rotate the bits in each packed 32-bit integer in "a" to the right by the number of bits specified in "imm8", and store the results in "dst". + + MOVBE + Load + + + Load 64 bits from memory, perform a byte swap operation, and store the result in "dst". -RIGHT_ROTATE_DWORDS(src, count_src){ - count := count_src modulo 32 - RETURN (src >>count) OR (src << (32 - count)) -} FOR j := 0 to 7 - i := j*32 - dst[i+31:i] := RIGHT_ROTATE_DWORDS(a[i+31:i], imm8[7:0]) + i := j*8 + dst[i+7:i] := MEM[ptr+63-i:ptr+56-i] ENDFOR -dst[MAX:256] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Shift - - - - - Rotate the bits in each packed 32-bit integer in "a" to the right by the number of bits specified in "imm8", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -RIGHT_ROTATE_DWORDS(src, count_src){ - count := count_src modulo 32 - RETURN (src >>count) OR (src << (32 - count)) -} -FOR j := 0 to 3 - i := j*32 - IF k[j] - dst[i+31:i] := RIGHT_ROTATE_DWORDS(a[i+31:i], imm8[7:0]) - ELSE - dst[i+31:i] := src[i+31:i] - FI + + MOVBE + Store + + + + Perform a bit swap operation of the 16 bits in "data", and store the results to memory. + +FOR j := 0 to 1 + i := j*8 + MEM[ptr+i+7:ptr+i] := data[15-i:8-i] ENDFOR -dst[MAX:128] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Shift - - - - Rotate the bits in each packed 32-bit integer in "a" to the right by the number of bits specified in "imm8", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + MOVBE + Store + + + + Perform a bit swap operation of the 32 bits in "data", and store the results to memory. -RIGHT_ROTATE_DWORDS(src, count_src){ - count := count_src modulo 32 - RETURN (src >>count) OR (src << (32 - count)) -} +addr := MEM[ptr] FOR j := 0 to 3 - i := j*32 - IF k[j] - dst[i+31:i] := RIGHT_ROTATE_DWORDS(a[i+31:i], imm8[7:0]) - ELSE - dst[i+31:i] := 0 - FI + i := j*8 + MEM[ptr+i+7:ptr+i] := data[31-i:24-i] ENDFOR -dst[MAX:128] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Shift - - - Rotate the bits in each packed 32-bit integer in "a" to the right by the number of bits specified in "imm8", and store the results in "dst". + + MOVBE + Store + + + + Perform a bit swap operation of the 64 bits in "data", and store the results to memory. -RIGHT_ROTATE_DWORDS(src, count_src){ - count := count_src modulo 32 - RETURN (src >>count) OR (src << (32 - count)) -} -FOR j := 0 to 3 - i := j*32 - dst[i+31:i] := RIGHT_ROTATE_DWORDS(a[i+31:i], imm8[7:0]) +addr := MEM[ptr] +FOR j := 0 to 7 + i := j*8 + MEM[ptr+i+7:ptr+i] := data[63-i:56-i] ENDFOR -dst[MAX:128] := 0 - +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Shift - - - - - Rotate the bits in each packed 64-bit integer in "a" to the right by the number of bits specified in "imm8", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -RIGHT_ROTATE_QWORDS(src, count_src){ - count := count_src modulo 64 - RETURN (src >> count) OR (src << (64 - count)) -} -FOR j := 0 to 3 - i := j*64 - IF k[j] - dst[i+63:i] := RIGHT_ROTATE_QWORDS(a[i+63:i], imm8[7:0]) - ELSE - dst[i+63:i] := src[i+63:i] - FI -ENDFOR -dst[MAX:256] := 0 + + MOVDIR64B + Store + + + + Move 64-byte (512-bit) value using direct store from source memory address "src" to destination memory address "dst". + +MEM[dst+511:dst] := MEM[src+511:src] - +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Shift - - - - Rotate the bits in each packed 64-bit integer in "a" to the right by the number of bits specified in "imm8", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + MOVDIRI + Store + + + + Store 64-bit integer from "val" into memory using direct store. -RIGHT_ROTATE_QWORDS(src, count_src){ - count := count_src modulo 64 - RETURN (src >> count) OR (src << (64 - count)) -} -FOR j := 0 to 3 - i := j*64 - IF k[j] - dst[i+63:i] := RIGHT_ROTATE_QWORDS(a[i+63:i], imm8[7:0]) - ELSE - dst[i+63:i] := 0 - FI -ENDFOR -dst[MAX:256] := 0 +MEM[dst+63:dst] := val[63:0] - +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Shift - - - Rotate the bits in each packed 64-bit integer in "a" to the right by the number of bits specified in "imm8", and store the results in "dst". + + MOVDIRI + Store + + + + Store 32-bit integer from "val" into memory using direct store. -RIGHT_ROTATE_QWORDS(src, count_src){ - count := count_src modulo 64 - RETURN (src >> count) OR (src << (64 - count)) -} -FOR j := 0 to 3 - i := j*64 - dst[i+63:i] := RIGHT_ROTATE_QWORDS(a[i+63:i], imm8[7:0]) -ENDFOR -dst[MAX:256] := 0 +MEM[dst+31:dst] := val[31:0] - +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Shift - - - - - Rotate the bits in each packed 64-bit integer in "a" to the right by the number of bits specified in "imm8", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -RIGHT_ROTATE_QWORDS(src, count_src){ - count := count_src modulo 64 - RETURN (src >> count) OR (src << (64 - count)) -} -FOR j := 0 to 1 - i := j*64 - IF k[j] - dst[i+63:i] := RIGHT_ROTATE_QWORDS(a[i+63:i], imm8[7:0]) - ELSE - dst[i+63:i] := src[i+63:i] - FI -ENDFOR -dst[MAX:128] := 0 + + MPX + Miscellaneous + + + + Make a pointer with the value of "srcmem" and bounds set to ["srcmem", "srcmem" + "size" - 1], and store the result in "dst". + dst := srcmem +dst.LB := srcmem.LB +dst.UB := srcmem + size - 1 - +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Shift - - - - Rotate the bits in each packed 64-bit integer in "a" to the right by the number of bits specified in "imm8", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - -RIGHT_ROTATE_QWORDS(src, count_src){ - count := count_src modulo 64 - RETURN (src >> count) OR (src << (64 - count)) -} -FOR j := 0 to 1 - i := j*64 - IF k[j] - dst[i+63:i] := RIGHT_ROTATE_QWORDS(a[i+63:i], imm8[7:0]) - ELSE - dst[i+63:i] := 0 - FI -ENDFOR -dst[MAX:128] := 0 + + MPX + Miscellaneous + + + + + Narrow the bounds for pointer "q" to the intersection of the bounds of "r" and the bounds ["q", "q" + "size" - 1], and store the result in "dst". + dst := q +IF r.LB > (q + size - 1) OR r.UB < q + dst.LB := 1 + dst.UB := 0 +ELSE + dst.LB := MAX(r.LB, q) + dst.UB := MIN(r.UB, (q + size - 1)) +FI -
immintrin.h
 - - Integer - AVX512VL - AVX512F - Shift - - - Rotate the bits in each packed 64-bit integer in "a" to the right by the number of bits specified in "imm8", and store the results in "dst". - -RIGHT_ROTATE_QWORDS(src, count_src){ - count := count_src modulo 64 - RETURN (src >> count) OR (src << (64 - count)) -} -FOR j := 0 to 1 - i := j*64 - dst[i+63:i] := RIGHT_ROTATE_QWORDS(a[i+63:i], imm8[7:0]) -ENDFOR -dst[MAX:128] := 0 + + MPX + Miscellaneous + + + + Make a pointer with the value of "q" and bounds set to the bounds of "r" (e.g. copy the bounds of "r" to pointer "q"), and store the result in "dst". + dst := q +dst.LB := r.LB +dst.UB := r.UB -
immintrin.h
 - - Integer - AVX512VL - AVX512F - Shift - - - - - Rotate the bits in each packed 32-bit integer in "a" to the right by the number of bits specified in the corresponding element of "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -RIGHT_ROTATE_DWORDS(src, count_src){ - count := count_src modulo 32 - RETURN (src >>count) OR (src << (32 - count)) -} -FOR j := 0 to 7 - i := j*32 - IF k[j] - dst[i+31:i] := RIGHT_ROTATE_DWORDS(a[i+31:i], b[i+31:i]) - ELSE - dst[i+31:i] := src[i+31:i] - FI -ENDFOR -dst[MAX:256] := 0 + + MPX + Miscellaneous + + + Make a pointer with the value of "q" and open bounds, which allow the pointer to access the entire virtual address space, and store the result in "dst". + dst := q +dst.LB := 0 +dst.UB := 0 + +
immintrin.h
+ + + MPX + Miscellaneous + + + + Stores the bounds of "ptr_val" pointer in memory at address "ptr_addr". + MEM[ptr_addr].LB := ptr_val.LB +MEM[ptr_addr].UB := ptr_val.UB + + +
immintrin.h
+ + + MPX + Miscellaneous + + + Checks if "q" is within its lower bound, and throws a #BR if not. + IF q < q.LB + #BR +FI + + +
immintrin.h
+ + + MPX + Miscellaneous + + + Checks if "q" is within its upper bound, and throws a #BR if not. + IF q > q.UB + #BR +FI + + + +
immintrin.h
+ + + MPX + Miscellaneous + + + + Checks if ["q", "q" + "size" - 1] is within the lower and upper bounds of "q" and throws a #BR if not. + IF (q + size - 1) < q.LB OR (q + size - 1) > q.UB + #BR +FI + + + +
immintrin.h
+ + + MPX + Miscellaneous + + + Return the lower bound of "q". + dst := q.LB + +
immintrin.h
+ + + MPX + Miscellaneous + + + Return the upper bound of "q". + dst := q.UB -
immintrin.h
 - + Integer - AVX512VL - AVX512F - Shift - - - - Rotate the bits in each packed 32-bit integer in "a" to the right by the number of bits specified in the corresponding element of "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Bit Manipulation + + + Set "dst" to the index of the lowest set bit in 32-bit integer "a". If no bits are set in "a" then "dst" is undefined. -RIGHT_ROTATE_DWORDS(src, count_src){ - count := count_src modulo 32 - RETURN (src >>count) OR (src << (32 - count)) -} -FOR j := 0 to 7 - i := j*32 - IF k[j] - dst[i+31:i] := RIGHT_ROTATE_DWORDS(a[i+31:i], b[i+31:i]) - ELSE - dst[i+31:i] := 0 - FI -ENDFOR -dst[MAX:256] := 0 +tmp := 0 +IF a == 0 + // dst is undefined +ELSE + DO WHILE ((tmp < 32) AND a[tmp] == 0) + tmp := tmp + 1 + OD +FI +dst := tmp - +
immintrin.h
 - + Integer - AVX512VL - AVX512F - Shift - - - Rotate the bits in each packed 32-bit integer in "a" to the right by the number of bits specified in the corresponding element of "b", and store the results in "dst". + Bit Manipulation + + + Set "dst" to the index of the highest set bit in 32-bit integer "a". If no bits are set in "a" then "dst" is undefined. -RIGHT_ROTATE_DWORDS(src, count_src){ - count := count_src modulo 32 - RETURN (src >>count) OR (src << (32 - count)) -} -FOR j := 0 to 7 - i := j*32 - dst[i+31:i] := RIGHT_ROTATE_DWORDS(a[i+31:i], b[i+31:i]) -ENDFOR -dst[MAX:256] := 0 +tmp := 31 +IF a == 0 + // dst is undefined +ELSE + DO WHILE ((tmp > 0) AND a[tmp] == 0) + tmp := tmp - 1 + OD +FI +dst := tmp - +
immintrin.h
 - + Integer - AVX512VL - AVX512F - Shift - - - - - Rotate the bits in each packed 32-bit integer in "a" to the right by the number of bits specified in the corresponding element of "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -RIGHT_ROTATE_DWORDS(src, count_src){ - count := count_src modulo 32 - RETURN (src >>count) OR (src << (32 - count)) -} -FOR j := 0 to 3 - i := j*32 - IF k[j] - dst[i+31:i] := RIGHT_ROTATE_DWORDS(a[i+31:i], b[i+31:i]) - ELSE - dst[i+31:i] := src[i+31:i] - FI -ENDFOR -dst[MAX:128] := 0 + Flag + Bit Manipulation + + + + Set "index" to the index of the lowest set bit in 32-bit integer "mask". If no bits are set in "a", then "index" is undefined and "dst" is set to 0, otherwise "dst" is set to 1. + +tmp := 0 +IF a == 0 + // MEM[index+31:index] is undefined + dst := 0 +ELSE + DO WHILE ((tmp < 32) AND a[tmp] == 0) + tmp := tmp + 1 + OD + MEM[index+31:index] := tmp + dst := (tmp == 31) ? 0 : 1 +FI - +
immintrin.h
 - + Integer - AVX512VL - AVX512F - Shift - - - - Rotate the bits in each packed 32-bit integer in "a" to the right by the number of bits specified in the corresponding element of "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Flag + Bit Manipulation + + + + Set "index" to the index of the highest set bit in 32-bit integer "mask". If no bits are set in "a", then "index" is undefined and "dst" is set to 0, otherwise "dst" is set to 1. -RIGHT_ROTATE_DWORDS(src, count_src){ - count := count_src modulo 32 - RETURN (src >>count) OR (src << (32 - count)) -} -FOR j := 0 to 3 - i := j*32 - IF k[j] - dst[i+31:i] := RIGHT_ROTATE_DWORDS(a[i+31:i], b[i+31:i]) - ELSE - dst[i+31:i] := 0 - FI -ENDFOR -dst[MAX:128] := 0 +tmp := 31 +IF a == 0 + // MEM[index+31:index] is undefined + dst := 0 +ELSE + DO WHILE ((tmp > 0) AND a[tmp] == 0) + tmp := tmp - 1 + OD + MEM[index+31:index] := tmp + dst := (tmp == 0) ? 0 : 1 +FI - +
immintrin.h
 - + Integer - AVX512VL - AVX512F - Shift - - - Rotate the bits in each packed 32-bit integer in "a" to the right by the number of bits specified in the corresponding element of "b", and store the results in "dst". + Flag + Bit Manipulation + + + + Set "index" to the index of the lowest set bit in 32-bit integer "mask". If no bits are set in "a", then "index" is undefined and "dst" is set to 0, otherwise "dst" is set to 1. -RIGHT_ROTATE_DWORDS(src, count_src){ - count := count_src modulo 32 - RETURN (src >>count) OR (src << (32 - count)) -} -FOR j := 0 to 3 - i := j*32 - dst[i+31:i] := RIGHT_ROTATE_DWORDS(a[i+31:i], b[i+31:i]) -ENDFOR -dst[MAX:128] := 0 +tmp := 0 +IF a == 0 + // MEM[index+31:index] is undefined + dst := 0 +ELSE + DO WHILE ((tmp < 64) AND a[tmp] == 0) + tmp := tmp + 1 + OD + MEM[index+31:index] := tmp + dst := (tmp == 63) ? 0 : 1 +FI - +
immintrin.h
 - + Integer - AVX512VL - AVX512F - Shift - - - - - Rotate the bits in each packed 64-bit integer in "a" to the right by the number of bits specified in the corresponding element of "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -RIGHT_ROTATE_QWORDS(src, count_src){ - count := count_src modulo 64 - RETURN (src >> count) OR (src << (64 - count)) -} -FOR j := 0 to 3 - i := j*64 - IF k[j] - dst[i+63:i] := RIGHT_ROTATE_QWORDS(a[i+63:i], b[i+63:i]) - ELSE - dst[i+63:i] := src[i+63:i] - FI -ENDFOR -dst[MAX:256] := 0 + Flag + Bit Manipulation + + + + Set "index" to the index of the highest set bit in 32-bit integer "mask". If no bits are set in "a", then "index" is undefined and "dst" is set to 0, otherwise "dst" is set to 1. + +tmp := 63 +IF a == 0 + // MEM[index+31:index] is undefined + dst := 0 +ELSE + DO WHILE ((tmp > 0) AND a[tmp] == 0) + tmp := tmp - 1 + OD + MEM[index+31:index] := tmp + dst := (tmp == 0) ? 0 : 1 +FI - +
immintrin.h
 - + Integer - AVX512VL - AVX512F - Shift - - - - Rotate the bits in each packed 64-bit integer in "a" to the right by the number of bits specified in the corresponding element of "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Flag + Bit Manipulation + + + + Return the bit at index "b" of 32-bit integer "a". -RIGHT_ROTATE_QWORDS(src, count_src){ - count := count_src modulo 64 - RETURN (src >> count) OR (src << (64 - count)) -} -FOR j := 0 to 3 - i := j*64 - IF k[j] - dst[i+63:i] := RIGHT_ROTATE_QWORDS(a[i+63:i], b[i+63:i]) - ELSE - dst[i+63:i] := 0 - FI -ENDFOR -dst[MAX:256] := 0 +addr := a + ZeroExtend64(b) +dst[0] := MEM[addr] - +
immintrin.h
 - + Integer - AVX512VL - AVX512F - Shift - - - Rotate the bits in each packed 64-bit integer in "a" to the right by the number of bits specified in the corresponding element of "b", and store the results in "dst". + Flag + Bit Manipulation + + + + Return the bit at index "b" of 32-bit integer "a", and set that bit to its complement. -RIGHT_ROTATE_QWORDS(src, count_src){ - count := count_src modulo 64 - RETURN (src >> count) OR (src << (64 - count)) -} -FOR j := 0 to 3 - i := j*64 - dst[i+63:i] := RIGHT_ROTATE_QWORDS(a[i+63:i], b[i+63:i]) -ENDFOR -dst[MAX:256] := 0 +addr := a + ZeroExtend64(b) +dst[0] := MEM[addr] +MEM[addr] := ~dst[0] - +
immintrin.h
 - + Integer - AVX512VL - AVX512F - Shift - - - - - Rotate the bits in each packed 64-bit integer in "a" to the right by the number of bits specified in the corresponding element of "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -RIGHT_ROTATE_QWORDS(src, count_src){ - count := count_src modulo 64 - RETURN (src >> count) OR (src << (64 - count)) -} -FOR j := 0 to 1 - i := j*64 - IF k[j] - dst[i+63:i] := RIGHT_ROTATE_QWORDS(a[i+63:i], b[i+63:i]) - ELSE - dst[i+63:i] := src[i+63:i] - FI -ENDFOR -dst[MAX:128] := 0 + Flag + Bit Manipulation + + + + Return the bit at index "b" of 32-bit integer "a", and set that bit to zero. + +addr := a + ZeroExtend64(b) +dst[0] := MEM[addr] +MEM[addr] := 0 - +
immintrin.h
 - + Integer - AVX512VL - AVX512F - Shift - - - - Rotate the bits in each packed 64-bit integer in "a" to the right by the number of bits specified in the corresponding element of "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Flag + Bit Manipulation + + + + Return the bit at index "b" of 32-bit integer "a", and set that bit to one. -RIGHT_ROTATE_QWORDS(src, count_src){ - count := count_src modulo 64 - RETURN (src >> count) OR (src << (64 - count)) -} -FOR j := 0 to 1 - i := j*64 - IF k[j] - dst[i+63:i] := RIGHT_ROTATE_QWORDS(a[i+63:i], b[i+63:i]) - ELSE - dst[i+63:i] := 0 - FI -ENDFOR -dst[MAX:128] := 0 +addr := a + ZeroExtend64(b) +dst[0] := MEM[addr] +MEM[addr] := 1 - +
immintrin.h
 - + Integer - AVX512VL - AVX512F - Shift - - - Rotate the bits in each packed 64-bit integer in "a" to the right by the number of bits specified in the corresponding element of "b", and store the results in "dst". + Flag + Bit Manipulation + + + + Return the bit at index "b" of 64-bit integer "a". -RIGHT_ROTATE_QWORDS(src, count_src){ - count := count_src modulo 64 - RETURN (src >> count) OR (src << (64 - count)) -} -FOR j := 0 to 1 - i := j*64 - dst[i+63:i] := RIGHT_ROTATE_QWORDS(a[i+63:i], b[i+63:i]) -ENDFOR -dst[MAX:128] := 0 +addr := a + b +dst[0] := MEM[addr] - +
immintrin.h
 - + Integer - AVX512BW - Miscellaneous - - - Compute the absolute differences of packed unsigned 8-bit integers in "a" and "b", then horizontally sum each consecutive 8 differences to produce eight unsigned 16-bit integers, and pack these unsigned 16-bit integers in the low 16 bits of 64-bit elements in "dst". + Flag + Bit Manipulation + + + + Return the bit at index "b" of 64-bit integer "a", and set that bit to its complement. -FOR j := 0 to 63 - i := j*8 - tmp[i+7:i] := ABS(a[i+7:i] - b[i+7:i]) -ENDFOR -FOR j := 0 to 7 - i := j*64 - dst[i+15:i] := tmp[i+7:i] + tmp[i+15:i+8] + tmp[i+23:i+16] + tmp[i+31:i+24] + tmp[i+39:i+32] + tmp[i+47:i+40] + tmp[i+55:i+48] + tmp[i+63:i+56] - dst[i+63:i+16] := 0 -ENDFOR -dst[MAX:512] := 0 +addr := a + b +dst[0] := MEM[addr] +MEM[addr] := ~dst[0] - +
immintrin.h
 - + Integer - AVX512VL - AVX512F - Store - - - - - Scatter 32-bit integers from "a" into memory using 32-bit indices. 32-bit elements are stored at addresses starting at "base_addr" and offset by each 32-bit element in "vindex" (each index is scaled by the factor in "scale"). "scale" should be 1, 2, 4 or 8. + Flag + Bit Manipulation + + + + Return the bit at index "b" of 64-bit integer "a", and set that bit to zero. -FOR j := 0 to 7 - i := j*32 - MEM[base_addr + SignExtend(vindex[i+31:i])*scale] := a[i+31:i] -ENDFOR +addr := a + b +dst[0] := MEM[addr] +MEM[addr] := 0 - +
immintrin.h
 - + Integer - AVX512VL - AVX512F - Store - - - - - - Scatter 32-bit integers from "a" into memory using 32-bit indices. 32-bit elements are stored at addresses starting at "base_addr" and offset by each 32-bit element in "vindex" (each index is scaled by the factor in "scale") subject to mask "k" (elements are not stored when the corresponding mask bit is not set). "scale" should be 1, 2, 4 or 8. - + Flag + Bit Manipulation + + + + Return the bit at index "b" of 64-bit integer "a", and set that bit to one. -FOR j := 0 to 7 - i := j*32 - IF k[j] - MEM[base_addr + SignExtend(vindex[i+31:i])*scale] := a[i+31:i] - k[j] := 0 - FI -ENDFOR -k[MAX:8] := 0 +addr := a + b +dst[0] := MEM[addr] +MEM[addr] := 1 - +
immintrin.h
 - + Integer - AVX512VL - AVX512F - Store - - - - - Scatter 32-bit integers from "a" into memory using 32-bit indices. 32-bit elements are stored at addresses starting at "base_addr" and offset by each 32-bit element in "vindex" (each index is scaled by the factor in "scale"). "scale" should be 1, 2, 4 or 8. + Bit Manipulation + + + Reverse the byte order of 32-bit integer "a", and store the result in "dst". This intrinsic is provided for conversion between little and big endian values. -FOR j := 0 to 3 - i := j*32 - MEM[base_addr + SignExtend(vindex[i+31:i])*scale] := a[i+31:i] -ENDFOR +dst[7:0] := a[31:24] +dst[15:8] := a[23:16] +dst[23:16] := a[15:8] +dst[31:24] := a[7:0] - +
immintrin.h
 - + Integer - AVX512VL - AVX512F - Store - - - - - - Scatter 32-bit integers from "a" into memory using 32-bit indices. 32-bit elements are stored at addresses starting at "base_addr" and offset by each 32-bit element in "vindex" (each index is scaled by the factor in "scale") subject to mask "k" (elements are not stored when the corresponding mask bit is not set). "scale" should be 1, 2, 4 or 8. - + Bit Manipulation + + + Reverse the byte order of 64-bit integer "a", and store the result in "dst". This intrinsic is provided for conversion between little and big endian values. -FOR j := 0 to 3 - i := j*32 - IF k[j] - MEM[base_addr + SignExtend(vindex[i+31:i])*scale] := a[i+31:i] - k[j] := 0 - FI -ENDFOR -k[MAX:4] := 0 +dst[7:0] := a[63:56] +dst[15:8] := a[55:48] +dst[23:16] := a[47:40] +dst[31:24] := a[39:32] +dst[39:32] := a[31:24] +dst[47:40] := a[23:16] +dst[55:48] := a[15:8] +dst[63:56] := a[7:0] - + +
immintrin.h
+ + + Floating Point + Integer + Cast + + + Cast from type float to type unsigned __int32 without conversion. + This intrinsic is only used for compilation and does not generate any instructions, thus it has zero latency. +
immintrin.h
+ + + Floating Point + Integer + Cast + + + Cast from type double to type unsigned __int64 without conversion. + This intrinsic is only used for compilation and does not generate any instructions, thus it has zero latency.
immintrin.h
 - + + Floating Point Integer - AVX512VL - AVX512F - Store - - - - - Scatter 64-bit integers from "a" into memory using 32-bit indices. 64-bit elements are stored at addresses starting at "base_addr" and offset by each 32-bit element in "vindex" (each index is scaled by the factor in "scale"). "scale" should be 1, 2, 4 or 8. - - -FOR j := 0 to 3 - i := j*64 - l := j*32 - MEM[base_addr + SignExtend(vindex[l+31:l])*scale] := a[i+63:i] -ENDFOR - - + Cast + + + Cast from type unsigned __int32 to type float without conversion. + This intrinsic is only used for compilation and does not generate any instructions, thus it has zero latency.
immintrin.h
 - + + Floating Point Integer - AVX512VL - AVX512F - Store - - - - - - Scatter 64-bit integers from "a" into memory using 32-bit indices. 64-bit elements are stored at addresses starting at "base_addr" and offset by each 32-bit element in "vindex" (each index is scaled by the factor in "scale") subject to mask "k" (elements are not stored when the corresponding mask bit is not set). "scale" should be 1, 2, 4 or 8. - - -FOR j := 0 to 3 - i := j*64 - l := j*32 - IF k[j] - MEM[base_addr + SignExtend(vindex[l+31:l])*scale] := a[i+63:i] - k[j] := 0 - FI -ENDFOR -k[MAX:4] := 0 - - + Cast + + + Cast from type unsigned __int64 to type double without conversion. + This intrinsic is only used for compilation and does not generate any instructions, thus it has zero latency.
immintrin.h
 - + Integer - AVX512VL - AVX512F - Store - - - - - Scatter 64-bit integers from "a" into memory using 32-bit indices. 64-bit elements are stored at addresses starting at "base_addr" and offset by each 32-bit element in "vindex" (each index is scaled by the factor in "scale"). "scale" should be 1, 2, 4 or 8. - - -FOR j := 0 to 1 - i := j*64 - l := j*32 - MEM[base_addr + SignExtend(vindex[l+31:l])*scale] := a[i+63:i] -ENDFOR + Shift + + + + Shift the bits of unsigned long integer "a" left by the number of bits specified in "shift", rotating the most-significant bit to the least-significant bit location, and store the unsigned result in "dst". + // size := 32 or 64 +dst := a +count := shift AND (size - 1) +DO WHILE (count > 0) + tmp[0] := dst[size - 1] + dst := (dst << 1) OR tmp[0] + count := count - 1 +OD - + +
immintrin.h
 - + Integer - AVX512VL - AVX512F - Store - - - - - - Scatter 64-bit integers from "a" into memory using 32-bit indices. 64-bit elements are stored at addresses starting at "base_addr" and offset by each 32-bit element in "vindex" (each index is scaled by the factor in "scale") subject to mask "k" (elements are not stored when the corresponding mask bit is not set). "scale" should be 1, 2, 4 or 8. - - -FOR j := 0 to 1 - i := j*64 - l := j*32 - IF k[j] - MEM[base_addr + SignExtend(vindex[l+31:l])*scale] := a[i+63:i] - k[j] := 0 - FI -ENDFOR -k[MAX:2] := 0 + Shift + + + + Shift the bits of unsigned long integer "a" right by the number of bits specified in "shift", rotating the least-significant bit to the most-significant bit location, and store the unsigned result in "dst". + // size := 32 or 64 +dst := a +count := shift AND (size - 1) +DO WHILE (count > 0) + tmp[size - 1] := dst[0] + dst := (dst >> 1) OR tmp[size - 1] + count := count - 1 +OD - +
immintrin.h
 - - Integer - AVX512VL - AVX512F - Store - - - - - Scatter 32-bit integers from "a" into memory using 64-bit indices. 32-bit elements are stored at addresses starting at "base_addr" and offset by each 64-bit element in "vindex" (each index is scaled by the factor in "scale"). "scale" should be 1, 2, 4 or 8. - + + General Support + + + Treat the processor-specific feature(s) specified in "a" as available. Multiple features may be OR'd together. See the valid feature flags below: -FOR j := 0 to 3 - i := j*32 - l := j*64 - MEM[base_addr + SignExtend(vindex[l+63:l])*scale] := a[i+31:i] -ENDFOR +_FEATURE_GENERIC_IA32 +_FEATURE_FPU +_FEATURE_CMOV +_FEATURE_MMX +_FEATURE_FXSAVE +_FEATURE_SSE +_FEATURE_SSE2 +_FEATURE_SSE3 +_FEATURE_SSSE3 +_FEATURE_SSE4_1 +_FEATURE_SSE4_2 +_FEATURE_MOVBE +_FEATURE_POPCNT +_FEATURE_PCLMULQDQ +_FEATURE_AES +_FEATURE_F16C +_FEATURE_AVX +_FEATURE_RDRND +_FEATURE_FMA +_FEATURE_BMI +_FEATURE_LZCNT +_FEATURE_HLE +_FEATURE_RTM +_FEATURE_AVX2 +_FEATURE_KNCNI +_FEATURE_AVX512F +_FEATURE_ADX +_FEATURE_RDSEED +_FEATURE_AVX512ER +_FEATURE_AVX512PF +_FEATURE_AVX512CD +_FEATURE_SHA +_FEATURE_MPX +_FEATURE_AVX512BW +_FEATURE_AVX512VL +_FEATURE_AVX512VBMI +_FEATURE_AVX512_4FMAPS +_FEATURE_AVX512_4VNNIW +_FEATURE_AVX512_VPOPCNTDQ +_FEATURE_AVX512_BITALG +_FEATURE_AVX512_VBMI2 +_FEATURE_GFNI +_FEATURE_VAES +_FEATURE_VPCLMULQDQ +_FEATURE_AVX512_VNNI +_FEATURE_CLWB +_FEATURE_RDPID +_FEATURE_IBT +_FEATURE_SHSTK +_FEATURE_SGX +_FEATURE_WBNOINVD +_FEATURE_PCONFIG +_FEATURE_AXV512_4VNNIB +_FEATURE_AXV512_4FMAPH +_FEATURE_AXV512_BITALG2 +_FEATURE_AXV512_VP2INTERSECT + +
immintrin.h
+ + + General Support + + + Dynamically query the processor to determine if the processor-specific feature(s) specified in "a" are available, and return true or false (1 or 0) if the set of features is available. Multiple features may be OR'd together. This intrinsic does not check the processor vendor. See the valid feature flags below: + +_FEATURE_GENERIC_IA32 +_FEATURE_FPU +_FEATURE_CMOV +_FEATURE_MMX +_FEATURE_FXSAVE +_FEATURE_SSE +_FEATURE_SSE2 +_FEATURE_SSE3 +_FEATURE_SSSE3 +_FEATURE_SSE4_1 +_FEATURE_SSE4_2 +_FEATURE_MOVBE +_FEATURE_POPCNT +_FEATURE_PCLMULQDQ +_FEATURE_AES +_FEATURE_F16C +_FEATURE_AVX +_FEATURE_RDRND +_FEATURE_FMA +_FEATURE_BMI +_FEATURE_LZCNT +_FEATURE_HLE +_FEATURE_RTM +_FEATURE_AVX2 +_FEATURE_KNCNI +_FEATURE_AVX512F +_FEATURE_ADX +_FEATURE_RDSEED +_FEATURE_AVX512ER +_FEATURE_AVX512PF +_FEATURE_AVX512CD +_FEATURE_SHA +_FEATURE_MPX +_FEATURE_AVX512BW +_FEATURE_AVX512VL +_FEATURE_AVX512VBMI +_FEATURE_AVX512_4FMAPS +_FEATURE_AVX512_4VNNIW +_FEATURE_AVX512_VPOPCNTDQ +_FEATURE_AVX512_BITALG +_FEATURE_AVX512_VBMI2 +_FEATURE_GFNI +_FEATURE_VAES +_FEATURE_VPCLMULQDQ +_FEATURE_AVX512_VNNI +_FEATURE_CLWB +_FEATURE_RDPID +_FEATURE_IBT +_FEATURE_SHSTK +_FEATURE_SGX +_FEATURE_WBNOINVD +_FEATURE_PCONFIG +_FEATURE_AXV512_4VNNIB +_FEATURE_AXV512_4FMAPH +_FEATURE_AXV512_BITALG2 +_FEATURE_AXV512_VP2INTERSECT + +
immintrin.h
+ + + General Support + + + Read the Performance Monitor Counter (PMC) specified by "a", and store up to 64-bits in "dst". The width of performance counters is implementation specific. + dst[63:0] := ReadPMC(a) - +
immintrin.h
 - + Integer - AVX512VL - AVX512F - Store - - - - - - Scatter 32-bit integers from "a" into memory using 64-bit indices. 32-bit elements are stored at addresses starting at "base_addr" and offset by each 64-bit element in "vindex" (each index is scaled by the factor in "scale") subject to mask "k" (elements are not stored when the corresponding mask bit is not set). "scale" should be 1, 2, 4 or 8. - + Shift + + + + Shift the bits of unsigned 32-bit integer "a" left by the number of bits specified in "shift", rotating the most-significant bit to the least-significant bit location, and store the unsigned result in "dst". -FOR j := 0 to 3 - i := j*32 - l := j*64 - IF k[j] - MEM[base_addr + SignExtend(vindex[l+63:l])*scale] := a[i+31:i] - k[j] := 0 - FI -ENDFOR -k[MAX:4] := 0 +dst := a +count := shift AND 31 +DO WHILE (count > 0) + tmp[0] := dst[31] + dst := (dst << 1) OR tmp[0] + count := count - 1 +OD - +
immintrin.h
 - + Integer - AVX512VL - AVX512F - Store - - - - - Scatter 32-bit integers from "a" into memory using 64-bit indices. 32-bit elements are stored at addresses starting at "base_addr" and offset by each 64-bit element in "vindex" (each index is scaled by the factor in "scale"). "scale" should be 1, 2, 4 or 8. - + Shift + + + + Shift the bits of unsigned 32-bit integer "a" right by the number of bits specified in "shift", rotating the least-significant bit to the most-significant bit location, and store the unsigned result in "dst". -FOR j := 0 to 1 - i := j*32 - l := j*64 - MEM[base_addr + SignExtend(vindex[l+63:l])*scale] := a[i+31:i] -ENDFOR +dst := a +count := shift AND 31 +DO WHILE (count > 0) + tmp[31] := dst[0] + dst := (dst >> 1) OR tmp + count := count - 1 +OD - +
immintrin.h
 - + Integer - AVX512VL - AVX512F - Store - - - - - - Scatter 32-bit integers from "a" into memory using 64-bit indices. 32-bit elements are stored at addresses starting at "base_addr" and offset by each 64-bit element in "vindex" (each index is scaled by the factor in "scale") subject to mask "k" (elements are not stored when the corresponding mask bit is not set). "scale" should be 1, 2, 4 or 8. - + Shift + + + + Shift the bits of unsigned 16-bit integer "a" left by the number of bits specified in "shift", rotating the most-significant bit to the least-significant bit location, and store the unsigned result in "dst". -FOR j := 0 to 1 - i := j*32 - l := j*64 - IF k[j] - MEM[base_addr + SignExtend(vindex[l+63:l])*scale] := a[i+31:i] - k[j] := 0 - FI -ENDFOR -k[MAX:2] := 0 +dst := a +count := shift AND 15 +DO WHILE (count > 0) + tmp[0] := dst[15] + dst := (dst << 1) OR tmp[0] + count := count - 1 +OD - +
immintrin.h
 - + Integer - AVX512VL - AVX512F - Store - - - - - Scatter 64-bit integers from "a" into memory using 64-bit indices. 64-bit elements are stored at addresses starting at "base_addr" and offset by each 64-bit element in "vindex" (each index is scaled by the factor in "scale"). "scale" should be 1, 2, 4 or 8. - + Shift + + + + Shift the bits of unsigned 16-bit integer "a" right by the number of bits specified in "shift", rotating the least-significant bit to the most-significant bit location, and store the unsigned result in "dst". -FOR j := 0 to 3 - i := j*64 - MEM[base_addr + SignExtend(vindex[i+63:i])*scale] := a[i+63:i] -ENDFOR +dst := a +count := shift AND 15 +DO WHILE (count > 0) + tmp[15] := dst[0] + dst := (dst >> 1) OR tmp + count := count - 1 +OD - +
immintrin.h
 - + Integer - AVX512VL - AVX512F - Store - - - - - - Scatter 64-bit integers from "a" into memory using 64-bit indices. 64-bit elements are stored at addresses starting at "base_addr" and offset by each 64-bit element in "vindex" (each index is scaled by the factor in "scale") subject to mask "k" (elements are not stored when the corresponding mask bit is not set). "scale" should be 1, 2, 4 or 8. - + Shift + + + + Shift the bits of unsigned 64-bit integer "a" left by the number of bits specified in "shift", rotating the most-significant bit to the least-significant bit location, and store the unsigned result in "dst". -FOR j := 0 to 3 - i := j*64 - IF k[j] - MEM[base_addr + SignExtend(vindex[i+63:i])*scale] := a[i+63:i] - k[j] := 0 - FI -ENDFOR -k[MAX:4] := 0 +dst := a +count := shift AND 63 +DO WHILE (count > 0) + tmp[0] := dst[63] + dst := (dst << 1) OR tmp[0] + count := count - 1 +OD - +
immintrin.h
 - + Integer - AVX512VL - AVX512F - Store - - - - - Scatter 64-bit integers from "a" into memory using 64-bit indices. 64-bit elements are stored at addresses starting at "base_addr" and offset by each 64-bit element in "vindex" (each index is scaled by the factor in "scale"). "scale" should be 1, 2, 4 or 8. - + Shift + + + + Shift the bits of unsigned 64-bit integer "a" right by the number of bits specified in "shift", rotating the least-significant bit to the most-significant bit location, and store the unsigned result in "dst". -FOR j := 0 to 1 - i := j*64 - MEM[base_addr + SignExtend(vindex[i+63:i])*scale] := a[i+63:i] -ENDFOR +dst := a +count := shift AND 63 +DO WHILE (count > 0) + tmp[63] := dst[0] + dst := (dst >> 1) OR tmp[63] + count := count - 1 +OD - +
immintrin.h
 - + Integer - AVX512VL - AVX512F - Store - - - - - - Scatter 64-bit integers from "a" into memory using 64-bit indices. 64-bit elements are stored at addresses starting at "base_addr" and offset by each 64-bit element in "vindex" (each index is scaled by the factor in "scale") subject to mask "k" (elements are not stored when the corresponding mask bit is not set). "scale" should be 1, 2, 4 or 8. - + Flag + Arithmetic + + + + + + Add unsigned 32-bit integers "a" and "b" with unsigned 8-bit carry-in "c_in" (carry flag), and store the unsigned 32-bit result in "out", and the carry-out in "dst" (carry or overflow flag). -FOR j := 0 to 1 - i := j*64 - IF k[j] - MEM[base_addr + SignExtend(vindex[i+63:i])*scale] := a[i+63:i] - k[j] := 0 - FI -ENDFOR -k[MAX:2] := 0 +tmp[32:0] := a[31:0] + b[31:0] + (c_in > 0 ? 1 : 0) +MEM[out+31:out] := tmp[31:0] +dst[0] := tmp[32] +dst[7:1] := 0 - +
immintrin.h
 - + Integer - AVX512VL - AVX512BW - Miscellaneous - - - - - Shuffle packed 8-bit integers in "a" according to shuffle control mask in the corresponding 8-bit element of "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Flag + Arithmetic + + + + + + Add unsigned 64-bit integers "a" and "b" with unsigned 8-bit carry-in "c_in" (carry flag), and store the unsigned 64-bit result in "out", and the carry-out in "dst" (carry or overflow flag). -FOR j := 0 to 31 - i := j*8 - IF k[j] - IF b[i+7] == 1 - dst[i+7:i] := 0 - ELSE - index[4:0] := b[i+3:i] + (j & 0x10) - dst[i+7:i] := a[index*8+7:index*8] - FI - ELSE - dst[i+7:i] := src[i+7:i] - FI -ENDFOR -dst[MAX:256] := 0 +tmp[64:0] := a[63:0] + b[63:0] + (c_in > 0 ? 1 : 0) +MEM[out+63:out] := tmp[63:0] +dst[0] := tmp[64] +dst[7:1] := 0 - +
immintrin.h
 - + Integer - AVX512VL - AVX512BW - Miscellaneous - - - - Shuffle packed 8-bit integers in "a" according to shuffle control mask in the corresponding 8-bit element of "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Flag + Arithmetic + + + + + + Add unsigned 8-bit borrow "c_in" (carry flag) to unsigned 32-bit integer "b", and subtract the result from unsigned 32-bit integer "a". Store the unsigned 32-bit result in "out", and the carry-out in "dst" (carry or overflow flag). -FOR j := 0 to 31 - i := j*8 - IF k[j] - IF b[i+7] == 1 - dst[i+7:i] := 0 - ELSE - index[4:0] := b[i+3:i] + (j & 0x10) - dst[i+7:i] := a[index*8+7:index*8] - FI - ELSE - dst[i+7:i] := 0 - FI -ENDFOR -dst[MAX:256] := 0 +tmp[32:0] := a[31:0] - (b[31:0] + (c_in > 0 ? 1 : 0)) +MEM[out+31:out] := tmp[31:0] +dst[0] := tmp[32] +dst[7:1] := 0 - +
immintrin.h
 - + Integer - AVX512BW - Miscellaneous - - - - - Shuffle 8-bit integers in "a" within 128-bit lanes using the control in the corresponding 8-bit element of "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Flag + Arithmetic + + + + + + Add unsigned 8-bit borrow "c_in" (carry flag) to unsigned 64-bit integer "b", and subtract the result from unsigned 64-bit integer "a". Store the unsigned 64-bit result in "out", and the carry-out in "dst" (carry or overflow flag). -FOR j := 0 to 63 - i := j*8 - IF k[j] - IF b[i+7] == 1 - dst[i+7:i] := 0 - ELSE - index[5:0] := b[i+3:i] + (j & 0x30) - dst[i+7:i] := a[index*8+7:index*8] - FI - ELSE - dst[i+7:i] := src[i+7:i] - FI -ENDFOR -dst[MAX:512] := 0 +tmp[64:0] := a[63:0] - (b[63:0] + (c_in > 0 ? 1 : 0)) +MEM[out+63:out] := tmp[63:0] +dst[0] := tmp[64] +dst[7:1] := 0 - +
immintrin.h
 - - Integer - AVX512BW + Miscellaneous - - - - Shuffle packed 8-bit integers in "a" according to shuffle control mask in the corresponding 8-bit element of "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - -FOR j := 0 to 63 - i := j*8 - IF k[j] - IF b[i+7] == 1 - dst[i+7:i] := 0 - ELSE - index[5:0] := b[i+3:i] + (j & 0x30) - dst[i+7:i] := a[index*8+7:index*8] - FI - ELSE - dst[i+7:i] := 0 - FI -ENDFOR -dst[MAX:512] := 0 - - + + + Insert the 32-bit data from "a" into a Processor Trace stream via a PTW packet. The PTW packet will be inserted if tracing is currently enabled and ptwrite is currently enabled. The current IP will also be inserted via a FUP packet if FUPonPTW is enabled. +
immintrin.h
 - - Integer - AVX512BW + Miscellaneous - - - Shuffle packed 8-bit integers in "a" according to shuffle control mask in the corresponding 8-bit element of "b", and store the results in "dst". - -FOR j := 0 to 63 - i := j*8 - IF b[i+7] == 1 - dst[i+7:i] := 0 - ELSE - index[5:0] := b[i+3:i] + (j & 0x30) - dst[i+7:i] := a[index*8+7:index*8] - FI -ENDFOR -dst[MAX:512] := 0 - - + + + Insert the 64-bit data from "a" into a Processor Trace stream via a PTW packet. The PTW packet will be inserted if tracing is currently enabled and ptwrite is currently enabled. The current IP will also be inserted via a FUP packet if FUPonPTW is enabled. +
immintrin.h
 - - Integer - AVX512VL - AVX512BW + Miscellaneous - - - - - Shuffle packed 8-bit integers in "a" according to shuffle control mask in the corresponding 8-bit element of "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -FOR j := 0 to 15 - i := j*8 - IF k[j] - IF b[i+7] == 1 - dst[i+7:i] := 0 - ELSE - index[3:0] := b[i+3:i] - dst[i+7:i] := a[index*8+7:index*8] - FI - ELSE - dst[i+7:i] := src[i+7:i] - FI -ENDFOR -dst[MAX:128] := 0 - - + + + + Invoke the Intel SGX enclave user (non-privilege) leaf function specified by "a", and return the error code. The "__data" array contains 3 32-bit elements that may act as input, output, or be unused, depending on the semantics of the specified leaf function; these correspond to ebx, ecx, and edx. +
immintrin.h
 - - Integer - AVX512VL - AVX512BW + Miscellaneous - - - - Shuffle packed 8-bit integers in "a" according to shuffle control mask in the corresponding 8-bit element of "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - -FOR j := 0 to 15 - i := j*8 - IF k[j] - IF b[i+7] == 1 - dst[i+7:i] := 0 - ELSE - index[3:0] := b[i+3:i] - dst[i+7:i] := a[index*8+7:index*8] - FI - ELSE - dst[i+7:i] := 0 - FI -ENDFOR -dst[MAX:128] := 0 - - + + + + Invoke the Intel SGX enclave system (privileged) leaf function specified by "a", and return the error code. The "__data" array contains 3 32-bit elements that may act as input, output, or be unused, depending on the semantics of the specified leaf function; these correspond to ebx, ecx, and edx. +
immintrin.h
 - - Integer - AVX512VL - AVX512F + Miscellaneous - - - - - Shuffle 32-bit integers in "a" within 128-bit lanes using the control in "imm8", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -SELECT4(src, control){ - CASE(control[1:0]) - 0: tmp[31:0] := src[31:0] - 1: tmp[31:0] := src[63:32] - 2: tmp[31:0] := src[95:64] - 3: tmp[31:0] := src[127:96] - ESAC - RETURN tmp[31:0] -} - -tmp_dst[31:0] := SELECT4(a[127:0], imm8[1:0]) -tmp_dst[63:32] := SELECT4(a[127:0], imm8[3:2]) -tmp_dst[95:64] := SELECT4(a[127:0], imm8[5:4]) -tmp_dst[127:96] := SELECT4(a[127:0], imm8[7:6]) -tmp_dst[159:128] := SELECT4(a[255:128], imm8[1:0]) -tmp_dst[191:160] := SELECT4(a[255:128], imm8[3:2]) -tmp_dst[223:192] := SELECT4(a[255:128], imm8[5:4]) -tmp_dst[255:224] := SELECT4(a[255:128], imm8[7:6]) -FOR j := 0 to 7 - i := j*32 - IF k[j] - dst[i+31:i] := tmp_dst[i+31:i] - ELSE - dst[i+31:i] := src[i+31:i] - FI -ENDFOR -dst[MAX:256] := 0 - - + + + + Invoke the Intel SGX enclave virtualized (VMM) leaf function specified by "a", and return the error code. The "__data" array contains 3 32-bit elements that may act as input, output, or be unused, depending on the semantics of the specified leaf function; these correspond to ebx, ecx, and edx. +
immintrin.h
 - - Integer - AVX512VL - AVX512F + Miscellaneous - - - - Shuffle 32-bit integers in "a" within 128-bit lanes using the control in "imm8", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + + Write back and flush internal caches. + Initiate writing-back and flushing of external + caches. + +
immintrin.h
+ + + Floating Point + Convert + + + Convert the half-precision (16-bit) floating-point value "a" to a single-precision (32-bit) floating-point value, and store the result in "dst". -SELECT4(src, control){ - CASE(control[1:0]) - 0: tmp[31:0] := src[31:0] - 1: tmp[31:0] := src[63:32] - 2: tmp[31:0] := src[95:64] - 3: tmp[31:0] := src[127:96] - ESAC - RETURN tmp[31:0] -} - -tmp_dst[31:0] := SELECT4(a[127:0], imm8[1:0]) -tmp_dst[63:32] := SELECT4(a[127:0], imm8[3:2]) -tmp_dst[95:64] := SELECT4(a[127:0], imm8[5:4]) -tmp_dst[127:96] := SELECT4(a[127:0], imm8[7:6]) -tmp_dst[159:128] := SELECT4(a[255:128], imm8[1:0]) -tmp_dst[191:160] := SELECT4(a[255:128], imm8[3:2]) -tmp_dst[223:192] := SELECT4(a[255:128], imm8[5:4]) -tmp_dst[255:224] := SELECT4(a[255:128], imm8[7:6]) -FOR j := 0 to 7 - i := j*32 - IF k[j] - dst[i+31:i] := tmp_dst[i+31:i] - ELSE - dst[i+31:i] := 0 - FI -ENDFOR -dst[MAX:256] := 0 +dst[31:0] := Convert_FP16_To_FP32(a[15:0]) - -
immintrin.h
+
emmintrin.h
 - - Integer - AVX512VL - AVX512F - Miscellaneous - - - - - Shuffle 32-bit integers in "a" using the control in "imm8", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Floating Point + Convert + + + + Convert the single-precision (32-bit) floating-point value "a" to a half-precision (16-bit) floating-point value, and store the result in "dst". + [round_note] -SELECT4(src, control){ - CASE(control[1:0]) - 0: tmp[31:0] := src[31:0] - 1: tmp[31:0] := src[63:32] - 2: tmp[31:0] := src[95:64] - 3: tmp[31:0] := src[127:96] - ESAC - RETURN tmp[31:0] -} - -tmp_dst[31:0] := SELECT4(a[127:0], imm8[1:0]) -tmp_dst[63:32] := SELECT4(a[127:0], imm8[3:2]) -tmp_dst[95:64] := SELECT4(a[127:0], imm8[5:4]) -tmp_dst[127:96] := SELECT4(a[127:0], imm8[7:6]) -FOR j := 0 to 3 - i := j*32 - IF k[j] - dst[i+31:i] := tmp_dst[i+31:i] - ELSE - dst[i+31:i] := src[i+31:i] - FI -ENDFOR -dst[MAX:128] := 0 +dst[15:0] := Convert_FP32_To_FP16(a[31:0]) - -
immintrin.h
+
emmintrin.h
 - + Integer - AVX512VL - AVX512F - Miscellaneous - - - - Shuffle 32-bit integers in "a" using the control in "imm8", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - -SELECT4(src, control){ - CASE(control[1:0]) - 0: tmp[31:0] := src[31:0] - 1: tmp[31:0] := src[63:32] - 2: tmp[31:0] := src[95:64] - 3: tmp[31:0] := src[127:96] - ESAC - RETURN tmp[31:0] -} - -tmp_dst[31:0] := SELECT4(a[127:0], imm8[1:0]) -tmp_dst[63:32] := SELECT4(a[127:0], imm8[3:2]) -tmp_dst[95:64] := SELECT4(a[127:0], imm8[5:4]) -tmp_dst[127:96] := SELECT4(a[127:0], imm8[7:6]) -FOR j := 0 to 3 - i := j*32 - IF k[j] - dst[i+31:i] := tmp_dst[i+31:i] - ELSE - dst[i+31:i] := 0 - FI + PCLMULQDQ + Application-Targeted + + + + + Perform a carry-less multiplication of two 64-bit integers, selected from "a" and "b" according to "imm8", and store the results in "dst". + +IF (imm8[0] == 0) + TEMP1 := a[63:0] +ELSE + TEMP1 := a[127:64] +FI +IF (imm8[4] == 0) + TEMP2 := b[63:0] +ELSE + TEMP2 := b[127:64] +FI +FOR i := 0 to 63 + TEMP[i] := (TEMP1[0] and TEMP2[i]) + FOR j := 1 to i + TEMP[i] := TEMP[i] XOR (TEMP1[j] AND TEMP2[i-j]) + ENDFOR + dst[i] := TEMP[i] ENDFOR -dst[MAX:128] := 0 +FOR i := 64 to 127 + TEMP[i] := 0 + FOR j := (i - 63) to 63 + TEMP[i] := TEMP[i] XOR (TEMP1[j] AND TEMP2[i-j]) + ENDFOR + dst[i] := TEMP[i] +ENDFOR +dst[127] := 0 - + +
wmmintrin.h
+ + + PCONFIG + Miscellaneous + + + + Invoke the PCONFIG leaf function specified by "a". The "__data" array contains 3 32-bit elements that may act as input, output, or be unused, depending on the semantics of the specified leaf function; these correspond to rbx, rcx, and rdx. May return the value in eax, depending on the semantics of the specified leaf function. +
immintrin.h
 - + Integer - AVX512VL - AVX512BW - Miscellaneous - - - - - Shuffle 16-bit integers in the high 64 bits of 128-bit lanes of "a" using the control in "imm8". Store the results in the high 64 bits of 128-bit lanes of "dst", with the low 64 bits of 128-bit lanes being copied from from "a" to "dst", using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Flag + POPCNT + Bit Manipulation + + + Count the number of bits set to 1 in unsigned 32-bit integer "a", and return that count in "dst". -tmp_dst[63:0] := a[63:0] -tmp_dst[79:64] := (a >> (imm8[1:0] * 16))[79:64] -tmp_dst[95:80] := (a >> (imm8[3:2] * 16))[79:64] -tmp_dst[111:96] := (a >> (imm8[5:4] * 16))[79:64] -tmp_dst[127:112] := (a >> (imm8[7:6] * 16))[79:64] -tmp_dst[191:128] := a[191:128] -tmp_dst[207:192] := (a >> (imm8[1:0] * 16))[207:192] -tmp_dst[223:208] := (a >> (imm8[3:2] * 16))[207:192] -tmp_dst[239:224] := (a >> (imm8[5:4] * 16))[207:192] -tmp_dst[255:240] := (a >> (imm8[7:6] * 16))[207:192] - -FOR j := 0 to 15 - i := j*16 - IF k[j] - dst[i+15:i] := tmp_dst[i+15:i] - ELSE - dst[i+15:i] := src[i+15:i] +dst := 0 +FOR i := 0 to 31 + IF a[i] + dst := dst + 1 FI ENDFOR -dst[MAX:256] := 0 - -
immintrin.h
+ +
nmmintrin.h
 - + Integer - AVX512VL - AVX512BW - Miscellaneous - - - - Shuffle 16-bit integers in the high 64 bits of 128-bit lanes of "a" using the control in "imm8". Store the results in the high 64 bits of 128-bit lanes of "dst", with the low 64 bits of 128-bit lanes being copied from from "a" to "dst", using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Flag + POPCNT + Bit Manipulation + + + Count the number of bits set to 1 in unsigned 64-bit integer "a", and return that count in "dst". -tmp_dst[63:0] := a[63:0] -tmp_dst[79:64] := (a >> (imm8[1:0] * 16))[79:64] -tmp_dst[95:80] := (a >> (imm8[3:2] * 16))[79:64] -tmp_dst[111:96] := (a >> (imm8[5:4] * 16))[79:64] -tmp_dst[127:112] := (a >> (imm8[7:6] * 16))[79:64] -tmp_dst[191:128] := a[191:128] -tmp_dst[207:192] := (a >> (imm8[1:0] * 16))[207:192] -tmp_dst[223:208] := (a >> (imm8[3:2] * 16))[207:192] -tmp_dst[239:224] := (a >> (imm8[5:4] * 16))[207:192] -tmp_dst[255:240] := (a >> (imm8[7:6] * 16))[207:192] - -FOR j := 0 to 15 - i := j*16 - IF k[j] - dst[i+15:i] := tmp_dst[i+15:i] - ELSE - dst[i+15:i] := 0 +dst := 0 +FOR i := 0 to 63 + IF a[i] + dst := dst + 1 FI ENDFOR -dst[MAX:256] := 0 - -
immintrin.h
+ +
nmmintrin.h
 - + Integer - AVX512BW - Miscellaneous - - - - - Shuffle 16-bit integers in the high 64 bits of 128-bit lanes of "a" using the control in "imm8". Store the results in the high 64 bits of 128-bit lanes of "dst", with the low 64 bits of 128-bit lanes being copied from from "a" to "dst", using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Flag + POPCNT + Bit Manipulation + + + Count the number of bits set to 1 in 32-bit integer "a", and return that count in "dst". -tmp_dst[63:0] := a[63:0] -tmp_dst[79:64] := (a >> (imm8[1:0] * 16))[79:64] -tmp_dst[95:80] := (a >> (imm8[3:2] * 16))[79:64] -tmp_dst[111:96] := (a >> (imm8[5:4] * 16))[79:64] -tmp_dst[127:112] := (a >> (imm8[7:6] * 16))[79:64] -tmp_dst[191:128] := a[191:128] -tmp_dst[207:192] := (a >> (imm8[1:0] * 16))[207:192] -tmp_dst[223:208] := (a >> (imm8[3:2] * 16))[207:192] -tmp_dst[239:224] := (a >> (imm8[5:4] * 16))[207:192] -tmp_dst[255:240] := (a >> (imm8[7:6] * 16))[207:192] -tmp_dst[319:256] := a[319:256] -tmp_dst[335:320] := (a >> (imm8[1:0] * 16))[335:320] -tmp_dst[351:336] := (a >> (imm8[3:2] * 16))[335:320] -tmp_dst[367:352] := (a >> (imm8[5:4] * 16))[335:320] -tmp_dst[383:368] := (a >> (imm8[7:6] * 16))[335:320] -tmp_dst[447:384] := a[447:384] -tmp_dst[463:448] := (a >> (imm8[1:0] * 16))[463:448] -tmp_dst[479:464] := (a >> (imm8[3:2] * 16))[463:448] -tmp_dst[495:480] := (a >> (imm8[5:4] * 16))[463:448] -tmp_dst[511:496] := (a >> (imm8[7:6] * 16))[463:448] - -FOR j := 0 to 31 - i := j*16 - IF k[j] - dst[i+15:i] := tmp_dst[i+15:i] - ELSE - dst[i+15:i] := src[i+15:i] +dst := 0 +FOR i := 0 to 31 + IF a[i] + dst := dst + 1 FI ENDFOR -dst[MAX:512] := 0 - +
immintrin.h
 - + Integer - AVX512BW - Miscellaneous - - - - Shuffle 16-bit integers in the high 64 bits of 128-bit lanes of "a" using the control in "imm8". Store the results in the high 64 bits of 128-bit lanes of "dst", with the low 64 bits of 128-bit lanes being copied from from "a" to "dst", using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Flag + POPCNT + Bit Manipulation + + + Count the number of bits set to 1 in 64-bit integer "a", and return that count in "dst". -tmp_dst[63:0] := a[63:0] -tmp_dst[79:64] := (a >> (imm8[1:0] * 16))[79:64] -tmp_dst[95:80] := (a >> (imm8[3:2] * 16))[79:64] -tmp_dst[111:96] := (a >> (imm8[5:4] * 16))[79:64] -tmp_dst[127:112] := (a >> (imm8[7:6] * 16))[79:64] -tmp_dst[191:128] := a[191:128] -tmp_dst[207:192] := (a >> (imm8[1:0] * 16))[207:192] -tmp_dst[223:208] := (a >> (imm8[3:2] * 16))[207:192] -tmp_dst[239:224] := (a >> (imm8[5:4] * 16))[207:192] -tmp_dst[255:240] := (a >> (imm8[7:6] * 16))[207:192] -tmp_dst[319:256] := a[319:256] -tmp_dst[335:320] := (a >> (imm8[1:0] * 16))[335:320] -tmp_dst[351:336] := (a >> (imm8[3:2] * 16))[335:320] -tmp_dst[367:352] := (a >> (imm8[5:4] * 16))[335:320] -tmp_dst[383:368] := (a >> (imm8[7:6] * 16))[335:320] -tmp_dst[447:384] := a[447:384] -tmp_dst[463:448] := (a >> (imm8[1:0] * 16))[463:448] -tmp_dst[479:464] := (a >> (imm8[3:2] * 16))[463:448] -tmp_dst[495:480] := (a >> (imm8[5:4] * 16))[463:448] -tmp_dst[511:496] := (a >> (imm8[7:6] * 16))[463:448] - -FOR j := 0 to 31 - i := j*16 - IF k[j] - dst[i+15:i] := tmp_dst[i+15:i] - ELSE - dst[i+15:i] := 0 +dst := 0 +FOR i := 0 to 63 + IF a[i] + dst := dst + 1 FI ENDFOR -dst[MAX:512] := 0 - + +
immintrin.h
+ + + PREFETCHWT1 + General Support + + + + Fetch the line of data from memory that contains address "p" to a location in the cache heirarchy specified by the locality hint "i". + +
xmmintrin.h
+ + + RDPID + General Support + + + Copy the IA32_TSC_AUX MSR (signature value) into "dst". + dst[31:0] := IA32_TSC_AUX[31:0] + +
immintrin.h
 - + Integer - AVX512BW - Miscellaneous - - - Shuffle 16-bit integers in the high 64 bits of 128-bit lanes of "a" using the control in "imm8". Store the results in the high 64 bits of 128-bit lanes of "dst", with the low 64 bits of 128-bit lanes being copied from from "a" to "dst". - -dst[63:0] := a[63:0] -dst[79:64] := (a >> (imm8[1:0] * 16))[79:64] -dst[95:80] := (a >> (imm8[3:2] * 16))[79:64] -dst[111:96] := (a >> (imm8[5:4] * 16))[79:64] -dst[127:112] := (a >> (imm8[7:6] * 16))[79:64] -dst[191:128] := a[191:128] -dst[207:192] := (a >> (imm8[1:0] * 16))[207:192] -dst[223:208] := (a >> (imm8[3:2] * 16))[207:192] -dst[239:224] := (a >> (imm8[5:4] * 16))[207:192] -dst[255:240] := (a >> (imm8[7:6] * 16))[207:192] -dst[319:256] := a[319:256] -dst[335:320] := (a >> (imm8[1:0] * 16))[335:320] -dst[351:336] := (a >> (imm8[3:2] * 16))[335:320] -dst[367:352] := (a >> (imm8[5:4] * 16))[335:320] -dst[383:368] := (a >> (imm8[7:6] * 16))[335:320] -dst[447:384] := a[447:384] -dst[463:448] := (a >> (imm8[1:0] * 16))[463:448] -dst[479:464] := (a >> (imm8[3:2] * 16))[463:448] -dst[495:480] := (a >> (imm8[5:4] * 16))[463:448] -dst[511:496] := (a >> (imm8[7:6] * 16))[463:448] -dst[MAX:512] := 0 + Flag + RDRAND + Random + + + Read a hardware generated 16-bit random value and store the result in "val". Return 1 if a random value was generated, and 0 otherwise. + IF HW_RND_GEN.ready == 1 + val[15:0] := HW_RND_GEN.data + dst := 1 +ELSE + val[15:0] := 0 + dst := 0 +FI - +
immintrin.h
 - + Integer - AVX512VL - AVX512BW - Miscellaneous - - - - - Shuffle 16-bit integers in the high 64 bits of "a" using the control in "imm8". Store the results in the high 64 bits of "dst", with the low 64 bits being copied from from "a" to "dst", using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -tmp_dst[63:0] := a[63:0] -tmp_dst[79:64] := (a >> (imm8[1:0] * 16))[79:64] -tmp_dst[95:80] := (a >> (imm8[3:2] * 16))[79:64] -tmp_dst[111:96] := (a >> (imm8[5:4] * 16))[79:64] -tmp_dst[127:112] := (a >> (imm8[7:6] * 16))[79:64] - -FOR j := 0 to 7 - i := j*16 - IF k[j] - dst[i+15:i] := tmp_dst[i+15:i] - ELSE - dst[i+15:i] := src[i+15:i] - FI -ENDFOR -dst[MAX:128] := 0 + Flag + RDRAND + Random + + + Read a hardware generated 32-bit random value and store the result in "val". Return 1 if a random value was generated, and 0 otherwise. + IF HW_RND_GEN.ready == 1 + val[31:0] := HW_RND_GEN.data + dst := 1 +ELSE + val[31:0] := 0 + dst := 0 +FI - +
immintrin.h
 - + Integer - AVX512VL - AVX512BW - Miscellaneous - - - - Shuffle 16-bit integers in the high 64 bits of "a" using the control in "imm8". Store the results in the high 64 bits of "dst", with the low 64 bits being copied from from "a" to "dst", using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - -tmp_dst[63:0] := a[63:0] -tmp_dst[79:64] := (a >> (imm8[1:0] * 16))[79:64] -tmp_dst[95:80] := (a >> (imm8[3:2] * 16))[79:64] -tmp_dst[111:96] := (a >> (imm8[5:4] * 16))[79:64] -tmp_dst[127:112] := (a >> (imm8[7:6] * 16))[79:64] - -FOR j := 0 to 7 - i := j*16 - IF k[j] - dst[i+15:i] := tmp_dst[i+15:i] + Flag + RDRAND + Random + + + Read a hardware generated 64-bit random value and store the result in "val". Return 1 if a random value was generated, and 0 otherwise. + IF HW_RND_GEN.ready == 1 + val[63:0] := HW_RND_GEN.data + dst := 1 +ELSE + val[63:0] := 0 + dst := 0 +FI + + +
immintrin.h
+ + + Flag + RDSEED + Random + + + Read a 16-bit NIST SP800-90B and SP800-90C compliant random value and store in "val". Return 1 if a random value was generated, and 0 otherwise. + IF HW_NRND_GEN.ready == 1 + val[15:0] := HW_NRND_GEN.data + dst := 1 +ELSE + val[15:0] := 0 + dst := 0 +FI + + +
immintrin.h
+ + + Flag + RDSEED + Random + + + Read a 32-bit NIST SP800-90B and SP800-90C compliant random value and store in "val". Return 1 if a random value was generated, and 0 otherwise. + IF HW_NRND_GEN.ready == 1 + val[31:0] := HW_NRND_GEN.data + dst := 1 +ELSE + val[31:0] := 0 + dst := 0 +FI + + +
immintrin.h
+ + + Flag + RDSEED + Random + + + Read a 64-bit NIST SP800-90B and SP800-90C compliant random value and store in "val". Return 1 if a random value was generated, and 0 otherwise. + IF HW_NRND_GEN.ready == 1 + val[63:0] := HW_NRND_GEN.data + dst := 1 +ELSE + val[63:0] := 0 + dst := 0 +FI + + +
immintrin.h
+ + + RDTSCP + General Support + + + Copy the current 64-bit value of the processor's time-stamp counter into "dst", and store the IA32_TSC_AUX MSR (signature value) into memory at "mem_addr". + dst[63:0] := TimeStampCounter +MEM[mem_addr+31:mem_addr] := IA32_TSC_AUX[31:0] + + +
immintrin.h
+ + + RTM + General Support + + + Force an RTM abort. The EAX register is updated to reflect an XABORT instruction caused the abort, and the "imm8" parameter will be provided in bits [31:24] of EAX. + Following an RTM abort, the logical processor resumes execution at the fallback address computed through the outermost XBEGIN instruction. + IF RTM_ACTIVE == 0 + // nop +ELSE + // restore architectural register state + // discard memory updates performed in transaction + // update EAX with status and imm8 value + eax[31:24] := imm8[7:0] + RTM_NEST_COUNT := 0 + RTM_ACTIVE := 0 + IF _64_BIT_MODE + RIP := fallbackRIP ELSE - dst[i+15:i] := 0 + EIP := fallbackEIP FI -ENDFOR -dst[MAX:128] := 0 +FI - +
immintrin.h
 - - Integer - AVX512VL - AVX512BW - Miscellaneous - - - - - Shuffle 16-bit integers in the low 64 bits of 128-bit lanes of "a" using the control in "imm8". Store the results in the low 64 bits of 128-bit lanes of "dst", with the high 64 bits of 128-bit lanes being copied from from "a" to "dst", using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -tmp_dst[15:0] := (a >> (imm8[1:0] * 16))[15:0] -tmp_dst[31:16] := (a >> (imm8[3:2] * 16))[15:0] -tmp_dst[47:32] := (a >> (imm8[5:4] * 16))[15:0] -tmp_dst[63:48] := (a >> (imm8[7:6] * 16))[15:0] -tmp_dst[127:64] := a[127:64] -tmp_dst[143:128] := (a >> (imm8[1:0] * 16))[143:128] -tmp_dst[159:144] := (a >> (imm8[3:2] * 16))[143:128] -tmp_dst[175:160] := (a >> (imm8[5:4] * 16))[143:128] -tmp_dst[191:176] := (a >> (imm8[7:6] * 16))[143:128] -tmp_dst[255:192] := a[255:192] - -FOR j := 0 to 15 - i := j*16 - IF k[j] - dst[i+15:i] := tmp_dst[i+15:i] - ELSE - dst[i+15:i] := src[i+15:i] + + RTM + General Support + + + Specify the start of an RTM code region. + If the logical processor was not already in transactional execution, then this call causes the logical processor to transition into transactional execution. + On an RTM abort, the logical processor discards all architectural register and memory updates performed during the RTM execution, restores architectural state, and starts execution beginning at the fallback address computed from the outermost XBEGIN instruction. Return status of ~0 (0xFFFF) if continuing inside transaction; all other codes are aborts. + IF RTM_NEST_COUNT < MAX_RTM_NEST_COUNT + RTM_NEST_COUNT := RTM_NEST_COUNT + 1 + IF RTM_NEST_COUNT == 1 + IF _64_BIT_MODE + fallbackRIP := RIP + ELSE IF _32_BIT_MODE + fallbackEIP := EIP + FI + + RTM_ACTIVE := 1 + // enter RTM execution, record register state, start tracking memory state FI -ENDFOR -dst[MAX:256] := 0 +ELSE + // RTM abort (see _xabort) +FI + + +
immintrin.h
+ + + RTM + General Support + + + Specify the end of an RTM code region. + If this corresponds to the outermost scope, the logical processor will attempt to commit the logical processor state atomically. + If the commit fails, the logical processor will perform an RTM abort. + IF RTM_ACTIVE == 1 + RTM_NEST_COUNT := RTM_NEST_COUNT - 1 + IF RTM_NEST_COUNT == 0 + // try to commit transaction + IF FAIL_TO_COMMIT_TRANSACTION + // RTM abort (see _xabort) + ELSE + RTM_ACTIVE := 0 + FI + FI +FI + + +
immintrin.h
+ + + RTM + General Support + + + Query the transactional execution status, return 1 if inside a transactionally executing RTM or HLE region, and return 0 otherwise. + IF (RTM_ACTIVE == 1 OR HLE_ACTIVE == 1) + dst := 1 +ELSE + dst := 0 +FI - + +
immintrin.h
+ + + SERIALIZE + General Support + + Serialize instruction execution, ensuring all modifications to flags, registers, and memory by previous instructions are completed before the next instruction is fetched. +
immintrin.h
 - + Integer - AVX512VL - AVX512BW - Miscellaneous - - - - Shuffle 16-bit integers in the low 64 bits of 128-bit lanes of "a" using the control in "imm8". Store the results in the low 64 bits of 128-bit lanes of "dst", with the high 64 bits of 128-bit lanes being copied from from "a" to "dst", using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + SHA + Cryptography + + + + Perform an intermediate calculation for the next four SHA1 message values (unsigned 32-bit integers) using previous message values from "a" and "b", and store the result in "dst". -tmp_dst[15:0] := (a >> (imm8[1:0] * 16))[15:0] -tmp_dst[31:16] := (a >> (imm8[3:2] * 16))[15:0] -tmp_dst[47:32] := (a >> (imm8[5:4] * 16))[15:0] -tmp_dst[63:48] := (a >> (imm8[7:6] * 16))[15:0] -tmp_dst[127:64] := a[127:64] -tmp_dst[143:128] := (a >> (imm8[1:0] * 16))[143:128] -tmp_dst[159:144] := (a >> (imm8[3:2] * 16))[143:128] -tmp_dst[175:160] := (a >> (imm8[5:4] * 16))[143:128] -tmp_dst[191:176] := (a >> (imm8[7:6] * 16))[143:128] -tmp_dst[255:192] := a[255:192] - -FOR j := 0 to 15 - i := j*16 - IF k[j] - dst[i+15:i] := tmp_dst[i+15:i] - ELSE - dst[i+15:i] := 0 - FI -ENDFOR -dst[MAX:256] := 0 +W0 := a[127:96] +W1 := a[95:64] +W2 := a[63:32] +W3 := a[31:0] +W4 := b[127:96] +W5 := b[95:64] +dst[127:96] := W2 XOR W0 +dst[95:64] := W3 XOR W1 +dst[63:32] := W4 XOR W2 +dst[31:0] := W5 XOR W3 - +
immintrin.h
 - + Integer - AVX512BW - Miscellaneous - - - - - Shuffle 16-bit integers in the low 64 bits of 128-bit lanes of "a" using the control in "imm8". Store the results in the low 64 bits of 128-bit lanes of "dst", with the high 64 bits of 128-bit lanes being copied from from "a" to "dst", using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + SHA + Cryptography + + + + Perform the final calculation for the next four SHA1 message values (unsigned 32-bit integers) using the intermediate result in "a" and the previous message values in "b", and store the result in "dst". -tmp_dst[15:0] := (a >> (imm8[1:0] * 16))[15:0] -tmp_dst[31:16] := (a >> (imm8[3:2] * 16))[15:0] -tmp_dst[47:32] := (a >> (imm8[5:4] * 16))[15:0] -tmp_dst[63:48] := (a >> (imm8[7:6] * 16))[15:0] -tmp_dst[127:64] := a[127:64] -tmp_dst[143:128] := (a >> (imm8[1:0] * 16))[143:128] -tmp_dst[159:144] := (a >> (imm8[3:2] * 16))[143:128] -tmp_dst[175:160] := (a >> (imm8[5:4] * 16))[143:128] -tmp_dst[191:176] := (a >> (imm8[7:6] * 16))[143:128] -tmp_dst[255:192] := a[255:192] -tmp_dst[271:256] := (a >> (imm8[1:0] * 16))[271:256] -tmp_dst[287:272] := (a >> (imm8[3:2] * 16))[271:256] -tmp_dst[303:288] := (a >> (imm8[5:4] * 16))[271:256] -tmp_dst[319:304] := (a >> (imm8[7:6] * 16))[271:256] -tmp_dst[383:320] := a[383:320] -tmp_dst[399:384] := (a >> (imm8[1:0] * 16))[399:384] -tmp_dst[415:400] := (a >> (imm8[3:2] * 16))[399:384] -tmp_dst[431:416] := (a >> (imm8[5:4] * 16))[399:384] -tmp_dst[447:432] := (a >> (imm8[7:6] * 16))[399:384] -tmp_dst[511:448] := a[511:448] - -FOR j := 0 to 31 - i := j*16 - IF k[j] - dst[i+15:i] := tmp_dst[i+15:i] - ELSE - dst[i+15:i] := src[i+15:i] - FI -ENDFOR -dst[MAX:512] := 0 +W13 := b[95:64] +W14 := b[63:32] +W15 := b[31:0] +W16 := (a[127:96] XOR W13) <<< 1 +W17 := (a[95:64] XOR W14) <<< 1 +W18 := (a[63:32] XOR W15) <<< 1 +W19 := (a[31:0] XOR W16) <<< 1 +dst[127:96] := W16 +dst[95:64] := W17 +dst[63:32] := W18 +dst[31:0] := W19 - +
immintrin.h
 - + Integer - AVX512BW - Miscellaneous - - - - Shuffle 16-bit integers in the low 64 bits of 128-bit lanes of "a" using the control in "imm8". Store the results in the low 64 bits of 128-bit lanes of "dst", with the high 64 bits of 128-bit lanes being copied from from "a" to "dst", using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + SHA + Cryptography + + + + Calculate SHA1 state variable E after four rounds of operation from the current SHA1 state variable "a", add that value to the scheduled values (unsigned 32-bit integers) in "b", and store the result in "dst". -tmp_dst[15:0] := (a >> (imm8[1:0] * 16))[15:0] -tmp_dst[31:16] := (a >> (imm8[3:2] * 16))[15:0] -tmp_dst[47:32] := (a >> (imm8[5:4] * 16))[15:0] -tmp_dst[63:48] := (a >> (imm8[7:6] * 16))[15:0] -tmp_dst[127:64] := a[127:64] -tmp_dst[143:128] := (a >> (imm8[1:0] * 16))[143:128] -tmp_dst[159:144] := (a >> (imm8[3:2] * 16))[143:128] -tmp_dst[175:160] := (a >> (imm8[5:4] * 16))[143:128] -tmp_dst[191:176] := (a >> (imm8[7:6] * 16))[143:128] -tmp_dst[255:192] := a[255:192] -tmp_dst[271:256] := (a >> (imm8[1:0] * 16))[271:256] -tmp_dst[287:272] := (a >> (imm8[3:2] * 16))[271:256] -tmp_dst[303:288] := (a >> (imm8[5:4] * 16))[271:256] -tmp_dst[319:304] := (a >> (imm8[7:6] * 16))[271:256] -tmp_dst[383:320] := a[383:320] -tmp_dst[399:384] := (a >> (imm8[1:0] * 16))[399:384] -tmp_dst[415:400] := (a >> (imm8[3:2] * 16))[399:384] -tmp_dst[431:416] := (a >> (imm8[5:4] * 16))[399:384] -tmp_dst[447:432] := (a >> (imm8[7:6] * 16))[399:384] -tmp_dst[511:448] := a[511:448] - -FOR j := 0 to 31 - i := j*16 - IF k[j] - dst[i+15:i] := tmp_dst[i+15:i] - ELSE - dst[i+15:i] := 0 - FI +tmp := (a[127:96] <<< 30) +dst[127:96] := b[127:96] + tmp +dst[95:64] := b[95:64] +dst[63:32] := b[63:32] +dst[31:0] := b[31:0] + + +
immintrin.h
+ + + Integer + SHA + Cryptography + + + + + Perform four rounds of SHA1 operation using an initial SHA1 state (A,B,C,D) from "a" and some pre-computed sum of the next 4 round message values (unsigned 32-bit integers), and state variable E from "b", and store the updated SHA1 state (A,B,C,D) in "dst". "func" contains the logic functions and round constants. + IF (func[1:0] == 0) + f := f0() + K := K0 +ELSE IF (func[1:0] == 1) + f := f1() + K := K1 +ELSE IF (func[1:0] == 2) + f := f2() + K := K2 +ELSE IF (func[1:0] == 3) + f := f3() + K := K3 +FI +A := a[127:96] +B := a[95:64] +C := a[63:32] +D := a[31:0] +W[0] := b[127:96] +W[1] := b[95:64] +W[2] := b[63:32] +W[3] := b[31:0] +A[1] := f(B, C, D) + (A <<< 5) + W[0] + K +B[1] := A +C[1] := B <<< 30 +D[1] := C +E[1] := D +FOR i := 1 to 3 + A[i+1] := f(B[i], C[i], D[i]) + (A[i] <<< 5) + W[i] + E[i] + K + B[i+1] := A[i] + C[i+1] := B[i] <<< 30 + D[i+1] := C[i] + E[i+1] := D[i] +ENDFOR +dst[127:96] := A[4] +dst[95:64] := B[4] +dst[63:32] := C[4] +dst[31:0] := D[4] + + +
immintrin.h
+ + + Integer + SHA + Cryptography + + + + Perform an intermediate calculation for the next four SHA256 message values (unsigned 32-bit integers) using previous message values from "a" and "b", and store the result in "dst". + W4 := b[31:0] +W3 := a[127:96] +W2 := a[95:64] +W1 := a[63:32] +W0 := a[31:0] +dst[127:96] := W3 + sigma0(W4) +dst[95:64] := W2 + sigma0(W3) +dst[63:32] := W1 + sigma0(W2) +dst[31:0] := W0 + sigma0(W1) + + +
immintrin.h
+ + + Integer + SHA + Cryptography + + + + Perform the final calculation for the next four SHA256 message values (unsigned 32-bit integers) using previous message values from "a" and "b", and store the result in "dst"." + W14 := b[95:64] +W15 := b[127:96] +W16 := a[31:0] + sigma1(W14) +W17 := a[63:32] + sigma1(W15) +W18 := a[95:64] + sigma1(W16) +W19 := a[127:96] + sigma1(W17) +dst[127:96] := W19 +dst[95:64] := W18 +dst[63:32] := W17 +dst[31:0] := W16 + + +
immintrin.h
+ + + Integer + SHA + Cryptography + + + + + Perform 2 rounds of SHA256 operation using an initial SHA256 state (C,D,G,H) from "a", an initial SHA256 state (A,B,E,F) from "b", and a pre-computed sum of the next 2 round message values (unsigned 32-bit integers) and the corresponding round constants from "k", and store the updated SHA256 state (A,B,E,F) in "dst". + A[0] := b[127:96] +B[0] := b[95:64] +C[0] := a[127:96] +D[0] := a[95:64] +E[0] := b[63:32] +F[0] := b[31:0] +G[0] := a[63:32] +H[0] := a[31:0] +W_K[0] := k[31:0] +W_K[1] := k[63:32] +FOR i := 0 to 1 + A[i+1] := Ch(E[i], F[i], G[i]) + sum1(E[i]) + W_K[i] + H[i] + Maj(A[i], B[i], C[i]) + sum0(A[i]) + B[i+1] := A[i] + C[i+1] := B[i] + D[i+1] := C[i] + E[i+1] := Ch(E[i], F[i], G[i]) + sum1(E[i]) + W_K[i] + H[i] + D[i] + F[i+1] := E[i] + G[i+1] := F[i] + H[i+1] := G[i] ENDFOR -dst[MAX:512] := 0 +dst[127:96] := A[2] +dst[95:64] := B[2] +dst[63:32] := E[2] +dst[31:0] := F[2] + + +
immintrin.h
+ + + SSE + Swizzle + + + + + + Macro: Transpose the 4x4 matrix formed by the 4 rows of single-precision (32-bit) floating-point elements in "row0", "row1", "row2", and "row3", and store the transposed matrix in these vectors ("row0" now contains column 0, etc.). + +__m128 tmp3, tmp2, tmp1, tmp0; +tmp0 := _mm_unpacklo_ps(row0, row1); +tmp2 := _mm_unpacklo_ps(row2, row3); +tmp1 := _mm_unpackhi_ps(row0, row1); +tmp3 := _mm_unpackhi_ps(row2, row3); +row0 := _mm_movelh_ps(tmp0, tmp2); +row1 := _mm_movehl_ps(tmp2, tmp0); +row2 := _mm_movelh_ps(tmp1, tmp3); +row3 := _mm_movehl_ps(tmp3, tmp1); + +
xmmintrin.h
+ + + SSE + General Support + + + Get the unsigned 32-bit value of the MXCSR control and status register. + dst[31:0] := MXCSR + + +
xmmintrin.h
+ + + SSE + General Support + + + Set the MXCSR control and status register with the value in unsigned 32-bit integer "a". + +MXCSR := a[31:0] + + +
xmmintrin.h
+ + + SSE + General Support + + Macro: Get the exception state bits from the MXCSR control and status register. The exception state may contain any of the following flags: _MM_EXCEPT_INVALID, _MM_EXCEPT_DIV_ZERO, _MM_EXCEPT_DENORM, _MM_EXCEPT_OVERFLOW, _MM_EXCEPT_UNDERFLOW, _MM_EXCEPT_INEXACT + dst[31:0] := MXCSR & _MM_EXCEPT_MASK + +
xmmintrin.h
+ + + SSE + General Support + + + Macro: Set the exception state bits of the MXCSR control and status register to the value in unsigned 32-bit integer "a". The exception state may contain any of the following flags: _MM_EXCEPT_INVALID, _MM_EXCEPT_DIV_ZERO, _MM_EXCEPT_DENORM, _MM_EXCEPT_OVERFLOW, _MM_EXCEPT_UNDERFLOW, _MM_EXCEPT_INEXACT + MXCSR := a[31:0] AND ~_MM_EXCEPT_MASK + +
xmmintrin.h
+ + + SSE + General Support + + Macro: Get the exception mask bits from the MXCSR control and status register. The exception mask may contain any of the following flags: _MM_MASK_INVALID, _MM_MASK_DIV_ZERO, _MM_MASK_DENORM, _MM_MASK_OVERFLOW, _MM_MASK_UNDERFLOW, _MM_MASK_INEXACT + dst[31:0] := MXCSR & _MM_MASK_MASK - -
immintrin.h
+
xmmintrin.h
 - - Integer - AVX512BW - Miscellaneous - - - Shuffle 16-bit integers in the low 64 bits of 128-bit lanes of "a" using the control in "imm8". Store the results in the low 64 bits of 128-bit lanes of "dst", with the high 64 bits of 128-bit lanes being copied from from "a" to "dst". - -dst[15:0] := (a >> (imm8[1:0] * 16))[15:0] -dst[31:16] := (a >> (imm8[3:2] * 16))[15:0] -dst[47:32] := (a >> (imm8[5:4] * 16))[15:0] -dst[63:48] := (a >> (imm8[7:6] * 16))[15:0] -dst[127:64] := a[127:64] -dst[143:128] := (a >> (imm8[1:0] * 16))[143:128] -dst[159:144] := (a >> (imm8[3:2] * 16))[143:128] -dst[175:160] := (a >> (imm8[5:4] * 16))[143:128] -dst[191:176] := (a >> (imm8[7:6] * 16))[143:128] -dst[255:192] := a[255:192] -dst[271:256] := (a >> (imm8[1:0] * 16))[271:256] -dst[287:272] := (a >> (imm8[3:2] * 16))[271:256] -dst[303:288] := (a >> (imm8[5:4] * 16))[271:256] -dst[319:304] := (a >> (imm8[7:6] * 16))[271:256] -dst[383:320] := a[383:320] -dst[399:384] := (a >> (imm8[1:0] * 16))[399:384] -dst[415:400] := (a >> (imm8[3:2] * 16))[399:384] -dst[431:416] := (a >> (imm8[5:4] * 16))[399:384] -dst[447:432] := (a >> (imm8[7:6] * 16))[399:384] -dst[511:448] := a[511:448] -dst[MAX:512] := 0 + + SSE + General Support + + + Macro: Set the exception mask bits of the MXCSR control and status register to the value in unsigned 32-bit integer "a". The exception mask may contain any of the following flags: _MM_MASK_INVALID, _MM_MASK_DIV_ZERO, _MM_MASK_DENORM, _MM_MASK_OVERFLOW, _MM_MASK_UNDERFLOW, _MM_MASK_INEXACT + MXCSR := a[31:0] AND ~_MM_MASK_MASK - -
immintrin.h
+
xmmintrin.h
+ + + SSE + General Support + + Macro: Get the rounding mode bits from the MXCSR control and status register. The rounding mode may contain any of the following flags: _MM_ROUND_NEAREST, _MM_ROUND_DOWN, _MM_ROUND_UP, _MM_ROUND_TOWARD_ZERO + dst[31:0] := MXCSR & _MM_ROUND_MASK + +
xmmintrin.h
+ + + SSE + General Support + + + Macro: Set the rounding mode bits of the MXCSR control and status register to the value in unsigned 32-bit integer "a". The rounding mode may contain any of the following flags: _MM_ROUND_NEAREST, _MM_ROUND_DOWN, _MM_ROUND_UP, _MM_ROUND_TOWARD_ZERO + MXCSR := a[31:0] AND ~_MM_ROUND_MASK + +
xmmintrin.h
+ + + SSE + General Support + + Macro: Get the flush zero bits from the MXCSR control and status register. The flush zero may contain any of the following flags: _MM_FLUSH_ZERO_ON or _MM_FLUSH_ZERO_OFF + dst[31:0] := MXCSR & _MM_FLUSH_MASK + +
xmmintrin.h
+ + + SSE + General Support + + + Macro: Set the flush zero bits of the MXCSR control and status register to the value in unsigned 32-bit integer "a". The flush zero may contain any of the following flags: _MM_FLUSH_ZERO_ON or _MM_FLUSH_ZERO_OFF + MXCSR := a[31:0] AND ~_MM_FLUSH_MASK + +
xmmintrin.h
+ + + SSE + General Support + + + + Fetch the line of data from memory that contains address "p" to a location in the cache heirarchy specified by the locality hint "i". + + + + +
xmmintrin.h
 - + + SSE + General Support + + + Perform a serializing operation on all store-to-memory instructions that were issued prior to this instruction. Guarantees that every store instruction that precedes, in program order, is globally visible before any store instruction which follows the fence in program order. + +
xmmintrin.h
+ + Integer - AVX512VL - AVX512BW - Miscellaneous - - - - - Shuffle 16-bit integers in the low 64 bits of "a" using the control in "imm8". Store the results in the low 64 bits of "dst", with the high 64 bits being copied from from "a" to "dst", using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + SSE + Special Math Functions + + + + Compare packed signed 16-bit integers in "a" and "b", and store packed maximum values in "dst". -tmp_dst[15:0] := (a >> (imm8[1:0] * 16))[15:0] -tmp_dst[31:16] := (a >> (imm8[3:2] * 16))[15:0] -tmp_dst[47:32] := (a >> (imm8[5:4] * 16))[15:0] -tmp_dst[63:48] := (a >> (imm8[7:6] * 16))[15:0] -tmp_dst[127:64] := a[127:64] - -FOR j := 0 to 7 +FOR j := 0 to 3 i := j*16 - IF k[j] - dst[i+15:i] := tmp_dst[i+15:i] - ELSE - dst[i+15:i] := src[i+15:i] - FI + dst[i+15:i] := MAX(a[i+15:i], b[i+15:i]) ENDFOR -dst[MAX:128] := 0 - -
immintrin.h
+ +
xmmintrin.h
 - + Integer - AVX512VL - AVX512BW - Miscellaneous - - - - Shuffle 16-bit integers in the low 64 bits of "a" using the control in "imm8". Store the results in the low 64 bits of "dst", with the high 64 bits being copied from from "a" to "dst", using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + SSE + Special Math Functions + + + + Compare packed signed 16-bit integers in "a" and "b", and store packed maximum values in "dst". -tmp_dst[15:0] := (a >> (imm8[1:0] * 16))[15:0] -tmp_dst[31:16] := (a >> (imm8[3:2] * 16))[15:0] -tmp_dst[47:32] := (a >> (imm8[5:4] * 16))[15:0] -tmp_dst[63:48] := (a >> (imm8[7:6] * 16))[15:0] -tmp_dst[127:64] := a[127:64] - -FOR j := 0 to 7 +FOR j := 0 to 3 i := j*16 - IF k[j] - dst[i+15:i] := tmp_dst[i+15:i] - ELSE - dst[i+15:i] := 0 - FI + dst[i+15:i] := MAX(a[i+15:i], b[i+15:i]) ENDFOR -dst[MAX:128] := 0 - -
immintrin.h
+ +
xmmintrin.h
 - + Integer - AVX512VL - AVX512F - Shift - - - - - Shift packed 32-bit integers in "a" left by "count" while shifting in zeros, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + SSE + Special Math Functions + + + + Compare packed unsigned 8-bit integers in "a" and "b", and store packed maximum values in "dst". FOR j := 0 to 7 - i := j*32 - IF k[j] - IF count[63:0] > 31 - dst[i+31:i] := 0 - ELSE - dst[i+31:i] := ZeroExtend(a[i+31:i] << count[63:0]) - FI - ELSE - dst[i+31:i] := src[i+31:i] - FI + i := j*8 + dst[i+7:i] := MAX(a[i+7:i], b[i+7:i]) ENDFOR -dst[MAX:256] := 0 - -
immintrin.h
+ +
xmmintrin.h
 - + Integer - AVX512VL - AVX512F - Shift - - - - - Shift packed 32-bit integers in "a" left by "imm8" while shifting in zeros, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + SSE + Special Math Functions + + + + Compare packed unsigned 8-bit integers in "a" and "b", and store packed maximum values in "dst". FOR j := 0 to 7 - i := j*32 - IF k[j] - IF imm8[7:0] > 31 - dst[i+31:i] := 0 - ELSE - dst[i+31:i] := ZeroExtend(a[i+31:i] << imm8[7:0]) - FI - ELSE - dst[i+31:i] := src[i+31:i] - FI + i := j*8 + dst[i+7:i] := MAX(a[i+7:i], b[i+7:i]) ENDFOR -dst[MAX:256] := 0 - -
immintrin.h
+ +
xmmintrin.h
 - + Integer - AVX512VL - AVX512F - Shift - - - - Shift packed 32-bit integers in "a" left by "count" while shifting in zeros, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + SSE + Special Math Functions + + + + Compare packed signed 16-bit integers in "a" and "b", and store packed minimum values in "dst". -FOR j := 0 to 7 - i := j*32 - IF k[j] - IF count[63:0] > 31 - dst[i+31:i] := 0 - ELSE - dst[i+31:i] := ZeroExtend(a[i+31:i] << count[63:0]) - FI - ELSE - dst[i+31:i] := 0 - FI +FOR j := 0 to 3 + i := j*16 + dst[i+15:i] := MIN(a[i+15:i], b[i+15:i]) ENDFOR -dst[MAX:256] := 0 - -
immintrin.h
+ +
xmmintrin.h
 - + Integer - AVX512VL - AVX512F - Shift - - - - Shift packed 32-bit integers in "a" left by "imm8" while shifting in zeros, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + SSE + Special Math Functions + + + + Compare packed signed 16-bit integers in "a" and "b", and store packed minimum values in "dst". -FOR j := 0 to 7 - i := j*32 - IF k[j] - IF imm8[7:0] > 31 - dst[i+31:i] := 0 - ELSE - dst[i+31:i] := ZeroExtend(a[i+31:i] << imm8[7:0]) - FI - ELSE - dst[i+31:i] := 0 - FI +FOR j := 0 to 3 + i := j*16 + dst[i+15:i] := MIN(a[i+15:i], b[i+15:i]) ENDFOR -dst[MAX:256] := 0 - -
immintrin.h
+ +
xmmintrin.h
 - + Integer - AVX512VL - AVX512F - Shift - - - - - Shift packed 32-bit integers in "a" left by "count" while shifting in zeros, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + SSE + Special Math Functions + + + + Compare packed unsigned 8-bit integers in "a" and "b", and store packed minimum values in "dst". -FOR j := 0 to 3 - i := j*32 - IF k[j] - IF count[63:0] > 31 - dst[i+31:i] := 0 - ELSE - dst[i+31:i] := ZeroExtend(a[i+31:i] << count[63:0]) - FI - ELSE - dst[i+31:i] := src[i+31:i] - FI +FOR j := 0 to 7 + i := j*8 + dst[i+7:i] := MIN(a[i+7:i], b[i+7:i]) ENDFOR -dst[MAX:128] := 0 - -
immintrin.h
+ +
xmmintrin.h
 - + Integer - AVX512VL - AVX512F - Shift - - - - - Shift packed 32-bit integers in "a" left by "imm8" while shifting in zeros, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + SSE + Special Math Functions + + + + Compare packed unsigned 8-bit integers in "a" and "b", and store packed minimum values in "dst". -FOR j := 0 to 3 - i := j*32 - IF k[j] - IF imm8[7:0] > 31 - dst[i+31:i] := 0 - ELSE - dst[i+31:i] := ZeroExtend(a[i+31:i] << imm8[7:0]) - FI - ELSE - dst[i+31:i] := src[i+31:i] - FI +FOR j := 0 to 7 + i := j*8 + dst[i+7:i] := MIN(a[i+7:i], b[i+7:i]) ENDFOR -dst[MAX:128] := 0 - -
immintrin.h
+ +
xmmintrin.h
 - + Integer - AVX512VL - AVX512F - Shift - - - - Shift packed 32-bit integers in "a" left by "count" while shifting in zeros, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + SSE + Arithmetic + + + + Multiply the packed unsigned 16-bit integers in "a" and "b", producing intermediate 32-bit integers, and store the high 16 bits of the intermediate integers in "dst". FOR j := 0 to 3 - i := j*32 - IF k[j] - IF count[63:0] > 31 - dst[i+31:i] := 0 - ELSE - dst[i+31:i] := ZeroExtend(a[i+31:i] << count[63:0]) - FI - ELSE - dst[i+31:i] := 0 - FI + i := j*16 + tmp[31:0] := a[i+15:i] * b[i+15:i] + dst[i+15:i] := tmp[31:16] ENDFOR -dst[MAX:128] := 0 - -
immintrin.h
+ +
xmmintrin.h
 - + Integer - AVX512VL - AVX512F - Shift - - - - Shift packed 32-bit integers in "a" left by "imm8" while shifting in zeros, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + SSE + Arithmetic + + + + Multiply the packed unsigned 16-bit integers in "a" and "b", producing intermediate 32-bit integers, and store the high 16 bits of the intermediate integers in "dst". FOR j := 0 to 3 - i := j*32 - IF k[j] - IF imm8[7:0] > 31 - dst[i+31:i] := 0 - ELSE - dst[i+31:i] := ZeroExtend(a[i+31:i] << imm8[7:0]) - FI - ELSE - dst[i+31:i] := 0 - FI + i := j*16 + tmp[31:0] := a[i+15:i] * b[i+15:i] + dst[i+15:i] := tmp[31:16] ENDFOR -dst[MAX:128] := 0 - -
immintrin.h
+ +
xmmintrin.h
 - + Integer - AVX512BW - Shift - - - Shift 128-bit lanes in "a" left by "imm8" bytes while shifting in zeros, and store the results in "dst". + SSE + Probability/Statistics + + + + Average packed unsigned 8-bit integers in "a" and "b", and store the results in "dst". -tmp := imm8[7:0] -IF tmp > 15 - tmp := 16 -FI -dst[127:0] := a[127:0] << (tmp*8) -dst[255:128] := a[255:128] << (tmp*8) -dst[383:256] := a[383:256] << (tmp*8) -dst[511:384] := a[511:384] << (tmp*8) -dst[MAX:512] := 0 +FOR j := 0 to 7 + i := j*8 + dst[i+7:i] := (a[i+7:i] + b[i+7:i] + 1) >> 1 +ENDFOR - -
immintrin.h
+ +
xmmintrin.h
 - + Integer - AVX512VL - AVX512F - Shift - - - - - Shift packed 64-bit integers in "a" left by "count" while shifting in zeros, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + SSE + Probability/Statistics + + + + Average packed unsigned 8-bit integers in "a" and "b", and store the results in "dst". -FOR j := 0 to 3 - i := j*64 - IF k[j] - IF count[63:0] > 63 - dst[i+63:i] := 0 - ELSE - dst[i+63:i] := ZeroExtend(a[i+63:i] << count[63:0]) - FI - ELSE - dst[i+63:i] := src[i+63:i] - FI +FOR j := 0 to 7 + i := j*8 + dst[i+7:i] := (a[i+7:i] + b[i+7:i] + 1) >> 1 ENDFOR -dst[MAX:256] := 0 - -
immintrin.h
+ +
xmmintrin.h
 - + Integer - AVX512VL - AVX512F - Shift - - - - - Shift packed 64-bit integers in "a" left by "imm8" while shifting in zeros, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + SSE + Probability/Statistics + + + + Average packed unsigned 16-bit integers in "a" and "b", and store the results in "dst". FOR j := 0 to 3 - i := j*64 - IF k[j] - IF imm8[7:0] > 63 - dst[i+63:i] := 0 - ELSE - dst[i+63:i] := ZeroExtend(a[i+63:i] << imm8[7:0]) - FI - ELSE - dst[i+63:i] := src[i+63:i] - FI + i := j*16 + dst[i+15:i] := (a[i+15:i] + b[i+15:i] + 1) >> 1 ENDFOR -dst[MAX:256] := 0 - -
immintrin.h
+ +
xmmintrin.h
 - + Integer - AVX512VL - AVX512F - Shift - - - - Shift packed 64-bit integers in "a" left by "count" while shifting in zeros, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + SSE + Probability/Statistics + + + + Average packed unsigned 16-bit integers in "a" and "b", and store the results in "dst". FOR j := 0 to 3 - i := j*64 - IF k[j] - IF count[63:0] > 63 - dst[i+63:i] := 0 - ELSE - dst[i+63:i] := ZeroExtend(a[i+63:i] << count[63:0]) - FI - ELSE - dst[i+63:i] := 0 - FI + i := j*16 + dst[i+15:i] := (a[i+15:i] + b[i+15:i] + 1) >> 1 ENDFOR -dst[MAX:256] := 0 - -
immintrin.h
+ +
xmmintrin.h
 - + Integer - AVX512VL - AVX512F - Shift - - - - Shift packed 64-bit integers in "a" left by "imm8" while shifting in zeros, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + SSE + Arithmetic + Miscellaneous + + + + Compute the absolute differences of packed unsigned 8-bit integers in "a" and "b", then horizontally sum each consecutive 8 differences to produce four unsigned 16-bit integers, and pack these unsigned 16-bit integers in the low 16 bits of "dst". -FOR j := 0 to 3 - i := j*64 - IF k[j] - IF imm8[7:0] > 63 - dst[i+63:i] := 0 - ELSE - dst[i+63:i] := ZeroExtend(a[i+63:i] << imm8[7:0]) - FI - ELSE - dst[i+63:i] := 0 - FI +FOR j := 0 to 7 + i := j*8 + tmp[i+7:i] := ABS(a[i+7:i] - b[i+7:i]) ENDFOR -dst[MAX:256] := 0 +dst[15:0] := tmp[7:0] + tmp[15:8] + tmp[23:16] + tmp[31:24] + tmp[39:32] + tmp[47:40] + tmp[55:48] + tmp[63:56] +dst[63:16] := 0 - -
immintrin.h
+ +
xmmintrin.h
 - + + Floating Point Integer - AVX512VL - AVX512F - Shift - - - - - Shift packed 64-bit integers in "a" left by "count" while shifting in zeros, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + SSE + Arithmetic + Miscellaneous + + + + Compute the absolute differences of packed unsigned 8-bit integers in "a" and "b", then horizontally sum each consecutive 8 differences to produce four unsigned 16-bit integers, and pack these unsigned 16-bit integers in the low 16 bits of "dst". -FOR j := 0 to 1 - i := j*64 - IF k[j] - IF count[63:0] > 63 - dst[i+63:i] := 0 - ELSE - dst[i+63:i] := ZeroExtend(a[i+63:i] << count[63:0]) - FI - ELSE - dst[i+63:i] := src[i+63:i] - FI +FOR j := 0 to 7 + i := j*8 + tmp[i+7:i] := ABS(a[i+7:i] - b[i+7:i]) ENDFOR -dst[MAX:128] := 0 +dst[15:0] := tmp[7:0] + tmp[15:8] + tmp[23:16] + tmp[31:24] + tmp[39:32] + tmp[47:40] + tmp[55:48] + tmp[63:56] +dst[63:16] := 0 - -
immintrin.h
+ +
xmmintrin.h
 - - Integer - AVX512VL - AVX512F - Shift - - - - - Shift packed 64-bit integers in "a" left by "imm8" while shifting in zeros, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Floating Point + SSE + Convert + + + + Convert the signed 32-bit integer "b" to a single-precision (32-bit) floating-point element, store the result in the lower element of "dst", and copy the upper 3 packed elements from "a" to the upper elements of "dst". -FOR j := 0 to 1 - i := j*64 - IF k[j] - IF imm8[7:0] > 63 - dst[i+63:i] := 0 - ELSE - dst[i+63:i] := ZeroExtend(a[i+63:i] << imm8[7:0]) - FI - ELSE - dst[i+63:i] := src[i+63:i] - FI -ENDFOR -dst[MAX:128] := 0 +dst[31:0] := Convert_Int32_To_FP32(b[31:0]) +dst[127:32] := a[127:32] - -
immintrin.h
+ +
xmmintrin.h
 - + Integer - AVX512VL - AVX512F - Shift - - - - Shift packed 64-bit integers in "a" left by "count" while shifting in zeros, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + SSE + Convert + + + + Convert the signed 32-bit integer "b" to a single-precision (32-bit) floating-point element, store the result in the lower element of "dst", and copy the upper 3 packed elements from "a" to the upper elements of "dst". -FOR j := 0 to 1 - i := j*64 - IF k[j] - IF count[63:0] > 63 - dst[i+63:i] := 0 - ELSE - dst[i+63:i] := ZeroExtend(a[i+63:i] << count[63:0]) - FI - ELSE - dst[i+63:i] := 0 - FI -ENDFOR -dst[MAX:128] := 0 +dst[31:0] := Convert_Int32_To_FP32(b[31:0]) +dst[127:32] := a[127:32] - -
immintrin.h
+ +
xmmintrin.h
 - + + Floating Point Integer - AVX512VL - AVX512F - Shift - - - - Shift packed 64-bit integers in "a" left by "imm8" while shifting in zeros, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + SSE + Convert + + + + Convert the signed 64-bit integer "b" to a single-precision (32-bit) floating-point element, store the result in the lower element of "dst", and copy the upper 3 packed elements from "a" to the upper elements of "dst". -FOR j := 0 to 1 - i := j*64 - IF k[j] - IF imm8[7:0] > 63 - dst[i+63:i] := 0 - ELSE - dst[i+63:i] := ZeroExtend(a[i+63:i] << imm8[7:0]) - FI - ELSE - dst[i+63:i] := 0 - FI -ENDFOR +dst[31:0] := Convert_Int64_To_FP32(b[63:0]) +dst[127:32] := a[127:32] dst[MAX:128] := 0 - -
immintrin.h
+ +
xmmintrin.h
 - - Integer - AVX512VL - AVX512F - Shift - - - - - Shift packed 32-bit integers in "a" left by the amount specified by the corresponding element in "count" while shifting in zeros, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Floating Point + SSE + Convert + + + + Convert packed 32-bit integers in "b" to packed single-precision (32-bit) floating-point elements, store the results in the lower 2 elements of "dst", and copy the upper 2 packed elements from "a" to the upper elements of "dst". -FOR j := 0 to 7 - i := j*32 - IF k[j] - dst[i+31:i] := ZeroExtend(a[i+31:i] << count[i+31:i]) - ELSE - dst[i+31:i] := src[i+31:i] - FI -ENDFOR -dst[MAX:256] := 0 +dst[31:0] := Convert_Int32_To_FP32(b[31:0]) +dst[63:32] := Convert_Int32_To_FP32(b[63:32]) +dst[95:64] := a[95:64] +dst[127:96] := a[127:96] - -
immintrin.h
+ +
xmmintrin.h
 - + Integer - AVX512VL - AVX512F - Shift - - - - Shift packed 32-bit integers in "a" left by the amount specified by the corresponding element in "count" while shifting in zeros, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + SSE + Convert + + + + Convert packed signed 32-bit integers in "b" to packed single-precision (32-bit) floating-point elements, store the results in the lower 2 elements of "dst", and copy the upper 2 packed elements from "a" to the upper elements of "dst". -FOR j := 0 to 7 - i := j*32 - IF k[j] - dst[i+31:i] := ZeroExtend(a[i+31:i] << count[i+31:i]) - ELSE - dst[i+31:i] := 0 - FI -ENDFOR -dst[MAX:256] := 0 +dst[31:0] := Convert_Int32_To_FP32(b[31:0]) +dst[63:32] := Convert_Int32_To_FP32(b[63:32]) +dst[95:64] := a[95:64] +dst[127:96] := a[127:96] - -
immintrin.h
+ +
xmmintrin.h
 - - Integer - AVX512VL - AVX512F - Shift - - - - - Shift packed 32-bit integers in "a" left by the amount specified by the corresponding element in "count" while shifting in zeros, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Floating Point + SSE + Convert + + + Convert packed 16-bit integers in "a" to packed single-precision (32-bit) floating-point elements, and store the results in "dst". FOR j := 0 to 3 - i := j*32 - IF k[j] - dst[i+31:i] := ZeroExtend(a[i+31:i] << count[i+31:i]) - ELSE - dst[i+31:i] := src[i+31:i] - FI + i := j*16 + m := j*32 + dst[m+31:m] := Convert_Int16_To_FP32(a[i+15:i]) ENDFOR -dst[MAX:128] := 0 - -
immintrin.h
+
xmmintrin.h
 - + + Floating Point Integer - AVX512VL - AVX512F - Shift - - - - Shift packed 32-bit integers in "a" left by the amount specified by the corresponding element in "count" while shifting in zeros, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + SSE + Convert + + + Convert packed unsigned 16-bit integers in "a" to packed single-precision (32-bit) floating-point elements, and store the results in "dst". FOR j := 0 to 3 - i := j*32 - IF k[j] - dst[i+31:i] := ZeroExtend(a[i+31:i] << count[i+31:i]) - ELSE - dst[i+31:i] := 0 - FI + i := j*16 + m := j*32 + dst[m+31:m] := Convert_Int16_To_FP32(a[i+15:i]) ENDFOR -dst[MAX:128] := 0 - -
immintrin.h
+
xmmintrin.h
 - - Integer - AVX512VL - AVX512F - Shift - - - - - Shift packed 64-bit integers in "a" left by the amount specified by the corresponding element in "count" while shifting in zeros, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Floating Point + SSE + Convert + + + Convert the lower packed 8-bit integers in "a" to packed single-precision (32-bit) floating-point elements, and store the results in "dst". FOR j := 0 to 3 - i := j*64 - IF k[j] - dst[i+63:i] := ZeroExtend(a[i+63:i] << count[i+63:i]) - ELSE - dst[i+63:i] := src[i+63:i] - FI + i := j*8 + m := j*32 + dst[m+31:m] := Convert_Int8_To_FP32(a[i+7:i]) ENDFOR -dst[MAX:256] := 0 - -
immintrin.h
+
xmmintrin.h
 - + + Floating Point Integer - AVX512VL - AVX512F - Shift - - - - Shift packed 64-bit integers in "a" left by the amount specified by the corresponding element in "count" while shifting in zeros, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + SSE + Convert + + + Convert the lower packed unsigned 8-bit integers in "a" to packed single-precision (32-bit) floating-point elements, and store the results in "dst". FOR j := 0 to 3 - i := j*64 - IF k[j] - dst[i+63:i] := ZeroExtend(a[i+63:i] << count[i+63:i]) - ELSE - dst[i+63:i] := 0 - FI + i := j*8 + m := j*32 + dst[m+31:m] := Convert_Int8_To_FP32(a[i+7:i]) ENDFOR -dst[MAX:256] := 0 - -
immintrin.h
+
xmmintrin.h
 - - Integer - AVX512VL - AVX512F - Shift - - - - - Shift packed 64-bit integers in "a" left by the amount specified by the corresponding element in "count" while shifting in zeros, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Floating Point + SSE + Convert + + + + Convert packed signed 32-bit integers in "a" to packed single-precision (32-bit) floating-point elements, store the results in the lower 2 elements of "dst", then covert the packed signed 32-bit integers in "b" to single-precision (32-bit) floating-point element, and store the results in the upper 2 elements of "dst". -FOR j := 0 to 1 - i := j*64 - IF k[j] - dst[i+63:i] := ZeroExtend(a[i+63:i] << count[i+63:i]) - ELSE - dst[i+63:i] := src[i+63:i] - FI -ENDFOR -dst[MAX:128] := 0 +dst[31:0] := Convert_Int32_To_FP32(a[31:0]) +dst[63:32] := Convert_Int32_To_FP32(a[63:32]) +dst[95:64] := Convert_Int32_To_FP32(b[31:0]) +dst[127:96] := Convert_Int32_To_FP32(b[63:32]) - -
immintrin.h
+
xmmintrin.h
 - + Integer - AVX512VL - AVX512F - Shift - - - - Shift packed 64-bit integers in "a" left by the amount specified by the corresponding element in "count" while shifting in zeros, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + SSE + Store + + + + Store 64-bits of integer data from "a" into memory using a non-temporal memory hint. -FOR j := 0 to 1 - i := j*64 - IF k[j] - dst[i+63:i] := ZeroExtend(a[i+63:i] << count[i+63:i]) - ELSE - dst[i+63:i] := 0 - FI -ENDFOR -dst[MAX:128] := 0 +MEM[mem_addr+63:mem_addr] := a[63:0] - -
immintrin.h
+ +
xmmintrin.h
 - + Integer - AVX512VL - AVX512BW - Shift - - - - - Shift packed 16-bit integers in "a" left by the amount specified by the corresponding element in "count" while shifting in zeros, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + SSE + Store + + + + + Conditionally store 8-bit integer elements from "a" into memory using "mask" (elements are not stored when the highest bit is not set in the corresponding element) and a non-temporal memory hint. -FOR j := 0 to 15 - i := j*16 - IF k[j] - dst[i+15:i] := ZeroExtend(a[i+15:i] << count[i+15:i]) - ELSE - dst[i+15:i] := src[i+15:i] - FI +FOR j := 0 to 7 + i := j*8 + IF mask[i+7] + MEM[mem_addr+i+7:mem_addr+i] := a[i+7:i] + FI ENDFOR -dst[MAX:256] := 0 - -
immintrin.h
+ +
xmmintrin.h
 - + Integer - AVX512VL - AVX512BW - Shift - - - - Shift packed 16-bit integers in "a" left by the amount specified by the corresponding element in "count" while shifting in zeros, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + SSE + Store + + + + + Conditionally store 8-bit integer elements from "a" into memory using "mask" (elements are not stored when the highest bit is not set in the corresponding element). -FOR j := 0 to 15 - i := j*16 - IF k[j] - dst[i+15:i] := ZeroExtend(a[i+15:i] << count[i+15:i]) - ELSE - dst[i+15:i] := 0 - FI +FOR j := 0 to 7 + i := j*8 + IF mask[i+7] + MEM[mem_addr+i+7:mem_addr+i] := a[i+7:i] + FI ENDFOR -dst[MAX:256] := 0 - -
immintrin.h
+ +
xmmintrin.h
 - + Integer - AVX512VL - AVX512BW - Shift - - - Shift packed 16-bit integers in "a" left by the amount specified by the corresponding element in "count" while shifting in zeros, and store the results in "dst". + SSE + Swizzle + + + + Extract a 16-bit integer from "a", selected with "imm8", and store the result in the lower element of "dst". -FOR j := 0 to 15 - i := j*16 - IF k[j] - dst[i+15:i] := ZeroExtend(a[i+15:i] << count[i+15:i]) - ELSE - dst[i+15:i] := 0 - FI -ENDFOR -dst[MAX:256] := 0 +dst[15:0] := (a[63:0] >> (imm8[1:0] * 16))[15:0] +dst[31:16] := 0 - -
immintrin.h
+ +
xmmintrin.h
 - + Integer - AVX512BW - Shift - - - - - Shift packed 16-bit integers in "a" left by the amount specified by the corresponding element in "count" while shifting in zeros, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + SSE + Swizzle + + + + Extract a 16-bit integer from "a", selected with "imm8", and store the result in the lower element of "dst". -FOR j := 0 to 31 - i := j*16 - IF k[j] - dst[i+15:i] := ZeroExtend(a[i+15:i] << count[i+15:i]) - ELSE - dst[i+15:i] := src[i+15:i] - FI -ENDFOR -dst[MAX:512] := 0 +dst[15:0] := (a[63:0] >> (imm8[1:0] * 16))[15:0] +dst[31:16] := 0 - -
immintrin.h
+ +
xmmintrin.h
 - + Integer - AVX512BW - Shift - - - - Shift packed 16-bit integers in "a" left by the amount specified by the corresponding element in "count" while shifting in zeros, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + SSE + Swizzle + + + + + Copy "a" to "dst", and insert the 16-bit integer "i" into "dst" at the location specified by "imm8". -FOR j := 0 to 31 - i := j*16 - IF k[j] - dst[i+15:i] := ZeroExtend(a[i+15:i] << count[i+15:i]) - ELSE - dst[i+15:i] := 0 - FI -ENDFOR -dst[MAX:512] := 0 +dst[63:0] := a[63:0] +sel := imm8[1:0]*16 +dst[sel+15:sel] := i[15:0] - -
immintrin.h
+ +
xmmintrin.h
 - + Integer - AVX512BW - Shift - - - Shift packed 16-bit integers in "a" left by the amount specified by the corresponding element in "count" while shifting in zeros, and store the results in "dst". + SSE + Swizzle + + + + + Copy "a" to "dst", and insert the 16-bit integer "i" into "dst" at the location specified by "imm8". -FOR j := 0 to 31 - i := j*16 - IF k[j] - dst[i+15:i] := ZeroExtend(a[i+15:i] << count[i+15:i]) - ELSE - dst[i+15:i] := 0 - FI -ENDFOR -dst[MAX:512] := 0 +dst[63:0] := a[63:0] +sel := imm8[1:0]*16 +dst[sel+15:sel] := i[15:0] - -
immintrin.h
+ +
xmmintrin.h
 - + Integer - AVX512VL - AVX512BW - Shift - - - - - Shift packed 16-bit integers in "a" left by the amount specified by the corresponding element in "count" while shifting in zeros, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + SSE + Miscellaneous + + + Create mask from the most significant bit of each 8-bit element in "a", and store the result in "dst". FOR j := 0 to 7 - i := j*16 - IF k[j] - dst[i+15:i] := ZeroExtend(a[i+15:i] << count[i+15:i]) - ELSE - dst[i+15:i] := src[i+15:i] - FI + i := j*8 + dst[j] := a[i+7] ENDFOR -dst[MAX:128] := 0 +dst[MAX:8] := 0 - -
immintrin.h
+ +
xmmintrin.h
 - + Integer - AVX512VL - AVX512BW - Shift - - - - Shift packed 16-bit integers in "a" left by the amount specified by the corresponding element in "count" while shifting in zeros, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + SSE + Miscellaneous + + + Create mask from the most significant bit of each 8-bit element in "a", and store the result in "dst". FOR j := 0 to 7 - i := j*16 - IF k[j] - dst[i+15:i] := ZeroExtend(a[i+15:i] << count[i+15:i]) - ELSE - dst[i+15:i] := 0 - FI + i := j*8 + dst[j] := a[i+7] ENDFOR -dst[MAX:128] := 0 +dst[MAX:8] := 0 - -
immintrin.h
+ +
xmmintrin.h
 - + Integer - AVX512VL - AVX512BW - Shift - - - Shift packed 16-bit integers in "a" left by the amount specified by the corresponding element in "count" while shifting in zeros, and store the results in "dst". + SSE + Swizzle + + + + Shuffle 16-bit integers in "a" using the control in "imm8", and store the results in "dst". -FOR j := 0 to 7 - i := j*16 - IF k[j] - dst[i+15:i] := ZeroExtend(a[i+15:i] << count[i+15:i]) - ELSE - dst[i+15:i] := 0 - FI -ENDFOR -dst[MAX:128] := 0 +DEFINE SELECT4(src, control) { + CASE(control[1:0]) OF + 0: tmp[15:0] := src[15:0] + 1: tmp[15:0] := src[31:16] + 2: tmp[15:0] := src[47:32] + 3: tmp[15:0] := src[63:48] + ESAC + RETURN tmp[15:0] +} +dst[15:0] := SELECT4(a[63:0], imm8[1:0]) +dst[31:16] := SELECT4(a[63:0], imm8[3:2]) +dst[47:32] := SELECT4(a[63:0], imm8[5:4]) +dst[63:48] := SELECT4(a[63:0], imm8[7:6]) - -
immintrin.h
+ +
xmmintrin.h
 - + + Floating Point Integer - AVX512VL - AVX512BW - Shift - - - - - Shift packed 16-bit integers in "a" left by "count" while shifting in zeros, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + SSE + Swizzle + + + + Shuffle 16-bit integers in "a" using the control in "imm8", and store the results in "dst". -FOR j := 0 to 15 - i := j*16 - IF k[j] - IF count[63:0] > 15 - dst[i+15:i] := 0 - ELSE - dst[i+15:i] := ZeroExtend(a[i+15:i] << count[63:0]) - FI - ELSE - dst[i+15:i] := src[i+15:i] - FI -ENDFOR -dst[MAX:256] := 0 +DEFINE SELECT4(src, control) { + CASE(control[1:0]) OF + 0: tmp[15:0] := src[15:0] + 1: tmp[15:0] := src[31:16] + 2: tmp[15:0] := src[47:32] + 3: tmp[15:0] := src[63:48] + ESAC + RETURN tmp[15:0] +} +dst[15:0] := SELECT4(a[63:0], imm8[1:0]) +dst[31:16] := SELECT4(a[63:0], imm8[3:2]) +dst[47:32] := SELECT4(a[63:0], imm8[5:4]) +dst[63:48] := SELECT4(a[63:0], imm8[7:6]) - -
immintrin.h
+ +
xmmintrin.h
 - - Integer - AVX512VL - AVX512BW - Shift - - - - - Shift packed 16-bit integers in "a" left by "imm8" while shifting in zeros, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Floating Point + SSE + Arithmetic + + + + Add the lower single-precision (32-bit) floating-point element in "a" and "b", store the result in the lower element of "dst", and copy the upper 3 packed elements from "a" to the upper elements of "dst". -FOR j := 0 to 15 - i := j*16 - IF k[j] - IF imm8[7:0] > 15 - dst[i+15:i] := 0 - ELSE - dst[i+15:i] := ZeroExtend(a[i+15:i] << imm8[7:0]) - FI - ELSE - dst[i+15:i] := src[i+15:i] - FI -ENDFOR -dst[MAX:256] := 0 +dst[31:0] := a[31:0] + b[31:0] +dst[127:32] := a[127:32] - -
immintrin.h
+ +
xmmintrin.h
 - - Integer - AVX512VL - AVX512BW - Shift - - - - Shift packed 16-bit integers in "a" left by "count" while shifting in zeros, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Floating Point + SSE + Arithmetic + + + + Add packed single-precision (32-bit) floating-point elements in "a" and "b", and store the results in "dst". -FOR j := 0 to 15 - i := j*16 - IF k[j] - IF count[63:0] > 15 - dst[i+15:i] := 0 - ELSE - dst[i+15:i] := ZeroExtend(a[i+15:i] << count[63:0]) - FI - ELSE - dst[i+15:i] := 0 - FI +FOR j := 0 to 3 + i := j*32 + dst[i+31:i] := a[i+31:i] + b[i+31:i] ENDFOR -dst[MAX:256] := 0 - -
immintrin.h
+ +
xmmintrin.h
 - - Integer - AVX512VL - AVX512BW - Shift - - - - Shift packed 16-bit integers in "a" left by "imm8" while shifting in zeros, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Floating Point + SSE + Arithmetic + + + + Subtract the lower single-precision (32-bit) floating-point element in "b" from the lower single-precision (32-bit) floating-point element in "a", store the result in the lower element of "dst", and copy the upper 3 packed elements from "a" to the upper elements of "dst". -FOR j := 0 to 15 - i := j*16 - IF k[j] - IF imm8[7:0] > 15 - dst[i+15:i] := 0 - ELSE - dst[i+15:i] := ZeroExtend(a[i+15:i] << imm8[7:0]) - FI - ELSE - dst[i+15:i] := 0 - FI -ENDFOR -dst[MAX:256] := 0 +dst[31:0] := a[31:0] - b[31:0] +dst[127:32] := a[127:32] - -
immintrin.h
+ +
xmmintrin.h
 - - Integer - AVX512BW - Shift - - - - - Shift packed 16-bit integers in "a" left by "count" while shifting in zeros, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Floating Point + SSE + Arithmetic + + + + Subtract packed single-precision (32-bit) floating-point elements in "b" from packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst". -FOR j := 0 to 31 - i := j*16 - IF k[j] - IF count[63:0] > 15 - dst[i+15:i] := 0 - ELSE - dst[i+15:i] := ZeroExtend(a[i+15:i] << count[63:0]) - FI - ELSE - dst[i+15:i] := src[i+15:i] - FI +FOR j := 0 to 3 + i := j*32 + dst[i+31:i] := a[i+31:i] - b[i+31:i] ENDFOR -dst[MAX:512] := 0 - -
immintrin.h
+ +
xmmintrin.h
 - - Integer - AVX512BW - Shift - - - - - Shift packed 16-bit integers in "a" left by "imm8" while shifting in zeros, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Floating Point + SSE + Arithmetic + + + + Multiply the lower single-precision (32-bit) floating-point element in "a" and "b", store the result in the lower element of "dst", and copy the upper 3 packed elements from "a" to the upper elements of "dst". -FOR j := 0 to 31 - i := j*16 - IF k[j] - IF imm8[7:0] > 15 - dst[i+15:i] := 0 - ELSE - dst[i+15:i] := ZeroExtend(a[i+15:i] << imm8[7:0]) - FI - ELSE - dst[i+15:i] := src[i+15:i] - FI +dst[31:0] := a[31:0] * b[31:0] +dst[127:32] := a[127:32] + + +
xmmintrin.h
+ + + Floating Point + SSE + Arithmetic + + + + Multiply packed single-precision (32-bit) floating-point elements in "a" and "b", and store the results in "dst". + +FOR j := 0 to 3 + i := j*32 + dst[i+31:i] := a[i+31:i] * b[i+31:i] ENDFOR -dst[MAX:512] := 0 - -
immintrin.h
+ +
xmmintrin.h
 - - Integer - AVX512BW - Shift - - - - Shift packed 16-bit integers in "a" left by "count" while shifting in zeros, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Floating Point + SSE + Arithmetic + + + + Divide the lower single-precision (32-bit) floating-point element in "a" by the lower single-precision (32-bit) floating-point element in "b", store the result in the lower element of "dst", and copy the upper 3 packed elements from "a" to the upper elements of "dst". -FOR j := 0 to 31 - i := j*16 - IF k[j] - IF count[63:0] > 15 - dst[i+15:i] := 0 - ELSE - dst[i+15:i] := ZeroExtend(a[i+15:i] << count[63:0]) - FI - ELSE - dst[i+15:i] := 0 - FI +dst[31:0] := a[31:0] / b[31:0] +dst[127:32] := a[127:32] + + +
xmmintrin.h
+ + + Floating Point + SSE + Arithmetic + + + + Divide packed single-precision (32-bit) floating-point elements in "a" by packed elements in "b", and store the results in "dst". + +FOR j := 0 to 3 + i := 32*j + dst[i+31:i] := a[i+31:i] / b[i+31:i] ENDFOR -dst[MAX:512] := 0 - -
immintrin.h
+ +
xmmintrin.h
 - - Integer - AVX512BW - Shift - - - - Shift packed 16-bit integers in "a" left by "imm8" while shifting in zeros, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Floating Point + SSE + Elementary Math Functions + + + Compute the square root of the lower single-precision (32-bit) floating-point element in "a", store the result in the lower element of "dst", and copy the upper 3 packed elements from "a" to the upper elements of "dst". -FOR j := 0 to 31 - i := j*16 - IF k[j] - IF imm8[7:0] > 15 - dst[i+15:i] := 0 - ELSE - dst[i+15:i] := ZeroExtend(a[i+15:i] << imm8[7:0]) - FI - ELSE - dst[i+15:i] := 0 - FI +dst[31:0] := SQRT(a[31:0]) +dst[127:32] := a[127:32] + + +
xmmintrin.h
+ + + Floating Point + SSE + Elementary Math Functions + + + Compute the square root of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst". + +FOR j := 0 to 3 + i := j*32 + dst[i+31:i] := SQRT(a[i+31:i]) ENDFOR -dst[MAX:512] := 0 - -
immintrin.h
+ +
xmmintrin.h
 - - Integer - AVX512BW - Shift - - - Shift packed 16-bit integers in "a" left by "count" while shifting in zeros, and store the results in "dst". + + Floating Point + SSE + Elementary Math Functions + + + Compute the approximate reciprocal of the lower single-precision (32-bit) floating-point element in "a", store the result in the lower element of "dst", and copy the upper 3 packed elements from "a" to the upper elements of "dst". The maximum relative error for this approximation is less than 1.5*2^-12. + +dst[31:0] := (1.0 / a[31:0]) +dst[127:32] := a[127:32] + + +
xmmintrin.h
+ + + Floating Point + SSE + Elementary Math Functions + + + Compute the approximate reciprocal of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst". The maximum relative error for this approximation is less than 1.5*2^-12. -FOR j := 0 to 31 - i := j*16 - IF count[63:0] > 15 - dst[i+15:i] := 0 - ELSE - dst[i+15:i] := ZeroExtend(a[i+15:i] << count[63:0]) - FI +FOR j := 0 to 3 + i := j*32 + dst[i+31:i] := (1.0 / a[i+31:i]) ENDFOR -dst[MAX:512] := 0 - -
immintrin.h
+ +
xmmintrin.h
 - - Integer - AVX512BW - Shift - - - Shift packed 16-bit integers in "a" left by "imm8" while shifting in zeros, and store the results in "dst". + + Floating Point + SSE + Elementary Math Functions + + + Compute the approximate reciprocal square root of the lower single-precision (32-bit) floating-point element in "a", store the result in the lower element of "dst", and copy the upper 3 packed elements from "a" to the upper elements of "dst". The maximum relative error for this approximation is less than 1.5*2^-12. -FOR j := 0 to 31 - i := j*16 - IF imm8[7:0] > 15 - dst[i+15:i] := 0 - ELSE - dst[i+15:i] := ZeroExtend(a[i+15:i] << imm8[7:0]) - FI -ENDFOR -dst[MAX:512] := 0 +dst[31:0] := (1.0 / SQRT(a[31:0])) +dst[127:32] := a[127:32] - -
immintrin.h
+ +
xmmintrin.h
 - - Integer - AVX512VL - AVX512BW - Shift - - - - - Shift packed 16-bit integers in "a" left by "count" while shifting in zeros, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Floating Point + SSE + Elementary Math Functions + + + Compute the approximate reciprocal square root of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst". The maximum relative error for this approximation is less than 1.5*2^-12. -FOR j := 0 to 7 - i := j*16 - IF k[j] - IF count[63:0] > 15 - dst[i+15:i] := 0 - ELSE - dst[i+15:i] := ZeroExtend(a[i+15:i] << count[63:0]) - FI - ELSE - dst[i+15:i] := src[i+15:i] - FI +FOR j := 0 to 3 + i := j*32 + dst[i+31:i] := (1.0 / SQRT(a[i+31:i])) ENDFOR -dst[MAX:128] := 0 - -
immintrin.h
+ +
xmmintrin.h
 - - Integer - AVX512VL - AVX512BW - Shift - - - - - Shift packed 16-bit integers in "a" left by "imm8" while shifting in zeros, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Floating Point + SSE + Special Math Functions + + + + Compare the lower single-precision (32-bit) floating-point elements in "a" and "b", store the minimum value in the lower element of "dst", and copy the upper 3 packed elements from "a" to the upper element of "dst". -FOR j := 0 to 7 - i := j*16 - IF k[j] - IF imm8[7:0] > 15 - dst[i+15:i] := 0 - ELSE - dst[i+15:i] := ZeroExtend(a[i+15:i] << imm8[7:0]) - FI - ELSE - dst[i+15:i] := src[i+15:i] - FI -ENDFOR -dst[MAX:128] := 0 +dst[31:0] := MIN(a[31:0], b[31:0]) +dst[127:32] := a[127:32] - -
immintrin.h
+ +
xmmintrin.h
 - - Integer - AVX512VL - AVX512BW - Shift - - - - Shift packed 16-bit integers in "a" left by "count" while shifting in zeros, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Floating Point + SSE + Special Math Functions + + + + Compare packed single-precision (32-bit) floating-point elements in "a" and "b", and store packed minimum values in "dst". -FOR j := 0 to 7 - i := j*16 - IF k[j] - IF count[63:0] > 15 - dst[i+15:i] := 0 - ELSE - dst[i+15:i] := ZeroExtend(a[i+15:i] << count[63:0]) - FI - ELSE - dst[i+15:i] := 0 - FI +FOR j := 0 to 3 + i := j*32 + dst[i+31:i] := MIN(a[i+31:i], b[i+31:i]) ENDFOR -dst[MAX:128] := 0 - -
immintrin.h
+ +
xmmintrin.h
 - - Integer - AVX512VL - AVX512BW - Shift - - - - Shift packed 16-bit integers in "a" left by "imm8" while shifting in zeros, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Floating Point + SSE + Special Math Functions + + + + Compare the lower single-precision (32-bit) floating-point elements in "a" and "b", store the maximum value in the lower element of "dst", and copy the upper 3 packed elements from "a" to the upper element of "dst". -FOR j := 0 to 7 - i := j*16 - IF k[j] - IF imm8[7:0] > 15 - dst[i+15:i] := 0 - ELSE - dst[i+15:i] := ZeroExtend(a[i+15:i] << imm8[7:0]) - FI - ELSE - dst[i+15:i] := 0 - FI -ENDFOR -dst[MAX:128] := 0 +dst[31:0] := MAX(a[31:0], b[31:0]) +dst[127:32] := a[127:32] - -
immintrin.h
+ +
xmmintrin.h
 - - Integer - AVX512VL - AVX512F - Shift - - - - - Shift packed 32-bit integers in "a" right by "count" while shifting in sign bits, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Floating Point + SSE + Special Math Functions + + + + Compare packed single-precision (32-bit) floating-point elements in "a" and "b", and store packed maximum values in "dst". -FOR j := 0 to 7 +FOR j := 0 to 3 i := j*32 - IF k[j] - IF count[63:0] > 31 - dst[i+31:i] := SignBit - ELSE - dst[i+31:i] := SignExtend(a[i+31:i] >> count[63:0]) - FI - ELSE - dst[i+31:i] := src[i+31:i] - FI + dst[i+31:i] := MAX(a[i+31:i], b[i+31:i]) ENDFOR -dst[MAX:256] := 0 - -
immintrin.h
+ +
xmmintrin.h
 - - Integer - AVX512VL - AVX512F - Shift - - - - - Shift packed 32-bit integers in "a" right by "imm8" while shifting in sign bits, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Floating Point + SSE + Logical + + + + Compute the bitwise AND of packed single-precision (32-bit) floating-point elements in "a" and "b", and store the results in "dst". -FOR j := 0 to 7 +FOR j := 0 to 3 i := j*32 - IF k[j] - IF imm8[7:0] > 31 - dst[i+31:i] := SignBit - ELSE - dst[i+31:i] := SignExtend(a[i+31:i] >> imm8[7:0]) - FI - ELSE - dst[i+31:i] := src[i+31:i] - FI + dst[i+31:i] := (a[i+31:i] AND b[i+31:i]) ENDFOR -dst[MAX:256] := 0 - -
immintrin.h
+ +
xmmintrin.h
 - - Integer - AVX512VL - AVX512F - Shift - - - - Shift packed 32-bit integers in "a" right by "count" while shifting in sign bits, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Floating Point + SSE + Logical + + + + Compute the bitwise NOT of packed single-precision (32-bit) floating-point elements in "a" and then AND with "b", and store the results in "dst". -FOR j := 0 to 7 +FOR j := 0 to 3 i := j*32 - IF k[j] - IF count[63:0] > 31 - dst[i+31:i] := SignBit - ELSE - dst[i+31:i] := SignExtend(a[i+31:i] >> count[63:0]) - FI - ELSE - dst[i+31:i] := 0 - FI + dst[i+31:i] := ((NOT a[i+31:i]) AND b[i+31:i]) ENDFOR -dst[MAX:256] := 0 - -
immintrin.h
+ +
xmmintrin.h
 - - Integer - AVX512VL - AVX512F - Shift - - - - Shift packed 32-bit integers in "a" right by "imm8" while shifting in sign bits, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Floating Point + SSE + Logical + + + + Compute the bitwise OR of packed single-precision (32-bit) floating-point elements in "a" and "b", and store the results in "dst". -FOR j := 0 to 7 +FOR j := 0 to 3 i := j*32 - IF k[j] - IF imm8[7:0] > 31 - dst[i+31:i] := SignBit - ELSE - dst[i+31:i] := SignExtend(a[i+31:i] >> imm8[7:0]) - FI - ELSE - dst[i+31:i] := 0 - FI + dst[i+31:i] := a[i+31:i] OR b[i+31:i] ENDFOR -dst[MAX:256] := 0 - -
immintrin.h
+ +
xmmintrin.h
 - - Integer - AVX512VL - AVX512F - Shift - - - - - Shift packed 32-bit integers in "a" right by "count" while shifting in sign bits, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Floating Point + SSE + Logical + + + + Compute the bitwise XOR of packed single-precision (32-bit) floating-point elements in "a" and "b", and store the results in "dst". FOR j := 0 to 3 i := j*32 - IF k[j] - IF count[63:0] > 31 - dst[i+31:i] := SignBit - ELSE - dst[i+31:i] := SignExtend(a[i+31:i] >> count[63:0]) - FI - ELSE - dst[i+31:i] := src[i+31:i] - FI + dst[i+31:i] := a[i+31:i] XOR b[i+31:i] ENDFOR -dst[MAX:128] := 0 - -
immintrin.h
+ +
xmmintrin.h
 - - Integer - AVX512VL - AVX512F - Shift - - - - - Shift packed 32-bit integers in "a" right by "imm8" while shifting in sign bits, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Floating Point + SSE + Compare + + + + Compare the lower single-precision (32-bit) floating-point elements in "a" and "b" for equality, store the result in the lower element of "dst", and copy the upper 3 packed elements from "a" to the upper elements of "dst". + +dst[31:0] := ( a[31:0] == b[31:0] ) ? 0xFFFFFFFF : 0 +dst[127:32] := a[127:32] + + +
xmmintrin.h
+ + + Floating Point + SSE + Compare + + + + Compare packed single-precision (32-bit) floating-point elements in "a" and "b" for equality, and store the results in "dst". FOR j := 0 to 3 i := j*32 - IF k[j] - IF imm8[7:0] > 31 - dst[i+31:i] := SignBit - ELSE - dst[i+31:i] := SignExtend(a[i+31:i] >> imm8[7:0]) - FI - ELSE - dst[i+31:i] := src[i+31:i] - FI + dst[i+31:i] := ( a[i+31:i] == b[i+31:i] ) ? 0xFFFFFFFF : 0 ENDFOR -dst[MAX:128] := 0 - -
immintrin.h
+ +
xmmintrin.h
 - - Integer - AVX512VL - AVX512F - Shift - - - - Shift packed 32-bit integers in "a" right by "count" while shifting in sign bits, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Floating Point + SSE + Compare + + + + Compare the lower single-precision (32-bit) floating-point elements in "a" and "b" for less-than, store the result in the lower element of "dst", and copy the upper 3 packed elements from "a" to the upper elements of "dst". + +dst[31:0] := ( a[31:0] < b[31:0] ) ? 0xFFFFFFFF : 0 +dst[127:32] := a[127:32] + + +
xmmintrin.h
+ + + Floating Point + SSE + Compare + + + + Compare packed single-precision (32-bit) floating-point elements in "a" and "b" for less-than, and store the results in "dst". FOR j := 0 to 3 i := j*32 - IF k[j] - IF count[63:0] > 31 - dst[i+31:i] := SignBit - ELSE - dst[i+31:i] := SignExtend(a[i+31:i] >> count[63:0]) - FI - ELSE - dst[i+31:i] := 0 - FI + dst[i+31:i] := ( a[i+31:i] < b[i+31:i] ) ? 0xFFFFFFFF : 0 ENDFOR -dst[MAX:128] := 0 - -
immintrin.h
+ +
xmmintrin.h
 - - Integer - AVX512VL - AVX512F - Shift - - - - Shift packed 32-bit integers in "a" right by "imm8" while shifting in sign bits, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Floating Point + SSE + Compare + + + + Compare the lower single-precision (32-bit) floating-point elements in "a" and "b" for less-than-or-equal, store the result in the lower element of "dst", and copy the upper 3 packed elements from "a" to the upper elements of "dst". + +dst[31:0] := ( a[31:0] <= b[31:0] ) ? 0xFFFFFFFF : 0 +dst[127:32] := a[127:32] + + +
xmmintrin.h
+ + + Floating Point + SSE + Compare + + + + Compare packed single-precision (32-bit) floating-point elements in "a" and "b" for less-than-or-equal, and store the results in "dst". FOR j := 0 to 3 i := j*32 - IF k[j] - IF imm8[7:0] > 31 - dst[i+31:i] := SignBit - ELSE - dst[i+31:i] := SignExtend(a[i+31:i] >> imm8[7:0]) - FI - ELSE - dst[i+31:i] := 0 - FI + dst[i+31:i] := ( a[i+31:i] <= b[i+31:i] ) ? 0xFFFFFFFF : 0 ENDFOR -dst[MAX:128] := 0 - -
immintrin.h
+ +
xmmintrin.h
 - - Integer - AVX512VL - AVX512F - Shift - - - - - Shift packed 64-bit integers in "a" right by "count" while shifting in sign bits, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Floating Point + SSE + Compare + + + + Compare the lower single-precision (32-bit) floating-point elements in "a" and "b" for greater-than, store the result in the lower element of "dst", and copy the upper 3 packed elements from "a" to the upper elements of "dst". -FOR j := 0 to 3 - i := j*64 - IF k[j] - IF count[63:0] > 63 - dst[i+63:i] := SignBit - ELSE - dst[i+63:i] := SignExtend(a[i+63:i] >> count[63:0]) - FI - ELSE - dst[i+63:i] := src[i+63:i] - FI -ENDFOR -dst[MAX:256] := 0 +dst[31:0] := ( a[31:0] > b[31:0] ) ? 0xFFFFFFFF : 0 +dst[127:32] := a[127:32] - -
immintrin.h
+ +
xmmintrin.h
 - - Integer - AVX512VL - AVX512F - Shift - - - - - Shift packed 64-bit integers in "a" right by "imm8" while shifting in sign bits, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Floating Point + SSE + Compare + + + + Compare packed single-precision (32-bit) floating-point elements in "a" and "b" for greater-than, and store the results in "dst". FOR j := 0 to 3 - i := j*64 - IF k[j] - IF imm8[7:0] > 63 - dst[i+63:i] := SignBit - ELSE - dst[i+63:i] := SignExtend(a[i+63:i] << imm8[7:0]) - FI - ELSE - dst[i+63:i] := src[i+63:i] - FI + i := j*32 + dst[i+31:i] := ( a[i+31:i] > b[i+31:i] ) ? 0xFFFFFFFF : 0 ENDFOR -dst[MAX:256] := 0 - -
immintrin.h
+ +
xmmintrin.h
 - - Integer - AVX512VL - AVX512F - Shift - - - - Shift packed 64-bit integers in "a" right by "count" while shifting in sign bits, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Floating Point + SSE + Compare + + + + Compare the lower single-precision (32-bit) floating-point elements in "a" and "b" for greater-than-or-equal, store the result in the lower element of "dst", and copy the upper 3 packed elements from "a" to the upper elements of "dst". -FOR j := 0 to 3 - i := j*64 - IF k[j] - IF count[63:0] > 63 - dst[i+63:i] := SignBit - ELSE - dst[i+63:i] := SignExtend(a[i+63:i] >> count[63:0]) - FI - ELSE - dst[i+63:i] := 0 - FI -ENDFOR -dst[MAX:256] := 0 +dst[31:0] := ( a[31:0] >= b[31:0] ) ? 0xFFFFFFFF : 0 +dst[127:32] := a[127:32] - -
immintrin.h
+ +
xmmintrin.h
 - - Integer - AVX512VL - AVX512F - Shift - - - - Shift packed 64-bit integers in "a" right by "imm8" while shifting in sign bits, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Floating Point + SSE + Compare + + + + Compare packed single-precision (32-bit) floating-point elements in "a" and "b" for greater-than-or-equal, and store the results in "dst". FOR j := 0 to 3 - i := j*64 - IF k[j] - IF imm8[7:0] > 63 - dst[i+63:i] := SignBit - ELSE - dst[i+63:i] := SignExtend(a[i+63:i] << imm8[7:0]) - FI - ELSE - dst[i+63:i] := 0 - FI + i := j*32 + dst[i+31:i] := ( a[i+31:i] >= b[i+31:i] ) ? 0xFFFFFFFF : 0 ENDFOR -dst[MAX:256] := 0 - -
immintrin.h
+ +
xmmintrin.h
 - - Integer - AVX512VL - AVX512F - Shift - - - Shift packed 64-bit integers in "a" right by "count" while shifting in sign bits, and store the results in "dst". + + Floating Point + SSE + Compare + + + + Compare the lower single-precision (32-bit) floating-point elements in "a" and "b" for not-equal, store the result in the lower element of "dst", and copy the upper 3 packed elements from "a" to the upper elements of "dst". -FOR j := 0 to 3 - i := j*64 - IF count[63:0] > 63 - dst[i+63:i] := SignBit - ELSE - dst[i+63:i] := SignExtend(a[i+63:i] >> count[63:0]) - FI -ENDFOR -dst[MAX:256] := 0 +dst[31:0] := ( a[31:0] != b[31:0] ) ? 0xFFFFFFFF : 0 +dst[127:32] := a[127:32] - -
immintrin.h
+ +
xmmintrin.h
 - - Integer - AVX512VL - AVX512F - Shift - - - Shift packed 64-bit integers in "a" right by "imm8" while shifting in sign bits, and store the results in "dst". + + Floating Point + SSE + Compare + + + + Compare packed single-precision (32-bit) floating-point elements in "a" and "b" for not-equal, and store the results in "dst". FOR j := 0 to 3 - i := j*64 - IF imm8[7:0] > 63 - dst[i+63:i] := SignBit - ELSE - dst[i+63:i] := SignExtend(a[i+63:i] << imm8[7:0]) - FI + i := j*32 + dst[i+31:i] := ( a[i+31:i] != b[i+31:i] ) ? 0xFFFFFFFF : 0 ENDFOR -dst[MAX:256] := 0 - -
immintrin.h
+ +
xmmintrin.h
 - - Integer - AVX512VL - AVX512F - Shift - - - - - Shift packed 64-bit integers in "a" right by "count" while shifting in sign bits, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Floating Point + SSE + Compare + + + + Compare the lower single-precision (32-bit) floating-point elements in "a" and "b" for not-less-than, store the result in the lower element of "dst", and copy the upper 3 packed elements from "a" to the upper elements of "dst". -FOR j := 0 to 1 - i := j*64 - IF k[j] - IF count[63:0] > 63 - dst[i+63:i] := SignBit - ELSE - dst[i+63:i] := SignExtend(a[i+63:i] >> count[63:0]) - FI - ELSE - dst[i+63:i] := src[i+63:i] - FI -ENDFOR -dst[MAX:128] := 0 +dst[31:0] := (!( a[31:0] < b[31:0] )) ? 0xFFFFFFFF : 0 +dst[127:32] := a[127:32] - -
immintrin.h
+ +
xmmintrin.h
 - - Integer - AVX512VL - AVX512F - Shift - - - - - Shift packed 64-bit integers in "a" right by "imm8" while shifting in sign bits, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Floating Point + SSE + Compare + + + + Compare packed single-precision (32-bit) floating-point elements in "a" and "b" for not-less-than, and store the results in "dst". -FOR j := 0 to 1 - i := j*64 - IF k[j] - IF imm8[7:0] > 63 - dst[i+63:i] := SignBit - ELSE - dst[i+63:i] := SignExtend(a[i+63:i] << imm8[7:0]) - FI - ELSE - dst[i+63:i] := src[i+63:i] - FI +FOR j := 0 to 3 + i := j*32 + dst[i+31:i] := !( a[i+31:i] < b[i+31:i] ) ? 0xFFFFFFFF : 0 ENDFOR -dst[MAX:128] := 0 - -
immintrin.h
+ +
xmmintrin.h
 - - Integer - AVX512VL - AVX512F - Shift - - - - Shift packed 64-bit integers in "a" right by "count" while shifting in sign bits, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Floating Point + SSE + Compare + + + + Compare the lower single-precision (32-bit) floating-point elements in "a" and "b" for not-less-than-or-equal, store the result in the lower element of "dst", and copy the upper 3 packed elements from "a" to the upper elements of "dst". -FOR j := 0 to 1 - i := j*64 - IF k[j] - IF count[63:0] > 63 - dst[i+63:i] := SignBit - ELSE - dst[i+63:i] := SignExtend(a[i+63:i] >> count[63:0]) - FI - ELSE - dst[i+63:i] := 0 - FI -ENDFOR -dst[MAX:128] := 0 +dst[31:0] := (!( a[31:0] <= b[31:0] )) ? 0xFFFFFFFF : 0 +dst[127:32] := a[127:32] - -
immintrin.h
+ +
xmmintrin.h
 - - Integer - AVX512VL - AVX512F - Shift - - - - Shift packed 64-bit integers in "a" right by "imm8" while shifting in sign bits, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Floating Point + SSE + Compare + + + + Compare packed single-precision (32-bit) floating-point elements in "a" and "b" for not-less-than-or-equal, and store the results in "dst". -FOR j := 0 to 1 - i := j*64 - IF k[j] - IF imm8[7:0] > 63 - dst[i+63:i] := SignBit - ELSE - dst[i+63:i] := SignExtend(a[i+63:i] << imm8[7:0]) - FI - ELSE - dst[i+63:i] := 0 - FI +FOR j := 0 to 3 + i := j*32 + dst[i+31:i] := (!( a[i+31:i] <= b[i+31:i] )) ? 0xFFFFFFFF : 0 ENDFOR -dst[MAX:128] := 0 - -
immintrin.h
+ +
xmmintrin.h
 - - Integer - AVX512VL - AVX512F - Shift - - - Shift packed 64-bit integers in "a" right by "count" while shifting in sign bits, and store the results in "dst". + + Floating Point + SSE + Compare + + + + Compare the lower single-precision (32-bit) floating-point elements in "a" and "b" for not-greater-than, store the result in the lower element of "dst", and copy the upper 3 packed elements from "a" to the upper elements of "dst". -FOR j := 0 to 1 - i := j*64 - IF count[63:0] > 63 - dst[i+63:i] := SignBit - ELSE - dst[i+63:i] := SignExtend(a[i+63:i] >> count[63:0]) - FI -ENDFOR -dst[MAX:128] := 0 +dst[31:0] := (!( a[31:0] > b[31:0] )) ? 0xFFFFFFFF : 0 +dst[127:32] := a[127:32] - -
immintrin.h
+ +
xmmintrin.h
 - - Integer - AVX512VL - AVX512F - Shift - - - Shift packed 64-bit integers in "a" right by "imm8" while shifting in sign bits, and store the results in "dst". + + Floating Point + SSE + Compare + + + + Compare packed single-precision (32-bit) floating-point elements in "a" and "b" for not-greater-than, and store the results in "dst". -FOR j := 0 to 1 - i := j*64 - IF imm8[7:0] > 63 - dst[i+63:i] := SignBit - ELSE - dst[i+63:i] := SignExtend(a[i+63:i] << imm8[7:0]) - FI +FOR j := 0 to 3 + i := j*32 + dst[i+31:i] := (!( a[i+31:i] > b[i+31:i] )) ? 0xFFFFFFFF : 0 ENDFOR -dst[MAX:128] := 0 - -
immintrin.h
+ +
xmmintrin.h
 - - Integer - AVX512VL - AVX512F - Shift - - - - - Shift packed 32-bit integers in "a" right by the amount specified by the corresponding element in "count" while shifting in sign bits, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Floating Point + SSE + Compare + + + + Compare the lower single-precision (32-bit) floating-point elements in "a" and "b" for not-greater-than-or-equal, store the result in the lower element of "dst", and copy the upper 3 packed elements from "a" to the upper elements of "dst". -FOR j := 0 to 7 - i := j*32 - IF k[j] - dst[i+31:i] := SignExtend(a[i+31:i] >> count[i+31:i]) - ELSE - dst[i+31:i] := src[i+31:i] - FI -ENDFOR -dst[MAX:256] := 0 +dst[31:0] := (!( a[31:0] >= b[31:0] )) ? 0xFFFFFFFF : 0 +dst[127:32] := a[127:32] - -
immintrin.h
+ +
xmmintrin.h
 - - Integer - AVX512VL - AVX512F - Shift - - - - Shift packed 32-bit integers in "a" right by the amount specified by the corresponding element in "count" while shifting in sign bits, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Floating Point + SSE + Compare + + + + Compare packed single-precision (32-bit) floating-point elements in "a" and "b" for not-greater-than-or-equal, and store the results in "dst". -FOR j := 0 to 7 +FOR j := 0 to 3 i := j*32 - IF k[j] - dst[i+31:i] := SignExtend(a[i+31:i] >> count[i+31:i]) - ELSE - dst[i+31:i] := 0 - FI + dst[i+31:i] := (!( a[i+31:i] >= b[i+31:i] )) ? 0xFFFFFFFF : 0 ENDFOR -dst[MAX:256] := 0 - -
immintrin.h
+ +
xmmintrin.h
 - - Integer - AVX512VL - AVX512F - Shift - - - - - Shift packed 32-bit integers in "a" right by the amount specified by the corresponding element in "count" while shifting in sign bits, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -FOR j := 0 to 3 + + Floating Point + SSE + Compare + + + + Compare the lower single-precision (32-bit) floating-point elements in "a" and "b" to see if neither is NaN, store the result in the lower element of "dst", and copy the upper 3 packed elements from "a" to the upper elements of "dst". + dst[31:0] := ( a[31:0] != NaN AND b[31:0] != NaN ) ? 0xFFFFFFFF : 0 +dst[127:32] := a[127:32] + + +
xmmintrin.h
+ + + Floating Point + SSE + Compare + + + + Compare packed single-precision (32-bit) floating-point elements in "a" and "b" to see if neither is NaN, and store the results in "dst". + FOR j := 0 to 3 i := j*32 - IF k[j] - dst[i+31:i] := SignExtend(a[i+31:i] >> count[i+31:i]) - ELSE - dst[i+31:i] := src[i+31:i] - FI + dst[i+31:i] := ( a[i+31:i] != NaN AND b[i+31:i] != NaN ) ? 0xFFFFFFFF : 0 ENDFOR -dst[MAX:128] := 0 - -
immintrin.h
+ +
xmmintrin.h
 - - Integer - AVX512VL - AVX512F - Shift - - - - Shift packed 32-bit integers in "a" right by the amount specified by the corresponding element in "count" while shifting in sign bits, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - -FOR j := 0 to 3 + + Floating Point + SSE + Compare + + + + Compare the lower single-precision (32-bit) floating-point elements in "a" and "b" to see if either is NaN, store the result in the lower element of "dst", and copy the upper 3 packed elements from "a" to the upper elements of "dst". + dst[31:0] := ( a[31:0] == NaN OR b[31:0] == NaN ) ? 0xFFFFFFFF : 0 +dst[127:32] := a[127:32] + + +
xmmintrin.h
+ + + Floating Point + SSE + Compare + + + + Compare packed single-precision (32-bit) floating-point elements in "a" and "b" to see if either is NaN, and store the results in "dst". + FOR j := 0 to 3 i := j*32 - IF k[j] - dst[i+31:i] := SignExtend(a[i+31:i] >> count[i+31:i]) - ELSE - dst[i+31:i] := 0 - FI + dst[i+31:i] := ( a[i+31:i] == NaN OR b[i+31:i] == NaN ) ? 0xFFFFFFFF : 0 ENDFOR -dst[MAX:128] := 0 - -
immintrin.h
+ +
xmmintrin.h
 - - Integer - AVX512VL - AVX512F - Shift - - - - - Shift packed 64-bit integers in "a" right by the amount specified by the corresponding element in "count" while shifting in sign bits, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Floating Point + Flag + SSE + Compare + + + + Compare the lower single-precision (32-bit) floating-point element in "a" and "b" for equality, and return the boolean result (0 or 1). -FOR j := 0 to 3 - i := j*64 - IF k[j] - dst[i+63:i] := SignExtend(a[i+63:i] >> count[i+63:i]) - ELSE - dst[i+63:i] := src[i+63:i] - FI -ENDFOR -dst[MAX:256] := 0 +RETURN ( a[31:0] == b[31:0] ) ? 1 : 0 - -
immintrin.h
+ +
xmmintrin.h
 - - Integer - AVX512VL - AVX512F - Shift - - - - Shift packed 64-bit integers in "a" right by the amount specified by the corresponding element in "count" while shifting in sign bits, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Floating Point + Flag + SSE + Compare + + + + Compare the lower single-precision (32-bit) floating-point element in "a" and "b" for less-than, and return the boolean result (0 or 1). -FOR j := 0 to 3 - i := j*64 - IF k[j] - dst[i+63:i] := SignExtend(a[i+63:i] >> count[i+63:i]) - ELSE - dst[i+63:i] := 0 - FI -ENDFOR -dst[MAX:256] := 0 +RETURN ( a[31:0] < b[31:0] ) ? 1 : 0 - -
immintrin.h
+ +
xmmintrin.h
 - - Integer - AVX512VL - AVX512F - Shift - - - Shift packed 64-bit integers in "a" right by the amount specified by the corresponding element in "count" while shifting in sign bits, and store the results in "dst". + + Floating Point + Flag + SSE + Compare + + + + Compare the lower single-precision (32-bit) floating-point element in "a" and "b" for less-than-or-equal, and return the boolean result (0 or 1). -FOR j := 0 to 3 - i := j*64 - dst[i+63:i] := SignExtend(a[i+63:i] >> count[i+63:i]) -ENDFOR -dst[MAX:256] := 0 +RETURN ( a[31:0] <= b[31:0] ) ? 1 : 0 - -
immintrin.h
+ +
xmmintrin.h
 - - Integer - AVX512VL - AVX512F - Shift - - - - - Shift packed 64-bit integers in "a" right by the amount specified by the corresponding element in "count" while shifting in sign bits, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Floating Point + Flag + SSE + Compare + + + + Compare the lower single-precision (32-bit) floating-point element in "a" and "b" for greater-than, and return the boolean result (0 or 1). -FOR j := 0 to 1 - i := j*64 - IF k[j] - dst[i+63:i] := SignExtend(a[i+63:i] >> count[i+63:i]) - ELSE - dst[i+63:i] := src[i+63:i] - FI -ENDFOR -dst[MAX:128] := 0 +RETURN ( a[31:0] > b[31:0] ) ? 1 : 0 - -
immintrin.h
+ +
xmmintrin.h
 - - Integer - AVX512VL - AVX512F - Shift - - - - Shift packed 64-bit integers in "a" right by the amount specified by the corresponding element in "count" while shifting in sign bits, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Floating Point + Flag + SSE + Compare + + + + Compare the lower single-precision (32-bit) floating-point element in "a" and "b" for greater-than-or-equal, and return the boolean result (0 or 1). -FOR j := 0 to 1 - i := j*64 - IF k[j] - dst[i+63:i] := SignExtend(a[i+63:i] >> count[i+63:i]) - ELSE - dst[i+63:i] := 0 - FI -ENDFOR -dst[MAX:128] := 0 +RETURN ( a[31:0] >= b[31:0] ) ? 1 : 0 - -
immintrin.h
+ +
xmmintrin.h
 - - Integer - AVX512VL - AVX512F - Shift - - - Shift packed 64-bit integers in "a" right by the amount specified by the corresponding element in "count" while shifting in sign bits, and store the results in "dst". + + Floating Point + Flag + SSE + Compare + + + + Compare the lower single-precision (32-bit) floating-point element in "a" and "b" for not-equal, and return the boolean result (0 or 1). -FOR j := 0 to 1 - i := j*64 - dst[i+63:i] := SignExtend(a[i+63:i] >> count[i+63:i]) -ENDFOR -dst[MAX:128] := 0 +RETURN ( a[31:0] != b[31:0] ) ? 1 : 0 - -
immintrin.h
+ +
xmmintrin.h
 - - Integer - AVX512VL - AVX512BW - Shift - - - - - Shift packed 16-bit integers in "a" right by the amount specified by the corresponding element in "count" while shifting in sign bits, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Floating Point + Flag + SSE + Compare + + + + Compare the lower single-precision (32-bit) floating-point element in "a" and "b" for equality, and return the boolean result (0 or 1). This instruction will not signal an exception for QNaNs. -FOR j := 0 to 15 - i := j*16 - IF k[j] - dst[i+15:i] := SignExtend(a[i+15:i] >> count[i+15:i]) - ELSE - dst[i+15:i] := src[i+15:i] - FI -ENDFOR -dst[MAX:256] := 0 +RETURN ( a[31:0] == b[31:0] ) ? 1 : 0 - -
immintrin.h
+ +
xmmintrin.h
 - - Integer - AVX512VL - AVX512BW - Shift - - - - Shift packed 16-bit integers in "a" right by the amount specified by the corresponding element in "count" while shifting in sign bits, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Floating Point + Flag + SSE + Compare + + + + Compare the lower single-precision (32-bit) floating-point element in "a" and "b" for less-than, and return the boolean result (0 or 1). This instruction will not signal an exception for QNaNs. -FOR j := 0 to 15 - i := j*16 - IF k[j] - dst[i+15:i] := SignExtend(a[i+15:i] >> count[i+15:i]) - ELSE - dst[i+15:i] := 0 - FI -ENDFOR -dst[MAX:256] := 0 +RETURN ( a[31:0] < b[31:0] ) ? 1 : 0 - -
immintrin.h
+ +
xmmintrin.h
 - - Integer - AVX512VL - AVX512BW - Shift - - - Shift packed 16-bit integers in "a" right by the amount specified by the corresponding element in "count" while shifting in sign bits, and store the results in "dst". + + Floating Point + Flag + SSE + Compare + + + + Compare the lower single-precision (32-bit) floating-point element in "a" and "b" for less-than-or-equal, and return the boolean result (0 or 1). This instruction will not signal an exception for QNaNs. -FOR j := 0 to 15 - i := j*16 - dst[i+15:i] := SignExtend(a[i+15:i] >> count[i+15:i]) -ENDFOR -dst[MAX:256] := 0 +RETURN ( a[31:0] <= b[31:0] ) ? 1 : 0 - -
immintrin.h
+ +
xmmintrin.h
 - - Integer - AVX512BW - Shift - - - - - Shift packed 16-bit integers in "a" right by the amount specified by the corresponding element in "count" while shifting in sign bits, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Floating Point + Flag + SSE + Compare + + + + Compare the lower single-precision (32-bit) floating-point element in "a" and "b" for greater-than, and return the boolean result (0 or 1). This instruction will not signal an exception for QNaNs. -FOR j := 0 to 31 - i := j*16 - IF k[j] - dst[i+15:i] := SignExtend(a[i+15:i] >> count[i+15:i]) - ELSE - dst[i+15:i] := src[i+15:i] - FI -ENDFOR -dst[MAX:512] := 0 +RETURN ( a[31:0] > b[31:0] ) ? 1 : 0 - -
immintrin.h
+ +
xmmintrin.h
+ + + Floating Point + Flag + SSE + Compare + + + + Compare the lower single-precision (32-bit) floating-point element in "a" and "b" for greater-than-or-equal, and return the boolean result (0 or 1). This instruction will not signal an exception for QNaNs. + +RETURN ( a[31:0] >= b[31:0] ) ? 1 : 0 + + +
xmmintrin.h
+ + + Floating Point + Flag + SSE + Compare + + + + Compare the lower single-precision (32-bit) floating-point element in "a" and "b" for not-equal, and return the boolean result (0 or 1). This instruction will not signal an exception for QNaNs. + +RETURN ( a[31:0] != b[31:0] ) ? 1 : 0 + + +
xmmintrin.h
 - + + Floating Point Integer - AVX512BW - Shift - - - - Shift packed 16-bit integers in "a" right by the amount specified by the corresponding element in "count" while shifting in sign bits, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + SSE + Convert + + + Convert the lower single-precision (32-bit) floating-point element in "a" to a 32-bit integer, and store the result in "dst". -FOR j := 0 to 31 - i := j*16 - IF k[j] - dst[i+15:i] := SignExtend(a[i+15:i] >> count[i+15:i]) - ELSE - dst[i+15:i] := 0 - FI -ENDFOR -dst[MAX:512] := 0 +dst[31:0] := Convert_FP32_To_Int32(a[31:0]) - -
immintrin.h
+ +
xmmintrin.h
+ + + Floating Point + SSE + Convert + + + Convert the lower single-precision (32-bit) floating-point element in "a" to a 32-bit integer, and store the result in "dst". + +dst[31:0] := Convert_FP32_To_Int32(a[31:0]) + + +
xmmintrin.h
 - + + Floating Point Integer - AVX512BW - Shift - - - Shift packed 16-bit integers in "a" right by the amount specified by the corresponding element in "count" while shifting in sign bits, and store the results in "dst". + SSE + Convert + + + Convert the lower single-precision (32-bit) floating-point element in "a" to a 64-bit integer, and store the result in "dst". -FOR j := 0 to 31 - i := j*16 - dst[i+15:i] := SignExtend(a[i+15:i] >> count[i+15:i]) -ENDFOR -dst[MAX:512] := 0 +dst[63:0] := Convert_FP32_To_Int64(a[31:0]) - -
immintrin.h
+ +
xmmintrin.h
+ + + Floating Point + SSE + Convert + + + Copy the lower single-precision (32-bit) floating-point element of "a" to "dst". + +dst[31:0] := a[31:0] + + +
xmmintrin.h
 - + + Floating Point Integer - AVX512VL - AVX512BW - Shift - - - - - Shift packed 16-bit integers in "a" right by the amount specified by the corresponding element in "count" while shifting in sign bits, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + SSE + Convert + + + Convert packed single-precision (32-bit) floating-point elements in "a" to packed 32-bit integers, and store the results in "dst". -FOR j := 0 to 7 - i := j*16 - IF k[j] - dst[i+15:i] := SignExtend(a[i+15:i] >> count[i+15:i]) - ELSE - dst[i+15:i] := src[i+15:i] - FI +FOR j := 0 to 1 + i := 32*j + dst[i+31:i] := Convert_FP32_To_Int32(a[i+31:i]) ENDFOR -dst[MAX:128] := 0 - -
immintrin.h
+ +
xmmintrin.h
 - - Integer - AVX512VL - AVX512BW - Shift - - - - Shift packed 16-bit integers in "a" right by the amount specified by the corresponding element in "count" while shifting in sign bits, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Floating Point + SSE + Convert + + + Convert packed single-precision (32-bit) floating-point elements in "a" to packed 32-bit integers, and store the results in "dst". -FOR j := 0 to 7 - i := j*16 - IF k[j] - dst[i+15:i] := SignExtend(a[i+15:i] >> count[i+15:i]) - ELSE - dst[i+15:i] := 0 - FI +FOR j := 0 to 1 + i := 32*j + dst[i+31:i] := Convert_FP32_To_Int32(a[i+31:i]) ENDFOR -dst[MAX:128] := 0 - -
immintrin.h
+ +
xmmintrin.h
 - + + Floating Point Integer - AVX512VL - AVX512BW - Shift - - - Shift packed 16-bit integers in "a" right by the amount specified by the corresponding element in "count" while shifting in sign bits, and store the results in "dst". + SSE + Convert + + + Convert the lower single-precision (32-bit) floating-point element in "a" to a 32-bit integer with truncation, and store the result in "dst". -FOR j := 0 to 7 - i := j*16 - dst[i+15:i] := SignExtend(a[i+15:i] >> count[i+15:i]) -ENDFOR -dst[MAX:128] := 0 +dst[31:0] := Convert_FP32_To_Int32_Truncate(a[31:0]) - -
immintrin.h
+ +
xmmintrin.h
+ + + Floating Point + SSE + Convert + + + Convert the lower single-precision (32-bit) floating-point element in "a" to a 32-bit integer with truncation, and store the result in "dst". + +dst[31:0] := Convert_FP32_To_Int32_Truncate(a[31:0]) + + +
xmmintrin.h
 - + + Floating Point Integer - AVX512VL - AVX512BW - Shift - - - - - Shift packed 16-bit integers in "a" right by "count" while shifting in sign bits, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + SSE + Convert + + + Convert the lower single-precision (32-bit) floating-point element in "a" to a 64-bit integer with truncation, and store the result in "dst". -FOR j := 0 to 15 - i := j*16 - IF k[j] - IF count[63:0] > 15 - dst[i+15:i] := SignBit - ELSE - dst[i+15:i] := SignExtend(a[i+15:i] >> count[63:0]) - FI - ELSE - dst[i+15:i] := src[i+15:i] - FI -ENDFOR -dst[MAX:256] := 0 +dst[63:0] := Convert_FP32_To_Int64_Truncate(a[31:0]) - -
immintrin.h
+ +
xmmintrin.h
 - + + Floating Point Integer - AVX512VL - AVX512BW - Shift - - - - - Shift packed 16-bit integers in "a" right by "imm8" while shifting in sign bits, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + SSE + Convert + + + Convert packed single-precision (32-bit) floating-point elements in "a" to packed 32-bit integers with truncation, and store the results in "dst". -FOR j := 0 to 15 - i := j*16 - IF k[j] - IF imm8[7:0] > 15 - dst[i+15:i] := SignBit - ELSE - dst[i+15:i] := SignExtend(a[i+15:i] >> imm8[7:0]) - FI - ELSE - dst[i+15:i] := src[i+15:i] - FI +FOR j := 0 to 1 + i := 32*j + dst[i+31:i] := Convert_FP32_To_Int32_Truncate(a[i+31:i]) ENDFOR -dst[MAX:256] := 0 - -
immintrin.h
+ +
xmmintrin.h
 - - Integer - AVX512VL - AVX512BW - Shift - - - - Shift packed 16-bit integers in "a" right by "count" while shifting in sign bits, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Floating Point + SSE + Convert + + + Convert packed single-precision (32-bit) floating-point elements in "a" to packed 32-bit integers with truncation, and store the results in "dst". -FOR j := 0 to 15 - i := j*16 - IF k[j] - IF count[63:0] > 15 - dst[i+15:i] := SignBit - ELSE - dst[i+15:i] := SignExtend(a[i+15:i] >> count[63:0]) - FI - ELSE - dst[i+15:i] := 0 - FI +FOR j := 0 to 1 + i := 32*j + dst[i+31:i] := Convert_FP32_To_Int32_Truncate(a[i+31:i]) ENDFOR -dst[MAX:256] := 0 - -
immintrin.h
+ +
xmmintrin.h
 - + + Floating Point Integer - AVX512VL - AVX512BW - Shift - - - - Shift packed 16-bit integers in "a" right by "imm8" while shifting in sign bits, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + SSE + Convert + + + Convert packed single-precision (32-bit) floating-point elements in "a" to packed 16-bit integers, and store the results in "dst". Note: this intrinsic will generate 0x7FFF, rather than 0x8000, for input values between 0x7FFF and 0x7FFFFFFF. -FOR j := 0 to 15 - i := j*16 - IF k[j] - IF imm8[7:0] > 15 - dst[i+15:i] := SignBit - ELSE - dst[i+15:i] := SignExtend(a[i+15:i] >> imm8[7:0]) - FI +FOR j := 0 to 3 + i := 16*j + k := 32*j + IF a[k+31:k] >= FP32(0x7FFF) && a[k+31:k] <= FP32(0x7FFFFFFF) + dst[i+15:i] := 0x7FFF ELSE - dst[i+15:i] := 0 + dst[i+15:i] := Convert_FP32_To_Int16(a[k+31:k]) FI ENDFOR -dst[MAX:256] := 0 - -
immintrin.h
+
xmmintrin.h
 - + + Floating Point Integer - AVX512BW - Shift - - - - - Shift packed 16-bit integers in "a" right by "count" while shifting in sign bits, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + SSE + Convert + + + Convert packed single-precision (32-bit) floating-point elements in "a" to packed 8-bit integers, and store the results in lower 4 elements of "dst". Note: this intrinsic will generate 0x7F, rather than 0x80, for input values between 0x7F and 0x7FFFFFFF. -FOR j := 0 to 31 - i := j*16 - IF k[j] - IF count[63:0] > 15 - dst[i+15:i] := SignBit - ELSE - dst[i+15:i] := SignExtend(a[i+15:i] >> count[63:0]) - FI +FOR j := 0 to 3 + i := 8*j + k := 32*j + IF a[k+31:k] >= FP32(0x7F) && a[k+31:k] <= FP32(0x7FFFFFFF) + dst[i+7:i] := 0x7F ELSE - dst[i+15:i] := src[i+15:i] + dst[i+7:i] := Convert_FP32_To_Int8(a[k+31:k]) FI ENDFOR -dst[MAX:512] := 0 - -
immintrin.h
+
xmmintrin.h
 - - Integer - AVX512BW - Shift - - - - - Shift packed 16-bit integers in "a" right by "imm8" while shifting in sign bits, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Floating Point + SSE + Set + + + Copy single-precision (32-bit) floating-point element "a" to the lower element of "dst", and zero the upper 3 elements. -FOR j := 0 to 31 - i := j*16 - IF k[j] - IF imm8[7:0] > 15 - dst[i+15:i] := SignBit - ELSE - dst[i+15:i] := SignExtend(a[i+15:i] >> imm8[7:0]) - FI - ELSE - dst[i+15:i] := src[i+15:i] - FI -ENDFOR -dst[MAX:512] := 0 +dst[31:0] := a[31:0] +dst[127:32] := 0 - -
immintrin.h
+
xmmintrin.h
 - - Integer - AVX512BW - Shift - - - - Shift packed 16-bit integers in "a" right by "count" while shifting in sign bits, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Floating Point + SSE + Set + + + Broadcast single-precision (32-bit) floating-point value "a" to all elements of "dst". -FOR j := 0 to 31 - i := j*16 - IF k[j] - IF count[63:0] > 15 - dst[i+15:i] := SignBit - ELSE - dst[i+15:i] := SignExtend(a[i+15:i] >> count[63:0]) - FI - ELSE - dst[i+15:i] := 0 - FI +FOR j := 0 to 3 + i := j*32 + dst[i+31:i] := a[31:0] ENDFOR -dst[MAX:512] := 0 - -
immintrin.h
+
xmmintrin.h
 - - Integer - AVX512BW - Shift - - - - Shift packed 16-bit integers in "a" right by "imm8" while shifting in sign bits, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Floating Point + SSE + Set + + + Broadcast single-precision (32-bit) floating-point value "a" to all elements of "dst". -FOR j := 0 to 31 - i := j*16 - IF k[j] - IF imm8[7:0] > 15 - dst[i+15:i] := SignBit - ELSE - dst[i+15:i] := SignExtend(a[i+15:i] >> imm8[7:0]) - FI - ELSE - dst[i+15:i] := 0 - FI +FOR j := 0 to 3 + i := j*32 + dst[i+31:i] := a[31:0] ENDFOR -dst[MAX:512] := 0 - -
immintrin.h
+
xmmintrin.h
 - - Integer - AVX512BW - Shift - - - Shift packed 16-bit integers in "a" right by "count" while shifting in sign bits, and store the results in "dst". + + Floating Point + SSE + Set + + + + + + Set packed single-precision (32-bit) floating-point elements in "dst" with the supplied values. -FOR j := 0 to 31 - i := j*16 - IF count[63:0] > 15 - dst[i+15:i] := SignBit - ELSE - dst[i+15:i] := SignExtend(a[i+15:i] >> count[63:0]) - FI -ENDFOR -dst[MAX:512] := 0 +dst[31:0] := e0 +dst[63:32] := e1 +dst[95:64] := e2 +dst[127:96] := e3 - -
immintrin.h
+
xmmintrin.h
 - - Integer - AVX512BW - Shift - - - Shift packed 16-bit integers in "a" right by "imm8" while shifting in sign bits, and store the results in "dst". + + Floating Point + SSE + Set + + + + + + Set packed single-precision (32-bit) floating-point elements in "dst" with the supplied values in reverse order. -FOR j := 0 to 31 - i := j*16 - IF imm8[7:0] > 15 - dst[i+15:i] := SignBit - ELSE - dst[i+15:i] := SignExtend(a[i+15:i] >> imm8[7:0]) - FI -ENDFOR -dst[MAX:512] := 0 +dst[31:0] := e3 +dst[63:32] := e2 +dst[95:64] := e1 +dst[127:96] := e0 - -
immintrin.h
+
xmmintrin.h
 - - Integer - AVX512VL - AVX512BW - Shift - - - - - Shift packed 16-bit integers in "a" right by "count" while shifting in sign bits, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Floating Point + SSE + Set + + + Return vector of type __m128 with all elements set to zero. -FOR j := 0 to 7 - i := j*16 - IF k[j] - IF count[63:0] > 15 - dst[i+15:i] := SignBit - ELSE - dst[i+15:i] := SignExtend(a[i+15:i] >> count[63:0]) - FI - ELSE - dst[i+15:i] := src[i+15:i] - FI -ENDFOR -dst[MAX:128] := 0 +dst[MAX:0] := 0 - -
immintrin.h
+ +
xmmintrin.h
 - + Integer - AVX512VL - AVX512BW - Shift - - - - - Shift packed 16-bit integers in "a" right by "imm8" while shifting in sign bits, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + SSE + Load + + + + Load 2 single-precision (32-bit) floating-point elements from memory into the upper 2 elements of "dst", and copy the lower 2 elements from "a" to "dst". "mem_addr" does not need to be aligned on any particular boundary. -FOR j := 0 to 7 - i := j*16 - IF k[j] - IF imm8[7:0] > 15 - dst[i+15:i] := SignBit - ELSE - dst[i+15:i] := SignExtend(a[i+15:i] >> imm8[7:0]) - FI - ELSE - dst[i+15:i] := src[i+15:i] - FI -ENDFOR -dst[MAX:128] := 0 +dst[31:0] := a[31:0] +dst[63:32] := a[63:32] +dst[95:64] := MEM[mem_addr+31:mem_addr] +dst[127:96] := MEM[mem_addr+63:mem_addr+32] - -
immintrin.h
+ +
xmmintrin.h
 - + Integer - AVX512VL - AVX512BW - Shift - - - - Shift packed 16-bit integers in "a" right by "count" while shifting in sign bits, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + SSE + Load + + + + Load 2 single-precision (32-bit) floating-point elements from memory into the lower 2 elements of "dst", and copy the upper 2 elements from "a" to "dst". "mem_addr" does not need to be aligned on any particular boundary. -FOR j := 0 to 7 - i := j*16 - IF k[j] - IF count[63:0] > 15 - dst[i+15:i] := SignBit - ELSE - dst[i+15:i] := SignExtend(a[i+15:i] >> count[63:0]) - FI - ELSE - dst[i+15:i] := 0 - FI -ENDFOR -dst[MAX:128] := 0 +dst[31:0] := MEM[mem_addr+31:mem_addr] +dst[63:32] := MEM[mem_addr+63:mem_addr+32] +dst[95:64] := a[95:64] +dst[127:96] := a[127:96] - -
immintrin.h
+ +
xmmintrin.h
 - - Integer - AVX512VL - AVX512BW - Shift - - - - Shift packed 16-bit integers in "a" right by "imm8" while shifting in sign bits, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Floating Point + SSE + Load + + + Load a single-precision (32-bit) floating-point element from memory into the lower of "dst", and zero the upper 3 elements. "mem_addr" does not need to be aligned on any particular boundary. -FOR j := 0 to 7 - i := j*16 - IF k[j] - IF imm8[7:0] > 15 - dst[i+15:i] := SignBit - ELSE - dst[i+15:i] := SignExtend(a[i+15:i] >> imm8[7:0]) - FI - ELSE - dst[i+15:i] := 0 - FI -ENDFOR -dst[MAX:128] := 0 +dst[31:0] := MEM[mem_addr+31:mem_addr] +dst[127:32] := 0 - -
immintrin.h
+ +
xmmintrin.h
 - - Integer - AVX512VL - AVX512F - Shift - - - - - Shift packed 32-bit integers in "a" right by "count" while shifting in zeros, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Floating Point + SSE + Load + + + Load a single-precision (32-bit) floating-point element from memory into all elements of "dst". -FOR j := 0 to 7 - i := j*32 - IF k[j] - IF count[63:0] > 31 - dst[i+31:i] := 0 - ELSE - dst[i+31:i] := ZeroExtend(a[i+31:i] >> count[63:0]) - FI - ELSE - dst[i+31:i] := src[i+31:i] - FI -ENDFOR -dst[MAX:256] := 0 +dst[31:0] := MEM[mem_addr+31:mem_addr] +dst[63:32] := MEM[mem_addr+31:mem_addr] +dst[95:64] := MEM[mem_addr+31:mem_addr] +dst[127:96] := MEM[mem_addr+31:mem_addr] - -
immintrin.h
+
xmmintrin.h
 - - Integer - AVX512VL - AVX512F - Shift - - - - - Shift packed 32-bit integers in "a" right by "imm8" while shifting in zeros, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Floating Point + SSE + Load + + + Load a single-precision (32-bit) floating-point element from memory into all elements of "dst". -FOR j := 0 to 7 - i := j*32 - IF k[j] - IF imm8[7:0] > 31 - dst[i+31:i] := 0 - ELSE - dst[i+31:i] := ZeroExtend(a[i+31:i] >> imm8[7:0]) - FI - ELSE - dst[i+31:i] := src[i+31:i] - FI -ENDFOR -dst[MAX:256] := 0 +dst[31:0] := MEM[mem_addr+31:mem_addr] +dst[63:32] := MEM[mem_addr+31:mem_addr] +dst[95:64] := MEM[mem_addr+31:mem_addr] +dst[127:96] := MEM[mem_addr+31:mem_addr] - -
immintrin.h
+
xmmintrin.h
 - - Integer - AVX512VL - AVX512F - Shift - - - - Shift packed 32-bit integers in "a" right by "count" while shifting in zeros, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Floating Point + SSE + Load + + + Load 128-bits (composed of 4 packed single-precision (32-bit) floating-point elements) from memory into "dst". + "mem_addr" must be aligned on a 16-byte boundary or a general-protection exception may be generated. -FOR j := 0 to 7 - i := j*32 - IF k[j] - IF count[63:0] > 31 - dst[i+31:i] := 0 - ELSE - dst[i+31:i] := ZeroExtend(a[i+31:i] >> count[63:0]) - FI - ELSE - dst[i+31:i] := 0 - FI -ENDFOR -dst[MAX:256] := 0 +dst[127:0] := MEM[mem_addr+127:mem_addr] - -
immintrin.h
+ +
xmmintrin.h
 - - Integer - AVX512VL - AVX512F - Shift - - - - Shift packed 32-bit integers in "a" right by "imm8" while shifting in zeros, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Floating Point + SSE + Load + + + Load 128-bits (composed of 4 packed single-precision (32-bit) floating-point elements) from memory into "dst". + "mem_addr" does not need to be aligned on any particular boundary. -FOR j := 0 to 7 - i := j*32 - IF k[j] - IF imm8[7:0] > 31 - dst[i+31:i] := 0 - ELSE - dst[i+31:i] := ZeroExtend(a[i+31:i] >> imm8[7:0]) - FI - ELSE - dst[i+31:i] := 0 - FI -ENDFOR -dst[MAX:256] := 0 +dst[127:0] := MEM[mem_addr+127:mem_addr] - -
immintrin.h
+ +
xmmintrin.h
 - - Integer - AVX512VL - AVX512F - Shift - - - - - Shift packed 32-bit integers in "a" right by "count" while shifting in zeros, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Floating Point + SSE + Load + + + Load 4 single-precision (32-bit) floating-point elements from memory into "dst" in reverse order. mem_addr must be aligned on a 16-byte boundary or a general-protection exception may be generated. -FOR j := 0 to 3 - i := j*32 - IF k[j] - IF count[63:0] > 31 - dst[i+31:i] := 0 - ELSE - dst[i+31:i] := ZeroExtend(a[i+31:i] >> count[63:0]) - FI - ELSE - dst[i+31:i] := src[i+31:i] - FI -ENDFOR -dst[MAX:128] := 0 +dst[31:0] := MEM[mem_addr+127:mem_addr+96] +dst[63:32] := MEM[mem_addr+95:mem_addr+64] +dst[95:64] := MEM[mem_addr+63:mem_addr+32] +dst[127:96] := MEM[mem_addr+31:mem_addr] - -
immintrin.h
+
xmmintrin.h
 - - Integer - AVX512VL - AVX512F - Shift - - - - - Shift packed 32-bit integers in "a" right by "imm8" while shifting in zeros, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Floating Point + SSE + Store + + + + Store 128-bits (composed of 4 packed single-precision (32-bit) floating-point elements) from "a" into memory using a non-temporal memory hint. + "mem_addr" must be aligned on a 16-byte boundary or a general-protection exception may be generated. -FOR j := 0 to 3 - i := j*32 - IF k[j] - IF imm8[7:0] > 31 - dst[i+31:i] := 0 - ELSE - dst[i+31:i] := ZeroExtend(a[i+31:i] >> imm8[7:0]) - FI - ELSE - dst[i+31:i] := src[i+31:i] - FI -ENDFOR -dst[MAX:128] := 0 +MEM[mem_addr+127:mem_addr] := a[127:0] - -
immintrin.h
+ +
xmmintrin.h
 - + Integer - AVX512VL - AVX512F - Shift - - - - Shift packed 32-bit integers in "a" right by "count" while shifting in zeros, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + SSE + Store + + + + Store the upper 2 single-precision (32-bit) floating-point elements from "a" into memory. -FOR j := 0 to 3 - i := j*32 - IF k[j] - IF count[63:0] > 31 - dst[i+31:i] := 0 - ELSE - dst[i+31:i] := ZeroExtend(a[i+31:i] >> count[63:0]) - FI - ELSE - dst[i+31:i] := 0 - FI -ENDFOR -dst[MAX:128] := 0 +MEM[mem_addr+31:mem_addr] := a[95:64] +MEM[mem_addr+63:mem_addr+32] := a[127:96] - -
immintrin.h
+ +
xmmintrin.h
 - + Integer - AVX512VL - AVX512F - Shift - - - - Shift packed 32-bit integers in "a" right by "imm8" while shifting in zeros, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + SSE + Store + + + + Store the lower 2 single-precision (32-bit) floating-point elements from "a" into memory. -FOR j := 0 to 3 - i := j*32 - IF k[j] - IF imm8[7:0] > 31 - dst[i+31:i] := 0 - ELSE - dst[i+31:i] := ZeroExtend(a[i+31:i] >> imm8[7:0]) - FI - ELSE - dst[i+31:i] := 0 - FI -ENDFOR -dst[MAX:128] := 0 +MEM[mem_addr+31:mem_addr] := a[31:0] +MEM[mem_addr+63:mem_addr+32] := a[63:32] - -
immintrin.h
+ +
xmmintrin.h
 - - Integer - AVX512BW - Shift - - - Shift 128-bit lanes in "a" right by "imm8" bytes while shifting in zeros, and store the results in "dst". + + Floating Point + SSE + Store + + + + Store the lower single-precision (32-bit) floating-point element from "a" into memory. "mem_addr" does not need to be aligned on any particular boundary. -tmp := imm8[7:0] -IF tmp > 15 - tmp := 16 -FI -dst[127:0] := a[127:0] >> (tmp*8) -dst[255:128] := a[255:128] >> (tmp*8) -dst[383:256] := a[383:256] >> (tmp*8) -dst[511:384] := a[511:384] >> (tmp*8) -dst[MAX:512] := 0 +MEM[mem_addr+31:mem_addr] := a[31:0] + + +
xmmintrin.h
+ + + Floating Point + SSE + Store + + + + Store the lower single-precision (32-bit) floating-point element from "a" into 4 contiguous elements in memory. "mem_addr" must be aligned on a 16-byte boundary or a general-protection exception may be generated. + +MEM[mem_addr+31:mem_addr] := a[31:0] +MEM[mem_addr+63:mem_addr+32] := a[31:0] +MEM[mem_addr+95:mem_addr+64] := a[31:0] +MEM[mem_addr+127:mem_addr+96] := a[31:0] + +
xmmintrin.h
+ + + Floating Point + SSE + Store + + + + Store the lower single-precision (32-bit) floating-point element from "a" into 4 contiguous elements in memory. "mem_addr" must be aligned on a 16-byte boundary or a general-protection exception may be generated. + +MEM[mem_addr+31:mem_addr] := a[31:0] +MEM[mem_addr+63:mem_addr+32] := a[31:0] +MEM[mem_addr+95:mem_addr+64] := a[31:0] +MEM[mem_addr+127:mem_addr+96] := a[31:0] + +
xmmintrin.h
+ + + Floating Point + SSE + Store + + + + Store 128-bits (composed of 4 packed single-precision (32-bit) floating-point elements) from "a" into memory. + "mem_addr" must be aligned on a 16-byte boundary or a general-protection exception may be generated. + +MEM[mem_addr+127:mem_addr] := a[127:0] + + +
xmmintrin.h
+ + + Floating Point + SSE + Store + + + + Store 128-bits (composed of 4 packed single-precision (32-bit) floating-point elements) from "a" into memory. + "mem_addr" does not need to be aligned on any particular boundary. + +MEM[mem_addr+127:mem_addr] := a[127:0] + + +
xmmintrin.h
+ + + Floating Point + SSE + Store + + + + Store 4 single-precision (32-bit) floating-point elements from "a" into memory in reverse order. + "mem_addr" must be aligned on a 16-byte boundary or a general-protection exception may be generated. + +MEM[mem_addr+31:mem_addr] := a[127:96] +MEM[mem_addr+63:mem_addr+32] := a[95:64] +MEM[mem_addr+95:mem_addr+64] := a[63:32] +MEM[mem_addr+127:mem_addr+96] := a[31:0] + + +
xmmintrin.h
+ + + Floating Point + SSE + Move + + + + Move the lower single-precision (32-bit) floating-point element from "b" to the lower element of "dst", and copy the upper 3 packed elements from "a" to the upper elements of "dst". + +dst[31:0] := b[31:0] +dst[127:32] := a[127:32] + + +
xmmintrin.h
+ + + Floating Point + SSE + Swizzle + + + + + Shuffle single-precision (32-bit) floating-point elements in "a" using the control in "imm8", and store the results in "dst". + +DEFINE SELECT4(src, control) { + CASE(control[1:0]) OF + 0: tmp[31:0] := src[31:0] + 1: tmp[31:0] := src[63:32] + 2: tmp[31:0] := src[95:64] + 3: tmp[31:0] := src[127:96] + ESAC + RETURN tmp[31:0] +} +dst[31:0] := SELECT4(a[127:0], imm8[1:0]) +dst[63:32] := SELECT4(a[127:0], imm8[3:2]) +dst[95:64] := SELECT4(b[127:0], imm8[5:4]) +dst[127:96] := SELECT4(b[127:0], imm8[7:6]) - -
immintrin.h
+ +
xmmintrin.h
 - - Integer - AVX512VL - AVX512F - Shift - - - - - Shift packed 64-bit integers in "a" right by "count" while shifting in zeros, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Floating Point + SSE + Swizzle + + + + Unpack and interleave single-precision (32-bit) floating-point elements from the high half "a" and "b", and store the results in "dst". -FOR j := 0 to 3 - i := j*64 - IF k[j] - IF count[63:0] > 63 - dst[i+63:i] := 0 - ELSE - dst[i+63:i] := ZeroExtend(a[i+63:i] >> count[63:0]) - FI - ELSE - dst[i+63:i] := src[i+63:i] - FI -ENDFOR -dst[MAX:256] := 0 +DEFINE INTERLEAVE_HIGH_DWORDS(src1[127:0], src2[127:0]) { + dst[31:0] := src1[95:64] + dst[63:32] := src2[95:64] + dst[95:64] := src1[127:96] + dst[127:96] := src2[127:96] + RETURN dst[127:0] +} +dst[127:0] := INTERLEAVE_HIGH_DWORDS(a[127:0], b[127:0]) - -
immintrin.h
+ +
xmmintrin.h
 - - Integer - AVX512VL - AVX512F - Shift - - - - - Shift packed 64-bit integers in "a" right by "imm8" while shifting in zeros, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Floating Point + SSE + Swizzle + + + + Unpack and interleave single-precision (32-bit) floating-point elements from the low half of "a" and "b", and store the results in "dst". -FOR j := 0 to 3 - i := j*64 - IF k[j] - IF imm8[7:0] > 63 - dst[i+63:i] := 0 - ELSE - dst[i+63:i] := ZeroExtend(a[i+63:i] >> imm8[7:0]) - FI - ELSE - dst[i+63:i] := src[i+63:i] - FI -ENDFOR -dst[MAX:256] := 0 +DEFINE INTERLEAVE_DWORDS(src1[127:0], src2[127:0]) { + dst[31:0] := src1[31:0] + dst[63:32] := src2[31:0] + dst[95:64] := src1[63:32] + dst[127:96] := src2[63:32] + RETURN dst[127:0] +} +dst[127:0] := INTERLEAVE_DWORDS(a[127:0], b[127:0]) - -
immintrin.h
+ +
xmmintrin.h
 - - Integer - AVX512VL - AVX512F - Shift - - - - Shift packed 64-bit integers in "a" right by "count" while shifting in zeros, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Floating Point + SSE + Move + + + + Move the upper 2 single-precision (32-bit) floating-point elements from "b" to the lower 2 elements of "dst", and copy the upper 2 elements from "a" to the upper 2 elements of "dst". -FOR j := 0 to 3 - i := j*64 - IF k[j] - IF count[63:0] > 63 - dst[i+63:i] := 0 - ELSE - dst[i+63:i] := ZeroExtend(a[i+63:i] >> count[63:0]) - FI - ELSE - dst[i+63:i] := 0 - FI -ENDFOR -dst[MAX:256] := 0 +dst[31:0] := b[95:64] +dst[63:32] := b[127:96] +dst[95:64] := a[95:64] +dst[127:96] := a[127:96] - -
immintrin.h
+ +
xmmintrin.h
 - - Integer - AVX512VL - AVX512F - Shift - - - - Shift packed 64-bit integers in "a" right by "imm8" while shifting in zeros, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Floating Point + SSE + Move + + + + Move the lower 2 single-precision (32-bit) floating-point elements from "b" to the upper 2 elements of "dst", and copy the lower 2 elements from "a" to the lower 2 elements of "dst". -FOR j := 0 to 3 - i := j*64 - IF k[j] - IF imm8[7:0] > 63 - dst[i+63:i] := 0 - ELSE - dst[i+63:i] := ZeroExtend(a[i+63:i] >> imm8[7:0]) - FI - ELSE - dst[i+63:i] := 0 - FI -ENDFOR -dst[MAX:256] := 0 +dst[31:0] := a[31:0] +dst[63:32] := a[63:32] +dst[95:64] := b[31:0] +dst[127:96] := b[63:32] - -
immintrin.h
+ +
xmmintrin.h
 - - Integer - AVX512VL - AVX512F - Shift - - - - - Shift packed 64-bit integers in "a" right by "count" while shifting in zeros, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Floating Point + SSE + Miscellaneous + + + Set each bit of mask "dst" based on the most significant bit of the corresponding packed single-precision (32-bit) floating-point element in "a". -FOR j := 0 to 1 - i := j*64 - IF k[j] - IF count[63:0] > 63 - dst[i+63:i] := 0 - ELSE - dst[i+63:i] := ZeroExtend(a[i+63:i] >> count[63:0]) - FI +FOR j := 0 to 3 + i := j*32 + IF a[i+31] + dst[j] := 1 ELSE - dst[i+63:i] := src[i+63:i] + dst[j] := 0 FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:4] := 0 - + +
xmmintrin.h
+ + + SSE + General Support + + + + Allocate "size" bytes of memory, aligned to the alignment specified in "align", and return a pointer to the allocated memory. "_mm_free" should be used to free memory that is allocated with "_mm_malloc". +
xmmintrin.h
+ + + SSE + General Support + + + Free aligned memory that was allocated with "_mm_malloc". +
xmmintrin.h
+ + + Floating Point + SSE + General Support + + + Return vector of type __m128 with undefined elements.
immintrin.h
 - - Integer - AVX512VL - AVX512F - Shift - - - - - Shift packed 64-bit integers in "a" right by "imm8" while shifting in zeros, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Floating Point + SSE + Trigonometry + + + Compute the inverse cosine of packed double-precision (64-bit) floating-point elements in "a" expressed in radians, and store the results in "dst". FOR j := 0 to 1 i := j*64 - IF k[j] - IF imm8[7:0] > 63 - dst[i+63:i] := 0 - ELSE - dst[i+63:i] := ZeroExtend(a[i+63:i] >> imm8[7:0]) - FI - ELSE - dst[i+63:i] := src[i+63:i] - FI + dst[i+63:i] := ACOS(a[i+63:i]) ENDFOR dst[MAX:128] := 0 -
immintrin.h
 - - Integer - AVX512VL - AVX512F - Shift - - - - Shift packed 64-bit integers in "a" right by "count" while shifting in zeros, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Floating Point + SSE + Trigonometry + + + Compute the inverse cosine of packed single-precision (32-bit) floating-point elements in "a" expressed in radians, and store the results in "dst". -FOR j := 0 to 1 - i := j*64 - IF k[j] - IF count[63:0] > 63 - dst[i+63:i] := 0 - ELSE - dst[i+63:i] := ZeroExtend(a[i+63:i] >> count[63:0]) - FI - ELSE - dst[i+63:i] := 0 - FI +FOR j := 0 to 3 + i := j*32 + dst[i+31:i] := ACOS(a[i+31:i]) ENDFOR dst[MAX:128] := 0 -
immintrin.h
 - - Integer - AVX512VL - AVX512F - Shift - - - - Shift packed 64-bit integers in "a" right by "imm8" while shifting in zeros, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Floating Point + SSE + Trigonometry + + + Compute the inverse hyperbolic cosine of packed double-precision (64-bit) floating-point elements in "a" expressed in radians, and store the results in "dst". FOR j := 0 to 1 i := j*64 - IF k[j] - IF imm8[7:0] > 63 - dst[i+63:i] := 0 - ELSE - dst[i+63:i] := ZeroExtend(a[i+63:i] >> imm8[7:0]) - FI - ELSE - dst[i+63:i] := 0 - FI + dst[i+63:i] := ACOSH(a[i+63:i]) ENDFOR dst[MAX:128] := 0 - -
immintrin.h
- - - Integer - AVX512VL - AVX512F - Shift - - - - - Shift packed 32-bit integers in "a" right by the amount specified by the corresponding element in "count" while shifting in zeros, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -FOR j := 0 to 7 - i := j*32 - IF k[j] - dst[i+31:i] := ZeroExtend(a[i+31:i] >> count[i+31:i]) - ELSE - dst[i+31:i] := src[i+31:i] - FI -ENDFOR -dst[MAX:256] := 0 - -
immintrin.h
 - - Integer - AVX512VL - AVX512F - Shift - - - - Shift packed 32-bit integers in "a" right by the amount specified by the corresponding element in "count" while shifting in zeros, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Floating Point + SSE + Trigonometry + + + Compute the inverse hyperbolic cosine of packed single-precision (32-bit) floating-point elements in "a" expressed in radians, and store the results in "dst". -FOR j := 0 to 7 +FOR j := 0 to 3 i := j*32 - IF k[j] - dst[i+31:i] := ZeroExtend(a[i+31:i] >> count[i+31:i]) - ELSE - dst[i+31:i] := 0 - FI + dst[i+31:i] := ACOSH(a[i+31:i]) ENDFOR -dst[MAX:256] := 0 +dst[MAX:128] := 0 -
immintrin.h
 - - Integer - AVX512VL - AVX512F - Shift - - - - - Shift packed 32-bit integers in "a" right by the amount specified by the corresponding element in "count" while shifting in zeros, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Floating Point + SSE + Trigonometry + + + Compute the inverse sine of packed double-precision (64-bit) floating-point elements in "a" expressed in radians, and store the results in "dst". -FOR j := 0 to 3 - i := j*32 - IF k[j] - dst[i+31:i] := ZeroExtend(a[i+31:i] >> count[i+31:i]) - ELSE - dst[i+31:i] := src[i+31:i] - FI +FOR j := 0 to 1 + i := j*64 + dst[i+63:i] := ASIN(a[i+63:i]) ENDFOR dst[MAX:128] := 0 -
immintrin.h
 - - Integer - AVX512VL - AVX512F - Shift - - - - Shift packed 32-bit integers in "a" right by the amount specified by the corresponding element in "count" while shifting in zeros, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Floating Point + SSE + Trigonometry + + + Compute the inverse sine of packed single-precision (32-bit) floating-point elements in "a" expressed in radians, and store the results in "dst". FOR j := 0 to 3 i := j*32 - IF k[j] - dst[i+31:i] := ZeroExtend(a[i+31:i] >> count[i+31:i]) - ELSE - dst[i+31:i] := 0 - FI + dst[i+31:i] := ASIN(a[i+31:i]) ENDFOR dst[MAX:128] := 0 -
immintrin.h
 - - Integer - AVX512VL - AVX512F - Shift - - - - - Shift packed 64-bit integers in "a" right by the amount specified by the corresponding element in "count" while shifting in zeros, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Floating Point + SSE + Trigonometry + + + Compute the inverse hyperbolic sine of packed double-precision (64-bit) floating-point elements in "a" expressed in radians, and store the results in "dst". -FOR j := 0 to 3 +FOR j := 0 to 1 i := j*64 - IF k[j] - dst[i+63:i] := ZeroExtend(a[i+63:i] >> count[i+63:i]) - ELSE - dst[i+63:i] := src[i+63:i] - FI + dst[i+63:i] := ASINH(a[i+63:i]) ENDFOR -dst[MAX:256] := 0 +dst[MAX:128] := 0 -
immintrin.h
 - - Integer - AVX512VL - AVX512F - Shift - - - - Shift packed 64-bit integers in "a" right by the amount specified by the corresponding element in "count" while shifting in zeros, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Floating Point + SSE + Trigonometry + + + Compute the inverse hyperbolic sine of packed single-precision (32-bit) floating-point elements in "a" expressed in radians, and store the results in "dst". FOR j := 0 to 3 - i := j*64 - IF k[j] - dst[i+63:i] := ZeroExtend(a[i+63:i] >> count[i+63:i]) - ELSE - dst[i+63:i] := 0 - FI + i := j*32 + dst[i+31:i] := ASINH(a[i+31:i]) ENDFOR -dst[MAX:256] := 0 +dst[MAX:128] := 0 -
immintrin.h
 - - Integer - AVX512VL - AVX512F - Shift - - - - - Shift packed 64-bit integers in "a" right by the amount specified by the corresponding element in "count" while shifting in zeros, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Floating Point + SSE + Trigonometry + + + Compute the inverse tangent of packed double-precision (64-bit) floating-point elements in "a" expressed in radians, and store the results in "dst". FOR j := 0 to 1 i := j*64 - IF k[j] - dst[i+63:i] := ZeroExtend(a[i+63:i] >> count[i+63:i]) - ELSE - dst[i+63:i] := src[i+63:i] - FI + dst[i+63:i] := ATAN(a[i+63:i]) ENDFOR dst[MAX:128] := 0 -
immintrin.h
 - - Integer - AVX512VL - AVX512F - Shift - - - - Shift packed 64-bit integers in "a" right by the amount specified by the corresponding element in "count" while shifting in zeros, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Floating Point + SSE + Trigonometry + + + Compute the inverse tangent of packed single-precision (32-bit) floating-point elements in "a" expressed in radians, and store the results in "dst". -FOR j := 0 to 1 - i := j*64 - IF k[j] - dst[i+63:i] := ZeroExtend(a[i+63:i] >> count[i+63:i]) - ELSE - dst[i+63:i] := 0 - FI +FOR j := 0 to 3 + i := j*32 + dst[i+31:i] := ATAN(a[i+31:i]) ENDFOR dst[MAX:128] := 0 -
immintrin.h
 - - Integer - AVX512VL - AVX512BW - Shift - - - - - Shift packed 16-bit integers in "a" right by the amount specified by the corresponding element in "count" while shifting in zeros, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Floating Point + SSE + Trigonometry + + + + Compute the inverse tangent of packed double-precision (64-bit) floating-point elements in "a" divided by packed elements in "b", and store the results in "dst" expressed in radians. -FOR j := 0 to 15 - i := j*16 - IF k[j] - dst[i+15:i] := ZeroExtend(a[i+15:i] >> count[i+63:i]) - ELSE - dst[i+15:i] := src[i+15:i] - FI +FOR j := 0 to 1 + i := j*64 + dst[i+63:i] := ATAN2(a[i+63:i], b[i+63:i]) ENDFOR -dst[MAX:256] := 0 +dst[MAX:128] := 0 -
immintrin.h
 - - Integer - AVX512VL - AVX512BW - Shift - - - - Shift packed 16-bit integers in "a" right by the amount specified by the corresponding element in "count" while shifting in zeros, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Floating Point + SSE + Trigonometry + + + + Compute the inverse tangent of packed single-precision (32-bit) floating-point elements in "a" divided by packed elements in "b", and store the results in "dst" expressed in radians. -FOR j := 0 to 15 - i := j*16 - IF k[j] - dst[i+15:i] := ZeroExtend(a[i+15:i] >> count[i+15:i]) - ELSE - dst[i+15:i] := 0 - FI +FOR j := 0 to 3 + i := j*32 + dst[i+31:i] := ATAN2(a[i+31:i], b[i+31:i]) ENDFOR -dst[MAX:256] := 0 +dst[MAX:128] := 0 -
immintrin.h
 - - Integer - AVX512VL - AVX512BW - Shift - - - Shift packed 16-bit integers in "a" right by the amount specified by the corresponding element in "count" while shifting in zeros, and store the results in "dst". + + Floating Point + SSE + Trigonometry + + + Compute the inverse hyperbolic tangent of packed double-precision (64-bit) floating-point elements in "a" expressed in radians, and store the results in "dst". -FOR j := 0 to 15 - i := j*16 - IF k[j] - dst[i+15:i] := ZeroExtend(a[i+15:i] >> count[i+15:i]) - ELSE - dst[i+15:i] := 0 - FI +FOR j := 0 to 1 + i := j*64 + dst[i+63:i] := ATANH(a[i+63:i]) ENDFOR -dst[MAX:256] := 0 +dst[MAX:128] := 0 -
immintrin.h
 - - Integer - AVX512BW - Shift - - - - - Shift packed 16-bit integers in "a" right by the amount specified by the corresponding element in "count" while shifting in zeros, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Floating Point + SSE + Trigonometry + + + Compute the inverse hyperbolic tangent of packed single-precision (32-bit) floating-point elements in "a" expressed in radians, and store the results in "dst". -FOR j := 0 to 31 - i := j*16 - IF k[j] - dst[i+15:i] := ZeroExtend(a[i+15:i] >> count[i+15:i]) - ELSE - dst[i+15:i] := src[i+15:i] - FI +FOR j := 0 to 3 + i := j*32 + dst[i+31:i] := ATANH(a[i+31:i]) ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 -
immintrin.h
 - - Integer - AVX512BW - Shift - - - - Shift packed 16-bit integers in "a" right by the amount specified by the corresponding element in "count" while shifting in zeros, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - -FOR j := 0 to 31 - i := j*16 - IF k[j] - dst[i+15:i] := ZeroExtend(a[i+15:i] >> count[i+15:i]) - ELSE - dst[i+15:i] := 0 - FI + + Floating Point + SSE + Elementary Math Functions + + + Compute the cube root of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst". + FOR j := 0 to 1 + i := j*64 + dst[i+63:i] := CubeRoot(a[i+63:i]) ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 -
immintrin.h
 - - Integer - AVX512BW - Shift - - - Shift packed 16-bit integers in "a" right by the amount specified by the corresponding element in "count" while shifting in zeros, and store the results in "dst". - -FOR j := 0 to 31 - i := j*16 - IF k[j] - dst[i+15:i] := ZeroExtend(a[i+15:i] >> count[i+15:i]) - ELSE - dst[i+15:i] := 0 - FI + + Floating Point + SSE + Elementary Math Functions + + + Compute the cube root of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst". + FOR j := 0 to 3 + i := j*32 + dst[i+31:i] := CubeRoot(a[i+31:i]) ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 -
immintrin.h
 - - Integer - AVX512VL - AVX512BW - Shift - - - - - Shift packed 16-bit integers in "a" right by the amount specified by the corresponding element in "count" while shifting in zeros, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -FOR j := 0 to 7 - i := j*16 - IF k[j] - dst[i+15:i] := ZeroExtend(a[i+15:i] >> count[i+15:i]) - ELSE - dst[i+15:i] := src[i+15:i] - FI + + Floating Point + SSE + Probability/Statistics + + + Compute the cumulative distribution function of packed double-precision (64-bit) floating-point elements in "a" using the normal distribution, and store the results in "dst". + FOR j := 0 to 1 + i := j*64 + dst[i+63:i] := CDFNormal(a[i+63:i]) ENDFOR dst[MAX:128] := 0 -
immintrin.h
 - - Integer - AVX512VL - AVX512BW - Shift - - - - Shift packed 16-bit integers in "a" right by the amount specified by the corresponding element in "count" while shifting in zeros, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - -FOR j := 0 to 7 - i := j*16 - IF k[j] - dst[i+15:i] := ZeroExtend(a[i+15:i] >> count[i+15:i]) - ELSE - dst[i+15:i] := 0 - FI + + Floating Point + SSE + Probability/Statistics + + + Compute the cumulative distribution function of packed single-precision (32-bit) floating-point elements in "a" using the normal distribution, and store the results in "dst". + FOR j := 0 to 3 + i := j*32 + dst[i+31:i] := CDFNormal(a[i+31:i]) ENDFOR dst[MAX:128] := 0 -
immintrin.h
 - - Integer - AVX512VL - AVX512BW - Shift - - - Shift packed 16-bit integers in "a" right by the amount specified by the corresponding element in "count" while shifting in zeros, and store the results in "dst". - -FOR j := 0 to 7 - i := j*16 - IF k[j] - dst[i+15:i] := ZeroExtend(a[i+15:i] >> count[i+15:i]) - ELSE - dst[i+15:i] := 0 - FI + + Floating Point + SSE + Probability/Statistics + + + Compute the inverse cumulative distribution function of packed double-precision (64-bit) floating-point elements in "a" using the normal distribution, and store the results in "dst". + FOR j := 0 to 1 + i := j*64 + dst[i+63:i] := InverseCDFNormal(a[i+63:i]) ENDFOR dst[MAX:128] := 0 - -
immintrin.h
- - - Integer - AVX512VL - AVX512BW - Shift - - - - - Shift packed 16-bit integers in "a" right by "count" while shifting in zeros, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -FOR j := 0 to 15 - i := j*16 - IF k[j] - IF count[63:0] > 15 - dst[i+15:i] := 0 - ELSE - dst[i+15:i] := ZeroExtend(a[i+15:i] >> count[63:0]) - FI - ELSE - dst[i+15:i] := src[i+15:i] - FI +
immintrin.h
+ + + Floating Point + SSE + Probability/Statistics + + + Compute the inverse cumulative distribution function of packed single-precision (32-bit) floating-point elements in "a" using the normal distribution, and store the results in "dst". + FOR j := 0 to 3 + i := j*32 + dst[i+31:i] := InverseCDFNormal(a[i+31:i]) ENDFOR -dst[MAX:256] := 0 +dst[MAX:128] := 0 -
immintrin.h
 - - Integer - AVX512VL - AVX512BW - Shift - - - - - Shift packed 16-bit integers in "a" right by "imm8" while shifting in zeros, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -FOR j := 0 to 15 - i := j*16 - IF k[j] - IF imm8[7:0] > 15 - dst[i+15:i] := 0 - ELSE - dst[i+15:i] := ZeroExtend(a[i+15:i] >> imm8[7:0]) - FI - ELSE - dst[i+15:i] := src[i+15:i] - FI + + Floating Point + SSE + Elementary Math Functions + + + Compute the exponential value of "e" raised to the power of packed complex numbers in "a", and store the complex results in "dst". Each complex number is composed of two adjacent single-precision (32-bit) floating-point elements, which defines the complex number "complex = vec.fp32[0] + i * vec.fp32[1]". + +DEFINE CEXP(a[31:0], b[31:0]) { + result[31:0] := POW(FP32(e), a[31:0]) * COS(b[31:0]) + result[63:32] := POW(FP32(e), a[31:0]) * SIN(b[31:0]) + RETURN result +} +FOR j := 0 to 1 + i := j*64 + dst[i+63:i] := CEXP(a[i+31:i], a[i+63:i+32]) ENDFOR -dst[MAX:256] := 0 +dst[MAX:128] := 0 -
immintrin.h
 - - Integer - AVX512VL - AVX512BW - Shift - - - - Shift packed 16-bit integers in "a" right by "count" while shifting in zeros, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - -FOR j := 0 to 15 - i := j*16 - IF k[j] - IF count[63:0] > 15 - dst[i+15:i] := 0 - ELSE - dst[i+15:i] := ZeroExtend(a[i+15:i] >> count[63:0]) - FI - ELSE - dst[i+15:i] := 0 - FI + + Floating Point + SSE + Elementary Math Functions + + + Compute the natural logarithm of packed complex numbers in "a", and store the complex results in "dst". Each complex number is composed of two adjacent single-precision (32-bit) floating-point elements, which defines the complex number "complex = vec.fp32[0] + i * vec.fp32[1]". + +DEFINE CLOG(a[31:0], b[31:0]) { + result[31:0] := LOG(SQRT(POW(a, 2.0) + POW(b, 2.0))) + result[63:32] := ATAN2(b, a) + RETURN result +} +FOR j := 0 to 1 + i := j*64 + dst[i+63:i] := CLOG(a[i+31:i], a[i+63:i+32]) ENDFOR -dst[MAX:256] := 0 +dst[MAX:128] := 0 -
immintrin.h
 - - Integer - AVX512VL - AVX512BW - Shift - - - - Shift packed 16-bit integers in "a" right by "imm8" while shifting in zeros, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Floating Point + SSE + Trigonometry + + + Compute the cosine of packed double-precision (64-bit) floating-point elements in "a" expressed in radians, and store the results in "dst". -FOR j := 0 to 15 - i := j*16 - IF k[j] - IF imm8[7:0] > 15 - dst[i+15:i] := 0 - ELSE - dst[i+15:i] := ZeroExtend(a[i+15:i] >> imm8[7:0]) - FI - ELSE - dst[i+15:i] := 0 - FI +FOR j := 0 to 1 + i := j*64 + dst[i+63:i] := COS(a[i+63:i]) ENDFOR -dst[MAX:256] := 0 +dst[MAX:128] := 0 -
immintrin.h
 - - Integer - AVX512BW - Shift - - - - - Shift packed 16-bit integers in "a" right by "count" while shifting in zeros, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Floating Point + SSE + Trigonometry + + + Compute the cosine of packed single-precision (32-bit) floating-point elements in "a" expressed in radians, and store the results in "dst". -FOR j := 0 to 31 - i := j*16 - IF k[j] - IF count[63:0] > 15 - dst[i+15:i] := 0 - ELSE - dst[i+15:i] := ZeroExtend(a[i+15:i] >> count[63:0]) - FI - ELSE - dst[i+15:i] := src[i+15:i] - FI +FOR j := 0 to 3 + i := j*32 + dst[i+31:i] := COS(a[i+31:i]) ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 -
immintrin.h
 - - Integer - AVX512BW - Shift - - - - - Shift packed 16-bit integers in "a" right by "imm8" while shifting in zeros, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -FOR j := 0 to 31 - i := j*16 - IF k[j] - IF imm8[7:0] > 15 - dst[i+15:i] := 0 - ELSE - dst[i+15:i] := ZeroExtend(a[i+15:i] >> imm8[7:0]) - FI - ELSE - dst[i+15:i] := src[i+15:i] - FI + + Floating Point + SSE + Trigonometry + + + Compute the cosine of packed double-precision (64-bit) floating-point elements in "a" expressed in degrees, and store the results in "dst". + FOR j := 0 to 1 + i := j*64 + dst[i+63:i] := COSD(a[i+63:i]) ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 -
immintrin.h
 - - Integer - AVX512BW - Shift - - - - Shift packed 16-bit integers in "a" right by "count" while shifting in zeros, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - -FOR j := 0 to 31 - i := j*16 - IF k[j] - IF count[63:0] > 15 - dst[i+15:i] := 0 - ELSE - dst[i+15:i] := ZeroExtend(a[i+15:i] >> count[63:0]) - FI - ELSE - dst[i+15:i] := 0 - FI + + Floating Point + SSE + Trigonometry + + + Compute the cosine of packed single-precision (32-bit) floating-point elements in "a" expressed in degrees, and store the results in "dst". + FOR j := 0 to 3 + i := j*32 + dst[i+31:i] := COSD(a[i+31:i]) ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 -
immintrin.h
 - - Integer - AVX512BW - Shift - - - - Shift packed 16-bit integers in "a" right by "imm8" while shifting in zeros, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Floating Point + SSE + Trigonometry + + + Compute the hyperbolic cosine of packed double-precision (64-bit) floating-point elements in "a" expressed in radians, and store the results in "dst". -FOR j := 0 to 31 - i := j*16 - IF k[j] - IF imm8[7:0] > 15 - dst[i+15:i] := 0 - ELSE - dst[i+15:i] := ZeroExtend(a[i+15:i] >> imm8[7:0]) - FI - ELSE - dst[i+15:i] := 0 - FI +FOR j := 0 to 1 + i := j*64 + dst[i+63:i] := COSH(a[i+63:i]) ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 -
immintrin.h
 - - Integer - AVX512BW - Shift - - - Shift packed 16-bit integers in "a" right by "count" while shifting in zeros, and store the results in "dst". + + Floating Point + SSE + Trigonometry + + + Compute the hyperbolic cosine of packed single-precision (32-bit) floating-point elements in "a" expressed in radians, and store the results in "dst". -FOR j := 0 to 31 - i := j*16 - IF count[63:0] > 15 - dst[i+15:i] := 0 - ELSE - dst[i+15:i] := ZeroExtend(a[i+15:i] >> count[63:0]) - FI +FOR j := 0 to 3 + i := j*32 + dst[i+31:i] := COSH(a[i+31:i]) ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 -
immintrin.h
 - - Integer - AVX512BW - Shift - - - Shift packed 16-bit integers in "a" right by "imm8" while shifting in zeros, and store the results in "dst". - -FOR j := 0 to 31 - i := j*16 - IF imm8[7:0] > 15 - dst[i+15:i] := 0 - ELSE - dst[i+15:i] := ZeroExtend(a[i+15:i] >> imm8[7:0]) - FI + + Floating Point + SSE + Elementary Math Functions + + + Compute the square root of packed complex snumbers in "a", and store the complex results in "dst". Each complex number is composed of two adjacent single-precision (32-bit) floating-point elements, which defines the complex number "complex = vec.fp32[0] + i * vec.fp32[1]". + +DEFINE CSQRT(a[31:0], b[31:0]) { + sign[31:0] := (b < 0.0) ? -FP32(1.0) : FP32(1.0) + result[31:0] := SQRT((a + SQRT(POW(a, 2.0) + POW(b, 2.0))) / 2.0) + result[63:32] := sign * SQRT((-a + SQRT(POW(a, 2.0) + POW(b, 2.0))) / 2.0) + RETURN result +} +FOR j := 0 to 1 + i := j*64 + dst[i+63:i] := CSQRT(a[i+31:i], a[i+63:i+32]) ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 -
immintrin.h
 - + Integer - AVX512VL - AVX512BW - Shift - - - - - Shift packed 16-bit integers in "a" right by "count" while shifting in zeros, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + SSE + Arithmetic + + + + Divide packed signed 8-bit integers in "a" by packed elements in "b", and store the truncated results in "dst". -FOR j := 0 to 7 - i := j*16 - IF k[j] - IF count[63:0] > 15 - dst[i+15:i] := 0 - ELSE - dst[i+15:i] := ZeroExtend(a[i+15:i] >> count[63:0]) - FI - ELSE - dst[i+15:i] := src[i+15:i] +FOR j := 0 to 15 + i := 8*j + IF b[i+7:i] == 0 + #DE FI + dst[i+7:i] := Truncate8(a[i+7:i] / b[i+7:i]) ENDFOR dst[MAX:128] := 0 -
immintrin.h
 - + Integer - AVX512VL - AVX512BW - Shift - - - - - Shift packed 16-bit integers in "a" right by "imm8" while shifting in zeros, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + SSE + Arithmetic + + + + Divide packed signed 16-bit integers in "a" by packed elements in "b", and store the truncated results in "dst". FOR j := 0 to 7 - i := j*16 - IF k[j] - IF imm8[7:0] > 15 - dst[i+15:i] := 0 - ELSE - dst[i+15:i] := ZeroExtend(a[i+15:i] >> imm8[7:0]) - FI - ELSE - dst[i+15:i] := src[i+15:i] + i := 16*j + IF b[i+15:i] == 0 + #DE FI + dst[i+15:i] := Truncate16(a[i+15:i] / b[i+15:i]) ENDFOR dst[MAX:128] := 0 -
immintrin.h
 - + Integer - AVX512VL - AVX512BW - Shift - - - - Shift packed 16-bit integers in "a" right by "count" while shifting in zeros, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + SSE + Arithmetic + + + + Divide packed 32-bit integers in "a" by packed elements in "b", and store the truncated results in "dst". -FOR j := 0 to 7 - i := j*16 - IF k[j] - IF count[63:0] > 15 - dst[i+15:i] := 0 - ELSE - dst[i+15:i] := ZeroExtend(a[i+15:i] >> count[63:0]) - FI - ELSE - dst[i+15:i] := 0 +FOR j := 0 to 3 + i := 32*j + IF b[i+31:i] == 0 + #DE FI + dst[i+31:i] := Truncate32(a[i+31:i] / b[i+31:i]) ENDFOR dst[MAX:128] := 0 -
immintrin.h
 - + Integer - AVX512VL - AVX512BW - Shift - - - - Shift packed 16-bit integers in "a" right by "imm8" while shifting in zeros, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + SSE + Arithmetic + + + + Divide packed signed 64-bit integers in "a" by packed elements in "b", and store the truncated results in "dst". -FOR j := 0 to 7 - i := j*16 - IF k[j] - IF imm8[7:0] > 15 - dst[i+15:i] := 0 - ELSE - dst[i+15:i] := ZeroExtend(a[i+15:i] >> imm8[7:0]) - FI - ELSE - dst[i+15:i] := 0 +FOR j := 0 to 1 + i := 64*j + IF b[i+63:i] == 0 + #DE FI + dst[i+63:i] := Truncate64(a[i+63:i] / b[i+63:i]) ENDFOR dst[MAX:128] := 0 -
immintrin.h
 - + Integer - AVX512VL - AVX512BW> + SSE Arithmetic - - - - - Subtract packed 8-bit integers in "b" from packed 8-bit integers in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + + + Divide packed unsigned 8-bit integers in "a" by packed elements in "b", and store the truncated results in "dst". -FOR j := 0 to 31 - i := j*8 - IF k[j] - dst[i+7:i] := a[i+7:i] - b[i+7:i] - ELSE - dst[i+7:i] := src[i+7:i] - FI +FOR j := 0 to 15 + i := 8*j + IF b[i+7:i] == 0 + #DE + FI + dst[i+7:i] := Truncate8(a[i+7:i] / b[i+7:i]) ENDFOR -dst[MAX:256] := 0 +dst[MAX:128] := 0 -
immintrin.h
 - + Integer - AVX512VL - AVX512BW> + SSE Arithmetic - - - - Subtract packed 8-bit integers in "b" from packed 8-bit integers in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + + + Divide packed unsigned 16-bit integers in "a" by packed elements in "b", and store the truncated results in "dst". -FOR j := 0 to 31 - i := j*8 - IF k[j] - dst[i+7:i] := a[i+7:i] - b[i+7:i] - ELSE - dst[i+7:i] := src[i+7:i] - FI +FOR j := 0 to 7 + i := 16*j + IF b[i+15:i] == 0 + #DE + FI + dst[i+15:i] := Truncate16(a[i+15:i] / b[i+15:i]) ENDFOR -dst[MAX:256] := 0 +dst[MAX:128] := 0 -
immintrin.h
 - + Integer - AVX512BW + SSE Arithmetic - - - - - Subtract packed 8-bit integers in "b" from packed 8-bit integers in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + + + Divide packed unsigned 32-bit integers in "a" by packed elements in "b", and store the truncated results in "dst". -FOR j := 0 to 63 - i := j*8 - IF k[j] - dst[i+7:i] := a[i+7:i] - b[i+7:i] - ELSE - dst[i+7:i] := src[i+7:i] - FI +FOR j := 0 to 3 + i := 32*j + IF b[i+31:i] == 0 + #DE + FI + dst[i+31:i] := Truncate32(a[i+31:i] / b[i+31:i]) ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 -
immintrin.h
 - + Integer - AVX512BW + SSE Arithmetic - - - - Subtract packed 8-bit integers in "b" from packed 8-bit integers in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + + + Divide packed unsigned 64-bit integers in "a" by packed elements in "b", and store the truncated results in "dst". -FOR j := 0 to 63 - i := j*8 - IF k[j] - dst[i+7:i] := a[i+7:i] - b[i+7:i] - ELSE - dst[i+7:i] := src[i+7:i] - FI +FOR j := 0 to 1 + i := 64*j + IF b[i+63:i] == 0 + #DE + FI + dst[i+63:i] := Truncate64(a[i+63:i] / b[i+63:i]) ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 -
immintrin.h
 - - Integer - AVX512BW + + Floating Point + SSE Arithmetic - - - Subtract packed 8-bit integers in "b" from packed 8-bit integers in "a", and store the results in "dst". - -FOR j := 0 to 63 - i := j*8 - dst[i+7:i] := a[i+7:i] - b[i+7:i] + + + Compute the error function of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst". + FOR j := 0 to 1 + i := j*64 + dst[i+63:i] := ERF(a[i+63:i]) ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 -
immintrin.h
 - - Integer - AVX512VL - AVX512BW> - Arithmetic - - - - - Subtract packed 8-bit integers in "b" from packed 8-bit integers in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -FOR j := 0 to 15 - i := j*8 - IF k[j] - dst[i+7:i] := a[i+7:i] - b[i+7:i] - ELSE - dst[i+7:i] := src[i+7:i] - FI + + Floating Point + SSE + Probability/Statistics + + + Compute the error function of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst". + FOR j := 0 to 3 + i := j*32 + dst[i+31:i] := ERF(a[i+31:i]) ENDFOR dst[MAX:128] := 0 -
immintrin.h
 - - Integer - AVX512VL - AVX512BW> - Arithmetic - - - - Subtract packed 8-bit integers in "b" from packed 8-bit integers in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - -FOR j := 0 to 15 - i := j*8 - IF k[j] - dst[i+7:i] := a[i+7:i] - b[i+7:i] - ELSE - dst[i+7:i] := src[i+7:i] - FI + + Floating Point + SSE + Probability/Statistics + + + Compute the complementary error function of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst". + FOR j := 0 to 1 + i := j*64 + dst[i+63:i] := 1.0 - ERF(a[i+63:i]) ENDFOR dst[MAX:128] := 0 -
immintrin.h
 - - Integer - AVX512VL - AVX512F - Arithmetic - - - - - Subtract packed 32-bit integers in "b" from packed 32-bit integers in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -FOR j := 0 to 7 + + Floating Point + SSE + Probability/Statistics + + + Compute the complementary error function of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst". + FOR j := 0 to 3 i := j*32 - IF k[j] - dst[i+31:i] := a[i+31:i] - b[i+31:i] - ELSE - dst[i+31:i] := src[i+31:i] - FI + dst[i+63:i] := 1.0 - ERF(a[i+31:i]) ENDFOR -dst[MAX:256] := 0 +dst[MAX:128] := 0 -
immintrin.h
 - - Integer - AVX512VL - AVX512F - Arithmetic - - - - Subtract packed 32-bit integers in "b" from packed 32-bit integers in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - -FOR j := 0 to 7 - i := j*32 - IF k[j] - dst[i+31:i] := a[i+31:i] - b[i+31:i] - ELSE - dst[i+31:i] := 0 - FI + + Floating Point + SSE + Probability/Statistics + + + Compute the inverse complementary error function of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst". + FOR j := 0 to 1 + i := j*64 + dst[i+63:i] := 1.0 / (1.0 - ERF(a[i+63:i])) ENDFOR -dst[MAX:256] := 0 +dst[MAX:128] := 0 -
immintrin.h
 - - Integer - AVX512VL - AVX512F - Arithmetic - - - - - Subtract packed 32-bit integers in "b" from packed 32-bit integers in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -FOR j := 0 to 3 + + Floating Point + SSE + Probability/Statistics + + + Compute the inverse complementary error function of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst". + FOR j := 0 to 3 i := j*32 - IF k[j] - dst[i+31:i] := a[i+31:i] - b[i+31:i] - ELSE - dst[i+31:i] := src[i+31:i] - FI + dst[i+63:i] := 1.0 / (1.0 - ERF(a[i+31:i])) ENDFOR dst[MAX:128] := 0 -
immintrin.h
 - - Integer - AVX512VL - AVX512F - Arithmetic - - - - Subtract packed 32-bit integers in "b" from packed 32-bit integers in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - -FOR j := 0 to 3 + + Floating Point + SSE + Probability/Statistics + + + Compute the inverse error function of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst". + FOR j := 0 to 1 + i := j*64 + dst[i+63:i] := 1.0 / ERF(a[i+63:i]) +ENDFOR +dst[MAX:128] := 0 + +
immintrin.h
+ + + Floating Point + SSE + Probability/Statistics + + + Compute the inverse error function of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst". + FOR j := 0 to 3 i := j*32 - IF k[j] - dst[i+31:i] := a[i+31:i] - b[i+31:i] - ELSE - dst[i+31:i] := 0 - FI + dst[i+63:i] := 1.0 / ERF(a[i+31:i]) ENDFOR dst[MAX:128] := 0 -
immintrin.h
 - - Integer - AVX512VL - AVX512F - Arithmetic - - - - - Subtract packed 64-bit integers in "b" from packed 64-bit integers in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Floating Point + SSE + Elementary Math Functions + + + Compute the exponential value of "e" raised to the power of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst". -FOR j := 0 to 3 +FOR j := 0 to 1 i := j*64 - IF k[j] - dst[i+63:i] := a[i+63:i] - b[i+63:i] - ELSE - dst[i+63:i] := src[i+63:i] - FI + dst[i+63:i] := POW(e, a[i+63:i]) ENDFOR -dst[MAX:256] := 0 +dst[MAX:128] := 0 -
immintrin.h
 - - Integer - AVX512VL - AVX512F - Arithmetic - - - - Subtract packed 64-bit integers in "b" from packed 64-bit integers in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Floating Point + SSE + Elementary Math Functions + + + Compute the exponential value of "e" raised to the power of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst". FOR j := 0 to 3 - i := j*64 - IF k[j] - dst[i+63:i] := a[i+63:i] - b[i+63:i] - ELSE - dst[i+63:i] := 0 - FI + i := j*32 + dst[i+31:i] := POW(FP32(e), a[i+31:i]) ENDFOR -dst[MAX:256] := 0 +dst[MAX:128] := 0 -
immintrin.h
 - - Integer - AVX512VL - AVX512F - Arithmetic - - - - - Subtract packed 64-bit integers in "b" from packed 64-bit integers in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Floating Point + SSE + Elementary Math Functions + + + Compute the exponential value of 10 raised to the power of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst". FOR j := 0 to 1 i := j*64 - IF k[j] - dst[i+63:i] := a[i+63:i] - b[i+63:i] - ELSE - dst[i+63:i] := src[i+63:i] - FI + dst[i+63:i] := POW(10.0, a[i+63:i]) ENDFOR dst[MAX:128] := 0 -
immintrin.h
 - - Integer - AVX512VL - AVX512F - Arithmetic - - - - Subtract packed 64-bit integers in "b" from packed 64-bit integers in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Floating Point + SSE + Elementary Math Functions + + + Compute the exponential value of 10 raised to the power of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst". -FOR j := 0 to 1 - i := j*64 - IF k[j] - dst[i+63:i] := a[i+63:i] - b[i+63:i] - ELSE - dst[i+63:i] := 0 - FI +FOR j := 0 to 3 + i := j*32 + dst[i+31:i] := POW(FP32(10.0), a[i+31:i]) ENDFOR dst[MAX:128] := 0 -
immintrin.h
 - - Integer - AVX512VL - AVX512BW - Arithmetic - - - - - Subtract packed 8-bit integers in "b" from packed 8-bit integers in "a" using saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -FOR j := 0 to 31 - i := j*8 - IF k[j] - dst[i+7:i] := Saturate_To_Int8(a[i+7:i] - b[i+7:i]) - ELSE - dst[i+7:i] := src[i+7:i] - FI + + Floating Point + SSE + Elementary Math Functions + + + Compute the exponential value of 2 raised to the power of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst". + +FOR j := 0 to 1 + i := j*64 + dst[i+63:i] := POW(2.0, a[i+63:i]) ENDFOR -dst[MAX:256] := 0 +dst[MAX:128] := 0 -
immintrin.h
 - - Integer - AVX512VL - AVX512BW - Arithmetic - - - - Subtract packed 8-bit integers in "b" from packed 8-bit integers in "a" using saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Floating Point + SSE + Elementary Math Functions + + + Compute the exponential value of 2 raised to the power of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst". -FOR j := 0 to 31 - i := j*8 - IF k[j] - dst[i+7:i] := Saturate_To_Int8(a[i+7:i] - b[i+7:i]) - ELSE - dst[i+7:i] := 0 - FI +FOR j := 0 to 3 + i := j*32 + dst[i+31:i] := POW(FP32(2.0), a[i+31:i]) ENDFOR -dst[MAX:256] := 0 +dst[MAX:128] := 0 -
immintrin.h
 - - Integer - AVX512BW - Arithmetic - - - - - Subtract packed 8-bit integers in "b" from packed 8-bit integers in "a" using saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Floating Point + SSE + Elementary Math Functions + + + Compute the exponential value of "e" raised to the power of packed double-precision (64-bit) floating-point elements in "a", subtract one from each element, and store the results in "dst". -FOR j := 0 to 63 - i := j*8 - IF k[j] - dst[i+7:i] := Saturate_To_Int8(a[i+7:i] - b[i+7:i]) - ELSE - dst[i+7:i] := src[i+7:i] - FI +FOR j := 0 to 1 + i := j*64 + dst[i+63:i] := POW(e, a[i+63:i]) - 1.0 ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 -
immintrin.h
 - - Integer - AVX512BW - Arithmetic - - - - Subtract packed 8-bit integers in "b" from packed 8-bit integers in "a" using saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Floating Point + SSE + Elementary Math Functions + + + Compute the exponential value of "e" raised to the power of packed single-precision (32-bit) floating-point elements in "a", subtract one from each element, and store the results in "dst". -FOR j := 0 to 63 - i := j*8 - IF k[j] - dst[i+7:i] := Saturate_To_Int8(a[i+7:i] - b[i+7:i]) - ELSE - dst[i+7:i] := 0 - FI +FOR j := 0 to 3 + i := j*32 + dst[i+31:i] := POW(FP32(e), a[i+31:i]) - 1.0 ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 -
immintrin.h
 - - Integer - AVX512BW - Arithmetic - - - Subtract packed 8-bit integers in "b" from packed 8-bit integers in "a" using saturation, and store the results in "dst". + + Floating Point + SSE + Trigonometry + + + + Compute the length of the hypotenous of a right triangle, with the lengths of the other two sides of the triangle stored as packed double-precision (64-bit) floating-point elements in "a" and "b", and store the results in "dst". -FOR j := 0 to 63 - i := j*8 - dst[i+7:i] := Saturate_To_Int8(a[i+7:i] - b[i+7:i]) +FOR j := 0 to 1 + i := j*64 + dst[i+63:i] := SQRT(POW(a[i+63:i], 2.0) + POW(b[i+63:i], 2.0)) ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 -
immintrin.h
 - - Integer - AVX512VL - AVX512BW - Arithmetic - - - - - Subtract packed 8-bit integers in "b" from packed 8-bit integers in "a" using saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Floating Point + SSE + Trigonometry + + + + Compute the length of the hypotenous of a right triangle, with the lengths of the other two sides of the triangle stored as packed single-precision (32-bit) floating-point elements in "a" and "b", and store the results in "dst". -FOR j := 0 to 15 - i := j*8 - IF k[j] - dst[i+7:i] := Saturate_To_Int8(a[i+7:i] - b[i+7:i]) - ELSE - dst[i+7:i] := src[i+7:i] - FI +FOR j := 0 to 3 + i := j*32 + dst[i+31:i] := SQRT(POW(a[i+31:i], 2.0) + POW(b[i+31:i], 2.0)) ENDFOR dst[MAX:128] := 0 -
immintrin.h
 - + Integer - AVX512VL - AVX512BW + SSE Arithmetic - - - - Subtract packed 8-bit integers in "b" from packed 8-bit integers in "a" using saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - -FOR j := 0 to 15 - i := j*8 - IF k[j] - dst[i+7:i] := Saturate_To_Int8(a[i+7:i] - b[i+7:i]) - ELSE - dst[i+7:i] := 0 - FI + + + + Divide packed 32-bit integers in "a" by packed elements in "b", and store the truncated results in "dst". + FOR j := 0 to 3 + i := 32*j + dst[i+31:i] := TRUNCATE(a[i+31:i] / b[i+31:i]) ENDFOR dst[MAX:128] := 0 -
immintrin.h
 - + Integer - AVX512VL - AVX512BW + SSE Arithmetic - - - - - Subtract packed 16-bit integers in "b" from packed 16-bit integers in "a" using saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -FOR j := 0 to 15 - i := j*16 - IF k[j] - dst[i+15:i] := Saturate_To_Int16(a[i+15:i] - b[i+15:i]) - ELSE - dst[i+15:i] := src[i+15:i] - FI + + + + + Divide packed 32-bit integers in "a" by packed elements in "b", store the truncated results in "dst", and store the remainders as packed 32-bit integers into memory at "mem_addr". + FOR j := 0 to 3 + i := 32*j + dst[i+31:i] := TRUNCATE(a[i+31:i] / b[i+31:i]) + MEM[mem_addr+i+31:mem_addr+i] := REMAINDER(a[i+31:i] / b[i+31:i]) ENDFOR -dst[MAX:256] := 0 +dst[MAX:128] := 0 -
immintrin.h
 - - Integer - AVX512VL - AVX512BW - Arithmetic - - - - Subtract packed 16-bit integers in "b" from packed 16-bit integers in "a" using saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - -FOR j := 0 to 15 - i := j*16 - IF k[j] - dst[i+15:i] := Saturate_To_Int16(a[i+15:i] - b[i+15:i]) - ELSE - dst[i+15:i] := 0 - FI + + Floating Point + SSE + Elementary Math Functions + + + Compute the inverse cube root of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst". + FOR j := 0 to 1 + i := j*64 + dst[i+63:i] := InvCubeRoot(a[i+63:i]) ENDFOR -dst[MAX:256] := 0 +dst[MAX:128] := 0 -
immintrin.h
 - - Integer - AVX512BW - Arithmetic - - - - - Subtract packed 16-bit integers in "b" from packed 16-bit integers in "a" using saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -FOR j := 0 to 31 - i := j*16 - IF k[j] - dst[i+15:i] := Saturate_To_Int16(a[i+15:i] - b[i+15:i]) - ELSE - dst[i+15:i] := src[i+15:i] - FI + + Floating Point + SSE + Elementary Math Functions + + + Compute the inverse cube root of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst". + FOR j := 0 to 3 + i := j*32 + dst[i+31:i] := InvCubeRoot(a[i+31:i]) ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 -
immintrin.h
 - - Integer - AVX512BW - Arithmetic - - - - Subtract packed 16-bit integers in "b" from packed 16-bit integers in "a" using saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - -FOR j := 0 to 31 - i := j*16 - IF k[j] - dst[i+15:i] := Saturate_To_Int16(a[i+15:i] - b[i+15:i]) - ELSE - dst[i+15:i] := 0 - FI + + Floating Point + SSE + Elementary Math Functions + + + Compute the inverse square root of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst". + FOR j := 0 to 1 + i := j*64 + dst[i+63:i] := InvSQRT(a[i+63:i]) ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 -
immintrin.h
 - - Integer - AVX512BW - Arithmetic - - - Subtract packed 16-bit integers in "b" from packed 16-bit integers in "a" using saturation, and store the results in "dst". - -FOR j := 0 to 31 - i := j*16 - dst[i+15:i] := Saturate_To_Int16(a[i+15:i] - b[i+15:i]) + + Floating Point + SSE + Elementary Math Functions + + + Compute the inverse square root of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst". + FOR j := 0 to 3 + i := j*32 + dst[i+31:i] := InvSQRT(a[i+31:i]) ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 -
immintrin.h
 - + Integer - AVX512VL - AVX512BW + SSE Arithmetic - - - - - Subtract packed 16-bit integers in "b" from packed 16-bit integers in "a" using saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -FOR j := 0 to 7 - i := j*16 - IF k[j] - dst[i+15:i] := Saturate_To_Int16(a[i+15:i] - b[i+15:i]) - ELSE - dst[i+15:i] := src[i+15:i] - FI + + + + Divide packed 32-bit integers in "a" by packed elements in "b", and store the remainders as packed 32-bit integers in "dst". + FOR j := 0 to 3 + i := 32*j + dst[i+31:i] := REMAINDER(a[i+31:i] / b[i+31:i]) ENDFOR dst[MAX:128] := 0 -
immintrin.h
 - - Integer - AVX512VL - AVX512BW - Arithmetic - - - - Subtract packed 16-bit integers in "b" from packed 16-bit integers in "a" using saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Floating Point + SSE + Elementary Math Functions + + + Compute the natural logarithm of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst". -FOR j := 0 to 7 - i := j*16 - IF k[j] - dst[i+15:i] := Saturate_To_Int16(a[i+15:i] - b[i+15:i]) - ELSE - dst[i+15:i] := 0 - FI +FOR j := 0 to 1 + i := j*64 + dst[i+63:i] := LOG(a[i+63:i]) ENDFOR dst[MAX:128] := 0 -
immintrin.h
 - - Integer - AVX512VL - AVX512BW - Arithmetic - - - - - Subtract packed unsigned 8-bit integers in "b" from packed unsigned 8-bit integers in "a" using saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Floating Point + SSE + Elementary Math Functions + + + Compute the natural logarithm of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst". -FOR j := 0 to 31 - i := j*8 - IF k[j] - dst[i+7:i] := Saturate_To_UnsignedInt8(a[i+7:i] - b[i+7:i]) - ELSE - dst[i+7:i] := src[i+7:i] - FI +FOR j := 0 to 3 + i := j*32 + dst[i+31:i] := LOG(a[i+31:i]) ENDFOR -dst[MAX:256] := 0 +dst[MAX:128] := 0 -
immintrin.h
 - - Integer - AVX512VL - AVX512BW - Arithmetic - - - - Subtract packed unsigned 8-bit integers in "b" from packed unsigned 8-bit integers in "a" using saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Floating Point + SSE + Elementary Math Functions + + + Compute the base-10 logarithm of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst". -FOR j := 0 to 31 - i := j*8 - IF k[j] - dst[i+7:i] := Saturate_To_UnsignedInt8(a[i+7:i] - b[i+7:i]) - ELSE - dst[i+7:i] := 0 - FI +FOR j := 0 to 1 + i := j*64 + dst[i+63:i] := LOG(a[i+63:i]) / LOG(10.0) ENDFOR -dst[MAX:256] := 0 +dst[MAX:128] := 0 -
immintrin.h
 - - Integer - AVX512BW - Arithmetic - - - - - Subtract packed unsigned 8-bit integers in "b" from packed unsigned 8-bit integers in "a" using saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Floating Point + SSE + Elementary Math Functions + + + Compute the base-10 logarithm of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst". -FOR j := 0 to 63 - i := j*8 - IF k[j] - dst[i+7:i] := Saturate_To_UnsignedInt8(a[i+7:i] - b[i+7:i]) - ELSE - dst[i+7:i] := src[i+7:i] - FI +FOR j := 0 to 3 + i := j*32 + dst[i+31:i] := LOG(a[i+31:i]) / LOG(10.0) ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 -
immintrin.h
 - - Integer - AVX512BW - Arithmetic - - - - Subtract packed unsigned 8-bit integers in "b" from packed unsigned 8-bit integers in "a" using saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Floating Point + SSE + Elementary Math Functions + + + Compute the natural logarithm of one plus packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst". -FOR j := 0 to 63 - i := j*8 - IF k[j] - dst[i+7:i] := Saturate_To_UnsignedInt8(a[i+7:i] - b[i+7:i]) - ELSE - dst[i+7:i] := 0 - FI +FOR j := 0 to 1 + i := j*64 + dst[i+63:i] := LOG(1.0 + a[i+63:i]) ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 -
immintrin.h
 - - Integer - AVX512BW - Arithmetic - - - Subtract packed unsigned 8-bit integers in "b" from packed unsigned 8-bit integers in "a" using saturation, and store the results in "dst". + + Floating Point + SSE + Elementary Math Functions + + + Compute the natural logarithm of one plus packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst". -FOR j := 0 to 63 - i := j*8 - dst[i+7:i] := Saturate_To_UnsignedInt8(a[i+7:i] - b[i+7:i]) +FOR j := 0 to 3 + i := j*32 + dst[i+31:i] := LOG(1.0 + a[i+31:i]) ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 -
immintrin.h
 - - Integer - AVX512VL - AVX512BW - Arithmetic - - - - - Subtract packed unsigned 8-bit integers in "b" from packed unsigned 8-bit integers in "a" using saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Floating Point + SSE + Elementary Math Functions + + + Compute the base-2 logarithm of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst". -FOR j := 0 to 15 - i := j*8 - IF k[j] - dst[i+7:i] := Saturate_To_UnsignedInt8(a[i+7:i] - b[i+7:i]) - ELSE - dst[i+7:i] := src[i+7:i] - FI +FOR j := 0 to 1 + i := j*64 + dst[i+63:i] := LOG(a[i+63:i]) / LOG(2.0) ENDFOR dst[MAX:128] := 0 -
immintrin.h
 - - Integer - AVX512VL - AVX512BW - Arithmetic - - - - Subtract packed unsigned 8-bit integers in "b" from packed unsigned 8-bit integers in "a" using saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Floating Point + SSE + Elementary Math Functions + + + Compute the base-2 logarithm of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst". -FOR j := 0 to 15 - i := j*8 - IF k[j] - dst[i+7:i] := Saturate_To_UnsignedInt8(a[i+7:i] - b[i+7:i]) - ELSE - dst[i+7:i] := 0 - FI +FOR j := 0 to 3 + i := j*32 + dst[i+31:i] := LOG(a[i+31:i]) / LOG(2.0) ENDFOR dst[MAX:128] := 0 -
immintrin.h
 - - Integer - AVX512VL - AVX512BW - Arithmetic - - - - - Subtract packed unsigned 16-bit integers in "b" from packed unsigned 16-bit integers in "a" using saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -FOR j := 0 to 15 - i := j*16 - IF k[j] - dst[i+15:i] := Saturate_To_UnsignedInt16(a[i+15:i] - b[i+15:i]) - ELSE - dst[i+15:i] := src[i+15:i] - FI + + Floating Point + SSE + Elementary Math Functions + + + Convert the exponent of each packed double-precision (64-bit) floating-point element in "a" to a double-precision floating-point number representing the integer exponent, and store the results in "dst". This intrinsic essentially calculates "floor(log2(x))" for each element. + FOR j := 0 to 1 + i := j*64 + dst[i+63:i] := ConvertExpFP64(a[i+63:i]) ENDFOR -dst[MAX:256] := 0 +dst[MAX:128] := 0 -
immintrin.h
 - - Integer - AVX512VL - AVX512BW - Arithmetic - - - - Subtract packed unsigned 16-bit integers in "b" from packed unsigned 16-bit integers in "a" using saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - -FOR j := 0 to 15 - i := j*16 - IF k[j] - dst[i+15:i] := Saturate_To_UnsignedInt16(a[i+15:i] - b[i+15:i]) - ELSE - dst[i+15:i] := 0 - FI + + Floating Point + SSE + Elementary Math Functions + + + Convert the exponent of each packed single-precision (32-bit) floating-point element in "a" to a single-precision floating-point number representing the integer exponent, and store the results in "dst". This intrinsic essentially calculates "floor(log2(x))" for each element. + FOR j := 0 to 3 + i := j*32 + dst[i+31:i] := ConvertExpFP32(a[i+31:i]) ENDFOR -dst[MAX:256] := 0 +dst[MAX:128] := 0 -
immintrin.h
 - - Integer - AVX512BW - Arithmetic - - - - - Subtract packed unsigned 16-bit integers in "b" from packed unsigned 16-bit integers in "a" using saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Floating Point + SSE + Elementary Math Functions + + + + Compute the exponential value of packed double-precision (64-bit) floating-point elements in "a" raised by packed elements in "b", and store the results in "dst". -FOR j := 0 to 31 - i := j*16 - IF k[j] - dst[i+15:i] := Saturate_To_UnsignedInt16(a[i+15:i] - b[i+15:i]) - ELSE - dst[i+15:i] := src[i+15:i] - FI +FOR j := 0 to 1 + i := j*64 + dst[i+63:i] := POW(a[i+63:i], b[i+63:i]) ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 -
immintrin.h
 - - Integer - AVX512BW - Arithmetic - - - - Subtract packed unsigned 16-bit integers in "b" from packed unsigned 16-bit integers in "a" using saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Floating Point + SSE + Elementary Math Functions + + + + Compute the exponential value of packed single-precision (32-bit) floating-point elements in "a" raised by packed elements in "b", and store the results in "dst". -FOR j := 0 to 31 - i := j*16 - IF k[j] - dst[i+15:i] := Saturate_To_UnsignedInt16(a[i+15:i] - b[i+15:i]) - ELSE - dst[i+15:i] := 0 - FI +FOR j := 0 to 3 + i := j*32 + dst[i+31:i] := POW(a[i+31:i], b[i+31:i]) ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 -
immintrin.h
 - + Integer - AVX512BW + SSE Arithmetic - - - Subtract packed unsigned 16-bit integers in "b" from packed unsigned 16-bit integers in "a" using saturation, and store the results in "dst". - -FOR j := 0 to 31 - i := j*16 - dst[i+15:i] := Saturate_To_UnsignedInt16(a[i+15:i] - b[i+15:i]) + + + + Divide packed 8-bit integers in "a" by packed elements in "b", and store the remainders as packed 32-bit integers in "dst". + FOR j := 0 to 15 + i := 8*j + dst[i+7:i] := REMAINDER(a[i+7:i] / b[i+7:i]) ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 -
immintrin.h
 - + Integer - AVX512VL - AVX512BW + SSE Arithmetic - - - - - Subtract packed unsigned 16-bit integers in "b" from packed unsigned 16-bit integers in "a" using saturation, and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -FOR j := 0 to 7 - i := j*16 - IF k[j] - dst[i+15:i] := Saturate_To_UnsignedInt16(a[i+15:i] - b[i+15:i]) - ELSE - dst[i+15:i] := src[i+15:i] - FI + + + + Divide packed 16-bit integers in "a" by packed elements in "b", and store the remainders as packed 32-bit integers in "dst". + FOR j := 0 to 7 + i := 16*j + dst[i+15:i] := REMAINDER(a[i+15:i] / b[i+15:i]) ENDFOR dst[MAX:128] := 0 -
immintrin.h
 - + Integer - AVX512VL - AVX512BW + SSE Arithmetic - - - - Subtract packed unsigned 16-bit integers in "b" from packed unsigned 16-bit integers in "a" using saturation, and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - -FOR j := 0 to 7 - i := j*16 - IF k[j] - dst[i+15:i] := Saturate_To_UnsignedInt16(a[i+15:i] - b[i+15:i]) - ELSE - dst[i+15:i] := 0 - FI + + + + Divide packed 32-bit integers in "a" by packed elements in "b", and store the remainders as packed 32-bit integers in "dst". + FOR j := 0 to 3 + i := 32*j + dst[i+31:i] := REMAINDER(a[i+31:i] / b[i+31:i]) ENDFOR dst[MAX:128] := 0 -
immintrin.h
 - + Integer - AVX512VL - AVX512BW + SSE Arithmetic - - - - - Subtract packed 16-bit integers in "b" from packed 16-bit integers in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -FOR j := 0 to 15 - i := j*16 - IF k[j] - dst[i+15:i] := a[i+15:i] - b[i+15:i] - ELSE - dst[i+15:i] := src[i+15:i] - FI + + + + Divide packed 64-bit integers in "a" by packed elements in "b", and store the remainders as packed 32-bit integers in "dst". + FOR j := 0 to 1 + i := 64*j + dst[i+63:i] := REMAINDER(a[i+63:i] / b[i+63:i]) ENDFOR -dst[MAX:256] := 0 +dst[MAX:128] := 0 -
immintrin.h
 - + Integer - AVX512VL - AVX512BW + SSE Arithmetic - - - - Subtract packed 16-bit integers in "b" from packed 16-bit integers in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - -FOR j := 0 to 15 - i := j*16 - IF k[j] - dst[i+15:i] := a[i+15:i] - b[i+15:i] - ELSE - dst[i+15:i] := src[i+15:i] - FI + + + + Divide packed unsigned 8-bit integers in "a" by packed elements in "b", and store the remainders as packed unsigned 32-bit integers in "dst". + FOR j := 0 to 15 + i := 8*j + dst[i+7:i] := REMAINDER(a[i+7:i] / b[i+7:i]) ENDFOR -dst[MAX:256] := 0 +dst[MAX:128] := 0 -
immintrin.h
 - + Integer - AVX512BW + SSE Arithmetic - - - - - Subtract packed 16-bit integers in "b" from packed 16-bit integers in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -FOR j := 0 to 31 - i := j*16 - IF k[j] - dst[i+15:i] := a[i+15:i] - b[i+15:i] - ELSE - dst[i+15:i] := src[i+15:i] - FI + + + + Divide packed unsigned 16-bit integers in "a" by packed elements in "b", and store the remainders as packed unsigned 32-bit integers in "dst". + FOR j := 0 to 7 + i := 16*j + dst[i+15:i] := REMAINDER(a[i+15:i] / b[i+15:i]) ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 -
immintrin.h
 - + Integer - AVX512BW + SSE Arithmetic - - - - Subtract packed 16-bit integers in "b" from packed 16-bit integers in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - -FOR j := 0 to 31 - i := j*16 - IF k[j] - dst[i+15:i] := a[i+15:i] - b[i+15:i] - ELSE - dst[i+15:i] := src[i+15:i] - FI + + + + Divide packed unsigned 32-bit integers in "a" by packed elements in "b", and store the remainders as packed unsigned 32-bit integers in "dst". + FOR j := 0 to 3 + i := 32*j + dst[i+31:i] := REMAINDER(a[i+31:i] / b[i+31:i]) ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 -
immintrin.h
 - + Integer - AVX512BW + SSE Arithmetic - - - Subtract packed 16-bit integers in "b" from packed 16-bit integers in "a", and store the results in "dst". + + + + Divide packed unsigned 64-bit integers in "a" by packed elements in "b", and store the remainders as packed unsigned 32-bit integers in "dst". + FOR j := 0 to 1 + i := 64*j + dst[i+63:i] := REMAINDER(a[i+63:i] / b[i+63:i]) +ENDFOR +dst[MAX:128] := 0 + +
immintrin.h
+ + + Floating Point + SSE + Trigonometry + + + Compute the sine of packed double-precision (64-bit) floating-point elements in "a" expressed in radians, and store the results in "dst". -FOR j := 0 to 31 - i := j*16 - dst[i+15:i] := a[i+15:i] - b[i+15:i] +FOR j := 0 to 1 + i := j*64 + dst[i+63:i] := SIN(a[i+63:i]) ENDFOR -dst[MAX:512] := 0 +dst[MAX:128] := 0 -
immintrin.h
 - - Integer - AVX512VL - AVX512BW - Arithmetic - - - - - Subtract packed 16-bit integers in "b" from packed 16-bit integers in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Floating Point + SSE + Trigonometry + + + Compute the sine of packed single-precision (32-bit) floating-point elements in "a" expressed in radians, and store the results in "dst". -FOR j := 0 to 7 - i := j*16 - IF k[j] - dst[i+15:i] := a[i+15:i] - b[i+15:i] - ELSE - dst[i+15:i] := src[i+15:i] - FI +FOR j := 0 to 3 + i := j*32 + dst[i+31:i] := SIN(a[i+31:i]) ENDFOR dst[MAX:128] := 0 -
immintrin.h
 - - Integer - AVX512VL - AVX512BW - Arithmetic - - - - Subtract packed 16-bit integers in "b" from packed 16-bit integers in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Floating Point + SSE + Trigonometry + + + + Compute the sine and cosine of packed double-precision (64-bit) floating-point elements in "a" expressed in radians, store the sine in "dst", and store the cosine into memory at "mem_addr". -FOR j := 0 to 7 - i := j*16 - IF k[j] - dst[i+15:i] := a[i+15:i] - b[i+15:i] - ELSE - dst[i+15:i] := src[i+15:i] - FI +FOR j := 0 to 1 + i := j*64 + dst[i+63:i] := SIN(a[i+63:i]) + MEM[mem_addr+i+63:mem_addr+i] := COS(a[i+63:i]) ENDFOR dst[MAX:128] := 0 -
immintrin.h
 - - Integer - AVX512VL - AVX512F - Logical - - - - - - Bitwise ternary logic that provides the capability to implement any three-operand binary function; the specific binary function is specified by value in "imm8". For each bit in each packed 32-bit integer, the corresponding bit from "src", "a", and "b" are used to form a 3 bit index into "imm8", and the value at that bit in "imm8" is written to the corresponding bit in "dst" using writemask "k" at 32-bit granularity (32-bit elements are copied from "src" when the corresponding mask bit is not set). + + Floating Point + SSE + Trigonometry + + + + Compute the sine and cosine of packed single-precision (32-bit) floating-point elements in "a" expressed in radians, store the sine in "dst", and store the cosine into memory at "mem_addr". -FOR j := 0 to 7 +FOR j := 0 to 3 i := j*32 - IF k[j] - FOR h := 0 to 31 - index[2:0] := (src[i+h] << 2) OR (a[i+h] << 1) OR b[i+h] - dst[i+h] := imm8[index[2:0]] - ENDFOR - ELSE - dst[i+31:i] := src[i+31:i] - FI + dst[i+31:i] := SIN(a[i+31:i]) + MEM[mem_addr+i+31:mem_addr+i] := COS(a[i+31:i]) ENDFOR -dst[MAX:256] := 0 +dst[MAX:128] := 0 -
immintrin.h
 - - Integer - AVX512VL - AVX512F - Logical - - - - - - Bitwise ternary logic that provides the capability to implement any three-operand binary function; the specific binary function is specified by value in "imm8". For each bit in each packed 32-bit integer, the corresponding bit from "a", "b", and "c" are used to form a 3 bit index into "imm8", and the value at that bit in "imm8" is written to the corresponding bit in "dst" using zeromask "k" at 32-bit granularity (32-bit elements are zeroed out when the corresponding mask bit is not set). - -FOR j := 0 to 7 + + Floating Point + SSE + Trigonometry + + + Compute the sine of packed double-precision (64-bit) floating-point elements in "a" expressed in degrees, and store the results in "dst". + FOR j := 0 to 1 + i := j*64 + dst[i+63:i] := SIND(a[i+63:i]) +ENDFOR +dst[MAX:128] := 0 + +
immintrin.h
+ + + Floating Point + SSE + Trigonometry + + + Compute the sine of packed single-precision (32-bit) floating-point elements in "a" expressed in degrees, and store the results in "dst". + FOR j := 0 to 3 i := j*32 - IF k[j] - FOR h := 0 to 31 - index[2:0] := (a[i+h] << 2) OR (b[i+h] << 1) OR c[i+h] - dst[i+h] := imm8[index[2:0]] - ENDFOR - ELSE - dst[i+31:i] := 0 - FI + dst[i+31:i] := SIND(a[i+31:i]) ENDFOR -dst[MAX:256] := 0 +dst[MAX:128] := 0 -
immintrin.h
 - - Integer - AVX512VL - AVX512F - Logical - - - - - Bitwise ternary logic that provides the capability to implement any three-operand binary function; the specific binary function is specified by value in "imm8". For each bit in each packed 32-bit integer, the corresponding bit from "a", "b", and "c" are used to form a 3 bit index into "imm8", and the value at that bit in "imm8" is written to the corresponding bit in "dst". + + Floating Point + SSE + Trigonometry + + + Compute the hyperbolic sine of packed double-precision (64-bit) floating-point elements in "a" expressed in radians, and store the results in "dst". -FOR j := 0 to 7 - i := j*32 - FOR h := 0 to 31 - index[2:0] := (a[i+h] << 2) OR (b[i+h] << 1) OR c[i+h] - dst[i+h] := imm8[index[2:0]] - ENDFOR +FOR j := 0 to 1 + i := j*64 + dst[i+63:i] := SINH(a[i+63:i]) ENDFOR -dst[MAX:256] := 0 +dst[MAX:128] := 0 -
immintrin.h
 - - Integer - AVX512VL - AVX512F - Logical - - - - - - Bitwise ternary logic that provides the capability to implement any three-operand binary function; the specific binary function is specified by value in "imm8". For each bit in each packed 32-bit integer, the corresponding bit from "src", "a", and "b" are used to form a 3 bit index into "imm8", and the value at that bit in "imm8" is written to the corresponding bit in "dst" using writemask "k" at 32-bit granularity (32-bit elements are copied from "src" when the corresponding mask bit is not set). + + Floating Point + SSE + Trigonometry + + + Compute the hyperbolic sine of packed single-precision (32-bit) floating-point elements in "a" expressed in radians, and store the results in "dst". FOR j := 0 to 3 i := j*32 - IF k[j] - FOR h := 0 to 31 - index[2:0] := (src[i+h] << 2) OR (a[i+h] << 1) OR b[i+h] - dst[i+h] := imm8[index[2:0]] - ENDFOR - ELSE - dst[i+31:i] := src[i+31:i] - FI + dst[i+31:i] := SINH(a[i+31:i]) ENDFOR dst[MAX:128] := 0 -
immintrin.h
 - - Integer - AVX512VL - AVX512F - Logical - - - - - - Bitwise ternary logic that provides the capability to implement any three-operand binary function; the specific binary function is specified by value in "imm8". For each bit in each packed 32-bit integer, the corresponding bit from "a", "b", and "c" are used to form a 3 bit index into "imm8", and the value at that bit in "imm8" is written to the corresponding bit in "dst" using zeromask "k" at 32-bit granularity (32-bit elements are zeroed out when the corresponding mask bit is not set). + + Floating Point + SSE + Special Math Functions + + + Round the packed double-precision (64-bit) floating-point elements in "a" up to an integer value, and store the results as packed double-precision floating-point elements in "dst". This intrinsic may generate the "roundpd"/"vroundpd" instruction. -FOR j := 0 to 3 - i := j*32 - IF k[j] - FOR h := 0 to 31 - index[2:0] := (a[i+h] << 2) OR (b[i+h] << 1) OR c[i+h] - dst[i+h] := imm8[index[2:0]] - ENDFOR - ELSE - dst[i+31:i] := 0 - FI +FOR j := 0 to 1 + i := j*64 + dst[i+63:i] := CEIL(a[i+63:i]) ENDFOR dst[MAX:128] := 0 -
immintrin.h
 - - Integer - AVX512VL - AVX512F - Logical - - - - - Bitwise ternary logic that provides the capability to implement any three-operand binary function; the specific binary function is specified by value in "imm8". For each bit in each packed 32-bit integer, the corresponding bit from "a", "b", and "c" are used to form a 3 bit index into "imm8", and the value at that bit in "imm8" is written to the corresponding bit in "dst". + + Floating Point + SSE + Special Math Functions + + + Round the packed single-precision (32-bit) floating-point elements in "a" up to an integer value, and store the results as packed single-precision floating-point elements in "dst". This intrinsic may generate the "roundps"/"vroundps" instruction. FOR j := 0 to 3 i := j*32 - FOR h := 0 to 31 - index[2:0] := (a[i+h] << 2) OR (b[i+h] << 1) OR c[i+h] - dst[i+h] := imm8[index[2:0]] - ENDFOR + dst[i+31:i] := CEIL(a[i+31:i]) ENDFOR dst[MAX:128] := 0 -
immintrin.h
 - - Integer - AVX512VL - AVX512F - Logical - - - - - - Bitwise ternary logic that provides the capability to implement any three-operand binary function; the specific binary function is specified by value in "imm8". For each bit in each packed 64-bit integer, the corresponding bit from "src", "a", and "b" are used to form a 3 bit index into "imm8", and the value at that bit in "imm8" is written to the corresponding bit in "dst" using writemask "k" at 64-bit granularity (64-bit elements are copied from "src" when the corresponding mask bit is not set). + + Floating Point + SSE + Special Math Functions + + + Round the packed double-precision (64-bit) floating-point elements in "a" down to an integer value, and store the results as packed double-precision floating-point elements in "dst". This intrinsic may generate the "roundpd"/"vroundpd" instruction. -FOR j := 0 to 3 +FOR j := 0 to 1 i := j*64 - IF k[j] - FOR h := 0 to 63 - index[2:0] := (src[i+h] << 2) OR (a[i+h] << 1) OR b[i+h] - dst[i+h] := imm8[index[2:0]] - ENDFOR - ELSE - dst[i+63:i] := src[i+63:i] - FI + dst[i+63:i] := FLOOR(a[i+63:i]) ENDFOR -dst[MAX:256] := 0 +dst[MAX:128] := 0 -
immintrin.h
 - - Integer - AVX512VL - AVX512F - Logical - - - - - - Bitwise ternary logic that provides the capability to implement any three-operand binary function; the specific binary function is specified by value in "imm8". For each bit in each packed 64-bit integer, the corresponding bit from "a", "b", and "c" are used to form a 3 bit index into "imm8", and the value at that bit in "imm8" is written to the corresponding bit in "dst" using zeromask "k" at 64-bit granularity (64-bit elements are zeroed out when the corresponding mask bit is not set). + + Floating Point + SSE + Special Math Functions + + + Round the packed single-precision (32-bit) floating-point elements in "a" down to an integer value, and store the results as packed single-precision floating-point elements in "dst". This intrinsic may generate the "roundps"/"vroundps" instruction. FOR j := 0 to 3 + i := j*32 + dst[i+31:i] := FLOOR(a[i+31:i]) +ENDFOR +dst[MAX:128] := 0 + +
immintrin.h
+ + + Floating Point + SSE + Special Math Functions + + + Round the packed double-precision (64-bit) floating-point elements in "a" to the nearest integer value, and store the results as packed double-precision floating-point elements in "dst". This intrinsic may generate the "roundpd"/"vroundpd" instruction. + +FOR j := 0 to 1 i := j*64 - IF k[j] - FOR h := 0 to 63 - index[2:0] := (a[i+h] << 2) OR (b[i+h] << 1) OR c[i+h] - dst[i+h] := imm8[index[2:0]] - ENDFOR - ELSE - dst[i+63:i] := 0 - FI + dst[i+63:i] := ROUND(a[i+63:i]) ENDFOR -dst[MAX:256] := 0 +dst[MAX:128] := 0 -
immintrin.h
 - - Integer - AVX512VL - AVX512F - Logical - - - - - Bitwise ternary logic that provides the capability to implement any three-operand binary function; the specific binary function is specified by value in "imm8". For each bit in each packed 64-bit integer, the corresponding bit from "a", "b", and "c" are used to form a 3 bit index into "imm8", and the value at that bit in "imm8" is written to the corresponding bit in "dst". + + Floating Point + SSE + Special Math Functions + + + Round the packed single-precision (32-bit) floating-point elements in "a" to the nearest integer value, and store the results as packed single-precision floating-point elements in "dst". This intrinsic may generate the "roundps"/"vroundps" instruction. FOR j := 0 to 3 - i := j*64 - FOR h := 0 to 63 - index[2:0] := (a[i+h] << 2) OR (b[i+h] << 1) OR c[i+h] - dst[i+h] := imm8[index[2:0]] - ENDFOR + i := j*32 + dst[i+31:i] := ROUND(a[i+31:i]) ENDFOR -dst[MAX:256] := 0 +dst[MAX:128] := 0 -
immintrin.h
 - - Integer - AVX512VL - AVX512F - Logical - - - - - - Bitwise ternary logic that provides the capability to implement any three-operand binary function; the specific binary function is specified by value in "imm8". For each bit in each packed 64-bit integer, the corresponding bit from "src", "a", and "b" are used to form a 3 bit index into "imm8", and the value at that bit in "imm8" is written to the corresponding bit in "dst" using writemask "k" at 64-bit granularity (64-bit elements are copied from "src" when the corresponding mask bit is not set). + + Floating Point + SSE + Elementary Math Functions + + + Compute the square root of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst". Note that this intrinsic is less efficient than "_mm_sqrt_pd". FOR j := 0 to 1 i := j*64 - IF k[j] - FOR h := 0 to 63 - index[2:0] := (src[i+h] << 2) OR (a[i+h] << 1) OR b[i+h] - dst[i+h] := imm8[index[2:0]] - ENDFOR - ELSE - dst[i+63:i] := src[i+63:i] - FI + dst[i+63:i] := SQRT(a[i+63:i]) ENDFOR dst[MAX:128] := 0 -
immintrin.h
 - - Integer - AVX512VL - AVX512F - Logical - - - - - - Bitwise ternary logic that provides the capability to implement any three-operand binary function; the specific binary function is specified by value in "imm8". For each bit in each packed 64-bit integer, the corresponding bit from "a", "b", and "c" are used to form a 3 bit index into "imm8", and the value at that bit in "imm8" is written to the corresponding bit in "dst" using zeromask "k" at 64-bit granularity (64-bit elements are zeroed out when the corresponding mask bit is not set). + + Floating Point + SSE + Elementary Math Functions + + + Compute the square root of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst". Note that this intrinsic is less efficient than "_mm_sqrt_ps". -FOR j := 0 to 1 - i := j*64 - IF k[j] - FOR h := 0 to 63 - index[2:0] := (a[i+h] << 2) OR (b[i+h] << 1) OR c[i+h] - dst[i+h] := imm8[index[2:0]] - ENDFOR - ELSE - dst[i+63:i] := 0 - FI +FOR j := 0 to 3 + i := j*32 + dst[i+31:i] := SQRT(a[i+31:i]) ENDFOR dst[MAX:128] := 0 -
immintrin.h
 - - Integer - AVX512VL - AVX512F - Logical - - - - - Bitwise ternary logic that provides the capability to implement any three-operand binary function; the specific binary function is specified by value in "imm8". For each bit in each packed 64-bit integer, the corresponding bit from "a", "b", and "c" are used to form a 3 bit index into "imm8", and the value at that bit in "imm8" is written to the corresponding bit in "dst". + + Floating Point + SSE + Trigonometry + + + Compute the tangent of packed double-precision (64-bit) floating-point elements in "a" expressed in radians, and store the results in "dst". FOR j := 0 to 1 i := j*64 - FOR h := 0 to 63 - index[2:0] := (a[i+h] << 2) OR (b[i+h] << 1) OR c[i+h] - dst[i+h] := imm8[index[2:0]] - ENDFOR + dst[i+63:i] := TAN(a[i+63:i]) ENDFOR dst[MAX:128] := 0 -
immintrin.h
 - - Integer - Mask - AVX512VL - AVX512BW - Compare - - - - Compute the bitwise AND of packed 8-bit integers in "a" and "b", producing intermediate 8-bit values, and set the corresponding bit in result mask "k" (subject to writemask "k") if the intermediate value is non-zero. + + Floating Point + SSE + Trigonometry + + + Compute the tangent of packed single-precision (32-bit) floating-point elements in "a" expressed in radians, and store the results in "dst". -FOR j := 0 to 31 - i := j*8 - IF k1[j] - k[j] := ((a[i+7:i] AND b[i+7:i]) != 0) ? 1 : 0 - ELSE - k[j] := 0 - FI +FOR j := 0 to 3 + i := j*32 + dst[i+31:i] := TAN(a[i+31:i]) ENDFOR -k[MAX:32] := 0 +dst[MAX:128] := 0 -
immintrin.h
 - - Integer - Mask - AVX512VL - AVX512BW - Compare - - - Compute the bitwise AND of packed 8-bit integers in "a" and "b", producing intermediate 8-bit values, and set the corresponding bit in result mask "k" if the intermediate value is non-zero. - -FOR j := 0 to 31 - i := j*8 - k[j] := ((a[i+7:i] AND b[i+7:i]) != 0) ? 1 : 0 + + Floating Point + SSE + Trigonometry + + + Compute the tangent of packed double-precision (64-bit) floating-point elements in "a" expressed in degrees, and store the results in "dst". + FOR j := 0 to 1 + i := j*64 + dst[i+63:i] := TAND(a[i+63:i]) ENDFOR -k[MAX:32] := 0 +dst[MAX:128] := 0 -
immintrin.h
 - - Integer - Mask - AVX512BW - Compare - - - - Compute the bitwise AND of packed 8-bit integers in "a" and "b", producing intermediate 8-bit values, and set the corresponding bit in result mask "k" (subject to writemask "k") if the intermediate value is non-zero. - -FOR j := 0 to 63 - i := j*8 - IF k1[j] - k[j] := ((a[i+7:i] AND b[i+7:i]) != 0) ? 1 : 0 - ELSE - k[j] := 0 - FI + + Floating Point + SSE + Trigonometry + + + Compute the tangent of packed single-precision (32-bit) floating-point elements in "a" expressed in degrees, and store the results in "dst". + FOR j := 0 to 3 + i := j*32 + dst[i+31:i] := TAND(a[i+31:i]) ENDFOR -k[MAX:64] := 0 +dst[MAX:128] := 0 -
immintrin.h
 - - Integer - Mask - AVX512BW - Compare - - - Compute the bitwise AND of packed 8-bit integers in "a" and "b", producing intermediate 8-bit values, and set the corresponding bit in result mask "k" if the intermediate value is non-zero. + + Floating Point + SSE + Trigonometry + + + Compute the hyperbolic tangent of packed double-precision (64-bit) floating-point elements in "a" expressed in radians, and store the results in "dst". -FOR j := 0 to 63 - i := j*8 - k[j] := ((a[i+7:i] AND b[i+7:i]) != 0) ? 1 : 0 +FOR j := 0 to 1 + i := j*64 + dst[i+63:i] := TANH(a[i+63:i]) ENDFOR -k[MAX:64] := 0 +dst[MAX:128] := 0 -
immintrin.h
 - - Integer - Mask - AVX512VL - AVX512BW - Compare - - - - Compute the bitwise AND of packed 8-bit integers in "a" and "b", producing intermediate 8-bit values, and set the corresponding bit in result mask "k" (subject to writemask "k") if the intermediate value is non-zero. + + Floating Point + SSE + Trigonometry + + + Compute the hyperbolic tangent of packed single-precision (32-bit) floating-point elements in "a" expressed in radians, and store the results in "dst". -FOR j := 0 to 15 - i := j*8 - IF k1[j] - k[j] := ((a[i+7:i] AND b[i+7:i]) != 0) ? 1 : 0 - ELSE - k[j] := 0 - FI +FOR j := 0 to 3 + i := j*32 + dst[i+31:i] := TANH(a[i+31:i]) ENDFOR -k[MAX:16] := 0 +dst[MAX:128] := 0 + +
immintrin.h
+ + + Floating Point + SSE + Miscellaneous + + + Truncate the packed double-precision (64-bit) floating-point elements in "a", and store the results as packed double-precision floating-point elements in "dst". This intrinsic may generate the "roundpd"/"vroundpd" instruction. + FOR j := 0 to 1 + i := j*64 + dst[i+63:i] := TRUNCATE(a[i+63:i]) +ENDFOR +dst[MAX:128] := 0 + +
immintrin.h
+ + + Floating Point + SSE + Miscellaneous + + + Truncate the packed single-precision (32-bit) floating-point elements in "a", and store the results as packed single-precision floating-point elements in "dst". This intrinsic may generate the "roundps"/"vroundps" instruction. + FOR j := 0 to 3 + i := j*32 + dst[i+31:i] := TRUNCATE(a[i+31:i]) +ENDFOR +dst[MAX:128] := 0 -
immintrin.h
 - + Integer - Mask - AVX512VL - AVX512BW - Compare - - - Compute the bitwise AND of packed 8-bit integers in "a" and "b", producing intermediate 8-bit values, and set the corresponding bit in result mask "k" if the intermediate value is non-zero. - -FOR j := 0 to 15 - i := j*8 - k[j] := ((a[i+7:i] AND b[i+7:i]) != 0) ? 1 : 0 + SSE + Arithmetic + + + + Divide packed unsigned 32-bit integers in "a" by packed elements in "b", and store the truncated results in "dst". + FOR j := 0 to 3 + i := 32*j + dst[i+31:i] := TRUNCATE(a[i+31:i] / b[i+31:i]) ENDFOR -k[MAX:16] := 0 +dst[MAX:128] := 0 -
immintrin.h
 - + Integer - Mask - AVX512VL - AVX512F - Compare - - - - Compute the bitwise AND of packed 32-bit integers in "a" and "b", producing intermediate 32-bit values, and set the corresponding bit in result mask "k" (subject to writemask "k") if the intermediate value is non-zero. - -FOR j := 0 to 7 - i := j*32 - IF k1[j] - k[j] := ((a[i+31:i] AND b[i+31:i]) != 0) ? 1 : 0 - ELSE - k[j] := 0 - FI + SSE + Arithmetic + + + + + Divide packed unsigned 32-bit integers in "a" by packed elements in "b", store the truncated results in "dst", and store the remainders as packed unsigned 32-bit integers into memory at "mem_addr". + FOR j := 0 to 3 + i := 32*j + dst[i+31:i] := TRUNCATE(a[i+31:i] / b[i+31:i]) + MEM[mem_addr+i+31:mem_addr+i] := REMAINDER(a[i+31:i] / b[i+31:i]) ENDFOR -k[MAX:8] := 0 +dst[MAX:128] := 0 -
immintrin.h
 - + Integer - Mask - AVX512VL - AVX512F - Compare - - - Compute the bitwise AND of packed 32-bit integers in "a" and "b", producing intermediate 32-bit values, and set the corresponding bit in result mask "k" if the intermediate value is non-zero. - -FOR j := 0 to 7 - i := j*32 - k[j] := ((a[i+31:i] AND b[i+31:i]) != 0) ? 1 : 0 + SSE + Arithmetic + + + + Divide packed unsigned 32-bit integers in "a" by packed elements in "b", and store the remainders as packed unsigned 32-bit integers in "dst". + FOR j := 0 to 3 + i := 32*j + dst[i+31:i] := REMAINDER(a[i+31:i] / b[i+31:i]) ENDFOR -k[MAX:8] := 0 +dst[MAX:128] := 0 -
immintrin.h
 - + Integer - Mask - AVX512VL - AVX512F - Compare - - - - Compute the bitwise AND of packed 32-bit integers in "a" and "b", producing intermediate 32-bit values, and set the corresponding bit in result mask "k" (subject to writemask "k") if the intermediate value is non-zero. + SSE + Store + + + + Store 16-bit integer from the first element of "a" into memory. "mem_addr" does not need to be aligned on any particular boundary. -FOR j := 0 to 3 - i := j*32 - IF k1[j] - k[j] := ((a[i+31:i] AND b[i+31:i]) != 0) ? 1 : 0 - ELSE - k[j] := 0 - FI -ENDFOR -k[MAX:4] := 0 +MEM[mem_addr+15:mem_addr] := a[15:0] -
immintrin.h
 - + Integer - Mask - AVX512VL - AVX512F - Compare - - - Compute the bitwise AND of packed 32-bit integers in "a" and "b", producing intermediate 32-bit values, and set the corresponding bit in result mask "k" if the intermediate value is non-zero. + SSE + Load + + + Load unaligned 64-bit integer from memory into the first element of "dst". -FOR j := 0 to 3 - i := j*32 - k[j] := ((a[i+31:i] AND b[i+31:i]) != 0) ? 1 : 0 -ENDFOR -k[MAX:4] := 0 +dst[63:0] := MEM[mem_addr+63:mem_addr] +dst[MAX:64] := 0 - +
immintrin.h
 - + Integer - Mask - AVX512VL - AVX512F - Compare - - - - Compute the bitwise AND of packed 64-bit integers in "a" and "b", producing intermediate 64-bit values, and set the corresponding bit in result mask "k" (subject to writemask "k") if the intermediate value is non-zero. + SSE + Store + + + + Store 64-bit integer from the first element of "a" into memory. "mem_addr" does not need to be aligned on any particular boundary. -FOR j := 0 to 3 - i := j*64 - IF k1[j] - k[j] := ((a[i+63:i] AND b[i+63:i]) != 0) ? 1 : 0 - ELSE - k[j] := 0 - FI -ENDFOR -k[MAX:4] := 0 +MEM[mem_addr+63:mem_addr] := a[63:0] - +
immintrin.h
 - + Integer - Mask - AVX512VL - AVX512F - Compare - - - Compute the bitwise AND of packed 64-bit integers in "a" and "b", producing intermediate 64-bit values, and set the corresponding bit in result mask "k" if the intermediate value is non-zero. + SSE + Load + + + Load unaligned 16-bit integer from memory into the first element of "dst". -FOR j := 0 to 3 - i := j*64 - k[j] := ((a[i+63:i] AND b[i+63:i]) != 0) ? 1 : 0 -ENDFOR -k[MAX:4] := 0 +dst[15:0] := MEM[mem_addr+15:mem_addr] +dst[MAX:16] := 0 -
immintrin.h
 - + + Floating Point + SSE2 + General Support + + + Return vector of type __m128d with undefined elements. +
immintrin.h
+ + Integer - Mask - AVX512VL - AVX512F - Compare - - - - Compute the bitwise AND of packed 64-bit integers in "a" and "b", producing intermediate 64-bit values, and set the corresponding bit in result mask "k" (subject to writemask "k") if the intermediate value is non-zero. + SSE2 + General Support + + + Return vector of type __m128i with undefined elements. +
immintrin.h
+ + + Integer + SSE2 + Load + + + Load unaligned 32-bit integer from memory into the first element of "dst". -FOR j := 0 to 1 - i := j*64 - IF k1[j] - k[j] := ((a[i+63:i] AND b[i+63:i]) != 0) ? 1 : 0 - ELSE - k[j] := 0 - FI -ENDFOR -k[MAX:2] := 0 +dst[31:0] := MEM[mem_addr+31:mem_addr] +dst[MAX:32] := 0 - +
immintrin.h
 - + Integer - Mask - AVX512VL - AVX512F - Compare - - - Compute the bitwise AND of packed 64-bit integers in "a" and "b", producing intermediate 64-bit values, and set the corresponding bit in result mask "k" if the intermediate value is non-zero. + SSE2 + Store + + + + Store 32-bit integer from the first element of "a" into memory. "mem_addr" does not need to be aligned on any particular boundary. -FOR j := 0 to 1 - i := j*64 - k[j] := ((a[i+63:i] AND b[i+63:i]) != 0) ? 1 : 0 -ENDFOR -k[MAX:2] := 0 +MEM[mem_addr+31:mem_addr] := a[31:0] - +
immintrin.h
 - + + SSE2 + General Support + + + Provide a hint to the processor that the code sequence is a spin-wait loop. This can help improve the performance and power consumption of spin-wait loops. + +
emmintrin.h
+ + + SSE2 + General Support + + + Invalidate and flush the cache line that contains "p" from all levels of the cache hierarchy. + +
emmintrin.h
+ + + SSE2 + General Support + + + Perform a serializing operation on all load-from-memory instructions that were issued prior to this instruction. Guarantees that every load instruction that precedes, in program order, is globally visible before any load instruction which follows the fence in program order. + +
emmintrin.h
+ + + SSE2 + General Support + + + Perform a serializing operation on all load-from-memory and store-to-memory instructions that were issued prior to this instruction. Guarantees that every memory access that precedes, in program order, the memory fence instruction is globally visible before any memory instruction which follows the fence in program order. + +
emmintrin.h
+ + Integer - Mask - AVX512VL - AVX512BW - Compare - - - - Compute the bitwise AND of packed 16-bit integers in "a" and "b", producing intermediate 16-bit values, and set the corresponding bit in result mask "k" (subject to writemask "k") if the intermediate value is non-zero. + SSE2 + Arithmetic + + + + Add packed 8-bit integers in "a" and "b", and store the results in "dst". FOR j := 0 to 15 - i := j*16 - IF k1[j] - k[j] := ((a[i+15:i] AND b[i+15:i]) != 0) ? 1 : 0 - ELSE - k[j] := 0 - FI + i := j*8 + dst[i+7:i] := a[i+7:i] + b[i+7:i] ENDFOR -k[MAX:16] := 0 - -
immintrin.h
+ +
emmintrin.h
 - + Integer - Mask - AVX512VL - AVX512BW - Compare - - - Compute the bitwise AND of packed 16-bit integers in "a" and "b", producing intermediate 16-bit values, and set the corresponding bit in result mask "k" if the intermediate value is non-zero. + SSE2 + Arithmetic + + + + Add packed 16-bit integers in "a" and "b", and store the results in "dst". -FOR j := 0 to 15 +FOR j := 0 to 7 i := j*16 - k[j] := ((a[i+15:i] AND b[i+15:i]) != 0) ? 1 : 0 + dst[i+15:i] := a[i+15:i] + b[i+15:i] ENDFOR -k[MAX:16] := 0 - -
immintrin.h
+ +
emmintrin.h
 - + Integer - Mask - AVX512BW - Compare - - - - Compute the bitwise AND of packed 16-bit integers in "a" and "b", producing intermediate 16-bit values, and set the corresponding bit in result mask "k" (subject to writemask "k") if the intermediate value is non-zero. + SSE2 + Arithmetic + + + + Add packed 32-bit integers in "a" and "b", and store the results in "dst". -FOR j := 0 to 31 - i := j*16 - IF k1[j] - k[j] := ((a[i+15:i] AND b[i+15:i]) != 0) ? 1 : 0 - ELSE - k[j] := 0 - FI +FOR j := 0 to 3 + i := j*32 + dst[i+31:i] := a[i+31:i] + b[i+31:i] ENDFOR -k[MAX:32] := 0 - -
immintrin.h
+ +
emmintrin.h
 - + Integer - Mask - AVX512BW - Compare - - - Compute the bitwise AND of packed 16-bit integers in "a" and "b", producing intermediate 16-bit values, and set the corresponding bit in result mask "k" if the intermediate value is non-zero. + SSE2 + Arithmetic + + + + Add 64-bit integers "a" and "b", and store the result in "dst". -FOR j := 0 to 31 - i := j*16 - k[j] := ((a[i+15:i] AND b[i+15:i]) != 0) ? 1 : 0 +dst[63:0] := a[63:0] + b[63:0] + + +
emmintrin.h
+ + + Integer + SSE2 + Arithmetic + + + + Add packed 64-bit integers in "a" and "b", and store the results in "dst". + +FOR j := 0 to 1 + i := j*64 + dst[i+63:i] := a[i+63:i] + b[i+63:i] ENDFOR -k[MAX:32] := 0 - -
immintrin.h
+ +
emmintrin.h
 - + Integer - Mask - AVX512VL - AVX512BW - Compare - - - - Compute the bitwise AND of packed 16-bit integers in "a" and "b", producing intermediate 16-bit values, and set the corresponding bit in result mask "k" (subject to writemask "k") if the intermediate value is non-zero. + SSE2 + Arithmetic + + + + Add packed signed 8-bit integers in "a" and "b" using saturation, and store the results in "dst". -FOR j := 0 to 7 - i := j*16 - IF k1[j] - k[j] := ((a[i+15:i] AND b[i+15:i]) != 0) ? 1 : 0 - ELSE - k[j] := 0 - FI +FOR j := 0 to 15 + i := j*8 + dst[i+7:i] := Saturate8( a[i+7:i] + b[i+7:i] ) ENDFOR -k[MAX:8] := 0 - -
immintrin.h
+ +
emmintrin.h
 - + Integer - Mask - AVX512VL - AVX512BW - Compare - - - Compute the bitwise AND of packed 16-bit integers in "a" and "b", producing intermediate 16-bit values, and set the corresponding bit in result mask "k" if the intermediate value is non-zero. + SSE2 + Arithmetic + + + + Add packed signed 16-bit integers in "a" and "b" using saturation, and store the results in "dst". FOR j := 0 to 7 i := j*16 - k[j] := ((a[i+15:i] AND b[i+15:i]) != 0) ? 1 : 0 + dst[i+15:i] := Saturate16( a[i+15:i] + b[i+15:i] ) ENDFOR -k[MAX:8] := 0 - -
immintrin.h
+ +
emmintrin.h
 - + Integer - Mask - AVX512VL - AVX512BW - Compare - - - - Compute the bitwise NAND of packed 8-bit integers in "a" and "b", producing intermediate 8-bit values, and set the corresponding bit in result mask "k" (subject to writemask "k") if the intermediate value is zero. + SSE2 + Arithmetic + + + + Add packed unsigned 8-bit integers in "a" and "b" using saturation, and store the results in "dst". -FOR j := 0 to 31 +FOR j := 0 to 15 i := j*8 - IF k1[j] - k[j] := ((a[i+7:i] AND b[i+7:i]) == 0) ? 1 : 0 - ELSE - k[j] := 0 - FI + dst[i+7:i] := SaturateU8( a[i+7:i] + b[i+7:i] ) ENDFOR -k[MAX:32] := 0 - -
immintrin.h
+ +
emmintrin.h
 - + Integer - Mask - AVX512VL - AVX512BW - Compare - - - Compute the bitwise NAND of packed 8-bit integers in "a" and "b", producing intermediate 8-bit values, and set the corresponding bit in result mask "k" if the intermediate value is zero. + SSE2 + Arithmetic + + + + Add packed unsigned 16-bit integers in "a" and "b" using saturation, and store the results in "dst". -FOR j := 0 to 31 - i := j*8 - k[j] := ((a[i+7:i] AND b[i+7:i]) == 0) ? 1 : 0 +FOR j := 0 to 7 + i := j*16 + dst[i+15:i] := SaturateU16( a[i+15:i] + b[i+15:i] ) ENDFOR -k[MAX:32] := 0 - -
immintrin.h
+ +
emmintrin.h
 - + Integer - Mask - AVX512BW - Compare - - - - Compute the bitwise NAND of packed 8-bit integers in "a" and "b", producing intermediate 8-bit values, and set the corresponding bit in result mask "k" (subject to writemask "k") if the intermediate value is zero. + SSE2 + Probability/Statistics + + + + Average packed unsigned 8-bit integers in "a" and "b", and store the results in "dst". -FOR j := 0 to 63 +FOR j := 0 to 15 i := j*8 - IF k1[j] - k[j] := ((a[i+7:i] AND b[i+7:i]) == 0) ? 1 : 0 - ELSE - k[j] := 0 - FI + dst[i+7:i] := (a[i+7:i] + b[i+7:i] + 1) >> 1 ENDFOR -k[MAX:64] := 0 - -
immintrin.h
+ +
emmintrin.h
 - + Integer - Mask - AVX512BW - Compare - - - Compute the bitwise NAND of packed 8-bit integers in "a" and "b", producing intermediate 8-bit values, and set the corresponding bit in result mask "k" if the intermediate value is zero. + SSE2 + Probability/Statistics + + + + Average packed unsigned 16-bit integers in "a" and "b", and store the results in "dst". -FOR j := 0 to 63 - i := j*8 - k[j] := ((a[i+7:i] AND b[i+7:i]) == 0) ? 1 : 0 +FOR j := 0 to 7 + i := j*16 + dst[i+15:i] := (a[i+15:i] + b[i+15:i] + 1) >> 1 ENDFOR -k[MAX:64] := 0 - -
immintrin.h
+ +
emmintrin.h
 - + Integer - Mask - AVX512VL - AVX512BW - Compare - - - - Compute the bitwise NAND of packed 8-bit integers in "a" and "b", producing intermediate 8-bit values, and set the corresponding bit in result mask "k" (subject to writemask "k") if the intermediate value is zero. + SSE2 + Arithmetic + + + + Multiply packed signed 16-bit integers in "a" and "b", producing intermediate signed 32-bit integers. Horizontally add adjacent pairs of intermediate 32-bit integers, and pack the results in "dst". -FOR j := 0 to 15 - i := j*8 - IF k1[j] - k[j] := ((a[i+7:i] AND b[i+7:i]) == 0) ? 1 : 0 - ELSE - k[j] := 0 - FI +FOR j := 0 to 3 + i := j*32 + dst[i+31:i] := SignExtend32(a[i+31:i+16]*b[i+31:i+16]) + SignExtend32(a[i+15:i]*b[i+15:i]) ENDFOR -k[MAX:16] := 0 - -
immintrin.h
+ +
emmintrin.h
 - + Integer - Mask - AVX512VL - AVX512BW - Compare - - - Compute the bitwise NAND of packed 8-bit integers in "a" and "b", producing intermediate 8-bit values, and set the corresponding bit in result mask "k" if the intermediate value is zero. + SSE2 + Special Math Functions + + + + Compare packed signed 16-bit integers in "a" and "b", and store packed maximum values in "dst". -FOR j := 0 to 15 - i := j*8 - k[j] := ((a[i+7:i] AND b[i+7:i]) == 0) ? 1 : 0 +FOR j := 0 to 7 + i := j*16 + dst[i+15:i] := MAX(a[i+15:i], b[i+15:i]) ENDFOR -k[MAX:16] := 0 - -
immintrin.h
+ +
emmintrin.h
 - + Integer - Mask - AVX512VL - AVX512F - Compare - - - - Compute the bitwise NAND of packed 32-bit integers in "a" and "b", producing intermediate 32-bit values, and set the corresponding bit in result mask "k" (subject to writemask "k") if the intermediate value is zero. + SSE2 + Special Math Functions + + + + Compare packed unsigned 8-bit integers in "a" and "b", and store packed maximum values in "dst". -FOR j := 0 to 7 - i := j*32 - IF k1[j] - k[j] := ((a[i+31:i] AND b[i+31:i]) == 0) ? 1 : 0 - ELSE - k[j] := 0 - FI +FOR j := 0 to 15 + i := j*8 + dst[i+7:i] := MAX(a[i+7:i], b[i+7:i]) ENDFOR -k[MAX:8] := 0 - -
immintrin.h
+ +
emmintrin.h
 - + Integer - Mask - AVX512VL - AVX512F - Compare - - - Compute the bitwise NAND of packed 32-bit integers in "a" and "b", producing intermediate 32-bit values, and set the corresponding bit in result mask "k" if the intermediate value is zero. + SSE2 + Special Math Functions + + + + Compare packed signed 16-bit integers in "a" and "b", and store packed minimum values in "dst". FOR j := 0 to 7 - i := j*32 - k[j] := ((a[i+31:i] AND b[i+31:i]) == 0) ? 1 : 0 + i := j*16 + dst[i+15:i] := MIN(a[i+15:i], b[i+15:i]) ENDFOR -k[MAX:8] := 0 - -
immintrin.h
+ +
emmintrin.h
 - + Integer - Mask - AVX512VL - AVX512F - Compare - - - - Compute the bitwise NAND of packed 32-bit integers in "a" and "b", producing intermediate 32-bit values, and set the corresponding bit in result mask "k" (subject to writemask "k") if the intermediate value is zero. + SSE2 + Special Math Functions + + + + Compare packed unsigned 8-bit integers in "a" and "b", and store packed minimum values in "dst". -FOR j := 0 to 3 - i := j*32 - IF k1[j] - k[j] := ((a[i+31:i] AND b[i+31:i]) == 0) ? 1 : 0 - ELSE - k[j] := 0 - FI +FOR j := 0 to 15 + i := j*8 + dst[i+7:i] := MIN(a[i+7:i], b[i+7:i]) ENDFOR -k[MAX:4] := 0 - -
immintrin.h
+ +
emmintrin.h
 - + Integer - Mask - AVX512VL - AVX512F - Compare - - - Compute the bitwise NAND of packed 32-bit integers in "a" and "b", producing intermediate 32-bit values, and set the corresponding bit in result mask "k" if the intermediate value is zero. + SSE2 + Arithmetic + + + + Multiply the packed signed 16-bit integers in "a" and "b", producing intermediate 32-bit integers, and store the high 16 bits of the intermediate integers in "dst". -FOR j := 0 to 3 - i := j*32 - k[j] := ((a[i+31:i] NAND b[i+31:i]) == 0) ? 1 : 0 +FOR j := 0 to 7 + i := j*16 + tmp[31:0] := SignExtend32(a[i+15:i]) * SignExtend32(b[i+15:i]) + dst[i+15:i] := tmp[31:16] ENDFOR -k[MAX:4] := 0 - -
immintrin.h
+ +
emmintrin.h
 - + Integer - Mask - AVX512VL - AVX512F - Compare - - - - Compute the bitwise NAND of packed 64-bit integers in "a" and "b", producing intermediate 64-bit values, and set the corresponding bit in result mask "k" (subject to writemask "k") if the intermediate value is zero. + SSE2 + Arithmetic + + + + Multiply the packed unsigned 16-bit integers in "a" and "b", producing intermediate 32-bit integers, and store the high 16 bits of the intermediate integers in "dst". -FOR j := 0 to 3 - i := j*64 - IF k1[j] - k[j] := ((a[i+63:i] AND b[i+63:i]) == 0) ? 1 : 0 - ELSE - k[j] := 0 - FI +FOR j := 0 to 7 + i := j*16 + tmp[31:0] := a[i+15:i] * b[i+15:i] + dst[i+15:i] := tmp[31:16] ENDFOR -k[MAX:4] := 0 - -
immintrin.h
+ +
emmintrin.h
 - + Integer - Mask - AVX512VL - AVX512F - Compare - - - Compute the bitwise NAND of packed 64-bit integers in "a" and "b", producing intermediate 64-bit values, and set the corresponding bit in result mask "k" if the intermediate value is zero. + SSE2 + Arithmetic + + + + Multiply the packed 16-bit integers in "a" and "b", producing intermediate 32-bit integers, and store the low 16 bits of the intermediate integers in "dst". -FOR j := 0 to 3 - i := j*64 - k[j] := ((a[i+63:i] AND b[i+63:i]) == 0) ? 1 : 0 +FOR j := 0 to 7 + i := j*16 + tmp[31:0] := SignExtend32(a[i+15:i]) * SignExtend32(b[i+15:i]) + dst[i+15:i] := tmp[15:0] ENDFOR -k[MAX:4] := 0 - -
immintrin.h
+ +
emmintrin.h
 - + Integer - Mask - AVX512VL - AVX512F - Compare - - - - Compute the bitwise NAND of packed 64-bit integers in "a" and "b", producing intermediate 64-bit values, and set the corresponding bit in result mask "k" (subject to writemask "k") if the intermediate value is zero. + SSE2 + Arithmetic + + + + Multiply the low unsigned 32-bit integers from "a" and "b", and store the unsigned 64-bit result in "dst". -FOR j := 0 to 1 - i := j*64 - IF k1[j] - k[j] := ((a[i+63:i] AND b[i+63:i]) == 0) ? 1 : 0 - ELSE - k[j] := 0 - FI -ENDFOR -k[MAX:2] := 0 +dst[63:0] := a[31:0] * b[31:0] - -
immintrin.h
+ +
emmintrin.h
 - + Integer - Mask - AVX512VL - AVX512F - Compare - - - Compute the bitwise NAND of packed 64-bit integers in "a" and "b", producing intermediate 64-bit values, and set the corresponding bit in result mask "k" if the intermediate value is zero. + SSE2 + Arithmetic + + + + Multiply the low unsigned 32-bit integers from each packed 64-bit element in "a" and "b", and store the unsigned 64-bit results in "dst". FOR j := 0 to 1 i := j*64 - k[j] := ((a[i+63:i] AND b[i+63:i]) == 0) ? 1 : 0 + dst[i+63:i] := a[i+31:i] * b[i+31:i] ENDFOR -k[MAX:2] := 0 - -
immintrin.h
+ +
emmintrin.h
 - + Integer - Mask - AVX512VL - AVX512BW - Compare - - - - Compute the bitwise NAND of packed 16-bit integers in "a" and "b", producing intermediate 16-bit values, and set the corresponding bit in result mask "k" (subject to writemask "k") if the intermediate value is zero. + SSE2 + Arithmetic + Miscellaneous + + + + Compute the absolute differences of packed unsigned 8-bit integers in "a" and "b", then horizontally sum each consecutive 8 differences to produce two unsigned 16-bit integers, and pack these unsigned 16-bit integers in the low 16 bits of 64-bit elements in "dst". -FOR j := 0 to 15 - i := j*16 - IF k1[j] - k[j] := ((a[i+15:i] AND b[i+15:i]) == 0) ? 1 : 0 - ELSE - k[j] := 0 - FI +FOR j := 0 to 15 + i := j*8 + tmp[i+7:i] := ABS(a[i+7:i] - b[i+7:i]) +ENDFOR +FOR j := 0 to 1 + i := j*64 + dst[i+15:i] := tmp[i+7:i] + tmp[i+15:i+8] + tmp[i+23:i+16] + tmp[i+31:i+24] + \ + tmp[i+39:i+32] + tmp[i+47:i+40] + tmp[i+55:i+48] + tmp[i+63:i+56] + dst[i+63:i+16] := 0 ENDFOR -k[MAX:16] := 0 - -
immintrin.h
+ +
emmintrin.h
 - + Integer - Mask - AVX512VL - AVX512BW - Compare - - - Compute the bitwise NAND of packed 16-bit integers in "a" and "b", producing intermediate 16-bit values, and set the corresponding bit in result mask "k" if the intermediate value is zero. + SSE2 + Arithmetic + + + + Subtract packed 8-bit integers in "b" from packed 8-bit integers in "a", and store the results in "dst". FOR j := 0 to 15 - i := j*16 - k[j] := ((a[i+15:i] AND b[i+15:i]) == 0) ? 1 : 0 + i := j*8 + dst[i+7:i] := a[i+7:i] - b[i+7:i] ENDFOR -k[MAX:16] := 0 - -
immintrin.h
+ +
emmintrin.h
 - + Integer - Mask - AVX512BW - Compare - - - - Compute the bitwise NAND of packed 16-bit integers in "a" and "b", producing intermediate 16-bit values, and set the corresponding bit in result mask "k" (subject to writemask "k") if the intermediate value is zero. + SSE2 + Arithmetic + + + + Subtract packed 16-bit integers in "b" from packed 16-bit integers in "a", and store the results in "dst". -FOR j := 0 to 31 +FOR j := 0 to 7 i := j*16 - IF k1[j] - k[j] := ((a[i+15:i] AND b[i+15:i]) == 0) ? 1 : 0 - ELSE - k[j] := 0 - FI + dst[i+15:i] := a[i+15:i] - b[i+15:i] ENDFOR -k[MAX:32] := 0 - -
immintrin.h
+ +
emmintrin.h
 - + Integer - Mask - AVX512BW - Compare - - - Compute the bitwise NAND of packed 16-bit integers in "a" and "b", producing intermediate 16-bit values, and set the corresponding bit in result mask "k" if the intermediate value is zero. + SSE2 + Arithmetic + + + + Subtract packed 32-bit integers in "b" from packed 32-bit integers in "a", and store the results in "dst". -FOR j := 0 to 31 - i := j*16 - k[j] := ((a[i+15:i] AND b[i+15:i]) == 0) ? 1 : 0 +FOR j := 0 to 3 + i := j*32 + dst[i+31:i] := a[i+31:i] - b[i+31:i] ENDFOR -k[MAX:32] := 0 - -
immintrin.h
+ +
emmintrin.h
 - + Integer - Mask - AVX512VL - AVX512BW - Compare - - - - Compute the bitwise NAND of packed 16-bit integers in "a" and "b", producing intermediate 16-bit values, and set the corresponding bit in result mask "k" (subject to writemask "k") if the intermediate value is zero. + SSE2 + Arithmetic + + + + Subtract 64-bit integer "b" from 64-bit integer "a", and store the result in "dst". -FOR j := 0 to 7 - i := j*16 - IF k1[j] - k[j] := ((a[i+15:i] AND b[i+15:i]) == 0) ? 1 : 0 - ELSE - k[j] := 0 - FI -ENDFOR -k[MAX:8] := 0 +dst[63:0] := a[63:0] - b[63:0] - -
immintrin.h
+ +
emmintrin.h
 - + Integer - Mask - AVX512VL - AVX512BW - Compare - - - Compute the bitwise NAND of packed 16-bit integers in "a" and "b", producing intermediate 16-bit values, and set the corresponding bit in result mask "k" if the intermediate value is zero. + SSE2 + Arithmetic + + + + Subtract packed 64-bit integers in "b" from packed 64-bit integers in "a", and store the results in "dst". -FOR j := 0 to 7 - i := j*16 - k[j] := ((a[i+15:i] AND b[i+15:i]) == 0) ? 1 : 0 +FOR j := 0 to 1 + i := j*64 + dst[i+63:i] := a[i+63:i] - b[i+63:i] ENDFOR -k[MAX:8] := 0 - -
immintrin.h
+ +
emmintrin.h
 - + Integer - AVX512VL - AVX512BW - Miscellaneous - - - - - Unpack and interleave 8-bit integers from the high half of each 128-bit lane in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + SSE2 + Arithmetic + + + + Subtract packed signed 8-bit integers in "b" from packed 8-bit integers in "a" using saturation, and store the results in "dst". -INTERLEAVE_HIGH_BYTES(src1[127:0], src2[127:0]){ - dst[7:0] := src1[71:64] - dst[15:8] := src2[71:64] - dst[23:16] := src1[79:72] - dst[31:24] := src2[79:72] - dst[39:32] := src1[87:80] - dst[47:40] := src2[87:80] - dst[55:48] := src1[95:88] - dst[63:56] := src2[95:88] - dst[71:64] := src1[103:96] - dst[79:72] := src2[103:96] - dst[87:80] := src1[111:104] - dst[95:88] := src2[111:104] - dst[103:96] := src1[119:112] - dst[111:104] := src2[119:112] - dst[119:112] := src1[127:120] - dst[127:120] := src2[127:120] - RETURN dst[127:0] -} - -tmp_dst[127:0] := INTERLEAVE_HIGH_BYTES(a[127:0], b[127:0]) -tmp_dst[255:128] := INTERLEAVE_HIGH_BYTES(a[255:128], b[255:128]) - -FOR j := 0 to 31 +FOR j := 0 to 15 i := j*8 - IF k[j] - dst[i+7:i] := tmp_dst[i+7:i] - ELSE - dst[i+7:i] := src[i+7:i] - FI + dst[i+7:i] := Saturate8(a[i+7:i] - b[i+7:i]) ENDFOR -dst[MAX:256] := 0 - -
immintrin.h
+ +
emmintrin.h
 - + Integer - AVX512VL - AVX512BW - Miscellaneous - - - - Unpack and interleave 8-bit integers from the high half of each 128-bit lane in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + SSE2 + Arithmetic + + + + Subtract packed signed 16-bit integers in "b" from packed 16-bit integers in "a" using saturation, and store the results in "dst". -INTERLEAVE_HIGH_BYTES(src1[127:0], src2[127:0]){ - dst[7:0] := src1[71:64] - dst[15:8] := src2[71:64] - dst[23:16] := src1[79:72] - dst[31:24] := src2[79:72] - dst[39:32] := src1[87:80] - dst[47:40] := src2[87:80] - dst[55:48] := src1[95:88] - dst[63:56] := src2[95:88] - dst[71:64] := src1[103:96] - dst[79:72] := src2[103:96] - dst[87:80] := src1[111:104] - dst[95:88] := src2[111:104] - dst[103:96] := src1[119:112] - dst[111:104] := src2[119:112] - dst[119:112] := src1[127:120] - dst[127:120] := src2[127:120] - RETURN dst[127:0] -} - -tmp_dst[127:0] := INTERLEAVE_HIGH_BYTES(a[127:0], b[127:0]) -tmp_dst[255:128] := INTERLEAVE_HIGH_BYTES(a[255:128], b[255:128]) - -FOR j := 0 to 31 - i := j*8 - IF k[j] - dst[i+7:i] := tmp_dst[i+7:i] - ELSE - dst[i+7:i] := 0 - FI +FOR j := 0 to 7 + i := j*16 + dst[i+15:i] := Saturate16(a[i+15:i] - b[i+15:i]) ENDFOR -dst[MAX:256] := 0 - -
immintrin.h
+ +
emmintrin.h
 - + Integer - AVX512BW - Miscellaneous - - - - - Unpack and interleave 8-bit integers from the high half of each 128-bit lane in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + SSE2 + Arithmetic + + + + Subtract packed unsigned 8-bit integers in "b" from packed unsigned 8-bit integers in "a" using saturation, and store the results in "dst". -INTERLEAVE_HIGH_BYTES(src1[127:0], src2[127:0]){ - dst[7:0] := src1[71:64] - dst[15:8] := src2[71:64] - dst[23:16] := src1[79:72] - dst[31:24] := src2[79:72] - dst[39:32] := src1[87:80] - dst[47:40] := src2[87:80] - dst[55:48] := src1[95:88] - dst[63:56] := src2[95:88] - dst[71:64] := src1[103:96] - dst[79:72] := src2[103:96] - dst[87:80] := src1[111:104] - dst[95:88] := src2[111:104] - dst[103:96] := src1[119:112] - dst[111:104] := src2[119:112] - dst[119:112] := src1[127:120] - dst[127:120] := src2[127:120] - RETURN dst[127:0] -} - -tmp_dst[127:0] := INTERLEAVE_HIGH_BYTES(a[127:0], b[127:0]) -tmp_dst[255:128] := INTERLEAVE_HIGH_BYTES(a[255:128], b[255:128]) -tmp_dst[383:256] := INTERLEAVE_HIGH_BYTES(a[383:256], b[383:256]) -tmp_dst[511:384] := INTERLEAVE_HIGH_BYTES(a[511:384], b[511:384]) - -FOR j := 0 to 63 +FOR j := 0 to 15 i := j*8 - IF k[j] - dst[i+7:i] := tmp_dst[i+7:i] - ELSE - dst[i+7:i] := src[i+7:i] - FI + dst[i+7:i] := SaturateU8(a[i+7:i] - b[i+7:i]) ENDFOR -dst[MAX:512] := 0 - -
immintrin.h
+ +
emmintrin.h
 - + Integer - AVX512BW - Miscellaneous - - - - Unpack and interleave 8-bit integers from the high half of each 128-bit lane in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + SSE2 + Arithmetic + + + + Subtract packed unsigned 16-bit integers in "b" from packed unsigned 16-bit integers in "a" using saturation, and store the results in "dst". -INTERLEAVE_HIGH_BYTES(src1[127:0], src2[127:0]){ - dst[7:0] := src1[71:64] - dst[15:8] := src2[71:64] - dst[23:16] := src1[79:72] - dst[31:24] := src2[79:72] - dst[39:32] := src1[87:80] - dst[47:40] := src2[87:80] - dst[55:48] := src1[95:88] - dst[63:56] := src2[95:88] - dst[71:64] := src1[103:96] - dst[79:72] := src2[103:96] - dst[87:80] := src1[111:104] - dst[95:88] := src2[111:104] - dst[103:96] := src1[119:112] - dst[111:104] := src2[119:112] - dst[119:112] := src1[127:120] - dst[127:120] := src2[127:120] - RETURN dst[127:0] -} - -tmp_dst[127:0] := INTERLEAVE_HIGH_BYTES(a[127:0], b[127:0]) -tmp_dst[255:128] := INTERLEAVE_HIGH_BYTES(a[255:128], b[255:128]) -tmp_dst[383:256] := INTERLEAVE_HIGH_BYTES(a[383:256], b[383:256]) -tmp_dst[511:384] := INTERLEAVE_HIGH_BYTES(a[511:384], b[511:384]) - -FOR j := 0 to 63 - i := j*8 - IF k[j] - dst[i+7:i] := tmp_dst[i+7:i] - ELSE - dst[i+7:i] := 0 - FI +FOR j := 0 to 7 + i := j*16 + dst[i+15:i] := SaturateU16(a[i+15:i] - b[i+15:i]) ENDFOR -dst[MAX:512] := 0 - -
immintrin.h
+ +
emmintrin.h
 - + Integer - AVX512BW - Miscellaneous - - - Unpack and interleave 8-bit integers from the high half of each 128-bit lane in "a" and "b", and store the results in "dst". + SSE2 + Shift + + + + Shift "a" left by "imm8" bytes while shifting in zeros, and store the results in "dst". -INTERLEAVE_HIGH_BYTES(src1[127:0], src2[127:0]){ - dst[7:0] := src1[71:64] - dst[15:8] := src2[71:64] - dst[23:16] := src1[79:72] - dst[31:24] := src2[79:72] - dst[39:32] := src1[87:80] - dst[47:40] := src2[87:80] - dst[55:48] := src1[95:88] - dst[63:56] := src2[95:88] - dst[71:64] := src1[103:96] - dst[79:72] := src2[103:96] - dst[87:80] := src1[111:104] - dst[95:88] := src2[111:104] - dst[103:96] := src1[119:112] - dst[111:104] := src2[119:112] - dst[119:112] := src1[127:120] - dst[127:120] := src2[127:120] - RETURN dst[127:0] -} - -dst[127:0] := INTERLEAVE_HIGH_BYTES(a[127:0], b[127:0]) -dst[255:128] := INTERLEAVE_HIGH_BYTES(a[255:128], b[255:128]) -dst[383:256] := INTERLEAVE_HIGH_BYTES(a[383:256], b[383:256]) -dst[511:384] := INTERLEAVE_HIGH_BYTES(a[511:384], b[511:384]) -dst[MAX:512] := 0 +tmp := imm8[7:0] +IF tmp > 15 + tmp := 16 +FI +dst[127:0] := a[127:0] << (tmp*8) - -
immintrin.h
+ +
emmintrin.h
 - + Integer - AVX512VL - AVX512BW - Miscellaneous - - - - - Unpack and interleave 8-bit integers from the high half of "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + SSE2 + Shift + + + + Shift "a" left by "imm8" bytes while shifting in zeros, and store the results in "dst". -INTERLEAVE_HIGH_BYTES(src1[127:0], src2[127:0]){ - dst[7:0] := src1[71:64] - dst[15:8] := src2[71:64] - dst[23:16] := src1[79:72] - dst[31:24] := src2[79:72] - dst[39:32] := src1[87:80] - dst[47:40] := src2[87:80] - dst[55:48] := src1[95:88] - dst[63:56] := src2[95:88] - dst[71:64] := src1[103:96] - dst[79:72] := src2[103:96] - dst[87:80] := src1[111:104] - dst[95:88] := src2[111:104] - dst[103:96] := src1[119:112] - dst[111:104] := src2[119:112] - dst[119:112] := src1[127:120] - dst[127:120] := src2[127:120] - RETURN dst[127:0] -} - -tmp_dst[127:0] := INTERLEAVE_HIGH_BYTES(a[127:0], b[127:0]) - -FOR j := 0 to 15 - i := j*8 - IF k[j] - dst[i+7:i] := tmp_dst[i+7:i] - ELSE - dst[i+7:i] := src[i+7:i] - FI -ENDFOR -dst[MAX:128] := 0 +tmp := imm8[7:0] +IF tmp > 15 + tmp := 16 +FI +dst[127:0] := a[127:0] << (tmp*8) - -
immintrin.h
+ +
emmintrin.h
 - + Integer - AVX512VL - AVX512BW - Miscellaneous - - - - Unpack and interleave 8-bit integers from the high half of "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + SSE2 + Shift + + + + Shift "a" right by "imm8" bytes while shifting in zeros, and store the results in "dst". -INTERLEAVE_HIGH_BYTES(src1[127:0], src2[127:0]){ - dst[7:0] := src1[71:64] - dst[15:8] := src2[71:64] - dst[23:16] := src1[79:72] - dst[31:24] := src2[79:72] - dst[39:32] := src1[87:80] - dst[47:40] := src2[87:80] - dst[55:48] := src1[95:88] - dst[63:56] := src2[95:88] - dst[71:64] := src1[103:96] - dst[79:72] := src2[103:96] - dst[87:80] := src1[111:104] - dst[95:88] := src2[111:104] - dst[103:96] := src1[119:112] - dst[111:104] := src2[119:112] - dst[119:112] := src1[127:120] - dst[127:120] := src2[127:120] - RETURN dst[127:0] -} - -tmp_dst[127:0] := INTERLEAVE_HIGH_BYTES(a[127:0], b[127:0]) - -FOR j := 0 to 15 - i := j*8 - IF k[j] - dst[i+7:i] := tmp_dst[i+7:i] - ELSE - dst[i+7:i] := 0 - FI -ENDFOR -dst[MAX:128] := 0 +tmp := imm8[7:0] +IF tmp > 15 + tmp := 16 +FI +dst[127:0] := a[127:0] >> (tmp*8) - -
immintrin.h
+ +
emmintrin.h
 - + Integer - AVX512VL - AVX512F - Miscellaneous - - - - - Unpack and interleave 32-bit integers from the high half of each 128-bit lane in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + SSE2 + Shift + + + + Shift packed 16-bit integers in "a" left by "imm8" while shifting in zeros, and store the results in "dst". -INTERLEAVE_HIGH_DWORDS(src1[127:0], src2[127:0]){ - dst[31:0] := src1[95:64] - dst[63:32] := src2[95:64] - dst[95:64] := src1[127:96] - dst[127:96] := src2[127:96] - RETURN dst[127:0] -} - -tmp_dst[127:0] := INTERLEAVE_HIGH_DWORDS(a[127:0], b[127:0]) -tmp_dst[255:128] := INTERLEAVE_HIGH_DWORDS(a[255:128], b[255:128]) - FOR j := 0 to 7 - i := j*32 - IF k[j] - dst[i+31:i] := tmp_dst[i+31:i] + i := j*16 + IF imm8[7:0] > 15 + dst[i+15:i] := 0 ELSE - dst[i+31:i] := src[i+31:i] + dst[i+15:i] := ZeroExtend16(a[i+15:i] << imm8[7:0]) FI ENDFOR -dst[MAX:256] := 0 - -
immintrin.h
+ +
emmintrin.h
 - + Integer - AVX512VL - AVX512F - Miscellaneous - - - - Unpack and interleave 32-bit integers from the high half of each 128-bit lane in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + SSE2 + Shift + + + + Shift packed 16-bit integers in "a" left by "count" while shifting in zeros, and store the results in "dst". -INTERLEAVE_HIGH_DWORDS(src1[127:0], src2[127:0]){ - dst[31:0] := src1[95:64] - dst[63:32] := src2[95:64] - dst[95:64] := src1[127:96] - dst[127:96] := src2[127:96] - RETURN dst[127:0] -} - -tmp_dst[127:0] := INTERLEAVE_HIGH_DWORDS(a[127:0], b[127:0]) -tmp_dst[255:128] := INTERLEAVE_HIGH_DWORDS(a[255:128], b[255:128]) - FOR j := 0 to 7 - i := j*32 - IF k[j] - dst[i+31:i] := tmp_dst[i+31:i] + i := j*16 + IF count[63:0] > 15 + dst[i+15:i] := 0 ELSE - dst[i+31:i] := 0 + dst[i+15:i] := ZeroExtend16(a[i+15:i] << count[63:0]) FI ENDFOR -dst[MAX:256] := 0 - -
immintrin.h
+ +
emmintrin.h
 - + Integer - AVX512VL - AVX512F - Miscellaneous - - - - - Unpack and interleave 32-bit integers from the high half of "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + SSE2 + Shift + + + + Shift packed 32-bit integers in "a" left by "imm8" while shifting in zeros, and store the results in "dst". -INTERLEAVE_HIGH_DWORDS(src1[127:0], src2[127:0]){ - dst[31:0] := src1[95:64] - dst[63:32] := src2[95:64] - dst[95:64] := src1[127:96] - dst[127:96] := src2[127:96] - RETURN dst[127:0] -} - -tmp_dst[127:0] := INTERLEAVE_HIGH_DWORDS(a[127:0], b[127:0]) - FOR j := 0 to 3 i := j*32 - IF k[j] - dst[i+31:i] := tmp_dst[i+31:i] + IF imm8[7:0] > 31 + dst[i+31:i] := 0 ELSE - dst[i+31:i] := src[i+31:i] + dst[i+31:i] := ZeroExtend32(a[i+31:i] << imm8[7:0]) FI ENDFOR -dst[MAX:128] := 0 - -
immintrin.h
+ +
emmintrin.h
 - + Integer - AVX512VL - AVX512F - Miscellaneous - - - - Unpack and interleave 32-bit integers from the high half of "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + SSE2 + Shift + + + + Shift packed 32-bit integers in "a" left by "count" while shifting in zeros, and store the results in "dst". -INTERLEAVE_HIGH_DWORDS(src1[127:0], src2[127:0]){ - dst[31:0] := src1[95:64] - dst[63:32] := src2[95:64] - dst[95:64] := src1[127:96] - dst[127:96] := src2[127:96] - RETURN dst[127:0] -} - -tmp_dst[127:0] := INTERLEAVE_HIGH_DWORDS(a[127:0], b[127:0]) - FOR j := 0 to 3 i := j*32 - IF k[j] - dst[i+31:i] := tmp_dst[i+31:i] - ELSE + IF count[63:0] > 31 dst[i+31:i] := 0 - FI -ENDFOR -dst[MAX:128] := 0 - - -
immintrin.h
- - - Integer - AVX512VL - AVX512F - Miscellaneous - - - - - Unpack and interleave 64-bit integers from the high half of each 128-bit lane in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -INTERLEAVE_HIGH_QWORDS(src1[127:0], src2[127:0]){ - dst[63:0] := src1[127:64] - dst[127:64] := src2[127:64] - RETURN dst[127:0] -} - -tmp_dst[127:0] := INTERLEAVE_HIGH_QWORDS(a[127:0], b[127:0]) -tmp_dst[255:128] := INTERLEAVE_HIGH_QWORDS(a[255:128], b[255:128]) - -FOR j := 0 to 3 - i := j*64 - IF k[j] - dst[i+63:i] := tmp_dst[i+63:i] - ELSE - dst[i+63:i] := src[i+63:i] - FI -ENDFOR -dst[MAX:256] := 0 - - -
immintrin.h
- - - Integer - AVX512VL - AVX512F - Miscellaneous - - - - Unpack and interleave 64-bit integers from the high half of each 128-bit lane in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - -INTERLEAVE_HIGH_QWORDS(src1[127:0], src2[127:0]){ - dst[63:0] := src1[127:64] - dst[127:64] := src2[127:64] - RETURN dst[127:0] -} - -tmp_dst[127:0] := INTERLEAVE_HIGH_QWORDS(a[127:0], b[127:0]) -tmp_dst[255:128] := INTERLEAVE_HIGH_QWORDS(a[255:128], b[255:128]) - -FOR j := 0 to 3 - i := j*64 - IF k[j] - dst[i+63:i] := tmp_dst[i+63:i] ELSE - dst[i+63:i] := 0 + dst[i+31:i] := ZeroExtend32(a[i+31:i] << count[63:0]) FI ENDFOR -dst[MAX:256] := 0 - -
immintrin.h
+ +
emmintrin.h
 - + Integer - AVX512VL - AVX512F - Miscellaneous - - - - - Unpack and interleave 64-bit integers from the high half of "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + SSE2 + Shift + + + + Shift packed 64-bit integers in "a" left by "imm8" while shifting in zeros, and store the results in "dst". -INTERLEAVE_HIGH_QWORDS(src1[127:0], src2[127:0]){ - dst[63:0] := src1[127:64] - dst[127:64] := src2[127:64] - RETURN dst[127:0] -} - -tmp_dst[127:0] := INTERLEAVE_HIGH_QWORDS(a[127:0], b[127:0]) - FOR j := 0 to 1 i := j*64 - IF k[j] - dst[i+63:i] := tmp_dst[i+63:i] + IF imm8[7:0] > 63 + dst[i+63:i] := 0 ELSE - dst[i+63:i] := src[i+63:i] + dst[i+63:i] := ZeroExtend64(a[i+63:i] << imm8[7:0]) FI ENDFOR -dst[MAX:128] := 0 - -
immintrin.h
+ +
emmintrin.h
 - + Integer - AVX512VL - AVX512F - Miscellaneous - - - - Unpack and interleave 64-bit integers from the high half of "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + SSE2 + Shift + + + + Shift packed 64-bit integers in "a" left by "count" while shifting in zeros, and store the results in "dst". -INTERLEAVE_HIGH_QWORDS(src1[127:0], src2[127:0]){ - dst[63:0] := src1[127:64] - dst[127:64] := src2[127:64] - RETURN dst[127:0] -} - -tmp_dst[127:0] := INTERLEAVE_HIGH_QWORDS(a[127:0], b[127:0]) - FOR j := 0 to 1 i := j*64 - IF k[j] - dst[i+63:i] := tmp_dst[i+63:i] - ELSE + IF count[63:0] > 63 dst[i+63:i] := 0 + ELSE + dst[i+63:i] := ZeroExtend64(a[i+63:i] << count[63:0]) FI ENDFOR -dst[MAX:128] := 0 - -
immintrin.h
+ +
emmintrin.h
 - + Integer - AVX512VL - AVX512BW - Miscellaneous - - - - - Unpack and interleave 16-bit integers from the high half of each 128-bit lane in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + SSE2 + Shift + + + + Shift packed 16-bit integers in "a" right by "imm8" while shifting in sign bits, and store the results in "dst". -INTERLEAVE_HIGH_WORDS(src1[127:0], src2[127:0]){ - dst[15:0] := src1[79:64] - dst[31:16] := src2[79:64] - dst[47:32] := src1[95:80] - dst[63:48] := src2[95:80] - dst[79:64] := src1[111:96] - dst[95:80] := src2[111:96] - dst[111:96] := src1[127:112] - dst[127:112] := src2[127:112] - RETURN dst[127:0] -} - -tmp_dst[127:0] := INTERLEAVE_HIGH_WORDS(a[127:0], b[127:0]) -tmp_dst[255:128] := INTERLEAVE_HIGH_WORDS(a[255:128], b[255:128]) - -FOR j := 0 to 15 +FOR j := 0 to 7 i := j*16 - IF k[j] - dst[i+15:i] := tmp_dst[i+15:i] + IF imm8[7:0] > 15 + dst[i+15:i] := (a[i+15] ? 0xFFFF : 0x0) ELSE - dst[i+15:i] := src[i+15:i] + dst[i+15:i] := SignExtend16(a[i+15:i] >> imm8[7:0]) FI ENDFOR -dst[MAX:256] := 0 - -
immintrin.h
+ +
emmintrin.h
 - + Integer - AVX512VL - AVX512BW - Miscellaneous - - - - Unpack and interleave 16-bit integers from the high half of each 128-bit lane in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + SSE2 + Shift + + + + Shift packed 16-bit integers in "a" right by "count" while shifting in sign bits, and store the results in "dst". -INTERLEAVE_HIGH_WORDS(src1[127:0], src2[127:0]){ - dst[15:0] := src1[79:64] - dst[31:16] := src2[79:64] - dst[47:32] := src1[95:80] - dst[63:48] := src2[95:80] - dst[79:64] := src1[111:96] - dst[95:80] := src2[111:96] - dst[111:96] := src1[127:112] - dst[127:112] := src2[127:112] - RETURN dst[127:0] -} - -tmp_dst[127:0] := INTERLEAVE_HIGH_WORDS(a[127:0], b[127:0]) -tmp_dst[255:128] := INTERLEAVE_HIGH_WORDS(a[255:128], b[255:128]) - -FOR j := 0 to 15 +FOR j := 0 to 7 i := j*16 - IF k[j] - dst[i+15:i] := tmp_dst[i+15:i] + IF count[63:0] > 15 + dst[i+15:i] := (a[i+15] ? 0xFFFF : 0x0) ELSE - dst[i+15:i] := 0 + dst[i+15:i] := SignExtend16(a[i+15:i] >> count[63:0]) FI ENDFOR -dst[MAX:256] := 0 - -
immintrin.h
+ +
emmintrin.h
 - + Integer - AVX512BW - Miscellaneous - - - - - Unpack and interleave 16-bit integers from the high half of each 128-bit lane in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + SSE2 + Shift + + + + Shift packed 32-bit integers in "a" right by "imm8" while shifting in sign bits, and store the results in "dst". -INTERLEAVE_HIGH_WORDS(src1[127:0], src2[127:0]){ - dst[15:0] := src1[79:64] - dst[31:16] := src2[79:64] - dst[47:32] := src1[95:80] - dst[63:48] := src2[95:80] - dst[79:64] := src1[111:96] - dst[95:80] := src2[111:96] - dst[111:96] := src1[127:112] - dst[127:112] := src2[127:112] - RETURN dst[127:0] -} - -tmp_dst[127:0] := INTERLEAVE_HIGH_WORDS(a[127:0], b[127:0]) -tmp_dst[255:128] := INTERLEAVE_HIGH_WORDS(a[255:128], b[255:128]) -tmp_dst[383:256] := INTERLEAVE_HIGH_WORDS(a[383:256], b[383:256]) -tmp_dst[511:384] := INTERLEAVE_HIGH_WORDS(a[511:384], b[511:384]) - -FOR j := 0 to 31 - i := j*16 - IF k[j] - dst[i+15:i] := tmp_dst[i+15:i] +FOR j := 0 to 3 + i := j*32 + IF imm8[7:0] > 31 + dst[i+31:i] := (a[i+31] ? 0xFFFFFFFF : 0x0) ELSE - dst[i+15:i] := src[i+15:i] + dst[i+31:i] := SignExtend32(a[i+31:i] >> imm8[7:0]) FI ENDFOR -dst[MAX:512] := 0 - -
immintrin.h
+ +
emmintrin.h
 - + Integer - AVX512BW - Miscellaneous - - - - Unpack and interleave 16-bit integers from the high half of each 128-bit lane in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + SSE2 + Shift + + + + Shift packed 32-bit integers in "a" right by "count" while shifting in sign bits, and store the results in "dst". -INTERLEAVE_HIGH_WORDS(src1[127:0], src2[127:0]){ - dst[15:0] := src1[79:64] - dst[31:16] := src2[79:64] - dst[47:32] := src1[95:80] - dst[63:48] := src2[95:80] - dst[79:64] := src1[111:96] - dst[95:80] := src2[111:96] - dst[111:96] := src1[127:112] - dst[127:112] := src2[127:112] - RETURN dst[127:0] -} - -tmp_dst[127:0] := INTERLEAVE_HIGH_WORDS(a[127:0], b[127:0]) -tmp_dst[255:128] := INTERLEAVE_HIGH_WORDS(a[255:128], b[255:128]) -tmp_dst[383:256] := INTERLEAVE_HIGH_WORDS(a[383:256], b[383:256]) -tmp_dst[511:384] := INTERLEAVE_HIGH_WORDS(a[511:384], b[511:384]) - -FOR j := 0 to 31 - i := j*16 - IF k[j] - dst[i+15:i] := tmp_dst[i+15:i] +FOR j := 0 to 3 + i := j*32 + IF count[63:0] > 31 + dst[i+31:i] := (a[i+31] ? 0xFFFFFFFF : 0x0) ELSE - dst[i+15:i] := 0 + dst[i+31:i] := SignExtend32(a[i+31:i] >> count[63:0]) FI ENDFOR -dst[MAX:512] := 0 - -
immintrin.h
+ +
emmintrin.h
 - + Integer - AVX512BW - Miscellaneous - - - Unpack and interleave 16-bit integers from the high half of each 128-bit lane in "a" and "b", and store the results in "dst". + SSE2 + Shift + + + + Shift "a" right by "imm8" bytes while shifting in zeros, and store the results in "dst". -INTERLEAVE_HIGH_WORDS(src1[127:0], src2[127:0]){ - dst[15:0] := src1[79:64] - dst[31:16] := src2[79:64] - dst[47:32] := src1[95:80] - dst[63:48] := src2[95:80] - dst[79:64] := src1[111:96] - dst[95:80] := src2[111:96] - dst[111:96] := src1[127:112] - dst[127:112] := src2[127:112] - RETURN dst[127:0] -} - -dst[127:0] := INTERLEAVE_HIGH_WORDS(a[127:0], b[127:0]) -dst[255:128] := INTERLEAVE_HIGH_WORDS(a[255:128], b[255:128]) -dst[383:256] := INTERLEAVE_HIGH_WORDS(a[383:256], b[383:256]) -dst[511:384] := INTERLEAVE_HIGH_WORDS(a[511:384], b[511:384]) -dst[MAX:512] := 0 +tmp := imm8[7:0] +IF tmp > 15 + tmp := 16 +FI +dst[127:0] := a[127:0] >> (tmp*8) - -
immintrin.h
+ +
emmintrin.h
 - + Integer - AVX512VL - AVX512BW - Miscellaneous - - - - - Unpack and interleave 16-bit integers from the high half of "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + SSE2 + Shift + + + + Shift packed 16-bit integers in "a" right by "imm8" while shifting in zeros, and store the results in "dst". -INTERLEAVE_HIGH_WORDS(src1[127:0], src2[127:0]){ - dst[15:0] := src1[79:64] - dst[31:16] := src2[79:64] - dst[47:32] := src1[95:80] - dst[63:48] := src2[95:80] - dst[79:64] := src1[111:96] - dst[95:80] := src2[111:96] - dst[111:96] := src1[127:112] - dst[127:112] := src2[127:112] - RETURN dst[127:0] -} - -tmp_dst[127:0] := INTERLEAVE_HIGH_WORDS(a[127:0], b[127:0]) - FOR j := 0 to 7 i := j*16 - IF k[j] - dst[i+15:i] := tmp_dst[i+15:i] + IF imm8[7:0] > 15 + dst[i+15:i] := 0 ELSE - dst[i+15:i] := src[i+15:i] + dst[i+15:i] := ZeroExtend16(a[i+15:i] >> imm8[7:0]) FI ENDFOR -dst[MAX:128] := 0 - -
immintrin.h
+ +
emmintrin.h
 - + Integer - AVX512VL - AVX512BW - Miscellaneous - - - - Unpack and interleave 16-bit integers from the high half of "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + SSE2 + Shift + + + + Shift packed 16-bit integers in "a" right by "count" while shifting in zeros, and store the results in "dst". -INTERLEAVE_HIGH_WORDS(src1[127:0], src2[127:0]){ - dst[15:0] := src1[79:64] - dst[31:16] := src2[79:64] - dst[47:32] := src1[95:80] - dst[63:48] := src2[95:80] - dst[79:64] := src1[111:96] - dst[95:80] := src2[111:96] - dst[111:96] := src1[127:112] - dst[127:112] := src2[127:112] - RETURN dst[127:0] -} - -tmp_dst[127:0] := INTERLEAVE_HIGH_WORDS(a[127:0], b[127:0]) - FOR j := 0 to 7 i := j*16 - IF k[j] - dst[i+15:i] := tmp_dst[i+15:i] - ELSE + IF count[63:0] > 15 dst[i+15:i] := 0 + ELSE + dst[i+15:i] := ZeroExtend16(a[i+15:i] >> count[63:0]) FI ENDFOR -dst[MAX:128] := 0 - -
immintrin.h
+ +
emmintrin.h
 - + Integer - AVX512VL - AVX512BW - Miscellaneous - - - - - Unpack and interleave 8-bit integers from the low half of each 128-bit lane in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + SSE2 + Shift + + + + Shift packed 32-bit integers in "a" right by "imm8" while shifting in zeros, and store the results in "dst". -INTERLEAVE_BYTES(src1[127:0], src2[127:0]){ - dst[7:0] := src1[7:0] - dst[15:8] := src2[7:0] - dst[23:16] := src1[15:8] - dst[31:24] := src2[15:8] - dst[39:32] := src1[23:16] - dst[47:40] := src2[23:16] - dst[55:48] := src1[31:24] - dst[63:56] := src2[31:24] - dst[71:64] := src1[39:32] - dst[79:72] := src2[39:32] - dst[87:80] := src1[47:40] - dst[95:88] := src2[47:40] - dst[103:96] := src1[55:48] - dst[111:104] := src2[55:48] - dst[119:112] := src1[63:56] - dst[127:120] := src2[63:56] - RETURN dst[127:0] -} - -tmp_dst[127:0] := INTERLEAVE_BYTES(a[127:0], b[127:0]) -tmp_dst[255:128] := INTERLEAVE_BYTES(a[255:128], b[255:128]) - -FOR j := 0 to 31 - i := j*8 - IF k[j] - dst[i+7:i] := tmp_dst[i+7:i] +FOR j := 0 to 3 + i := j*32 + IF imm8[7:0] > 31 + dst[i+31:i] := 0 ELSE - dst[i+7:i] := src[i+7:i] + dst[i+31:i] := ZeroExtend32(a[i+31:i] >> imm8[7:0]) FI ENDFOR -dst[MAX:256] := 0 - -
immintrin.h
+ +
emmintrin.h
 - + Integer - AVX512VL - AVX512BW - Miscellaneous - - - - Unpack and interleave 8-bit integers from the low half of each 128-bit lane in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + SSE2 + Shift + + + + Shift packed 32-bit integers in "a" right by "count" while shifting in zeros, and store the results in "dst". -INTERLEAVE_BYTES(src1[127:0], src2[127:0]){ - dst[7:0] := src1[7:0] - dst[15:8] := src2[7:0] - dst[23:16] := src1[15:8] - dst[31:24] := src2[15:8] - dst[39:32] := src1[23:16] - dst[47:40] := src2[23:16] - dst[55:48] := src1[31:24] - dst[63:56] := src2[31:24] - dst[71:64] := src1[39:32] - dst[79:72] := src2[39:32] - dst[87:80] := src1[47:40] - dst[95:88] := src2[47:40] - dst[103:96] := src1[55:48] - dst[111:104] := src2[55:48] - dst[119:112] := src1[63:56] - dst[127:120] := src2[63:56] - RETURN dst[127:0] -} - -tmp_dst[127:0] := INTERLEAVE_BYTES(a[127:0], b[127:0]) -tmp_dst[255:128] := INTERLEAVE_BYTES(a[255:128], b[255:128]) - -FOR j := 0 to 31 - i := j*8 - IF k[j] - dst[i+7:i] := tmp_dst[i+7:i] +FOR j := 0 to 3 + i := j*32 + IF count[63:0] > 31 + dst[i+31:i] := 0 ELSE - dst[i+7:i] := 0 + dst[i+31:i] := ZeroExtend32(a[i+31:i] >> count[63:0]) FI ENDFOR -dst[MAX:256] := 0 - -
immintrin.h
+ +
emmintrin.h
 - + Integer - AVX512BW - Miscellaneous - - - - - Unpack and interleave 8-bit integers from the low half of each 128-bit lane in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + SSE2 + Shift + + + + Shift packed 64-bit integers in "a" right by "imm8" while shifting in zeros, and store the results in "dst". -INTERLEAVE_BYTES(src1[127:0], src2[127:0]){ - dst[7:0] := src1[7:0] - dst[15:8] := src2[7:0] - dst[23:16] := src1[15:8] - dst[31:24] := src2[15:8] - dst[39:32] := src1[23:16] - dst[47:40] := src2[23:16] - dst[55:48] := src1[31:24] - dst[63:56] := src2[31:24] - dst[71:64] := src1[39:32] - dst[79:72] := src2[39:32] - dst[87:80] := src1[47:40] - dst[95:88] := src2[47:40] - dst[103:96] := src1[55:48] - dst[111:104] := src2[55:48] - dst[119:112] := src1[63:56] - dst[127:120] := src2[63:56] - RETURN dst[127:0] -} - -tmp_dst[127:0] := INTERLEAVE_BYTES(a[127:0], b[127:0]) -tmp_dst[255:128] := INTERLEAVE_BYTES(a[255:128], b[255:128]) -tmp_dst[383:256] := INTERLEAVE_BYTES(a[383:256], b[383:256]) -tmp_dst[511:384] := INTERLEAVE_BYTES(a[511:384], b[511:384]) - -FOR j := 0 to 63 - i := j*8 - IF k[j] - dst[i+7:i] := tmp_dst[i+7:i] +FOR j := 0 to 1 + i := j*64 + IF imm8[7:0] > 63 + dst[i+63:i] := 0 ELSE - dst[i+7:i] := src[i+7:i] + dst[i+63:i] := ZeroExtend64(a[i+63:i] >> imm8[7:0]) FI ENDFOR -dst[MAX:512] := 0 - -
immintrin.h
+ +
emmintrin.h
 - + Integer - AVX512BW - Miscellaneous - - - - Unpack and interleave 8-bit integers from the low half of each 128-bit lane in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + SSE2 + Shift + + + + Shift packed 64-bit integers in "a" right by "count" while shifting in zeros, and store the results in "dst". -INTERLEAVE_BYTES(src1[127:0], src2[127:0]){ - dst[7:0] := src1[7:0] - dst[15:8] := src2[7:0] - dst[23:16] := src1[15:8] - dst[31:24] := src2[15:8] - dst[39:32] := src1[23:16] - dst[47:40] := src2[23:16] - dst[55:48] := src1[31:24] - dst[63:56] := src2[31:24] - dst[71:64] := src1[39:32] - dst[79:72] := src2[39:32] - dst[87:80] := src1[47:40] - dst[95:88] := src2[47:40] - dst[103:96] := src1[55:48] - dst[111:104] := src2[55:48] - dst[119:112] := src1[63:56] - dst[127:120] := src2[63:56] - RETURN dst[127:0] -} - -tmp_dst[127:0] := INTERLEAVE_BYTES(a[127:0], b[127:0]) -tmp_dst[255:128] := INTERLEAVE_BYTES(a[255:128], b[255:128]) -tmp_dst[383:256] := INTERLEAVE_BYTES(a[383:256], b[383:256]) -tmp_dst[511:384] := INTERLEAVE_BYTES(a[511:384], b[511:384]) - -FOR j := 0 to 63 - i := j*8 - IF k[j] - dst[i+7:i] := tmp_dst[i+7:i] +FOR j := 0 to 1 + i := j*64 + IF count[63:0] > 63 + dst[i+63:i] := 0 ELSE - dst[i+7:i] := 0 + dst[i+63:i] := ZeroExtend64(a[i+63:i] >> count[63:0]) FI ENDFOR -dst[MAX:512] := 0 - -
immintrin.h
+ +
emmintrin.h
 - + Integer - AVX512BW - Miscellaneous - - - Unpack and interleave 8-bit integers from the low half of each 128-bit lane in "a" and "b", and store the results in "dst". + SSE2 + Logical + + + + Compute the bitwise AND of 128 bits (representing integer data) in "a" and "b", and store the result in "dst". -INTERLEAVE_BYTES(src1[127:0], src2[127:0]){ - dst[7:0] := src1[7:0] - dst[15:8] := src2[7:0] - dst[23:16] := src1[15:8] - dst[31:24] := src2[15:8] - dst[39:32] := src1[23:16] - dst[47:40] := src2[23:16] - dst[55:48] := src1[31:24] - dst[63:56] := src2[31:24] - dst[71:64] := src1[39:32] - dst[79:72] := src2[39:32] - dst[87:80] := src1[47:40] - dst[95:88] := src2[47:40] - dst[103:96] := src1[55:48] - dst[111:104] := src2[55:48] - dst[119:112] := src1[63:56] - dst[127:120] := src2[63:56] - RETURN dst[127:0] -} - -dst[127:0] := INTERLEAVE_BYTES(a[127:0], b[127:0]) -dst[255:128] := INTERLEAVE_BYTES(a[255:128], b[255:128]) -dst[383:256] := INTERLEAVE_BYTES(a[383:256], b[383:256]) -dst[511:384] := INTERLEAVE_BYTES(a[511:384], b[511:384]) -dst[MAX:512] := 0 +dst[127:0] := (a[127:0] AND b[127:0]) - -
immintrin.h
+ +
emmintrin.h
 - + Integer - AVX512VL - AVX512BW - Miscellaneous - - - - - Unpack and interleave 8-bit integers from the low half of "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + SSE2 + Logical + + + + Compute the bitwise NOT of 128 bits (representing integer data) in "a" and then AND with "b", and store the result in "dst". + +dst[127:0] := ((NOT a[127:0]) AND b[127:0]) + + +
emmintrin.h
+ + + Integer + SSE2 + Logical + + + + Compute the bitwise OR of 128 bits (representing integer data) in "a" and "b", and store the result in "dst". + +dst[127:0] := (a[127:0] OR b[127:0]) + + +
emmintrin.h
+ + + Integer + SSE2 + Logical + + + + Compute the bitwise XOR of 128 bits (representing integer data) in "a" and "b", and store the result in "dst". + +dst[127:0] := (a[127:0] XOR b[127:0]) + + +
emmintrin.h
+ + + Integer + SSE2 + Compare + + + + Compare packed 8-bit integers in "a" and "b" for equality, and store the results in "dst". -INTERLEAVE_BYTES(src1[127:0], src2[127:0]){ - dst[7:0] := src1[7:0] - dst[15:8] := src2[7:0] - dst[23:16] := src1[15:8] - dst[31:24] := src2[15:8] - dst[39:32] := src1[23:16] - dst[47:40] := src2[23:16] - dst[55:48] := src1[31:24] - dst[63:56] := src2[31:24] - dst[71:64] := src1[39:32] - dst[79:72] := src2[39:32] - dst[87:80] := src1[47:40] - dst[95:88] := src2[47:40] - dst[103:96] := src1[55:48] - dst[111:104] := src2[55:48] - dst[119:112] := src1[63:56] - dst[127:120] := src2[63:56] - RETURN dst[127:0] -} - -tmp_dst[127:0] := INTERLEAVE_BYTES(a[127:0], b[127:0]) - FOR j := 0 to 15 i := j*8 - IF k[j] - dst[i+7:i] := tmp_dst[i+7:i] - ELSE - dst[i+7:i] := src[i+7:i] - FI + dst[i+7:i] := ( a[i+7:i] == b[i+7:i] ) ? 0xFF : 0 ENDFOR -dst[MAX:128] := 0 - -
immintrin.h
+ +
emmintrin.h
 - + Integer - AVX512VL - AVX512BW - Miscellaneous - - - - Unpack and interleave 8-bit integers from the low half of "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + SSE2 + Compare + + + + Compare packed 16-bit integers in "a" and "b" for equality, and store the results in "dst". + +FOR j := 0 to 7 + i := j*16 + dst[i+15:i] := ( a[i+15:i] == b[i+15:i] ) ? 0xFFFF : 0 +ENDFOR + + +
emmintrin.h
+ + + Integer + SSE2 + Compare + + + + Compare packed 32-bit integers in "a" and "b" for equality, and store the results in "dst". + +FOR j := 0 to 3 + i := j*32 + dst[i+31:i] := ( a[i+31:i] == b[i+31:i] ) ? 0xFFFFFFFF : 0 +ENDFOR + + +
emmintrin.h
+ + + Integer + SSE2 + Compare + + + + Compare packed signed 8-bit integers in "a" and "b" for greater-than, and store the results in "dst". -INTERLEAVE_BYTES(src1[127:0], src2[127:0]){ - dst[7:0] := src1[7:0] - dst[15:8] := src2[7:0] - dst[23:16] := src1[15:8] - dst[31:24] := src2[15:8] - dst[39:32] := src1[23:16] - dst[47:40] := src2[23:16] - dst[55:48] := src1[31:24] - dst[63:56] := src2[31:24] - dst[71:64] := src1[39:32] - dst[79:72] := src2[39:32] - dst[87:80] := src1[47:40] - dst[95:88] := src2[47:40] - dst[103:96] := src1[55:48] - dst[111:104] := src2[55:48] - dst[119:112] := src1[63:56] - dst[127:120] := src2[63:56] - RETURN dst[127:0] -} - -tmp_dst[127:0] := INTERLEAVE_BYTES(a[127:0], b[127:0]) - FOR j := 0 to 15 i := j*8 - IF k[j] - dst[i+7:i] := tmp_dst[i+7:i] - ELSE - dst[i+7:i] := 0 - FI + dst[i+7:i] := ( a[i+7:i] > b[i+7:i] ) ? 0xFF : 0 ENDFOR -dst[MAX:128] := 0 - -
immintrin.h
+ +
emmintrin.h
 - + Integer - AVX512VL - AVX512F - Miscellaneous - - - - - Unpack and interleave 32-bit integers from the low half of each 128-bit lane in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + SSE2 + Compare + + + + Compare packed signed 16-bit integers in "a" and "b" for greater-than, and store the results in "dst". -INTERLEAVE_DWORDS(src1[127:0], src2[127:0]){ - dst[31:0] := src1[31:0] - dst[63:32] := src2[31:0] - dst[95:64] := src1[63:32] - dst[127:96] := src2[63:32] - RETURN dst[127:0] -} - -tmp_dst[127:0] := INTERLEAVE_DWORDS(a[127:0], b[127:0]) -tmp_dst[255:128] := INTERLEAVE_DWORDS(a[255:128], b[255:128]) - FOR j := 0 to 7 + i := j*16 + dst[i+15:i] := ( a[i+15:i] > b[i+15:i] ) ? 0xFFFF : 0 +ENDFOR + + +
emmintrin.h
+ + + Integer + SSE2 + Compare + + + + Compare packed signed 32-bit integers in "a" and "b" for greater-than, and store the results in "dst". + +FOR j := 0 to 3 i := j*32 - IF k[j] - dst[i+31:i] := tmp_dst[i+31:i] - ELSE - dst[i+31:i] := src[i+31:i] - FI + dst[i+31:i] := ( a[i+31:i] > b[i+31:i] ) ? 0xFFFFFFFF : 0 +ENDFOR + + +
emmintrin.h
+ + + Integer + SSE2 + Compare + + + + Compare packed signed 8-bit integers in "a" and "b" for less-than, and store the results in "dst". Note: This intrinsic emits the pcmpgtb instruction with the order of the operands switched. + +FOR j := 0 to 15 + i := j*8 + dst[i+7:i] := ( a[i+7:i] < b[i+7:i] ) ? 0xFF : 0 ENDFOR -dst[MAX:256] := 0 - -
immintrin.h
+ +
emmintrin.h
 - + Integer - AVX512VL - AVX512F - Miscellaneous - - - - Unpack and interleave 32-bit integers from the low half of each 128-bit lane in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + SSE2 + Compare + + + + Compare packed signed 16-bit integers in "a" and "b" for less-than, and store the results in "dst". Note: This intrinsic emits the pcmpgtw instruction with the order of the operands switched. -INTERLEAVE_DWORDS(src1[127:0], src2[127:0]){ - dst[31:0] := src1[31:0] - dst[63:32] := src2[31:0] - dst[95:64] := src1[63:32] - dst[127:96] := src2[63:32] - RETURN dst[127:0] -} - -tmp_dst[127:0] := INTERLEAVE_DWORDS(a[127:0], b[127:0]) -tmp_dst[255:128] := INTERLEAVE_DWORDS(a[255:128], b[255:128]) - FOR j := 0 to 7 - i := j*32 - IF k[j] - dst[i+31:i] := tmp_dst[i+31:i] - ELSE - dst[i+31:i] := 0 - FI + i := j*16 + dst[i+15:i] := ( a[i+15:i] < b[i+15:i] ) ? 0xFFFF : 0 ENDFOR -dst[MAX:256] := 0 - -
immintrin.h
+ +
emmintrin.h
 - + Integer - AVX512VL - AVX512F - Miscellaneous - - - - - Unpack and interleave 32-bit integers from the low half of "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + SSE2 + Compare + + + + Compare packed signed 32-bit integers in "a" and "b" for less-than, and store the results in "dst". Note: This intrinsic emits the pcmpgtd instruction with the order of the operands switched. -INTERLEAVE_DWORDS(src1[127:0], src2[127:0]){ - dst[31:0] := src1[31:0] - dst[63:32] := src2[31:0] - dst[95:64] := src1[63:32] - dst[127:96] := src2[63:32] - RETURN dst[127:0] -} - -tmp_dst[127:0] := INTERLEAVE_DWORDS(a[127:0], b[127:0]) - FOR j := 0 to 3 i := j*32 - IF k[j] - dst[i+31:i] := tmp_dst[i+31:i] - ELSE - dst[i+31:i] := src[i+31:i] - FI + dst[i+31:i] := ( a[i+31:i] < b[i+31:i] ) ? 0xFFFFFFFF : 0 ENDFOR -dst[MAX:128] := 0 - -
immintrin.h
+ +
emmintrin.h
 - + + Floating Point Integer - AVX512VL - AVX512F - Miscellaneous - - - - Unpack and interleave 32-bit integers from the low half of "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + SSE2 + Convert + + + Convert packed signed 32-bit integers in "a" to packed double-precision (64-bit) floating-point elements, and store the results in "dst". -INTERLEAVE_DWORDS(src1[127:0], src2[127:0]){ - dst[31:0] := src1[31:0] - dst[63:32] := src2[31:0] - dst[95:64] := src1[63:32] - dst[127:96] := src2[63:32] - RETURN dst[127:0] -} - -tmp_dst[127:0] := INTERLEAVE_DWORDS(a[127:0], b[127:0]) - -FOR j := 0 to 3 +FOR j := 0 to 1 i := j*32 - IF k[j] - dst[i+31:i] := tmp_dst[i+31:i] - ELSE - dst[i+31:i] := 0 - FI + m := j*64 + dst[m+63:m] := Convert_Int32_To_FP64(a[i+31:i]) ENDFOR + + +
emmintrin.h
+ + + Floating Point + SSE2 + Convert + + + + Convert the signed 32-bit integer "b" to a double-precision (64-bit) floating-point element, store the result in the lower element of "dst", and copy the upper element from "a" to the upper element of "dst". + +dst[63:0] := Convert_Int32_To_FP64(b[31:0]) +dst[127:64] := a[127:64] dst[MAX:128] := 0 - -
immintrin.h
+ +
emmintrin.h
 - + + Floating Point Integer - AVX512VL - AVX512F - Miscellaneous - - - - - Unpack and interleave 64-bit integers from the low half of each 128-bit lane in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + SSE2 + Convert + + + + Convert the signed 64-bit integer "b" to a double-precision (64-bit) floating-point element, store the result in the lower element of "dst", and copy the upper element from "a" to the upper element of "dst". -INTERLEAVE_QWORDS(src1[127:0], src2[127:0]){ - dst[63:0] := src1[63:0] - dst[127:64] := src2[63:0] - RETURN dst[127:0] -} - -tmp_dst[127:0] := INTERLEAVE_QWORDS(a[127:0], b[127:0]) -tmp_dst[255:128] := INTERLEAVE_QWORDS(a[255:128], b[255:128]) - -FOR j := 0 to 3 - i := j*64 - IF k[j] - dst[i+63:i] := tmp_dst[i+63:i] - ELSE - dst[i+63:i] := src[i+63:i] - FI -ENDFOR -dst[MAX:256] := 0 +dst[63:0] := Convert_Int64_To_FP64(b[63:0]) +dst[127:64] := a[127:64] +dst[MAX:128] := 0 - -
immintrin.h
+ +
emmintrin.h
 - + + Floating Point Integer - AVX512VL - AVX512F - Miscellaneous - - - - Unpack and interleave 64-bit integers from the low half of each 128-bit lane in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + SSE2 + Convert + + + + Convert the signed 64-bit integer "b" to a double-precision (64-bit) floating-point element, store the result in the lower element of "dst", and copy the upper element from "a" to the upper element of "dst". + +dst[63:0] := Convert_Int64_To_FP64(b[63:0]) +dst[127:64] := a[127:64] +dst[MAX:128] := 0 + + +
emmintrin.h
+ + + Floating Point + Integer + SSE2 + Convert + + + Convert packed signed 32-bit integers in "a" to packed single-precision (32-bit) floating-point elements, and store the results in "dst". -INTERLEAVE_QWORDS(src1[127:0], src2[127:0]){ - dst[63:0] := src1[63:0] - dst[127:64] := src2[63:0] - RETURN dst[127:0] -} - -tmp_dst[127:0] := INTERLEAVE_QWORDS(a[127:0], b[127:0]) -tmp_dst[255:128] := INTERLEAVE_QWORDS(a[255:128], b[255:128]) - FOR j := 0 to 3 - i := j*64 - IF k[j] - dst[i+63:i] := tmp_dst[i+63:i] - ELSE - dst[i+63:i] := 0 - FI + i := 32*j + dst[i+31:i] := Convert_Int32_To_FP32(a[i+31:i]) ENDFOR -dst[MAX:256] := 0 - -
immintrin.h
+ +
emmintrin.h
 - - Integer - AVX512VL - AVX512F - Miscellaneous - - - - - Unpack and interleave 64-bit integers from the low half of "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Floating Point + SSE2 + Convert + + + Convert packed signed 32-bit integers in "a" to packed double-precision (64-bit) floating-point elements, and store the results in "dst". -INTERLEAVE_QWORDS(src1[127:0], src2[127:0]){ - dst[63:0] := src1[63:0] - dst[127:64] := src2[63:0] - RETURN dst[127:0] -} - -tmp_dst[127:0] := INTERLEAVE_QWORDS(a[127:0], b[127:0]) - FOR j := 0 to 1 - i := j*64 - IF k[j] - dst[i+63:i] := tmp_dst[i+63:i] - ELSE - dst[i+63:i] := src[i+63:i] - FI + i := j*32 + m := j*64 + dst[m+63:m] := Convert_Int32_To_FP64(a[i+31:i]) ENDFOR -dst[MAX:128] := 0 - -
immintrin.h
+ +
emmintrin.h
 - + Integer - AVX512VL - AVX512F - Miscellaneous - - - - Unpack and interleave 64-bit integers from the low half of "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + SSE2 + Convert + + + Copy 32-bit integer "a" to the lower elements of "dst", and zero the upper elements of "dst". -INTERLEAVE_QWORDS(src1[127:0], src2[127:0]){ - dst[63:0] := src1[63:0] - dst[127:64] := src2[63:0] - RETURN dst[127:0] -} - -tmp_dst[127:0] := INTERLEAVE_QWORDS(a[127:0], b[127:0]) - -FOR j := 0 to 1 - i := j*64 - IF k[j] - dst[i+63:i] := tmp_dst[i+63:i] - ELSE - dst[i+63:i] := 0 - FI -ENDFOR -dst[MAX:128] := 0 +dst[31:0] := a[31:0] +dst[127:32] := 0 - -
immintrin.h
+ +
emmintrin.h
 - + Integer - AVX512VL - AVX512BW - Miscellaneous - - - - - Unpack and interleave 16-bit integers from the low half of each 128-bit lane in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + SSE2 + Convert + + + Copy 64-bit integer "a" to the lower element of "dst", and zero the upper element. -INTERLEAVE_WORDS(src1[127:0], src2[127:0]){ - dst[15:0] := src1[15:0] - dst[31:16] := src2[15:0] - dst[47:32] := src1[31:16] - dst[63:48] := src2[31:16] - dst[79:64] := src1[47:32] - dst[95:80] := src2[47:32] - dst[111:96] := src1[63:48] - dst[127:112] := src2[63:48] - RETURN dst[127:0] -} - -tmp_dst[127:0] := INTERLEAVE_WORDS(a[127:0], b[127:0]) -tmp_dst[255:128] := INTERLEAVE_WORDS(a[255:128], b[255:128]) - -FOR j := 0 to 15 - i := j*16 - IF k[j] - dst[i+15:i] := tmp_dst[i+15:i] - ELSE - dst[i+15:i] := src[i+15:i] - FI -ENDFOR -dst[MAX:256] := 0 +dst[63:0] := a[63:0] +dst[127:64] := 0 - -
immintrin.h
+ +
emmintrin.h
 - + Integer - AVX512VL - AVX512BW - Miscellaneous - - - - Unpack and interleave 16-bit integers from the low half of each 128-bit lane in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + SSE2 + Convert + + + Copy 64-bit integer "a" to the lower element of "dst", and zero the upper element. -INTERLEAVE_WORDS(src1[127:0], src2[127:0]){ - dst[15:0] := src1[15:0] - dst[31:16] := src2[15:0] - dst[47:32] := src1[31:16] - dst[63:48] := src2[31:16] - dst[79:64] := src1[47:32] - dst[95:80] := src2[47:32] - dst[111:96] := src1[63:48] - dst[127:112] := src2[63:48] - RETURN dst[127:0] -} - -tmp_dst[127:0] := INTERLEAVE_WORDS(a[127:0], b[127:0]) -tmp_dst[255:128] := INTERLEAVE_WORDS(a[255:128], b[255:128]) - -FOR j := 0 to 15 - i := j*16 - IF k[j] - dst[i+15:i] := tmp_dst[i+15:i] - ELSE - dst[i+15:i] := 0 - FI -ENDFOR -dst[MAX:256] := 0 +dst[63:0] := a[63:0] +dst[127:64] := 0 - -
immintrin.h
+ +
emmintrin.h
 - + Integer - AVX512BW - Miscellaneous - - - - - Unpack and interleave 16-bit integers from the low half of each 128-bit lane in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + SSE2 + Convert + + + Copy the lower 32-bit integer in "a" to "dst". -INTERLEAVE_WORDS(src1[127:0], src2[127:0]){ - dst[15:0] := src1[15:0] - dst[31:16] := src2[15:0] - dst[47:32] := src1[31:16] - dst[63:48] := src2[31:16] - dst[79:64] := src1[47:32] - dst[95:80] := src2[47:32] - dst[111:96] := src1[63:48] - dst[127:112] := src2[63:48] - RETURN dst[127:0] -} - -tmp_dst[127:0] := INTERLEAVE_WORDS(a[127:0], b[127:0]) -tmp_dst[255:128] := INTERLEAVE_WORDS(a[255:128], b[255:128]) -tmp_dst[383:256] := INTERLEAVE_WORDS(a[383:256], b[383:256]) -tmp_dst[511:384] := INTERLEAVE_WORDS(a[511:384], b[511:384]) - -FOR j := 0 to 31 - i := j*16 - IF k[j] - dst[i+15:i] := tmp_dst[i+15:i] - ELSE - dst[i+15:i] := src[i+15:i] - FI +dst[31:0] := a[31:0] + + +
emmintrin.h
+ + + Integer + SSE2 + Convert + + + Copy the lower 64-bit integer in "a" to "dst". + +dst[63:0] := a[63:0] + + +
emmintrin.h
+ + + Integer + SSE2 + Convert + + + Copy the lower 64-bit integer in "a" to "dst". + +dst[63:0] := a[63:0] + + +
emmintrin.h
+ + + Integer + SSE2 + Set + + + + Set packed 64-bit integers in "dst" with the supplied values. + +dst[63:0] := e0 +dst[127:64] := e1 + +
emmintrin.h
+ + + Integer + SSE2 + Set + + + + Set packed 64-bit integers in "dst" with the supplied values. + +dst[63:0] := e0 +dst[127:64] := e1 + +
emmintrin.h
+ + + Integer + SSE2 + Set + + + + + + Set packed 32-bit integers in "dst" with the supplied values. + +dst[31:0] := e0 +dst[63:32] := e1 +dst[95:64] := e2 +dst[127:96] := e3 + +
emmintrin.h
+ + + Integer + SSE2 + Set + + + + + + + + + + Set packed 16-bit integers in "dst" with the supplied values. + +dst[15:0] := e0 +dst[31:16] := e1 +dst[47:32] := e2 +dst[63:48] := e3 +dst[79:64] := e4 +dst[95:80] := e5 +dst[111:96] := e6 +dst[127:112] := e7 + +
emmintrin.h
+ + + Integer + SSE2 + Set + + + + + + + + + + + + + + + + + + Set packed 8-bit integers in "dst" with the supplied values. + +dst[7:0] := e0 +dst[15:8] := e1 +dst[23:16] := e2 +dst[31:24] := e3 +dst[39:32] := e4 +dst[47:40] := e5 +dst[55:48] := e6 +dst[63:56] := e7 +dst[71:64] := e8 +dst[79:72] := e9 +dst[87:80] := e10 +dst[95:88] := e11 +dst[103:96] := e12 +dst[111:104] := e13 +dst[119:112] := e14 +dst[127:120] := e15 + +
emmintrin.h
+ + + Integer + SSE2 + Set + + + Broadcast 64-bit integer "a" to all elements of "dst". + +FOR j := 0 to 1 + i := j*64 + dst[i+63:i] := a[63:0] ENDFOR -dst[MAX:512] := 0 - -
immintrin.h
+
emmintrin.h
 - + Integer - AVX512BW - Miscellaneous - - - - Unpack and interleave 16-bit integers from the low half of each 128-bit lane in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + SSE2 + Set + + + Broadcast 64-bit integer "a" to all elements of "dst". This intrinsic may generate the "vpbroadcastq". -INTERLEAVE_WORDS(src1[127:0], src2[127:0]){ - dst[15:0] := src1[15:0] - dst[31:16] := src2[15:0] - dst[47:32] := src1[31:16] - dst[63:48] := src2[31:16] - dst[79:64] := src1[47:32] - dst[95:80] := src2[47:32] - dst[111:96] := src1[63:48] - dst[127:112] := src2[63:48] - RETURN dst[127:0] -} - -tmp_dst[127:0] := INTERLEAVE_WORDS(a[127:0], b[127:0]) -tmp_dst[255:128] := INTERLEAVE_WORDS(a[255:128], b[255:128]) -tmp_dst[383:256] := INTERLEAVE_WORDS(a[383:256], b[383:256]) -tmp_dst[511:384] := INTERLEAVE_WORDS(a[511:384], b[511:384]) - -FOR j := 0 to 31 - i := j*16 - IF k[j] - dst[i+15:i] := tmp_dst[i+15:i] - ELSE - dst[i+15:i] := 0 - FI +FOR j := 0 to 1 + i := j*64 + dst[i+63:i] := a[63:0] ENDFOR -dst[MAX:512] := 0 - -
immintrin.h
+
emmintrin.h
 - + Integer - AVX512BW - Miscellaneous - - - Unpack and interleave 16-bit integers from the low half of each 128-bit lane in "a" and "b", and store the results in "dst". + SSE2 + Set + + + Broadcast 32-bit integer "a" to all elements of "dst". This intrinsic may generate "vpbroadcastd". -INTERLEAVE_WORDS(src1[127:0], src2[127:0]){ - dst[15:0] := src1[15:0] - dst[31:16] := src2[15:0] - dst[47:32] := src1[31:16] - dst[63:48] := src2[31:16] - dst[79:64] := src1[47:32] - dst[95:80] := src2[47:32] - dst[111:96] := src1[63:48] - dst[127:112] := src2[63:48] - RETURN dst[127:0] -} - -dst[127:0] := INTERLEAVE_WORDS(a[127:0], b[127:0]) -dst[255:128] := INTERLEAVE_WORDS(a[255:128], b[255:128]) -dst[383:256] := INTERLEAVE_WORDS(a[383:256], b[383:256]) -dst[511:384] := INTERLEAVE_WORDS(a[511:384], b[511:384]) -dst[MAX:512] := 0 +FOR j := 0 to 3 + i := j*32 + dst[i+31:i] := a[31:0] +ENDFOR - -
immintrin.h
+
emmintrin.h
 - + Integer - AVX512VL - AVX512BW - Miscellaneous - - - - - Unpack and interleave 16-bit integers from the low half of "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + SSE2 + Set + + + Broadcast 16-bit integer "a" to all all elements of "dst". This intrinsic may generate "vpbroadcastw". -INTERLEAVE_WORDS(src1[127:0], src2[127:0]){ - dst[15:0] := src1[15:0] - dst[31:16] := src2[15:0] - dst[47:32] := src1[31:16] - dst[63:48] := src2[31:16] - dst[79:64] := src1[47:32] - dst[95:80] := src2[47:32] - dst[111:96] := src1[63:48] - dst[127:112] := src2[63:48] - RETURN dst[127:0] -} - -tmp_dst[127:0] := INTERLEAVE_WORDS(a[127:0], b[127:0]) - FOR j := 0 to 7 i := j*16 - IF k[j] - dst[i+15:i] := tmp_dst[i+15:i] - ELSE - dst[i+15:i] := src[i+15:i] - FI + dst[i+15:i] := a[15:0] ENDFOR -dst[MAX:128] := 0 - -
immintrin.h
+
emmintrin.h
 - + Integer - AVX512VL - AVX512BW - Miscellaneous - - - - Unpack and interleave 16-bit integers from the low half of "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + SSE2 + Set + + + Broadcast 8-bit integer "a" to all elements of "dst". This intrinsic may generate "vpbroadcastb". -INTERLEAVE_WORDS(src1[127:0], src2[127:0]){ - dst[15:0] := src1[15:0] - dst[31:16] := src2[15:0] - dst[47:32] := src1[31:16] - dst[63:48] := src2[31:16] - dst[79:64] := src1[47:32] - dst[95:80] := src2[47:32] - dst[111:96] := src1[63:48] - dst[127:112] := src2[63:48] - RETURN dst[127:0] -} - -tmp_dst[127:0] := INTERLEAVE_WORDS(a[127:0], b[127:0]) - -FOR j := 0 to 7 - i := j*16 - IF k[j] - dst[i+15:i] := tmp_dst[i+15:i] - ELSE - dst[i+15:i] := 0 - FI +FOR j := 0 to 15 + i := j*8 + dst[i+7:i] := a[7:0] ENDFOR -dst[MAX:128] := 0 - -
immintrin.h
+
emmintrin.h
+ + + Integer + SSE2 + Set + + + + Set packed 64-bit integers in "dst" with the supplied values in reverse order. + +dst[63:0] := e1 +dst[127:64] := e0 + +
emmintrin.h
+ + + Integer + SSE2 + Set + + + + + + Set packed 32-bit integers in "dst" with the supplied values in reverse order. + +dst[31:0] := e3 +dst[63:32] := e2 +dst[95:64] := e1 +dst[127:96] := e0 + +
emmintrin.h
 - + Integer - AVX512VL - AVX512F - Logical - - - - - Compute the bitwise XOR of packed 32-bit integers in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + SSE2 + Set + + + + + + + + + + Set packed 16-bit integers in "dst" with the supplied values in reverse order. -FOR j := 0 to 7 - i := j*32 - IF k[j] - dst[i+31:i] := a[i+31:i] XOR b[i+31:i] - ELSE - dst[i+31:i] := src[i+31:i] - FI -ENDFOR -dst[MAX:256] := 0 +dst[15:0] := e7 +dst[31:16] := e6 +dst[47:32] := e5 +dst[63:48] := e4 +dst[79:64] := e3 +dst[95:80] := e2 +dst[111:96] := e1 +dst[127:112] := e0 - -
immintrin.h
+
emmintrin.h
 - + Integer - AVX512VL - AVX512F - Logical - - - - Compute the bitwise XOR of packed 32-bit integers in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - + SSE2 + Set + + + + + + + + + + + + + + + + + + Set packed 8-bit integers in "dst" with the supplied values in reverse order. -FOR j := 0 to 7 - i := j*32 - IF k[j] - dst[i+31:i] := a[i+31:i] XOR b[i+31:i] - ELSE - dst[i+31:i] := 0 - FI -ENDFOR -dst[MAX:256] := 0 +dst[7:0] := e15 +dst[15:8] := e14 +dst[23:16] := e13 +dst[31:24] := e12 +dst[39:32] := e11 +dst[47:40] := e10 +dst[55:48] := e9 +dst[63:56] := e8 +dst[71:64] := e7 +dst[79:72] := e6 +dst[87:80] := e5 +dst[95:88] := e4 +dst[103:96] := e3 +dst[111:104] := e2 +dst[119:112] := e1 +dst[127:120] := e0 - -
immintrin.h
+
emmintrin.h
 - + Integer - AVX512VL - AVX512F - Logical - - - - - Compute the bitwise XOR of packed 32-bit integers in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + SSE2 + Set + + Return vector of type __m128i with all elements set to zero. -FOR j := 0 to 3 - i := j*32 - IF k[j] - dst[i+31:i] := a[i+31:i] XOR b[i+31:i] - ELSE - dst[i+31:i] := src[i+31:i] - FI -ENDFOR -dst[MAX:128] := 0 +dst[MAX:0] := 0 - -
immintrin.h
+ +
emmintrin.h
 - + Integer - AVX512VL - AVX512F - Logical - - - - Compute the bitwise XOR of packed 32-bit integers in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - + SSE2 + Load + + + Load 64-bit integer from memory into the first element of "dst". -FOR j := 0 to 3 - i := j*32 - IF k[j] - dst[i+31:i] := a[i+31:i] XOR b[i+31:i] - ELSE - dst[i+31:i] := 0 - FI -ENDFOR -dst[MAX:128] := 0 +dst[63:0] := MEM[mem_addr+63:mem_addr] +dst[MAX:64] := 0 - -
immintrin.h
+ +
emmintrin.h
 - + Integer - AVX512VL - AVX512F - Logical - - - - - Compute the bitwise XOR of packed 64-bit integers in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + SSE2 + Load + + + Load 128-bits of integer data from memory into "dst". + "mem_addr" must be aligned on a 16-byte boundary or a general-protection exception may be generated. -FOR j := 0 to 3 - i := j*64 - IF k[j] - dst[i+63:i] := a[i+63:i] XOR b[i+63:i] - ELSE - dst[i+63:i] := src[i+63:i] - FI -ENDFOR -dst[MAX:256] := 0 +dst[127:0] := MEM[mem_addr+127:mem_addr] - -
immintrin.h
+ +
emmintrin.h
 - + Integer - AVX512VL - AVX512F - Logical - - - - Compute the bitwise XOR of packed 64-bit integers in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - + SSE2 + Load + + + Load 128-bits of integer data from memory into "dst". + "mem_addr" does not need to be aligned on any particular boundary. -FOR j := 0 to 3 - i := j*64 - IF k[j] - dst[i+63:i] := a[i+63:i] XOR b[i+63:i] - ELSE - dst[i+63:i] := 0 - FI -ENDFOR -dst[MAX:256] := 0 +dst[127:0] := MEM[mem_addr+127:mem_addr] - -
immintrin.h
+ +
emmintrin.h
 - + Integer - AVX512VL - AVX512F - Logical - - - - - Compute the bitwise XOR of packed 64-bit integers in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + SSE2 + Store + + + + + Conditionally store 8-bit integer elements from "a" into memory using "mask" (elements are not stored when the highest bit is not set in the corresponding element) and a non-temporal memory hint. "mem_addr" does not need to be aligned on any particular boundary. -FOR j := 0 to 1 - i := j*64 - IF k[j] - dst[i+63:i] := a[i+63:i] XOR b[i+63:i] - ELSE - dst[i+63:i] := src[i+63:i] +FOR j := 0 to 15 + i := j*8 + IF mask[i+7] + MEM[mem_addr+i+7:mem_addr+i] := a[i+7:i] FI ENDFOR -dst[MAX:128] := 0 - -
immintrin.h
+ +
emmintrin.h
 - + Integer - AVX512VL - AVX512F - Logical - - - - Compute the bitwise XOR of packed 64-bit integers in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - + SSE2 + Store + + + + Store 128-bits of integer data from "a" into memory. + "mem_addr" must be aligned on a 16-byte boundary or a general-protection exception may be generated. -FOR j := 0 to 1 - i := j*64 - IF k[j] - dst[i+63:i] := a[i+63:i] XOR b[i+63:i] - ELSE - dst[i+63:i] := 0 - FI -ENDFOR -dst[MAX:128] := 0 +MEM[mem_addr+127:mem_addr] := a[127:0] - -
immintrin.h
+ +
emmintrin.h
 - - Floating Point - AVX512VL - AVX512DQ - Miscellaneous - - - - - - Calculate the max, min, absolute max, or absolute min (depending on control in "imm8") for packed double-precision (64-bit) floating-point elements in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - imm8[1:0] specifies the operation control: 00 = min, 01 = max, 10 = absolute max, 11 = absolute min. - imm8[3:2] specifies the sign control: 00 = sign from a, 01 = sign from compare result, 10 = clear sign bit, 11 = set sign bit. - + + Integer + SSE2 + Store + + + + Store 128-bits of integer data from "a" into memory. + "mem_addr" does not need to be aligned on any particular boundary. -RANGE(src1[63:0], src2[63:0], opCtl[1:0], signSelCtl[1:0]) -{ - CASE opCtl[1:0] - 0: tmp[63:0] := (src1[63:0] <= src2[63:0]) ? src1[63:0] : src2[63:0] - 1: tmp[63:0] := (src1[63:0] <= src2[63:0]) ? src2[63:0] : src1[63:0] - 2: tmp[63:0] := (ABS(src1[63:0]) <= ABS(src2[63:0])) ? src1[63:0] : src2[63:0] - 3: tmp[63:0] := (ABS(src1[63:0]) <= ABS(src2[63:0])) ? src2[63:0] : src1[63:0] - ESAC - - CASE signSelCtl[1:0] - 0: dst[63:0] := (src1[63] << 63) OR (tmp[62:0]) - 1: dst[63:0] := tmp[63:0] - 2: dst[63:0] := (0 << 63) OR (tmp[62:0]) - 3: dst[63:0] := (1 << 63) OR (tmp[62:0]) - ESAC - - RETURN dst -} - -FOR j := 0 to 3 - i := j*64 - IF k[j] - dst[i+63:i] := RANGE(a[i+63:i], b[i+63:i], imm8[1:0], imm8[3:2]) - ELSE - dst[i+63:i] := src[i+63:i] - FI -ENDFOR -dst[MAX:256] := 0 +MEM[mem_addr+127:mem_addr] := a[127:0] - -
immintrin.h
+ +
emmintrin.h
 - - Floating Point - AVX512VL - AVX512DQ - Miscellaneous - - - - - Calculate the max, min, absolute max, or absolute min (depending on control in "imm8") for packed double-precision (64-bit) floating-point elements in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - imm8[1:0] specifies the operation control: 00 = min, 01 = max, 10 = absolute max, 11 = absolute min. - imm8[3:2] specifies the sign control: 00 = sign from a, 01 = sign from compare result, 10 = clear sign bit, 11 = set sign bit. - + + Integer + SSE2 + Store + + + + Store 64-bit integer from the first element of "a" into memory. -RANGE(src1[63:0], src2[63:0], opCtl[1:0], signSelCtl[1:0]) -{ - CASE opCtl[1:0] - 0: tmp[63:0] := (src1[63:0] <= src2[63:0]) ? src1[63:0] : src2[63:0] - 1: tmp[63:0] := (src1[63:0] <= src2[63:0]) ? src2[63:0] : src1[63:0] - 2: tmp[63:0] := (ABS(src1[63:0]) <= ABS(src2[63:0])) ? src1[63:0] : src2[63:0] - 3: tmp[63:0] := (ABS(src1[63:0]) <= ABS(src2[63:0])) ? src2[63:0] : src1[63:0] - ESAC - - CASE signSelCtl[1:0] - 0: dst[63:0] := (src1[63] << 63) OR (tmp[62:0]) - 1: dst[63:0] := tmp[63:0] - 2: dst[63:0] := (0 << 63) OR (tmp[62:0]) - 3: dst[63:0] := (1 << 63) OR (tmp[62:0]) - ESAC - - RETURN dst -} - -FOR j := 0 to 3 - i := j*64 - IF k[j] - dst[i+63:i] := RANGE(a[i+63:i], b[i+63:i], imm8[1:0], imm8[3:2]) - ELSE - dst[i+63:i] := 0 - FI -ENDFOR -dst[MAX:256] := 0 +MEM[mem_addr+63:mem_addr] := a[63:0] - -
immintrin.h
+ +
emmintrin.h
 - - Floating Point - AVX512VL - AVX512DQ - Miscellaneous - - - - Calculate the max, min, absolute max, or absolute min (depending on control in "imm8") for packed double-precision (64-bit) floating-point elements in "a" and "b", and store the results in "dst". - imm8[1:0] specifies the operation control: 00 = min, 01 = max, 10 = absolute max, 11 = absolute min. - imm8[3:2] specifies the sign control: 00 = sign from a, 01 = sign from compare result, 10 = clear sign bit, 11 = set sign bit. - + + Integer + SSE2 + Store + + + + Store 128-bits of integer data from "a" into memory using a non-temporal memory hint. + "mem_addr" must be aligned on a 16-byte boundary or a general-protection exception may be generated. -RANGE(src1[63:0], src2[63:0], opCtl[1:0], signSelCtl[1:0]) -{ - CASE opCtl[1:0] - 0: tmp[63:0] := (src1[63:0] <= src2[63:0]) ? src1[63:0] : src2[63:0] - 1: tmp[63:0] := (src1[63:0] <= src2[63:0]) ? src2[63:0] : src1[63:0] - 2: tmp[63:0] := (ABS(src1[63:0]) <= ABS(src2[63:0])) ? src1[63:0] : src2[63:0] - 3: tmp[63:0] := (ABS(src1[63:0]) <= ABS(src2[63:0])) ? src2[63:0] : src1[63:0] - ESAC - - CASE signSelCtl[1:0] - 0: dst[63:0] := (src1[63] << 63) OR (tmp[62:0]) - 1: dst[63:0] := tmp[63:0] - 2: dst[63:0] := (0 << 63) OR (tmp[62:0]) - 3: dst[63:0] := (1 << 63) OR (tmp[62:0]) - ESAC - - RETURN dst -} - -FOR j := 0 to 3 - i := j*64 - dst[i+63:i] := RANGE(a[i+63:i], b[i+63:i], imm8[1:0], imm8[3:2]) -ENDFOR -dst[MAX:256] := 0 +MEM[mem_addr+127:mem_addr] := a[127:0] - -
immintrin.h
+ +
emmintrin.h
 - - Floating Point - AVX512DQ - Miscellaneous - - - - - - Calculate the max, min, absolute max, or absolute min (depending on control in "imm8") for packed double-precision (64-bit) floating-point elements in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - imm8[1:0] specifies the operation control: 00 = min, 01 = max, 10 = absolute max, 11 = absolute min. - imm8[3:2] specifies the sign control: 00 = sign from a, 01 = sign from compare result, 10 = clear sign bit, 11 = set sign bit. - + + Integer + SSE2 + Store + + + + Store 32-bit integer "a" into memory using a non-temporal hint to minimize cache pollution. If the cache line containing address "mem_addr" is already in the cache, the cache will be updated. -RANGE(src1[63:0], src2[63:0], opCtl[1:0], signSelCtl[1:0]) -{ - CASE opCtl[1:0] - 0: tmp[63:0] := (src1[63:0] <= src2[63:0]) ? src1[63:0] : src2[63:0] - 1: tmp[63:0] := (src1[63:0] <= src2[63:0]) ? src2[63:0] : src1[63:0] - 2: tmp[63:0] := (ABS(src1[63:0]) <= ABS(src2[63:0])) ? src1[63:0] : src2[63:0] - 3: tmp[63:0] := (ABS(src1[63:0]) <= ABS(src2[63:0])) ? src2[63:0] : src1[63:0] - ESAC - - CASE signSelCtl[1:0] - 0: dst[63:0] := (src1[63] << 63) OR (tmp[62:0]) - 1: dst[63:0] := tmp[63:0] - 2: dst[63:0] := (0 << 63) OR (tmp[62:0]) - 3: dst[63:0] := (1 << 63) OR (tmp[62:0]) - ESAC - - RETURN dst -} - -FOR j := 0 to 7 - i := j*64 - IF k[j] - dst[i+63:i] := RANGE(a[i+63:i], b[i+63:i], imm8[1:0], imm8[3:2]) - ELSE - dst[i+63:i] := src[i+63:i] - FI -ENDFOR -dst[MAX:512] := 0 +MEM[mem_addr+31:mem_addr] := a[31:0] - -
immintrin.h
+ +
emmintrin.h
 - - Floating Point - AVX512DQ - Miscellaneous - - - - - - - Calculate the max, min, absolute max, or absolute min (depending on control in "imm8") for packed double-precision (64-bit) floating-point elements in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - imm8[1:0] specifies the operation control: 00 = min, 01 = max, 10 = absolute max, 11 = absolute min. - imm8[3:2] specifies the sign control: 00 = sign from a, 01 = sign from compare result, 10 = clear sign bit, 11 = set sign bit. - [round_note] - + + Integer + SSE2 + Store + + + + Store 64-bit integer "a" into memory using a non-temporal hint to minimize cache pollution. If the cache line containing address "mem_addr" is already in the cache, the cache will be updated. -RANGE(src1[63:0], src2[63:0], opCtl[1:0], signSelCtl[1:0]) -{ - CASE opCtl[1:0] - 0: tmp[63:0] := (src1[63:0] <= src2[63:0]) ? src1[63:0] : src2[63:0] - 1: tmp[63:0] := (src1[63:0] <= src2[63:0]) ? src2[63:0] : src1[63:0] - 2: tmp[63:0] := (ABS(src1[63:0]) <= ABS(src2[63:0])) ? src1[63:0] : src2[63:0] - 3: tmp[63:0] := (ABS(src1[63:0]) <= ABS(src2[63:0])) ? src2[63:0] : src1[63:0] - ESAC - - CASE signSelCtl[1:0] - 0: dst[63:0] := (src1[63] << 63) OR (tmp[62:0]) - 1: dst[63:0] := tmp[63:0] - 2: dst[63:0] := (0 << 63) OR (tmp[62:0]) - 3: dst[63:0] := (1 << 63) OR (tmp[62:0]) - ESAC - - RETURN dst -} - -FOR j := 0 to 7 - i := j*64 - IF k[j] - dst[i+63:i] := RANGE(a[i+63:i], b[i+63:i], imm8[1:0], imm8[3:2]) - ELSE - dst[i+63:i] := src[i+63:i] - FI -ENDFOR -dst[MAX:512] := 0 +MEM[mem_addr+63:mem_addr] := a[63:0] - -
immintrin.h
+ +
emmintrin.h
 - - Floating Point - AVX512DQ + + Integer + SSE2 Miscellaneous - - - - - Calculate the max, min, absolute max, or absolute min (depending on control in "imm8") for packed double-precision (64-bit) floating-point elements in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - imm8[1:0] specifies the operation control: 00 = min, 01 = max, 10 = absolute max, 11 = absolute min. - imm8[3:2] specifies the sign control: 00 = sign from a, 01 = sign from compare result, 10 = clear sign bit, 11 = set sign bit. - + + + Copy the lower 64-bit integer in "a" to "dst". -RANGE(src1[63:0], src2[63:0], opCtl[1:0], signSelCtl[1:0]) -{ - CASE opCtl[1:0] - 0: tmp[63:0] := (src1[63:0] <= src2[63:0]) ? src1[63:0] : src2[63:0] - 1: tmp[63:0] := (src1[63:0] <= src2[63:0]) ? src2[63:0] : src1[63:0] - 2: tmp[63:0] := (ABS(src1[63:0]) <= ABS(src2[63:0])) ? src1[63:0] : src2[63:0] - 3: tmp[63:0] := (ABS(src1[63:0]) <= ABS(src2[63:0])) ? src2[63:0] : src1[63:0] - ESAC - - CASE signSelCtl[1:0] - 0: dst[63:0] := (src1[63] << 63) OR (tmp[62:0]) - 1: dst[63:0] := tmp[63:0] - 2: dst[63:0] := (0 << 63) OR (tmp[62:0]) - 3: dst[63:0] := (1 << 63) OR (tmp[62:0]) - ESAC - - RETURN dst -} - -FOR j := 0 to 7 - i := j*64 - IF k[j] - dst[i+63:i] := RANGE(a[i+63:i], b[i+63:i], imm8[1:0], imm8[3:2]) - ELSE - dst[i+63:i] := 0 - FI -ENDFOR -dst[MAX:512] := 0 +dst[63:0] := a[63:0] - -
immintrin.h
+ +
emmintrin.h
 - - Floating Point - AVX512DQ - Miscellaneous - - - - - - Calculate the max, min, absolute max, or absolute min (depending on control in "imm8") for packed double-precision (64-bit) floating-point elements in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - imm8[1:0] specifies the operation control: 00 = min, 01 = max, 10 = absolute max, 11 = absolute min. - imm8[3:2] specifies the sign control: 00 = sign from a, 01 = sign from compare result, 10 = clear sign bit, 11 = set sign bit. - [round_note] - + + Integer + SSE2 + Move + + + Copy the 64-bit integer "a" to the lower element of "dst", and zero the upper element. -RANGE(src1[63:0], src2[63:0], opCtl[1:0], signSelCtl[1:0]) -{ - CASE opCtl[1:0] - 0: tmp[63:0] := (src1[63:0] <= src2[63:0]) ? src1[63:0] : src2[63:0] - 1: tmp[63:0] := (src1[63:0] <= src2[63:0]) ? src2[63:0] : src1[63:0] - 2: tmp[63:0] := (ABS(src1[63:0]) <= ABS(src2[63:0])) ? src1[63:0] : src2[63:0] - 3: tmp[63:0] := (ABS(src1[63:0]) <= ABS(src2[63:0])) ? src2[63:0] : src1[63:0] - ESAC - - CASE signSelCtl[1:0] - 0: dst[63:0] := (src1[63] << 63) OR (tmp[62:0]) - 1: dst[63:0] := tmp[63:0] - 2: dst[63:0] := (0 << 63) OR (tmp[62:0]) - 3: dst[63:0] := (1 << 63) OR (tmp[62:0]) - ESAC - - RETURN dst -} - -FOR j := 0 to 7 - i := j*64 - IF k[j] - dst[i+63:i] := RANGE(a[i+63:i], b[i+63:i], imm8[1:0], imm8[3:2]) - ELSE - dst[i+63:i] := 0 - FI -ENDFOR -dst[MAX:512] := 0 +dst[63:0] := a[63:0] +dst[127:64] := 0 - -
immintrin.h
+ +
emmintrin.h
 - - Floating Point - AVX512DQ - Miscellaneous - - - - Calculate the max, min, absolute max, or absolute min (depending on control in "imm8") for packed double-precision (64-bit) floating-point elements in "a" and "b", and store the results in "dst". - imm8[1:0] specifies the operation control: 00 = min, 01 = max, 10 = absolute max, 11 = absolute min. - imm8[3:2] specifies the sign control: 00 = sign from a, 01 = sign from compare result, 10 = clear sign bit, 11 = set sign bit. - + + Integer + SSE2 + Move + + + Copy the lower 64-bit integer in "a" to the lower element of "dst", and zero the upper element. -RANGE(src1[63:0], src2[63:0], opCtl[1:0], signSelCtl[1:0]) -{ - CASE opCtl[1:0] - 0: tmp[63:0] := (src1[63:0] <= src2[63:0]) ? src1[63:0] : src2[63:0] - 1: tmp[63:0] := (src1[63:0] <= src2[63:0]) ? src2[63:0] : src1[63:0] - 2: tmp[63:0] := (ABS(src1[63:0]) <= ABS(src2[63:0])) ? src1[63:0] : src2[63:0] - 3: tmp[63:0] := (ABS(src1[63:0]) <= ABS(src2[63:0])) ? src2[63:0] : src1[63:0] - ESAC - - CASE signSelCtl[1:0] - 0: dst[63:0] := (src1[63] << 63) OR (tmp[62:0]) - 1: dst[63:0] := tmp[63:0] - 2: dst[63:0] := (0 << 63) OR (tmp[62:0]) - 3: dst[63:0] := (1 << 63) OR (tmp[62:0]) - ESAC - - RETURN dst -} - -FOR j := 0 to 7 - i := j*64 - dst[i+63:i] := RANGE(a[i+63:i], b[i+63:i], imm8[1:0], imm8[3:2]) -ENDFOR -dst[MAX:512] := 0 +dst[63:0] := a[63:0] +dst[127:64] := 0 - -
immintrin.h
+ +
emmintrin.h
 - - Floating Point - AVX512DQ + + Integer + SSE2 Miscellaneous - - - - - Calculate the max, min, absolute max, or absolute min (depending on control in "imm8") for packed double-precision (64-bit) floating-point elements in "a" and "b", and store the results in "dst". - imm8[1:0] specifies the operation control: 00 = min, 01 = max, 10 = absolute max, 11 = absolute min. - imm8[3:2] specifies the sign control: 00 = sign from a, 01 = sign from compare result, 10 = clear sign bit, 11 = set sign bit. - [round_note] - - -RANGE(src1[63:0], src2[63:0], opCtl[1:0], signSelCtl[1:0]) -{ - CASE opCtl[1:0] - 0: tmp[63:0] := (src1[63:0] <= src2[63:0]) ? src1[63:0] : src2[63:0] - 1: tmp[63:0] := (src1[63:0] <= src2[63:0]) ? src2[63:0] : src1[63:0] - 2: tmp[63:0] := (ABS(src1[63:0]) <= ABS(src2[63:0])) ? src1[63:0] : src2[63:0] - 3: tmp[63:0] := (ABS(src1[63:0]) <= ABS(src2[63:0])) ? src2[63:0] : src1[63:0] - ESAC - - CASE signSelCtl[1:0] - 0: dst[63:0] := (src1[63] << 63) OR (tmp[62:0]) - 1: dst[63:0] := tmp[63:0] - 2: dst[63:0] := (0 << 63) OR (tmp[62:0]) - 3: dst[63:0] := (1 << 63) OR (tmp[62:0]) - ESAC - - RETURN dst -} - -FOR j := 0 to 7 - i := j*64 - dst[i+63:i] := RANGE(a[i+63:i], b[i+63:i], imm8[1:0], imm8[3:2]) -ENDFOR -dst[MAX:512] := 0 - - -
immintrin.h
+ + + + Convert packed signed 16-bit integers from "a" and "b" to packed 8-bit integers using signed saturation, and store the results in "dst". + +dst[7:0] := Saturate8(a[15:0]) +dst[15:8] := Saturate8(a[31:16]) +dst[23:16] := Saturate8(a[47:32]) +dst[31:24] := Saturate8(a[63:48]) +dst[39:32] := Saturate8(a[79:64]) +dst[47:40] := Saturate8(a[95:80]) +dst[55:48] := Saturate8(a[111:96]) +dst[63:56] := Saturate8(a[127:112]) +dst[71:64] := Saturate8(b[15:0]) +dst[79:72] := Saturate8(b[31:16]) +dst[87:80] := Saturate8(b[47:32]) +dst[95:88] := Saturate8(b[63:48]) +dst[103:96] := Saturate8(b[79:64]) +dst[111:104] := Saturate8(b[95:80]) +dst[119:112] := Saturate8(b[111:96]) +dst[127:120] := Saturate8(b[127:112]) + + +
emmintrin.h
 - - Floating Point - AVX512VL - AVX512DQ - Miscellaneous - - - - - - Calculate the max, min, absolute max, or absolute min (depending on control in "imm8") for packed double-precision (64-bit) floating-point elements in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - imm8[1:0] specifies the operation control: 00 = min, 01 = max, 10 = absolute max, 11 = absolute min. - imm8[3:2] specifies the sign control: 00 = sign from a, 01 = sign from compare result, 10 = clear sign bit, 11 = set sign bit. - - -RANGE(src1[63:0], src2[63:0], opCtl[1:0], signSelCtl[1:0]) -{ - CASE opCtl[1:0] - 0: tmp[63:0] := (src1[63:0] <= src2[63:0]) ? src1[63:0] : src2[63:0] - 1: tmp[63:0] := (src1[63:0] <= src2[63:0]) ? src2[63:0] : src1[63:0] - 2: tmp[63:0] := (ABS(src1[63:0]) <= ABS(src2[63:0])) ? src1[63:0] : src2[63:0] - 3: tmp[63:0] := (ABS(src1[63:0]) <= ABS(src2[63:0])) ? src2[63:0] : src1[63:0] - ESAC - - CASE signSelCtl[1:0] - 0: dst[63:0] := (src1[63] << 63) OR (tmp[62:0]) - 1: dst[63:0] := tmp[63:0] - 2: dst[63:0] := (0 << 63) OR (tmp[62:0]) - 3: dst[63:0] := (1 << 63) OR (tmp[62:0]) - ESAC - - RETURN dst -} - -FOR j := 0 to 1 - i := j*64 - IF k[j] - dst[i+63:i] := RANGE(a[i+63:i], b[i+63:i], imm8[1:0], imm8[3:2]) - ELSE - dst[i+63:i] := src[i+63:i] - FI -ENDFOR -dst[MAX:128] := 0 - - -
immintrin.h
+ + Integer + SSE2 + Miscellaneous + + + + Convert packed signed 32-bit integers from "a" and "b" to packed 16-bit integers using signed saturation, and store the results in "dst". + +dst[15:0] := Saturate16(a[31:0]) +dst[31:16] := Saturate16(a[63:32]) +dst[47:32] := Saturate16(a[95:64]) +dst[63:48] := Saturate16(a[127:96]) +dst[79:64] := Saturate16(b[31:0]) +dst[95:80] := Saturate16(b[63:32]) +dst[111:96] := Saturate16(b[95:64]) +dst[127:112] := Saturate16(b[127:96]) + + +
emmintrin.h
 - - Floating Point - AVX512VL - AVX512DQ + + Integer + SSE2 Miscellaneous - - - - - Calculate the max, min, absolute max, or absolute min (depending on control in "imm8") for packed double-precision (64-bit) floating-point elements in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - imm8[1:0] specifies the operation control: 00 = min, 01 = max, 10 = absolute max, 11 = absolute min. - imm8[3:2] specifies the sign control: 00 = sign from a, 01 = sign from compare result, 10 = clear sign bit, 11 = set sign bit. - + + + + Convert packed signed 16-bit integers from "a" and "b" to packed 8-bit integers using unsigned saturation, and store the results in "dst". + +dst[7:0] := SaturateU8(a[15:0]) +dst[15:8] := SaturateU8(a[31:16]) +dst[23:16] := SaturateU8(a[47:32]) +dst[31:24] := SaturateU8(a[63:48]) +dst[39:32] := SaturateU8(a[79:64]) +dst[47:40] := SaturateU8(a[95:80]) +dst[55:48] := SaturateU8(a[111:96]) +dst[63:56] := SaturateU8(a[127:112]) +dst[71:64] := SaturateU8(b[15:0]) +dst[79:72] := SaturateU8(b[31:16]) +dst[87:80] := SaturateU8(b[47:32]) +dst[95:88] := SaturateU8(b[63:48]) +dst[103:96] := SaturateU8(b[79:64]) +dst[111:104] := SaturateU8(b[95:80]) +dst[119:112] := SaturateU8(b[111:96]) +dst[127:120] := SaturateU8(b[127:112]) + + +
emmintrin.h
+ + + Integer + SSE2 + Swizzle + + + + Extract a 16-bit integer from "a", selected with "imm8", and store the result in the lower element of "dst". -RANGE(src1[63:0], src2[63:0], opCtl[1:0], signSelCtl[1:0]) -{ - CASE opCtl[1:0] - 0: tmp[63:0] := (src1[63:0] <= src2[63:0]) ? src1[63:0] : src2[63:0] - 1: tmp[63:0] := (src1[63:0] <= src2[63:0]) ? src2[63:0] : src1[63:0] - 2: tmp[63:0] := (ABS(src1[63:0]) <= ABS(src2[63:0])) ? src1[63:0] : src2[63:0] - 3: tmp[63:0] := (ABS(src1[63:0]) <= ABS(src2[63:0])) ? src2[63:0] : src1[63:0] - ESAC - - CASE signSelCtl[1:0] - 0: dst[63:0] := (src1[63] << 63) OR (tmp[62:0]) - 1: dst[63:0] := tmp[63:0] - 2: dst[63:0] := (0 << 63) OR (tmp[62:0]) - 3: dst[63:0] := (1 << 63) OR (tmp[62:0]) - ESAC - - RETURN dst -} - -FOR j := 0 to 1 - i := j*64 - IF k[j] - dst[i+63:i] := RANGE(a[i+63:i], b[i+63:i], imm8[1:0], imm8[3:2]) - ELSE - dst[i+63:i] := 0 - FI -ENDFOR -dst[MAX:128] := 0 +dst[15:0] := (a[127:0] >> (imm8[2:0] * 16))[15:0] +dst[31:16] := 0 - -
immintrin.h
+ +
emmintrin.h
 - - Floating Point - AVX512VL - AVX512DQ - Miscellaneous - - - - Calculate the max, min, absolute max, or absolute min (depending on control in "imm8") for packed double-precision (64-bit) floating-point elements in "a" and "b", and store the results in "dst". - imm8[1:0] specifies the operation control: 00 = min, 01 = max, 10 = absolute max, 11 = absolute min. - imm8[3:2] specifies the sign control: 00 = sign from a, 01 = sign from compare result, 10 = clear sign bit, 11 = set sign bit. - + + Integer + SSE2 + Swizzle + + + + + Copy "a" to "dst", and insert the 16-bit integer "i" into "dst" at the location specified by "imm8". -RANGE(src1[63:0], src2[63:0], opCtl[1:0], signSelCtl[1:0]) -{ - CASE opCtl[1:0] - 0: tmp[63:0] := (src1[63:0] <= src2[63:0]) ? src1[63:0] : src2[63:0] - 1: tmp[63:0] := (src1[63:0] <= src2[63:0]) ? src2[63:0] : src1[63:0] - 2: tmp[63:0] := (ABS(src1[63:0]) <= ABS(src2[63:0])) ? src1[63:0] : src2[63:0] - 3: tmp[63:0] := (ABS(src1[63:0]) <= ABS(src2[63:0])) ? src2[63:0] : src1[63:0] - ESAC - - CASE signSelCtl[1:0] - 0: dst[63:0] := (src1[63] << 63) OR (tmp[62:0]) - 1: dst[63:0] := tmp[63:0] - 2: dst[63:0] := (0 << 63) OR (tmp[62:0]) - 3: dst[63:0] := (1 << 63) OR (tmp[62:0]) - ESAC - - RETURN dst -} - -FOR j := 0 to 1 - i := j*64 - dst[i+63:i] := RANGE(a[i+63:i], b[i+63:i], imm8[1:0], imm8[3:2]) -ENDFOR -dst[MAX:128] := 0 +dst[127:0] := a[127:0] +sel := imm8[2:0]*16 +dst[sel+15:sel] := i[15:0] - -
immintrin.h
+ +
emmintrin.h
 - - Floating Point - AVX512VL - AVX512DQ + + Integer + SSE2 Miscellaneous - - - - - - Calculate the max, min, absolute max, or absolute min (depending on control in "imm8") for packed single-precision (32-bit) floating-point elements in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - imm8[1:0] specifies the operation control: 00 = min, 01 = max, 10 = absolute max, 11 = absolute min. - imm8[3:2] specifies the sign control: 00 = sign from a, 01 = sign from compare result, 10 = clear sign bit, 11 = set sign bit. - + + + Create mask from the most significant bit of each 8-bit element in "a", and store the result in "dst". -RANGE(src1[31:0], src2[31:0], opCtl[1:0], signSelCtl[1:0]) -{ - CASE opCtl[1:0] - 0: tmp[31:0] := (src1[31:0] <= src2[31:0]) ? src1[31:0] : src2[31:0] - 1: tmp[31:0] := (src1[31:0] <= src2[31:0]) ? src2[31:0] : src1[31:0] - 2: tmp[31:0] := (ABS(src1[31:0]) <= ABS(src2[31:0])) ? src1[31:0] : src2[31:0] - 3: tmp[31:0] := (ABS(src1[31:0]) <= ABS(src2[31:0])) ? src2[31:0] : src1[31:0] - ESAC - - CASE signSelCtl[1:0] - 0: dst[31:0] := (src1[31] << 31) OR (tmp[30:0]) - 1: dst[31:0] := tmp[63:0] - 2: dst[31:0] := (0 << 31) OR (tmp[30:0]) - 3: dst[31:0] := (1 << 31) OR (tmp[30:0]) - ESAC - - RETURN dst -} - -FOR j := 0 to 7 - i := j*32 - IF k[j] - dst[i+31:i] := RANGE(a[i+31:i], b[i+31:i], imm8[1:0], imm8[3:2]) - ELSE - dst[i+31:i] := src[i+31:i] - FI +FOR j := 0 to 15 + i := j*8 + dst[j] := a[i+7] ENDFOR -dst[MAX:256] := 0 +dst[MAX:16] := 0 - -
immintrin.h
+ +
emmintrin.h
 - - Floating Point - AVX512VL - AVX512DQ - Miscellaneous - - - - - Calculate the max, min, absolute max, or absolute min (depending on control in "imm8") for packed single-precision (32-bit) floating-point elements in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - imm8[1:0] specifies the operation control: 00 = min, 01 = max, 10 = absolute max, 11 = absolute min. - imm8[3:2] specifies the sign control: 00 = sign from a, 01 = sign from compare result, 10 = clear sign bit, 11 = set sign bit. - + + Integer + SSE2 + Swizzle + + + + Shuffle 32-bit integers in "a" using the control in "imm8", and store the results in "dst". -RANGE(src1[31:0], src2[31:0], opCtl[1:0], signSelCtl[1:0]) -{ - CASE opCtl[1:0] - 0: tmp[31:0] := (src1[31:0] <= src2[31:0]) ? src1[31:0] : src2[31:0] - 1: tmp[31:0] := (src1[31:0] <= src2[31:0]) ? src2[31:0] : src1[31:0] - 2: tmp[31:0] := (ABS(src1[31:0]) <= ABS(src2[31:0])) ? src1[31:0] : src2[31:0] - 3: tmp[31:0] := (ABS(src1[31:0]) <= ABS(src2[31:0])) ? src2[31:0] : src1[31:0] - ESAC - - CASE signSelCtl[1:0] - 0: dst[31:0] := (src1[31] << 31) OR (tmp[30:0]) - 1: dst[31:0] := tmp[63:0] - 2: dst[31:0] := (0 << 31) OR (tmp[30:0]) - 3: dst[31:0] := (1 << 31) OR (tmp[30:0]) +DEFINE SELECT4(src, control) { + CASE(control[1:0]) OF + 0: tmp[31:0] := src[31:0] + 1: tmp[31:0] := src[63:32] + 2: tmp[31:0] := src[95:64] + 3: tmp[31:0] := src[127:96] ESAC - - RETURN dst + RETURN tmp[31:0] } - -FOR j := 0 to 7 - i := j*32 - IF k[j] - dst[i+31:i] := RANGE(a[i+31:i], b[i+31:i], imm8[1:0], imm8[3:2]) - ELSE - dst[i+31:i] := 0 - FI -ENDFOR -dst[MAX:256] := 0 +dst[31:0] := SELECT4(a[127:0], imm8[1:0]) +dst[63:32] := SELECT4(a[127:0], imm8[3:2]) +dst[95:64] := SELECT4(a[127:0], imm8[5:4]) +dst[127:96] := SELECT4(a[127:0], imm8[7:6]) - -
immintrin.h
+ +
emmintrin.h
 - - Floating Point - AVX512VL - AVX512DQ - Miscellaneous - - - - Calculate the max, min, absolute max, or absolute min (depending on control in "imm8") for packed single-precision (32-bit) floating-point elements in "a" and "b", and store the results in "dst". - imm8[1:0] specifies the operation control: 00 = min, 01 = max, 10 = absolute max, 11 = absolute min. - imm8[3:2] specifies the sign control: 00 = sign from a, 01 = sign from compare result, 10 = clear sign bit, 11 = set sign bit. - + + Integer + SSE2 + Swizzle + + + + Shuffle 16-bit integers in the high 64 bits of "a" using the control in "imm8". Store the results in the high 64 bits of "dst", with the low 64 bits being copied from from "a" to "dst". -RANGE(src1[31:0], src2[31:0], opCtl[1:0], signSelCtl[1:0]) -{ - CASE opCtl[1:0] - 0: tmp[31:0] := (src1[31:0] <= src2[31:0]) ? src1[31:0] : src2[31:0] - 1: tmp[31:0] := (src1[31:0] <= src2[31:0]) ? src2[31:0] : src1[31:0] - 2: tmp[31:0] := (ABS(src1[31:0]) <= ABS(src2[31:0])) ? src1[31:0] : src2[31:0] - 3: tmp[31:0] := (ABS(src1[31:0]) <= ABS(src2[31:0])) ? src2[31:0] : src1[31:0] - ESAC - - CASE signSelCtl[1:0] - 0: dst[31:0] := (src1[31] << 31) OR (tmp[30:0]) - 1: dst[31:0] := tmp[63:0] - 2: dst[31:0] := (0 << 31) OR (tmp[30:0]) - 3: dst[31:0] := (1 << 31) OR (tmp[30:0]) - ESAC - - RETURN dst -} - -FOR j := 0 to 7 - i := j*32 - dst[i+31:i] := RANGE(a[i+31:i], b[i+31:i], imm8[1:0], imm8[3:2]) -ENDFOR -dst[MAX:256] := 0 +dst[63:0] := a[63:0] +dst[79:64] := (a >> (imm8[1:0] * 16))[79:64] +dst[95:80] := (a >> (imm8[3:2] * 16))[79:64] +dst[111:96] := (a >> (imm8[5:4] * 16))[79:64] +dst[127:112] := (a >> (imm8[7:6] * 16))[79:64] - -
immintrin.h
+ +
emmintrin.h
 - - Floating Point - AVX512DQ - Miscellaneous - - - - - - Calculate the max, min, absolute max, or absolute min (depending on control in "imm8") for packed single-precision (32-bit) floating-point elements in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - imm8[1:0] specifies the operation control: 00 = min, 01 = max, 10 = absolute max, 11 = absolute min. - imm8[3:2] specifies the sign control: 00 = sign from a, 01 = sign from compare result, 10 = clear sign bit, 11 = set sign bit. - + + Integer + SSE2 + Swizzle + + + + Shuffle 16-bit integers in the low 64 bits of "a" using the control in "imm8". Store the results in the low 64 bits of "dst", with the high 64 bits being copied from from "a" to "dst". -RANGE(src1[31:0], src2[31:0], opCtl[1:0], signSelCtl[1:0]) -{ - CASE opCtl[1:0] - 0: tmp[31:0] := (src1[31:0] <= src2[31:0]) ? src1[31:0] : src2[31:0] - 1: tmp[31:0] := (src1[31:0] <= src2[31:0]) ? src2[31:0] : src1[31:0] - 2: tmp[31:0] := (ABS(src1[31:0]) <= ABS(src2[31:0])) ? src1[31:0] : src2[31:0] - 3: tmp[31:0] := (ABS(src1[31:0]) <= ABS(src2[31:0])) ? src2[31:0] : src1[31:0] - ESAC - - CASE signSelCtl[1:0] - 0: dst[31:0] := (src1[31] << 31) OR (tmp[30:0]) - 1: dst[31:0] := tmp[63:0] - 2: dst[31:0] := (0 << 31) OR (tmp[30:0]) - 3: dst[31:0] := (1 << 31) OR (tmp[30:0]) - ESAC - - RETURN dst +dst[15:0] := (a >> (imm8[1:0] * 16))[15:0] +dst[31:16] := (a >> (imm8[3:2] * 16))[15:0] +dst[47:32] := (a >> (imm8[5:4] * 16))[15:0] +dst[63:48] := (a >> (imm8[7:6] * 16))[15:0] +dst[127:64] := a[127:64] + + +
emmintrin.h
+ + + Integer + SSE2 + Swizzle + + + + Unpack and interleave 8-bit integers from the high half of "a" and "b", and store the results in "dst". + +DEFINE INTERLEAVE_HIGH_BYTES(src1[127:0], src2[127:0]) { + dst[7:0] := src1[71:64] + dst[15:8] := src2[71:64] + dst[23:16] := src1[79:72] + dst[31:24] := src2[79:72] + dst[39:32] := src1[87:80] + dst[47:40] := src2[87:80] + dst[55:48] := src1[95:88] + dst[63:56] := src2[95:88] + dst[71:64] := src1[103:96] + dst[79:72] := src2[103:96] + dst[87:80] := src1[111:104] + dst[95:88] := src2[111:104] + dst[103:96] := src1[119:112] + dst[111:104] := src2[119:112] + dst[119:112] := src1[127:120] + dst[127:120] := src2[127:120] + RETURN dst[127:0] } - -FOR j := 0 to 15 - i := j*32 - IF k[j] - dst[i+31:i] := RANGE(a[i+31:i], b[i+31:i], imm8[1:0], imm8[3:2]) - ELSE - dst[i+31:i] := src[i+31:i] - FI -ENDFOR -dst[MAX:512] := 0 +dst[127:0] := INTERLEAVE_HIGH_BYTES(a[127:0], b[127:0]) - -
immintrin.h
+ +
emmintrin.h
 - - Floating Point - AVX512DQ - Miscellaneous - - - - - - - Calculate the max, min, absolute max, or absolute min (depending on control in "imm8") for packed single-precision (32-bit) floating-point elements in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - imm8[1:0] specifies the operation control: 00 = min, 01 = max, 10 = absolute max, 11 = absolute min. - imm8[3:2] specifies the sign control: 00 = sign from a, 01 = sign from compare result, 10 = clear sign bit, 11 = set sign bit. - [round_note] - + + Integer + SSE2 + Swizzle + + + + Unpack and interleave 16-bit integers from the high half of "a" and "b", and store the results in "dst". -RANGE(src1[31:0], src2[31:0], opCtl[1:0], signSelCtl[1:0]) -{ - CASE opCtl[1:0] - 0: tmp[31:0] := (src1[31:0] <= src2[31:0]) ? src1[31:0] : src2[31:0] - 1: tmp[31:0] := (src1[31:0] <= src2[31:0]) ? src2[31:0] : src1[31:0] - 2: tmp[31:0] := (ABS(src1[31:0]) <= ABS(src2[31:0])) ? src1[31:0] : src2[31:0] - 3: tmp[31:0] := (ABS(src1[31:0]) <= ABS(src2[31:0])) ? src2[31:0] : src1[31:0] - ESAC - - CASE signSelCtl[1:0] - 0: dst[31:0] := (src1[31] << 31) OR (tmp[30:0]) - 1: dst[31:0] := tmp[63:0] - 2: dst[31:0] := (0 << 31) OR (tmp[30:0]) - 3: dst[31:0] := (1 << 31) OR (tmp[30:0]) - ESAC - - RETURN dst +DEFINE INTERLEAVE_HIGH_WORDS(src1[127:0], src2[127:0]) { + dst[15:0] := src1[79:64] + dst[31:16] := src2[79:64] + dst[47:32] := src1[95:80] + dst[63:48] := src2[95:80] + dst[79:64] := src1[111:96] + dst[95:80] := src2[111:96] + dst[111:96] := src1[127:112] + dst[127:112] := src2[127:112] + RETURN dst[127:0] } - -FOR j := 0 to 15 - i := j*32 - IF k[j] - dst[i+31:i] := RANGE(a[i+31:i], b[i+31:i], imm8[1:0], imm8[3:2]) - ELSE - dst[i+31:i] := src[i+31:i] - FI -ENDFOR -dst[MAX:512] := 0 +dst[127:0] := INTERLEAVE_HIGH_WORDS(a[127:0], b[127:0]) - -
immintrin.h
+ +
emmintrin.h
 - - Floating Point - AVX512DQ - Miscellaneous - - - - - Calculate the max, min, absolute max, or absolute min (depending on control in "imm8") for packed single-precision (32-bit) floating-point elements in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - imm8[1:0] specifies the operation control: 00 = min, 01 = max, 10 = absolute max, 11 = absolute min. - imm8[3:2] specifies the sign control: 00 = sign from a, 01 = sign from compare result, 10 = clear sign bit, 11 = set sign bit. - + + Integer + SSE2 + Swizzle + + + + Unpack and interleave 32-bit integers from the high half of "a" and "b", and store the results in "dst". -RANGE(src1[31:0], src2[31:0], opCtl[1:0], signSelCtl[1:0]) -{ - CASE opCtl[1:0] - 0: tmp[31:0] := (src1[31:0] <= src2[31:0]) ? src1[31:0] : src2[31:0] - 1: tmp[31:0] := (src1[31:0] <= src2[31:0]) ? src2[31:0] : src1[31:0] - 2: tmp[31:0] := (ABS(src1[31:0]) <= ABS(src2[31:0])) ? src1[31:0] : src2[31:0] - 3: tmp[31:0] := (ABS(src1[31:0]) <= ABS(src2[31:0])) ? src2[31:0] : src1[31:0] - ESAC - - CASE signSelCtl[1:0] - 0: dst[31:0] := (src1[31] << 31) OR (tmp[30:0]) - 1: dst[31:0] := tmp[63:0] - 2: dst[31:0] := (0 << 31) OR (tmp[30:0]) - 3: dst[31:0] := (1 << 31) OR (tmp[30:0]) - ESAC - - RETURN dst +DEFINE INTERLEAVE_HIGH_DWORDS(src1[127:0], src2[127:0]) { + dst[31:0] := src1[95:64] + dst[63:32] := src2[95:64] + dst[95:64] := src1[127:96] + dst[127:96] := src2[127:96] + RETURN dst[127:0] } - -FOR j := 0 to 15 - i := j*32 - IF k[j] - dst[i+31:i] := RANGE(a[i+31:i], b[i+31:i], imm8[1:0], imm8[3:2]) - ELSE - dst[i+31:i] := 0 - FI -ENDFOR -dst[MAX:512] := 0 +dst[127:0] := INTERLEAVE_HIGH_DWORDS(a[127:0], b[127:0]) - -
immintrin.h
+ +
emmintrin.h
 - - Floating Point - AVX512DQ - Miscellaneous - - - - - - Calculate the max, min, absolute max, or absolute min (depending on control in "imm8") for packed single-precision (32-bit) floating-point elements in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - imm8[1:0] specifies the operation control: 00 = min, 01 = max, 10 = absolute max, 11 = absolute min. - imm8[3:2] specifies the sign control: 00 = sign from a, 01 = sign from compare result, 10 = clear sign bit, 11 = set sign bit. - [round_note] - + + Integer + SSE2 + Swizzle + + + + Unpack and interleave 64-bit integers from the high half of "a" and "b", and store the results in "dst". -RANGE(src1[31:0], src2[31:0], opCtl[1:0], signSelCtl[1:0]) -{ - CASE opCtl[1:0] - 0: tmp[31:0] := (src1[31:0] <= src2[31:0]) ? src1[31:0] : src2[31:0] - 1: tmp[31:0] := (src1[31:0] <= src2[31:0]) ? src2[31:0] : src1[31:0] - 2: tmp[31:0] := (ABS(src1[31:0]) <= ABS(src2[31:0])) ? src1[31:0] : src2[31:0] - 3: tmp[31:0] := (ABS(src1[31:0]) <= ABS(src2[31:0])) ? src2[31:0] : src1[31:0] - ESAC - - CASE signSelCtl[1:0] - 0: dst[31:0] := (src1[31] << 31) OR (tmp[30:0]) - 1: dst[31:0] := tmp[63:0] - 2: dst[31:0] := (0 << 31) OR (tmp[30:0]) - 3: dst[31:0] := (1 << 31) OR (tmp[30:0]) - ESAC - - RETURN dst +DEFINE INTERLEAVE_HIGH_QWORDS(src1[127:0], src2[127:0]) { + dst[63:0] := src1[127:64] + dst[127:64] := src2[127:64] + RETURN dst[127:0] } - -FOR j := 0 to 15 - i := j*32 - IF k[j] - dst[i+31:i] := RANGE(a[i+31:i], b[i+31:i], imm8[1:0], imm8[3:2]) - ELSE - dst[i+31:i] := 0 - FI -ENDFOR -dst[MAX:512] := 0 +dst[127:0] := INTERLEAVE_HIGH_QWORDS(a[127:0], b[127:0]) - -
immintrin.h
+ +
emmintrin.h
 - - Floating Point - AVX512DQ - Miscellaneous - - - - Calculate the max, min, absolute max, or absolute min (depending on control in "imm8") for packed single-precision (32-bit) floating-point elements in "a" and "b", and store the results in "dst". - imm8[1:0] specifies the operation control: 00 = min, 01 = max, 10 = absolute max, 11 = absolute min. - imm8[3:2] specifies the sign control: 00 = sign from a, 01 = sign from compare result, 10 = clear sign bit, 11 = set sign bit. - + + Integer + SSE2 + Swizzle + + + + Unpack and interleave 8-bit integers from the low half of "a" and "b", and store the results in "dst". -RANGE(src1[31:0], src2[31:0], opCtl[1:0], signSelCtl[1:0]) -{ - CASE opCtl[1:0] - 0: tmp[31:0] := (src1[31:0] <= src2[31:0]) ? src1[31:0] : src2[31:0] - 1: tmp[31:0] := (src1[31:0] <= src2[31:0]) ? src2[31:0] : src1[31:0] - 2: tmp[31:0] := (ABS(src1[31:0]) <= ABS(src2[31:0])) ? src1[31:0] : src2[31:0] - 3: tmp[31:0] := (ABS(src1[31:0]) <= ABS(src2[31:0])) ? src2[31:0] : src1[31:0] - ESAC - - CASE signSelCtl[1:0] - 0: dst[31:0] := (src1[31] << 31) OR (tmp[30:0]) - 1: dst[31:0] := tmp[63:0] - 2: dst[31:0] := (0 << 31) OR (tmp[30:0]) - 3: dst[31:0] := (1 << 31) OR (tmp[30:0]) - ESAC - - RETURN dst +DEFINE INTERLEAVE_BYTES(src1[127:0], src2[127:0]) { + dst[7:0] := src1[7:0] + dst[15:8] := src2[7:0] + dst[23:16] := src1[15:8] + dst[31:24] := src2[15:8] + dst[39:32] := src1[23:16] + dst[47:40] := src2[23:16] + dst[55:48] := src1[31:24] + dst[63:56] := src2[31:24] + dst[71:64] := src1[39:32] + dst[79:72] := src2[39:32] + dst[87:80] := src1[47:40] + dst[95:88] := src2[47:40] + dst[103:96] := src1[55:48] + dst[111:104] := src2[55:48] + dst[119:112] := src1[63:56] + dst[127:120] := src2[63:56] + RETURN dst[127:0] } - -FOR j := 0 to 15 - i := j*32 - dst[i+31:i] := RANGE(a[i+31:i], b[i+31:i], imm8[1:0], imm8[3:2]) -ENDFOR -dst[MAX:512] := 0 +dst[127:0] := INTERLEAVE_BYTES(a[127:0], b[127:0]) - -
immintrin.h
+ +
emmintrin.h
 - - Floating Point - AVX512DQ - Miscellaneous - - - - - Calculate the max, min, absolute max, or absolute min (depending on control in "imm8") for packed single-precision (32-bit) floating-point elements in "a" and "b", and store the results in "dst". - imm8[1:0] specifies the operation control: 00 = min, 01 = max, 10 = absolute max, 11 = absolute min. - imm8[3:2] specifies the sign control: 00 = sign from a, 01 = sign from compare result, 10 = clear sign bit, 11 = set sign bit. - [round_note] - + + Integer + SSE2 + Swizzle + + + + Unpack and interleave 16-bit integers from the low half of "a" and "b", and store the results in "dst". -RANGE(src1[31:0], src2[31:0], opCtl[1:0], signSelCtl[1:0]) -{ - CASE opCtl[1:0] - 0: tmp[31:0] := (src1[31:0] <= src2[31:0]) ? src1[31:0] : src2[31:0] - 1: tmp[31:0] := (src1[31:0] <= src2[31:0]) ? src2[31:0] : src1[31:0] - 2: tmp[31:0] := (ABS(src1[31:0]) <= ABS(src2[31:0])) ? src1[31:0] : src2[31:0] - 3: tmp[31:0] := (ABS(src1[31:0]) <= ABS(src2[31:0])) ? src2[31:0] : src1[31:0] - ESAC - - CASE signSelCtl[1:0] - 0: dst[31:0] := (src1[31] << 31) OR (tmp[30:0]) - 1: dst[31:0] := tmp[63:0] - 2: dst[31:0] := (0 << 31) OR (tmp[30:0]) - 3: dst[31:0] := (1 << 31) OR (tmp[30:0]) - ESAC - - RETURN dst +DEFINE INTERLEAVE_WORDS(src1[127:0], src2[127:0]) { + dst[15:0] := src1[15:0] + dst[31:16] := src2[15:0] + dst[47:32] := src1[31:16] + dst[63:48] := src2[31:16] + dst[79:64] := src1[47:32] + dst[95:80] := src2[47:32] + dst[111:96] := src1[63:48] + dst[127:112] := src2[63:48] + RETURN dst[127:0] } - -FOR j := 0 to 15 - i := j*32 - dst[i+31:i] := RANGE(a[i+31:i], b[i+31:i], imm8[1:0], imm8[3:2]) -ENDFOR -dst[MAX:512] := 0 +dst[127:0] := INTERLEAVE_WORDS(a[127:0], b[127:0]) - -
immintrin.h
+ +
emmintrin.h
 - - Floating Point - AVX512VL - AVX512DQ - Miscellaneous - - - - - - Calculate the max, min, absolute max, or absolute min (depending on control in "imm8") for packed single-precision (32-bit) floating-point elements in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - imm8[1:0] specifies the operation control: 00 = min, 01 = max, 10 = absolute max, 11 = absolute min. - imm8[3:2] specifies the sign control: 00 = sign from a, 01 = sign from compare result, 10 = clear sign bit, 11 = set sign bit. - + + Integer + SSE2 + Swizzle + + + + Unpack and interleave 32-bit integers from the low half of "a" and "b", and store the results in "dst". -RANGE(src1[31:0], src2[31:0], opCtl[1:0], signSelCtl[1:0]) -{ - CASE opCtl[1:0] - 0: tmp[31:0] := (src1[31:0] <= src2[31:0]) ? src1[31:0] : src2[31:0] - 1: tmp[31:0] := (src1[31:0] <= src2[31:0]) ? src2[31:0] : src1[31:0] - 2: tmp[31:0] := (ABS(src1[31:0]) <= ABS(src2[31:0])) ? src1[31:0] : src2[31:0] - 3: tmp[31:0] := (ABS(src1[31:0]) <= ABS(src2[31:0])) ? src2[31:0] : src1[31:0] - ESAC - - CASE signSelCtl[1:0] - 0: dst[31:0] := (src1[31] << 31) OR (tmp[30:0]) - 1: dst[31:0] := tmp[63:0] - 2: dst[31:0] := (0 << 31) OR (tmp[30:0]) - 3: dst[31:0] := (1 << 31) OR (tmp[30:0]) - ESAC - - RETURN dst +DEFINE INTERLEAVE_DWORDS(src1[127:0], src2[127:0]) { + dst[31:0] := src1[31:0] + dst[63:32] := src2[31:0] + dst[95:64] := src1[63:32] + dst[127:96] := src2[63:32] + RETURN dst[127:0] } - -FOR j := 0 to 3 - i := j*32 - IF k[j] - dst[i+31:i] := RANGE(a[i+31:i], b[i+31:i], imm8[1:0], imm8[3:2]) - ELSE - dst[i+31:i] := src[i+31:i] - FI -ENDFOR -dst[MAX:128] := 0 +dst[127:0] := INTERLEAVE_DWORDS(a[127:0], b[127:0]) - -
immintrin.h
+ +
emmintrin.h
 - - Floating Point - AVX512VL - AVX512DQ - Miscellaneous - - - - - Calculate the max, min, absolute max, or absolute min (depending on control in "imm8") for packed single-precision (32-bit) floating-point elements in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - imm8[1:0] specifies the operation control: 00 = min, 01 = max, 10 = absolute max, 11 = absolute min. - imm8[3:2] specifies the sign control: 00 = sign from a, 01 = sign from compare result, 10 = clear sign bit, 11 = set sign bit. - + + Integer + SSE2 + Swizzle + + + + Unpack and interleave 64-bit integers from the low half of "a" and "b", and store the results in "dst". -RANGE(src1[31:0], src2[31:0], opCtl[1:0], signSelCtl[1:0]) -{ - CASE opCtl[1:0] - 0: tmp[31:0] := (src1[31:0] <= src2[31:0]) ? src1[31:0] : src2[31:0] - 1: tmp[31:0] := (src1[31:0] <= src2[31:0]) ? src2[31:0] : src1[31:0] - 2: tmp[31:0] := (ABS(src1[31:0]) <= ABS(src2[31:0])) ? src1[31:0] : src2[31:0] - 3: tmp[31:0] := (ABS(src1[31:0]) <= ABS(src2[31:0])) ? src2[31:0] : src1[31:0] - ESAC - - CASE signSelCtl[1:0] - 0: dst[31:0] := (src1[31] << 31) OR (tmp[30:0]) - 1: dst[31:0] := tmp[63:0] - 2: dst[31:0] := (0 << 31) OR (tmp[30:0]) - 3: dst[31:0] := (1 << 31) OR (tmp[30:0]) - ESAC - - RETURN dst +DEFINE INTERLEAVE_QWORDS(src1[127:0], src2[127:0]) { + dst[63:0] := src1[63:0] + dst[127:64] := src2[63:0] + RETURN dst[127:0] } - -FOR j := 0 to 3 - i := j*32 - IF k[j] - dst[i+31:i] := RANGE(a[i+31:i], b[i+31:i], imm8[1:0], imm8[3:2]) - ELSE - dst[i+31:i] := 0 - FI -ENDFOR -dst[MAX:128] := 0 +dst[127:0] := INTERLEAVE_QWORDS(a[127:0], b[127:0]) - -
immintrin.h
+ +
emmintrin.h
 - + Floating Point - AVX512VL - AVX512DQ - Miscellaneous - - - - Calculate the max, min, absolute max, or absolute min (depending on control in "imm8") for packed single-precision (32-bit) floating-point elements in "a" and "b", and store the results in "dst". - imm8[1:0] specifies the operation control: 00 = min, 01 = max, 10 = absolute max, 11 = absolute min. - imm8[3:2] specifies the sign control: 00 = sign from a, 01 = sign from compare result, 10 = clear sign bit, 11 = set sign bit. - + SSE2 + Arithmetic + + + + Add the lower double-precision (64-bit) floating-point element in "a" and "b", store the result in the lower element of "dst", and copy the upper element from "a" to the upper element of "dst". -RANGE(src1[31:0], src2[31:0], opCtl[1:0], signSelCtl[1:0]) -{ - CASE opCtl[1:0] - 0: tmp[31:0] := (src1[31:0] <= src2[31:0]) ? src1[31:0] : src2[31:0] - 1: tmp[31:0] := (src1[31:0] <= src2[31:0]) ? src2[31:0] : src1[31:0] - 2: tmp[31:0] := (ABS(src1[31:0]) <= ABS(src2[31:0])) ? src1[31:0] : src2[31:0] - 3: tmp[31:0] := (ABS(src1[31:0]) <= ABS(src2[31:0])) ? src2[31:0] : src1[31:0] - ESAC - - CASE signSelCtl[1:0] - 0: dst[31:0] := (src1[31] << 31) OR (tmp[30:0]) - 1: dst[31:0] := tmp[63:0] - 2: dst[31:0] := (0 << 31) OR (tmp[30:0]) - 3: dst[31:0] := (1 << 31) OR (tmp[30:0]) - ESAC - - RETURN dst -} - -FOR j := 0 to 3 - i := j*32 - dst[i+31:i] := RANGE(a[i+31:i], b[i+31:i], imm8[1:0], imm8[3:2]) +dst[63:0] := a[63:0] + b[63:0] +dst[127:64] := a[127:64] + + +
emmintrin.h
+ + + Floating Point + SSE2 + Arithmetic + + + + Add packed double-precision (64-bit) floating-point elements in "a" and "b", and store the results in "dst". + +FOR j := 0 to 1 + i := j*64 + dst[i+63:i] := a[i+63:i] + b[i+63:i] ENDFOR -dst[MAX:128] := 0 - -
immintrin.h
+ +
emmintrin.h
 - + Floating Point - AVX512DQ - Miscellaneous - - - - - - - Calculate the max, min, absolute max, or absolute min (depending on control in "imm8") for the lower double-precision (64-bit) floating-point element in "a" and "b", store the result in the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". - imm8[1:0] specifies the operation control: 00 = min, 01 = max, 10 = absolute max, 11 = absolute min. - imm8[3:2] specifies the sign control: 00 = sign from a, 01 = sign from compare result, 10 = clear sign bit, 11 = set sign bit. - [round_note] - + SSE2 + Arithmetic + + + + Divide the lower double-precision (64-bit) floating-point element in "a" by the lower double-precision (64-bit) floating-point element in "b", store the result in the lower element of "dst", and copy the upper element from "a" to the upper element of "dst". -RANGE(src1[63:0], src2[63:0], opCtl[1:0], signSelCtl[1:0]) -{ - CASE opCtl[1:0] - 0: tmp[63:0] := (src1[63:0] <= src2[63:0]) ? src1[63:0] : src2[63:0] - 1: tmp[63:0] := (src1[63:0] <= src2[63:0]) ? src2[63:0] : src1[63:0] - 2: tmp[63:0] := (ABS(src1[63:0]) <= ABS(src2[63:0])) ? src1[63:0] : src2[63:0] - 3: tmp[63:0] := (ABS(src1[63:0]) <= ABS(src2[63:0])) ? src2[63:0] : src1[63:0] - ESAC - - CASE signSelCtl[1:0] - 0: dst[63:0] := (src1[63] << 63) OR (tmp[62:0]) - 1: dst[63:0] := tmp[63:0] - 2: dst[63:0] := (0 << 63) OR (tmp[62:0]) - 3: dst[63:0] := (1 << 63) OR (tmp[62:0]) - ESAC - - RETURN dst -} - -IF k[0] - dst[63:0]] := RANGE(a[63:0], b[63:0], imm8[1:0], imm8[3:2]) -ELSE - dst[63:0] := src[63:0] -FI +dst[63:0] := a[63:0] / b[63:0] dst[127:64] := a[127:64] -dst[MAX:128] := 0 - -
immintrin.h
+ +
emmintrin.h
 - + Floating Point - AVX512DQ - Miscellaneous - - - - - - Calculate the max, min, absolute max, or absolute min (depending on control in "imm8") for the lower double-precision (64-bit) floating-point element in "a" and "b", store the result in the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". - imm8[1:0] specifies the operation control: 00 = min, 01 = max, 10 = absolute max, 11 = absolute min. - imm8[3:2] specifies the sign control: 00 = sign from a, 01 = sign from compare result, 10 = clear sign bit, 11 = set sign bit. - + SSE2 + Arithmetic + + + + Divide packed double-precision (64-bit) floating-point elements in "a" by packed elements in "b", and store the results in "dst". -RANGE(src1[63:0], src2[63:0], opCtl[1:0], signSelCtl[1:0]) -{ - CASE opCtl[1:0] - 0: tmp[63:0] := (src1[63:0] <= src2[63:0]) ? src1[63:0] : src2[63:0] - 1: tmp[63:0] := (src1[63:0] <= src2[63:0]) ? src2[63:0] : src1[63:0] - 2: tmp[63:0] := (ABS(src1[63:0]) <= ABS(src2[63:0])) ? src1[63:0] : src2[63:0] - 3: tmp[63:0] := (ABS(src1[63:0]) <= ABS(src2[63:0])) ? src2[63:0] : src1[63:0] - ESAC - - CASE signSelCtl[1:0] - 0: dst[63:0] := (src1[63] << 63) OR (tmp[62:0]) - 1: dst[63:0] := tmp[63:0] - 2: dst[63:0] := (0 << 63) OR (tmp[62:0]) - 3: dst[63:0] := (1 << 63) OR (tmp[62:0]) - ESAC - - RETURN dst -} - -IF k[0] - dst[63:0]] := RANGE(a[63:0], b[63:0], imm8[1:0], imm8[3:2]) -ELSE - dst[63:0] := src[63:0] -FI -dst[127:64] := a[127:64] -dst[MAX:128] := 0 +FOR j := 0 to 1 + i := 64*j + dst[i+63:i] := a[i+63:i] / b[i+63:i] +ENDFOR - -
immintrin.h
+ +
emmintrin.h
 - + Floating Point - AVX512DQ - Miscellaneous - - - - - - Calculate the max, min, absolute max, or absolute min (depending on control in "imm8") for the lower double-precision (64-bit) floating-point element in "a" and "b", store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". - imm8[1:0] specifies the operation control: 00 = min, 01 = max, 10 = absolute max, 11 = absolute min. - imm8[3:2] specifies the sign control: 00 = sign from a, 01 = sign from compare result, 10 = clear sign bit, 11 = set sign bit. - [round_note] - + SSE2 + Special Math Functions + + + + Compare the lower double-precision (64-bit) floating-point elements in "a" and "b", store the maximum value in the lower element of "dst", and copy the upper element from "a" to the upper element of "dst". -RANGE(src1[63:0], src2[63:0], opCtl[1:0], signSelCtl[1:0]) -{ - CASE opCtl[1:0] - 0: tmp[63:0] := (src1[63:0] <= src2[63:0]) ? src1[63:0] : src2[63:0] - 1: tmp[63:0] := (src1[63:0] <= src2[63:0]) ? src2[63:0] : src1[63:0] - 2: tmp[63:0] := (ABS(src1[63:0]) <= ABS(src2[63:0])) ? src1[63:0] : src2[63:0] - 3: tmp[63:0] := (ABS(src1[63:0]) <= ABS(src2[63:0])) ? src2[63:0] : src1[63:0] - ESAC - - CASE signSelCtl[1:0] - 0: dst[63:0] := (src1[63] << 63) OR (tmp[62:0]) - 1: dst[63:0] := tmp[63:0] - 2: dst[63:0] := (0 << 63) OR (tmp[62:0]) - 3: dst[63:0] := (1 << 63) OR (tmp[62:0]) - ESAC - - RETURN dst -} - -IF k[0] - dst[63:0]] := RANGE(a[63:0], b[63:0], imm8[1:0], imm8[3:2]) -ELSE - dst[63:0] := 0 -FI +dst[63:0] := MAX(a[63:0], b[63:0]) dst[127:64] := a[127:64] -dst[MAX:128] := 0 - -
immintrin.h
+ +
emmintrin.h
 - + Floating Point - AVX512DQ - Miscellaneous - - - - - Calculate the max, min, absolute max, or absolute min (depending on control in "imm8") for the lower double-precision (64-bit) floating-point element in "a" and "b", store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper element from "a" to the upper element of "dst". - imm8[1:0] specifies the operation control: 00 = min, 01 = max, 10 = absolute max, 11 = absolute min. - imm8[3:2] specifies the sign control: 00 = sign from a, 01 = sign from compare result, 10 = clear sign bit, 11 = set sign bit. - + SSE2 + Special Math Functions + + + + Compare packed double-precision (64-bit) floating-point elements in "a" and "b", and store packed maximum values in "dst". -RANGE(src1[63:0], src2[63:0], opCtl[1:0], signSelCtl[1:0]) -{ - CASE opCtl[1:0] - 0: tmp[63:0] := (src1[63:0] <= src2[63:0]) ? src1[63:0] : src2[63:0] - 1: tmp[63:0] := (src1[63:0] <= src2[63:0]) ? src2[63:0] : src1[63:0] - 2: tmp[63:0] := (ABS(src1[63:0]) <= ABS(src2[63:0])) ? src1[63:0] : src2[63:0] - 3: tmp[63:0] := (ABS(src1[63:0]) <= ABS(src2[63:0])) ? src2[63:0] : src1[63:0] - ESAC - - CASE signSelCtl[1:0] - 0: dst[63:0] := (src1[63] << 63) OR (tmp[62:0]) - 1: dst[63:0] := tmp[63:0] - 2: dst[63:0] := (0 << 63) OR (tmp[62:0]) - 3: dst[63:0] := (1 << 63) OR (tmp[62:0]) - ESAC - - RETURN dst -} - -IF k[0] - dst[63:0]] := RANGE(a[63:0], b[63:0], imm8[1:0], imm8[3:2]) -ELSE - dst[63:0] := 0 -FI -dst[127:64] := a[127:64] -dst[MAX:128] := 0 +FOR j := 0 to 1 + i := j*64 + dst[i+63:i] := MAX(a[i+63:i], b[i+63:i]) +ENDFOR - -
immintrin.h
+ +
emmintrin.h
 - + Floating Point - AVX512DQ - Miscellaneous - - - - - Calculate the max, min, absolute max, or absolute min (depending on control in "imm8") for the lower double-precision (64-bit) floating-point element in "a" and "b", store the result in the lower element of "dst", and copy the upper element from "a" to the upper element of "dst". - imm8[1:0] specifies the operation control: 00 = min, 01 = max, 10 = absolute max, 11 = absolute min. - imm8[3:2] specifies the sign control: 00 = sign from a, 01 = sign from compare result, 10 = clear sign bit, 11 = set sign bit. - [round_note] - + SSE2 + Special Math Functions + + + + Compare the lower double-precision (64-bit) floating-point elements in "a" and "b", store the minimum value in the lower element of "dst", and copy the upper element from "a" to the upper element of "dst". -RANGE(src1[63:0], src2[63:0], opCtl[1:0], signSelCtl[1:0]) -{ - CASE opCtl[1:0] - 0: tmp[63:0] := (src1[63:0] <= src2[63:0]) ? src1[63:0] : src2[63:0] - 1: tmp[63:0] := (src1[63:0] <= src2[63:0]) ? src2[63:0] : src1[63:0] - 2: tmp[63:0] := (ABS(src1[63:0]) <= ABS(src2[63:0])) ? src1[63:0] : src2[63:0] - 3: tmp[63:0] := (ABS(src1[63:0]) <= ABS(src2[63:0])) ? src2[63:0] : src1[63:0] - ESAC - - CASE signSelCtl[1:0] - 0: dst[63:0] := (src1[63] << 63) OR (tmp[62:0]) - 1: dst[63:0] := tmp[63:0] - 2: dst[63:0] := (0 << 63) OR (tmp[62:0]) - 3: dst[63:0] := (1 << 63) OR (tmp[62:0]) - ESAC - - RETURN dst -} - -dst[63:0]] := RANGE(a[63:0], b[63:0], imm8[1:0], imm8[3:2]) +dst[63:0] := MIN(a[63:0], b[63:0]) dst[127:64] := a[127:64] -dst[MAX:128] := 0 - -
immintrin.h
+ +
emmintrin.h
 - + Floating Point - AVX512DQ - Miscellaneous - - - - - - - Calculate the max, min, absolute max, or absolute min (depending on control in "imm8") for the lower single-precision (32-bit) floating-point element in "a" and "b", store the result in the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". - imm8[1:0] specifies the operation control: 00 = min, 01 = max, 10 = absolute max, 11 = absolute min. - imm8[3:2] specifies the sign control: 00 = sign from a, 01 = sign from compare result, 10 = clear sign bit, 11 = set sign bit. - [round_note] - + SSE2 + Special Math Functions + + + + Compare packed double-precision (64-bit) floating-point elements in "a" and "b", and store packed minimum values in "dst". -RANGE(src1[31:0], src2[31:0], opCtl[1:0], signSelCtl[1:0]) -{ - CASE opCtl[1:0] - 0: tmp[31:0] := (src1[31:0] <= src2[31:0]) ? src1[31:0] : src2[31:0] - 1: tmp[31:0] := (src1[31:0] <= src2[31:0]) ? src2[31:0] : src1[31:0] - 2: tmp[31:0] := (ABS(src1[31:0]) <= ABS(src2[31:0])) ? src1[31:0] : src2[31:0] - 3: tmp[31:0] := (ABS(src1[31:0]) <= ABS(src2[31:0])) ? src2[31:0] : src1[31:0] - ESAC - - CASE signSelCtl[1:0] - 0: dst[31:0] := (src1[31] << 31) OR (tmp[30:0]) - 1: dst[31:0] := tmp[31:0] - 2: dst[31:0] := (0 << 31) OR (tmp[30:0]) - 3: dst[31:0] := (1 << 31) OR (tmp[30:0]) - ESAC - - RETURN dst -} - -IF k[0] - dst[31:0]] := RANGE(a[31:0], b[31:0], imm8[1:0], imm8[3:2]) -ELSE - dst[31:0] := src[31:0] -FI -dst[127:32] := a[127:32] -dst[MAX:128] := 0 +FOR j := 0 to 1 + i := j*64 + dst[i+63:i] := MIN(a[i+63:i], b[i+63:i]) +ENDFOR - -
immintrin.h
+ +
emmintrin.h
 - + Floating Point - AVX512DQ - Miscellaneous - - - - - - Calculate the max, min, absolute max, or absolute min (depending on control in "imm8") for the lower single-precision (32-bit) floating-point element in "a" and "b", store the result in the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". - imm8[1:0] specifies the operation control: 00 = min, 01 = max, 10 = absolute max, 11 = absolute min. - imm8[3:2] specifies the sign control: 00 = sign from a, 01 = sign from compare result, 10 = clear sign bit, 11 = set sign bit. - + SSE2 + Arithmetic + + + + Multiply the lower double-precision (64-bit) floating-point element in "a" and "b", store the result in the lower element of "dst", and copy the upper element from "a" to the upper element of "dst". -RANGE(src1[31:0], src2[31:0], opCtl[1:0], signSelCtl[1:0]) -{ - CASE opCtl[1:0] - 0: tmp[31:0] := (src1[31:0] <= src2[31:0]) ? src1[31:0] : src2[31:0] - 1: tmp[31:0] := (src1[31:0] <= src2[31:0]) ? src2[31:0] : src1[31:0] - 2: tmp[31:0] := (ABS(src1[31:0]) <= ABS(src2[31:0])) ? src1[31:0] : src2[31:0] - 3: tmp[31:0] := (ABS(src1[31:0]) <= ABS(src2[31:0])) ? src2[31:0] : src1[31:0] - ESAC - - CASE signSelCtl[1:0] - 0: dst[31:0] := (src1[31] << 31) OR (tmp[30:0]) - 1: dst[31:0] := tmp[31:0] - 2: dst[31:0] := (0 << 31) OR (tmp[30:0]) - 3: dst[31:0] := (1 << 31) OR (tmp[30:0]) - ESAC - - RETURN dst -} - -IF k[0] - dst[31:0]] := RANGE(a[31:0], b[31:0], imm8[1:0], imm8[3:2]) -ELSE - dst[31:0] := src[31:0] -FI -dst[127:32] := a[127:32] -dst[MAX:128] := 0 +dst[63:0] := a[63:0] * b[63:0] +dst[127:64] := a[127:64] - -
immintrin.h
+ +
emmintrin.h
 - + Floating Point - AVX512DQ - Miscellaneous - - - - - - Calculate the max, min, absolute max, or absolute min (depending on control in "imm8") for the lower single-precision (32-bit) floating-point element in "a" and "b", store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". - imm8[1:0] specifies the operation control: 00 = min, 01 = max, 10 = absolute max, 11 = absolute min. - imm8[3:2] specifies the sign control: 00 = sign from a, 01 = sign from compare result, 10 = clear sign bit, 11 = set sign bit. - [round_note] - + SSE2 + Arithmetic + + + + Multiply packed double-precision (64-bit) floating-point elements in "a" and "b", and store the results in "dst". -RANGE(src1[31:0], src2[31:0], opCtl[1:0], signSelCtl[1:0]) -{ - CASE opCtl[1:0] - 0: tmp[31:0] := (src1[31:0] <= src2[31:0]) ? src1[31:0] : src2[31:0] - 1: tmp[31:0] := (src1[31:0] <= src2[31:0]) ? src2[31:0] : src1[31:0] - 2: tmp[31:0] := (ABS(src1[31:0]) <= ABS(src2[31:0])) ? src1[31:0] : src2[31:0] - 3: tmp[31:0] := (ABS(src1[31:0]) <= ABS(src2[31:0])) ? src2[31:0] : src1[31:0] - ESAC - - CASE signSelCtl[1:0] - 0: dst[31:0] := (src1[31] << 31) OR (tmp[30:0]) - 1: dst[31:0] := tmp[31:0] - 2: dst[31:0] := (0 << 31) OR (tmp[30:0]) - 3: dst[31:0] := (1 << 31) OR (tmp[30:0]) - ESAC - - RETURN dst -} - -IF k[0] - dst[31:0]] := RANGE(a[31:0], b[31:0], imm8[1:0], imm8[3:2]) -ELSE - dst[31:0] := 0 -FI -dst[127:32] := a[127:32] -dst[MAX:128] := 0 +FOR j := 0 to 1 + i := j*64 + dst[i+63:i] := a[i+63:i] * b[i+63:i] +ENDFOR - -
immintrin.h
+ +
emmintrin.h
 - + Floating Point - AVX512DQ - Miscellaneous - - - - - Calculate the max, min, absolute max, or absolute min (depending on control in "imm8") for the lower single-precision (32-bit) floating-point element in "a" and "b", store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper 3 packed elements from "a" to the upper elements of "dst". - imm8[1:0] specifies the operation control: 00 = min, 01 = max, 10 = absolute max, 11 = absolute min. - imm8[3:2] specifies the sign control: 00 = sign from a, 01 = sign from compare result, 10 = clear sign bit, 11 = set sign bit. - + SSE2 + Elementary Math Functions + + + + Compute the square root of the lower double-precision (64-bit) floating-point element in "b", store the result in the lower element of "dst", and copy the upper element from "a" to the upper element of "dst". -RANGE(src1[31:0], src2[31:0], opCtl[1:0], signSelCtl[1:0]) -{ - CASE opCtl[1:0] - 0: tmp[31:0] := (src1[31:0] <= src2[31:0]) ? src1[31:0] : src2[31:0] - 1: tmp[31:0] := (src1[31:0] <= src2[31:0]) ? src2[31:0] : src1[31:0] - 2: tmp[31:0] := (ABS(src1[31:0]) <= ABS(src2[31:0])) ? src1[31:0] : src2[31:0] - 3: tmp[31:0] := (ABS(src1[31:0]) <= ABS(src2[31:0])) ? src2[31:0] : src1[31:0] - ESAC - - CASE signSelCtl[1:0] - 0: dst[31:0] := (src1[31] << 31) OR (tmp[30:0]) - 1: dst[31:0] := tmp[31:0] - 2: dst[31:0] := (0 << 31) OR (tmp[30:0]) - 3: dst[31:0] := (1 << 31) OR (tmp[30:0]) - ESAC - - RETURN dst -} - -IF k[0] - dst[31:0]] := RANGE(a[31:0], b[31:0], imm8[1:0], imm8[3:2]) -ELSE - dst[31:0] := 0 -FI -dst[127:32] := a[127:32] -dst[MAX:128] := 0 +dst[63:0] := SQRT(b[63:0]) +dst[127:64] := a[127:64] - -
immintrin.h
+ +
emmintrin.h
 - + Floating Point - AVX512DQ - Miscellaneous - - - - - Calculate the max, min, absolute max, or absolute min (depending on control in "imm8") for the lower single-precision (32-bit) floating-point element in "a" and "b", store the result in the lower element of "dst", and copy the upper 3 packed elements from "a" to the upper elements of "dst". - imm8[1:0] specifies the operation control: 00 = min, 01 = max, 10 = absolute max, 11 = absolute min. - imm8[3:2] specifies the sign control: 00 = sign from a, 01 = sign from compare result, 10 = clear sign bit, 11 = set sign bit. - [round_note] - + SSE2 + Elementary Math Functions + + + Compute the square root of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst". -RANGE(src1[31:0], src2[31:0], opCtl[1:0], signSelCtl[1:0]) -{ - CASE opCtl[1:0] - 0: tmp[31:0] := (src1[31:0] <= src2[31:0]) ? src1[31:0] : src2[31:0] - 1: tmp[31:0] := (src1[31:0] <= src2[31:0]) ? src2[31:0] : src1[31:0] - 2: tmp[31:0] := (ABS(src1[31:0]) <= ABS(src2[31:0])) ? src1[31:0] : src2[31:0] - 3: tmp[31:0] := (ABS(src1[31:0]) <= ABS(src2[31:0])) ? src2[31:0] : src1[31:0] - ESAC - - CASE signSelCtl[1:0] - 0: dst[31:0] := (src1[31] << 31) OR (tmp[30:0]) - 1: dst[31:0] := tmp[31:0] - 2: dst[31:0] := (0 << 31) OR (tmp[30:0]) - 3: dst[31:0] := (1 << 31) OR (tmp[30:0]) - ESAC - - RETURN dst -} - -dst[31:0]] := RANGE(a[31:0], b[31:0], imm8[1:0], imm8[3:2]) -dst[127:32] := a[127:32] -dst[MAX:128] := 0 +FOR j := 0 to 1 + i := j*64 + dst[i+63:i] := SQRT(a[i+63:i]) +ENDFOR - -
immintrin.h
+ +
emmintrin.h
 - + Floating Point - AVX512VL - AVX512F + SSE2 Arithmetic - - - - Compute the approximate reciprocal of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). The maximum relative error for this approximation is less than 2^-14. + + + + Subtract the lower double-precision (64-bit) floating-point element in "b" from the lower double-precision (64-bit) floating-point element in "a", store the result in the lower element of "dst", and copy the upper element from "a" to the upper element of "dst". -FOR j := 0 to 3 - i := j*64 - IF k[j] - dst[i+63:i] := APPROXIMATE(1.0/a[i+63:i]) - ELSE - dst[i+63:i] := src[i+63:i] - FI -ENDFOR -dst[MAX:256] := 0 +dst[63:0] := a[63:0] - b[63:0] +dst[127:64] := a[127:64] - -
immintrin.h
+ +
emmintrin.h
 - + Floating Point - AVX512VL - AVX512F + SSE2 Arithmetic - - - Compute the approximate reciprocal of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). The maximum relative error for this approximation is less than 2^-14. + + + + Subtract packed double-precision (64-bit) floating-point elements in "b" from packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst". -FOR j := 0 to 3 +FOR j := 0 to 1 i := j*64 - IF k[j] - dst[i+63:i] := APPROXIMATE(1.0/a[i+63:i]) - ELSE - dst[i+63:i] := 0 - FI + dst[i+63:i] := a[i+63:i] - b[i+63:i] ENDFOR -dst[MAX:256] := 0 - -
immintrin.h
+ +
emmintrin.h
 - + Floating Point - AVX512VL - AVX512F - Arithmetic - - Compute the approximate reciprocal of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst". The maximum relative error for this approximation is less than 2^-14. + SSE2 + Logical + + + + Compute the bitwise AND of packed double-precision (64-bit) floating-point elements in "a" and "b", and store the results in "dst". -FOR j := 0 to 3 +FOR j := 0 to 1 i := j*64 - dst[i+63:i] := APPROXIMATE(1.0/a[i+63:i]) + dst[i+63:i] := (a[i+63:i] AND b[i+63:i]) ENDFOR -dst[MAX:256] := 0 - -
immintrin.h
+ +
emmintrin.h
 - + Floating Point - AVX512VL - AVX512F - Arithmetic - - - - Compute the approximate reciprocal of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). The maximum relative error for this approximation is less than 2^-14. + SSE2 + Logical + + + + Compute the bitwise NOT of packed double-precision (64-bit) floating-point elements in "a" and then AND with "b", and store the results in "dst". FOR j := 0 to 1 i := j*64 - IF k[j] - dst[i+63:i] := APPROXIMATE(1.0/a[i+63:i]) - ELSE - dst[i+63:i] := src[i+63:i] - FI + dst[i+63:i] := ((NOT a[i+63:i]) AND b[i+63:i]) ENDFOR -dst[MAX:128] := 0 - -
immintrin.h
+ +
emmintrin.h
 - + Floating Point - AVX512VL - AVX512F - Arithmetic - - - Compute the approximate reciprocal of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). The maximum relative error for this approximation is less than 2^-14. + SSE2 + Logical + + + + Compute the bitwise OR of packed double-precision (64-bit) floating-point elements in "a" and "b", and store the results in "dst". FOR j := 0 to 1 i := j*64 - IF k[j] - dst[i+63:i] := APPROXIMATE(1.0/a[i+63:i]) - ELSE - dst[i+63:i] := 0 - FI + dst[i+63:i] := a[i+63:i] OR b[i+63:i] ENDFOR -dst[MAX:128] := 0 - -
immintrin.h
+ +
emmintrin.h
 - + Floating Point - AVX512VL - AVX512F - Arithmetic - - Compute the approximate reciprocal of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst". The maximum relative error for this approximation is less than 2^-14. + SSE2 + Logical + + + + Compute the bitwise XOR of packed double-precision (64-bit) floating-point elements in "a" and "b", and store the results in "dst". FOR j := 0 to 1 i := j*64 - dst[i+63:i] := APPROXIMATE(1.0/a[i+63:i]) + dst[i+63:i] := a[i+63:i] XOR b[i+63:i] ENDFOR -dst[MAX:128] := 0 - -
immintrin.h
+ +
emmintrin.h
 - + Floating Point - AVX512VL - AVX512F - Arithmetic - - - - Compute the approximate reciprocal of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). The maximum relative error for this approximation is less than 2^-14. + SSE2 + Compare + + + + Compare the lower double-precision (64-bit) floating-point elements in "a" and "b" for equality, store the result in the lower element of "dst", and copy the upper element from "a" to the upper element of "dst". -FOR j := 0 to 7 - i := j*32 - IF k[j] - dst[i+31:i] := APPROXIMATE(1.0/a[i+31:i]) - ELSE - dst[i+31:i] := src[i+31:i] - FI -ENDFOR -dst[MAX:256] := 0 +dst[63:0] := (a[63:0] == b[63:0]) ? 0xFFFFFFFFFFFFFFFF : 0 +dst[127:64] := a[127:64] + + +
emmintrin.h
+ + + Floating Point + SSE2 + Compare + + + + Compare the lower double-precision (64-bit) floating-point elements in "a" and "b" for less-than, store the result in the lower element of "dst", and copy the upper element from "a" to the upper element of "dst". + +dst[63:0] := (a[63:0] < b[63:0]) ? 0xFFFFFFFFFFFFFFFF : 0 +dst[127:64] := a[127:64] + + +
emmintrin.h
+ + + Floating Point + SSE2 + Compare + + + + Compare the lower double-precision (64-bit) floating-point elements in "a" and "b" for less-than-or-equal, store the result in the lower element of "dst", and copy the upper element from "a" to the upper element of "dst". + +dst[63:0] := (a[63:0] <= b[63:0]) ? 0xFFFFFFFFFFFFFFFF : 0 +dst[127:64] := a[127:64] + + +
emmintrin.h
+ + + Floating Point + SSE2 + Compare + + + + Compare the lower double-precision (64-bit) floating-point elements in "a" and "b" for greater-than, store the result in the lower element of "dst", and copy the upper element from "a" to the upper element of "dst". + +dst[63:0] := (a[63:0] > b[63:0]) ? 0xFFFFFFFFFFFFFFFF : 0 +dst[127:64] := a[127:64] + + +
emmintrin.h
+ + + Floating Point + SSE2 + Compare + + + + Compare the lower double-precision (64-bit) floating-point elements in "a" and "b" for greater-than-or-equal, store the result in the lower element of "dst", and copy the upper element from "a" to the upper element of "dst". + +dst[63:0] := (a[63:0] >= b[63:0]) ? 0xFFFFFFFFFFFFFFFF : 0 +dst[127:64] := a[127:64] + + +
emmintrin.h
+ + + Floating Point + SSE2 + Compare + + + + Compare the lower double-precision (64-bit) floating-point elements in "a" and "b" to see if neither is NaN, store the result in the lower element of "dst", and copy the upper element from "a" to the upper element of "dst". + dst[63:0] := (a[63:0] != NaN AND b[63:0] != NaN) ? 0xFFFFFFFFFFFFFFFF : 0 +dst[127:64] := a[127:64] + + +
emmintrin.h
+ + + Floating Point + SSE2 + Compare + + + + Compare the lower double-precision (64-bit) floating-point elements in "a" and "b" to see if either is NaN, store the result in the lower element of "dst", and copy the upper element from "a" to the upper element of "dst". + dst[63:0] := (a[63:0] == NaN OR b[63:0] == NaN) ? 0xFFFFFFFFFFFFFFFF : 0 +dst[127:64] := a[127:64] - -
immintrin.h
+ +
emmintrin.h
 - + Floating Point - AVX512VL - AVX512F - Arithmetic - - - Compute the approximate reciprocal of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). The maximum relative error for this approximation is less than 2^-14. + SSE2 + Compare + + + + Compare the lower double-precision (64-bit) floating-point elements in "a" and "b" for not-equal, store the result in the lower element of "dst", and copy the upper element from "a" to the upper element of "dst". -FOR j := 0 to 7 - i := j*32 - IF k[j] - dst[i+31:i] := APPROXIMATE(1.0/a[i+31:i]) - ELSE - dst[i+31:i] := 0 - FI -ENDFOR -dst[MAX:256] := 0 +dst[63:0] := (a[63:0] != b[63:0]) ? 0xFFFFFFFFFFFFFFFF : 0 +dst[127:64] := a[127:64] - -
immintrin.h
+ +
emmintrin.h
 - + Floating Point - AVX512VL - AVX512F - Arithmetic - - Compute the approximate reciprocal of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst". The maximum relative error for this approximation is less than 2^-14. + SSE2 + Compare + + + + Compare the lower double-precision (64-bit) floating-point elements in "a" and "b" for not-less-than, store the result in the lower element of "dst", and copy the upper element from "a" to the upper element of "dst". -FOR j := 0 to 7 - i := j*32 - dst[i+31:i] := APPROXIMATE(1.0/a[i+31:i]) -ENDFOR -dst[MAX:256] := 0 +dst[63:0] := (!(a[63:0] < b[63:0])) ? 0xFFFFFFFFFFFFFFFF : 0 +dst[127:64] := a[127:64] - -
immintrin.h
+ +
emmintrin.h
 - + Floating Point - AVX512VL - AVX512F - Arithmetic - - - - Compute the approximate reciprocal of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). The maximum relative error for this approximation is less than 2^-14. + SSE2 + Compare + + + + Compare the lower double-precision (64-bit) floating-point elements in "a" and "b" for not-less-than-or-equal, store the result in the lower element of "dst", and copy the upper element from "a" to the upper element of "dst". -FOR j := 0 to 3 - i := j*32 - IF k[j] - dst[i+31:i] := APPROXIMATE(1.0/a[i+31:i]) - ELSE - dst[i+31:i] := src[i+31:i] - FI -ENDFOR -dst[MAX:128] := 0 +dst[63:0] := (!(a[63:0] <= b[63:0])) ? 0xFFFFFFFFFFFFFFFF : 0 +dst[127:64] := a[127:64] - -
immintrin.h
+ +
emmintrin.h
 - + Floating Point - AVX512VL - AVX512F - Arithmetic - - - Compute the approximate reciprocal of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). The maximum relative error for this approximation is less than 2^-14. + SSE2 + Compare + + + + Compare the lower double-precision (64-bit) floating-point elements in "a" and "b" for not-greater-than, store the result in the lower element of "dst", and copy the upper element from "a" to the upper element of "dst". -FOR j := 0 to 3 - i := j*32 - IF k[j] - dst[i+31:i] := APPROXIMATE(1.0/a[i+31:i]) - ELSE - dst[i+31:i] := 0 - FI -ENDFOR -dst[MAX:128] := 0 +dst[63:0] := (!(a[63:0] > b[63:0])) ? 0xFFFFFFFFFFFFFFFF : 0 +dst[127:64] := a[127:64] - -
immintrin.h
+ +
emmintrin.h
 - + Floating Point - AVX512VL - AVX512F - Arithmetic - - Compute the approximate reciprocal of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst". The maximum relative error for this approximation is less than 2^-14. + SSE2 + Compare + + + + Compare the lower double-precision (64-bit) floating-point elements in "a" and "b" for not-greater-than-or-equal, store the result in the lower element of "dst", and copy the upper element from "a" to the upper element of "dst". -FOR j := 0 to 3 - i := j*32 - dst[i+31:i] := APPROXIMATE(1.0/a[i+31:i]) -ENDFOR -dst[MAX:128] := 0 +dst[63:0] := (!(a[63:0] >= b[63:0])) ? 0xFFFFFFFFFFFFFFFF : 0 +dst[127:64] := a[127:64] - -
immintrin.h
+ +
emmintrin.h
 - + Floating Point - AVX512VL - AVX512DQ - Miscellaneous - - - - - Extract the reduced argument of packed double-precision (64-bit) floating-point elements in "a" by the number of bits specified by "imm8", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -ReduceArgumentPD(src1[63:0], imm8[7:0]) -{ - m := imm8[7:4] // number of fraction bits after the binary point to be preserved - rc := imm8[1:0] // round control - rc_src := imm8[2] // round ccontrol source - spe := 0 - tmp[63:0] := pow(2, -m) * ROUND(pow(2, m) * src1[63:0], spe, rc_source, rc) - tmp[63:0] := src1[63:0] - tmp[63:0] - RETURN tmp[63:0] -} - -FOR j := 0 to 3 + SSE2 + Compare + + + + Compare packed double-precision (64-bit) floating-point elements in "a" and "b" for equality, and store the results in "dst". + +FOR j := 0 to 1 i := j*64 - IF k[j] - dst[i+63:i] := ReduceArgumentPD(src[i+63:i], imm8[7:0]) - ELSE - dst[i+63:i] := src[i+63:i] - FI + dst[i+63:i] := (a[i+63:i] == b[i+63:i]) ? 0xFFFFFFFFFFFFFFFF : 0 ENDFOR -dst[MAX:256] := 0 - -
immintrin.h
+ +
emmintrin.h
 - + Floating Point - AVX512VL - AVX512DQ - Miscellaneous - - - - Extract the reduced argument of packed double-precision (64-bit) floating-point elements in "a" by the number of bits specified by "imm8", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - -ReduceArgumentPD(src1[63:0], imm8[7:0]) -{ - m := imm8[7:4] // number of fraction bits after the binary point to be preserved - rc := imm8[1:0] // round control - rc_src := imm8[2] // round ccontrol source - spe := 0 - tmp[63:0] := pow(2, -m) * ROUND(pow(2, m) * src1[63:0], spe, rc_source, rc) - tmp[63:0] := src1[63:0] - tmp[63:0] - RETURN tmp[63:0] -} - -FOR j := 0 to 3 + SSE2 + Compare + + + + Compare packed double-precision (64-bit) floating-point elements in "a" and "b" for less-than, and store the results in "dst". + +FOR j := 0 to 1 i := j*64 - IF k[j] - dst[i+63:i] := ReduceArgumentPD(src[i+63:i], imm8[7:0]) - ELSE - dst[i+63:i] := 0 - FI + dst[i+63:i] := (a[i+63:i] < b[i+63:i]) ? 0xFFFFFFFFFFFFFFFF : 0 ENDFOR -dst[MAX:256] := 0 - -
immintrin.h
+ +
emmintrin.h
 - + Floating Point - AVX512VL - AVX512DQ - Miscellaneous - - - Extract the reduced argument of packed double-precision (64-bit) floating-point elements in "a" by the number of bits specified by "imm8", and store the results in "dst". - -ReduceArgumentPD(src1[63:0], imm8[7:0]) -{ - m := imm8[7:4] // number of fraction bits after the binary point to be preserved - rc := imm8[1:0] // round control - rc_src := imm8[2] // round ccontrol source - spe := 0 - tmp[63:0] := pow(2, -m) * ROUND(pow(2, m) * src1[63:0], spe, rc_source, rc) - tmp[63:0] := src1[63:0] - tmp[63:0] - RETURN tmp[63:0] -} - -FOR j := 0 to 3 + SSE2 + Compare + + + + Compare packed double-precision (64-bit) floating-point elements in "a" and "b" for less-than-or-equal, and store the results in "dst". + +FOR j := 0 to 1 i := j*64 - dst[i+63:i] := ReduceArgumentPD(src[i+63:i], imm8[7:0]) + dst[i+63:i] := (a[i+63:i] <= b[i+63:i]) ? 0xFFFFFFFFFFFFFFFF : 0 ENDFOR -dst[MAX:256] := 0 - -
immintrin.h
+ +
emmintrin.h
 - + Floating Point - AVX512DQ - Miscellaneous - - - - - Extract the reduced argument of packed double-precision (64-bit) floating-point elements in "a" by the number of bits specified by "imm8", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -ReduceArgumentPD(src1[63:0], imm8[7:0]) -{ - m := imm8[7:4] // number of fraction bits after the binary point to be preserved - rc := imm8[1:0] // round control - rc_src := imm8[2] // round ccontrol source - spe := 0 - tmp[63:0] := pow(2, -m) * ROUND(pow(2, m) * src1[63:0], spe, rc_source, rc) - tmp[63:0] := src1[63:0] - tmp[63:0] - RETURN tmp[63:0] -} - -FOR j := 0 to 7 + SSE2 + Compare + + + + Compare packed double-precision (64-bit) floating-point elements in "a" and "b" for greater-than, and store the results in "dst". + +FOR j := 0 to 1 i := j*64 - IF k[j] - dst[i+63:i] := ReduceArgumentPD(src[i+63:i], imm8[7:0]) - ELSE - dst[i+63:i] := src[i+63:i] - FI + dst[i+63:i] := (a[i+63:i] > b[i+63:i]) ? 0xFFFFFFFFFFFFFFFF : 0 ENDFOR -dst[MAX:512] := 0 - -
immintrin.h
+ +
emmintrin.h
 - + Floating Point - AVX512DQ - Miscellaneous - - - - - - Extract the reduced argument of packed double-precision (64-bit) floating-point elements in "a" by the number of bits specified by "imm8", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - [round_note] + SSE2 + Compare + + + + Compare packed double-precision (64-bit) floating-point elements in "a" and "b" for greater-than-or-equal, and store the results in "dst". -ReduceArgumentPD(src1[63:0], imm8[7:0]) -{ - m := imm8[7:4] // number of fraction bits after the binary point to be preserved - rc := imm8[1:0] // round control - rc_src := imm8[2] // round ccontrol source - spe := 0 - tmp[63:0] := pow(2, -m) * ROUND(pow(2, m) * src1[63:0], spe, rc_source, rc) - tmp[63:0] := src1[63:0] - tmp[63:0] - RETURN tmp[63:0] -} - -FOR j := 0 to 7 +FOR j := 0 to 1 i := j*64 - IF k[j] - dst[i+63:i] := ReduceArgumentPD(src[i+63:i], imm8[7:0]) - ELSE - dst[i+63:i] := src[i+63:i] - FI + dst[i+63:i] := (a[i+63:i] >= b[i+63:i]) ? 0xFFFFFFFFFFFFFFFF : 0 ENDFOR -dst[MAX:512] := 0 - -
immintrin.h
+ +
emmintrin.h
 - + Floating Point - AVX512DQ - Miscellaneous - - - - Extract the reduced argument of packed double-precision (64-bit) floating-point elements in "a" by the number of bits specified by "imm8", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - -ReduceArgumentPD(src1[63:0], imm8[7:0]) -{ - m := imm8[7:4] // number of fraction bits after the binary point to be preserved - rc := imm8[1:0] // round control - rc_src := imm8[2] // round ccontrol source - spe := 0 - tmp[63:0] := pow(2, -m) * ROUND(pow(2, m) * src1[63:0], spe, rc_source, rc) - tmp[63:0] := src1[63:0] - tmp[63:0] - RETURN tmp[63:0] -} - -FOR j := 0 to 7 + SSE2 + Compare + + + + Compare packed double-precision (64-bit) floating-point elements in "a" and "b" to see if neither is NaN, and store the results in "dst". + FOR j := 0 to 1 i := j*64 - IF k[j] - dst[i+63:i] := ReduceArgumentPD(src[i+63:i], imm8[7:0]) - ELSE - dst[i+63:i] := 0 - FI + dst[i+63:i] := (a[i+63:i] != NaN AND b[i+63:i] != NaN) ? 0xFFFFFFFFFFFFFFFF : 0 ENDFOR -dst[MAX:512] := 0 - -
immintrin.h
+ +
emmintrin.h
 - + Floating Point - AVX512DQ - Miscellaneous - - - - - Extract the reduced argument of packed double-precision (64-bit) floating-point elements in "a" by the number of bits specified by "imm8", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - [round_note] - -ReduceArgumentPD(src1[63:0], imm8[7:0]) -{ - m := imm8[7:4] // number of fraction bits after the binary point to be preserved - rc := imm8[1:0] // round control - rc_src := imm8[2] // round ccontrol source - spe := 0 - tmp[63:0] := pow(2, -m) * ROUND(pow(2, m) * src1[63:0], spe, rc_source, rc) - tmp[63:0] := src1[63:0] - tmp[63:0] - RETURN tmp[63:0] -} - -FOR j := 0 to 7 + SSE2 + Compare + + + + Compare packed double-precision (64-bit) floating-point elements in "a" and "b" to see if either is NaN, and store the results in "dst". + FOR j := 0 to 1 i := j*64 - IF k[j] - dst[i+63:i] := ReduceArgumentPD(src[i+63:i], imm8[7:0]) - ELSE - dst[i+63:i] := 0 - FI + dst[i+63:i] := (a[i+63:i] == NaN OR b[i+63:i] == NaN) ? 0xFFFFFFFFFFFFFFFF : 0 ENDFOR -dst[MAX:512] := 0 - -
immintrin.h
+ +
emmintrin.h
 - + Floating Point - AVX512DQ - Miscellaneous - - - Extract the reduced argument of packed double-precision (64-bit) floating-point elements in "a" by the number of bits specified by "imm8", and store the results in "dst". - -ReduceArgumentPD(src1[63:0], imm8[7:0]) -{ - m := imm8[7:4] // number of fraction bits after the binary point to be preserved - rc := imm8[1:0] // round control - rc_src := imm8[2] // round ccontrol source - spe := 0 - tmp[63:0] := pow(2, -m) * ROUND(pow(2, m) * src1[63:0], spe, rc_source, rc) - tmp[63:0] := src1[63:0] - tmp[63:0] - RETURN tmp[63:0] -} - -FOR j := 0 to 7 + SSE2 + Compare + + + + Compare packed double-precision (64-bit) floating-point elements in "a" and "b" for not-equal, and store the results in "dst". + +FOR j := 0 to 1 i := j*64 - dst[i+63:i] := ReduceArgumentPD(src[i+63:i], imm8[7:0]) + dst[i+63:i] := (a[i+63:i] != b[i+63:i]) ? 0xFFFFFFFFFFFFFFFF : 0 ENDFOR -dst[MAX:512] := 0 - -
immintrin.h
+ +
emmintrin.h
 - + Floating Point - AVX512DQ - Miscellaneous - - - - Extract the reduced argument of packed double-precision (64-bit) floating-point elements in "a" by the number of bits specified by "imm8", and store the results in "dst". - [round_note] + SSE2 + Compare + + + + Compare packed double-precision (64-bit) floating-point elements in "a" and "b" for not-less-than, and store the results in "dst". -ReduceArgumentPD(src1[63:0], imm8[7:0]) -{ - m := imm8[7:4] // number of fraction bits after the binary point to be preserved - rc := imm8[1:0] // round control - rc_src := imm8[2] // round ccontrol source - spe := 0 - tmp[63:0] := pow(2, -m) * ROUND(pow(2, m) * src1[63:0], spe, rc_source, rc) - tmp[63:0] := src1[63:0] - tmp[63:0] - RETURN tmp[63:0] -} - -FOR j := 0 to 7 +FOR j := 0 to 1 i := j*64 - dst[i+63:i] := ReduceArgumentPD(src[i+63:i], imm8[7:0]) + dst[i+63:i] := (!(a[i+63:i] < b[i+63:i])) ? 0xFFFFFFFFFFFFFFFF : 0 ENDFOR -dst[MAX:512] := 0 - -
immintrin.h
+ +
emmintrin.h
 - + Floating Point - AVX512VL - AVX512DQ - Miscellaneous - - - - - Extract the reduced argument of packed double-precision (64-bit) floating-point elements in "a" by the number of bits specified by "imm8", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -ReduceArgumentPD(src1[63:0], imm8[7:0]) -{ - m := imm8[7:4] // number of fraction bits after the binary point to be preserved - rc := imm8[1:0] // round control - rc_src := imm8[2] // round ccontrol source - spe := 0 - tmp[63:0] := pow(2, -m) * ROUND(pow(2, m) * src1[63:0], spe, rc_source, rc) - tmp[63:0] := src1[63:0] - tmp[63:0] - RETURN tmp[63:0] -} - + SSE2 + Compare + + + + Compare packed double-precision (64-bit) floating-point elements in "a" and "b" for not-less-than-or-equal, and store the results in "dst". + FOR j := 0 to 1 i := j*64 - IF k[j] - dst[i+63:i] := ReduceArgumentPD(src[i+63:i], imm8[7:0]) - ELSE - dst[i+63:i] := src[i+63:i] - FI + dst[i+63:i] := (!(a[i+63:i] <= b[i+63:i])) ? 0xFFFFFFFFFFFFFFFF : 0 ENDFOR -dst[MAX:128] := 0 - -
immintrin.h
+ +
emmintrin.h
 - + Floating Point - AVX512VL - AVX512DQ - Miscellaneous - - - - Extract the reduced argument of packed double-precision (64-bit) floating-point elements in "a" by the number of bits specified by "imm8", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - -ReduceArgumentPD(src1[63:0], imm8[7:0]) -{ - m := imm8[7:4] // number of fraction bits after the binary point to be preserved - rc := imm8[1:0] // round control - rc_src := imm8[2] // round ccontrol source - spe := 0 - tmp[63:0] := pow(2, -m) * ROUND(pow(2, m) * src1[63:0], spe, rc_source, rc) - tmp[63:0] := src1[63:0] - tmp[63:0] - RETURN tmp[63:0] -} - + SSE2 + Compare + + + + Compare packed double-precision (64-bit) floating-point elements in "a" and "b" for not-greater-than, and store the results in "dst". + FOR j := 0 to 1 i := j*64 - IF k[j] - dst[i+63:i] := ReduceArgumentPD(src[i+63:i], imm8[7:0]) - ELSE - dst[i+63:i] := 0 - FI + dst[i+63:i] := (!(a[i+63:i] > b[i+63:i])) ? 0xFFFFFFFFFFFFFFFF : 0 ENDFOR -dst[MAX:128] := 0 - -
immintrin.h
+ +
emmintrin.h
 - + Floating Point - AVX512VL - AVX512DQ - Miscellaneous - - - Extract the reduced argument of packed double-precision (64-bit) floating-point elements in "a" by the number of bits specified by "imm8", and store the results in "dst". - -ReduceArgumentPD(src1[63:0], imm8[7:0]) -{ - m := imm8[7:4] // number of fraction bits after the binary point to be preserved - rc := imm8[1:0] // round control - rc_src := imm8[2] // round ccontrol source - spe := 0 - tmp[63:0] := pow(2, -m) * ROUND(pow(2, m) * src1[63:0], spe, rc_source, rc) - tmp[63:0] := src1[63:0] - tmp[63:0] - RETURN tmp[63:0] -} - + SSE2 + Compare + + + + Compare packed double-precision (64-bit) floating-point elements in "a" and "b" for not-greater-than-or-equal, and store the results in "dst". + FOR j := 0 to 1 i := j*64 - dst[i+63:i] := ReduceArgumentPD(src[i+63:i], imm8[7:0]) + dst[i+63:i] := (!(a[i+63:i] >= b[i+63:i])) ? 0xFFFFFFFFFFFFFFFF : 0 ENDFOR -dst[MAX:128] := 0 - -
immintrin.h
+ +
emmintrin.h
 - + Floating Point - AVX512VL - AVX512DQ - Miscellaneous - - - - - Extract the reduced argument of packed single-precision (32-bit) floating-point elements in "a" by the number of bits specified by "imm8", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Flag + SSE2 + Compare + + + + Compare the lower double-precision (64-bit) floating-point element in "a" and "b" for equality, and return the boolean result (0 or 1). -ReduceArgumentPS(src1[31:0], imm8[7:0]) -{ - IF src1[31:0] == NAN - RETURN (convert src1[31:0] to QNaN) - FI - - m := imm8[7:4] // number of fraction bits after the binary point to be preserved - rc := imm8[1:0] // round control - rc_src := imm8[2] // round ccontrol source - spe := 0 - tmp[31:0] := pow(2, -m)*ROUND(pow(2, m)*src1[31:0], spe, rc_source, rc) - tmp[31:0] := src1[31:0] - tmp[31:0] - RETURN tmp[31:0] -} -FOR j := 0 to 7 - i := j*32 - IF k[j] - dst[i+31:i] := ReduceArgumentPS(src[i+31:i], imm8[7:0]) - ELSE - dst[i+31:i] := src[i+31:i] - FI -ENDFOR -dst[MAX:256] := 0 +RETURN ( a[63:0] == b[63:0] ) ? 1 : 0 - -
immintrin.h
+ +
emmintrin.h
 - + Floating Point - AVX512VL - AVX512DQ - Miscellaneous - - - - Extract the reduced argument of packed single-precision (32-bit) floating-point elements in "a" by the number of bits specified by "imm8", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Flag + SSE2 + Compare + + + + Compare the lower double-precision (64-bit) floating-point element in "a" and "b" for less-than, and return the boolean result (0 or 1). -ReduceArgumentPS(src1[31:0], imm8[7:0]) -{ - IF src1[31:0] == NAN - RETURN (convert src1[31:0] to QNaN) - FI - - m := imm8[7:4] // number of fraction bits after the binary point to be preserved - rc := imm8[1:0] // round control - rc_src := imm8[2] // round ccontrol source - spe := 0 - tmp[31:0] := pow(2, -m)*ROUND(pow(2, m)*src1[31:0], spe, rc_source, rc) - tmp[31:0] := src1[31:0] - tmp[31:0] - RETURN tmp[31:0] -} -FOR j := 0 to 7 - i := j*32 - IF k[j] - dst[i+31:i] := ReduceArgumentPS(src[i+31:i], imm8[7:0]) - ELSE - dst[i+31:i] := 0 - FI -ENDFOR -dst[MAX:256] := 0 +RETURN ( a[63:0] < b[63:0] ) ? 1 : 0 - -
immintrin.h
+ +
emmintrin.h
 - + Floating Point - AVX512VL - AVX512DQ - Miscellaneous - - - Extract the reduced argument of packed single-precision (32-bit) floating-point elements in "a" by the number of bits specified by "imm8", and store the results in "dst". + Flag + SSE2 + Compare + + + + Compare the lower double-precision (64-bit) floating-point element in "a" and "b" for less-than-or-equal, and return the boolean result (0 or 1). -ReduceArgumentPS(src1[31:0], imm8[7:0]) -{ - IF src1[31:0] == NAN - RETURN (convert src1[31:0] to QNaN) - FI - - m := imm8[7:4] // number of fraction bits after the binary point to be preserved - rc := imm8[1:0] // round control - rc_src := imm8[2] // round ccontrol source - spe := 0 - tmp[31:0] := pow(2, -m)*ROUND(pow(2, m)*src1[31:0], spe, rc_source, rc) - tmp[31:0] := src1[31:0] - tmp[31:0] - RETURN tmp[31:0] -} -FOR j := 0 to 7 - i := j*32 - dst[i+31:i] := ReduceArgumentPS(src[i+31:i], imm8[7:0]) -ENDFOR -dst[MAX:256] := 0 +RETURN ( a[63:0] <= b[63:0] ) ? 1 : 0 - -
immintrin.h
+ +
emmintrin.h
 - + Floating Point - AVX512DQ - Miscellaneous - - - - - Extract the reduced argument of packed single-precision (32-bit) floating-point elements in "a" by the number of bits specified by "imm8", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Flag + SSE2 + Compare + + + + Compare the lower double-precision (64-bit) floating-point element in "a" and "b" for greater-than, and return the boolean result (0 or 1). -ReduceArgumentPS(src1[31:0], imm8[7:0]) -{ - IF src1[31:0] == NAN - RETURN (convert src1[31:0] to QNaN) - FI - - m := imm8[7:4] // number of fraction bits after the binary point to be preserved - rc := imm8[1:0] // round control - rc_src := imm8[2] // round ccontrol source - spe := 0 - tmp[31:0] := pow(2, -m)*ROUND(pow(2, m)*src1[31:0], spe, rc_source, rc) - tmp[31:0] := src1[31:0] - tmp[31:0] - RETURN tmp[31:0] -} -FOR j := 0 to 15 - i := j*32 - IF k[j] - dst[i+31:i] := ReduceArgumentPS(src[i+31:i], imm8[7:0]) - ELSE - dst[i+31:i] := src[i+31:i] - FI -ENDFOR -dst[MAX:512] := 0 +RETURN ( a[63:0] > b[63:0] ) ? 1 : 0 - -
immintrin.h
+ +
emmintrin.h
 - + Floating Point - AVX512DQ - Miscellaneous - - - - - - Extract the reduced argument of packed single-precision (32-bit) floating-point elements in "a" by the number of bits specified by "imm8", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - [round_note] + Flag + SSE2 + Compare + + + + Compare the lower double-precision (64-bit) floating-point element in "a" and "b" for greater-than-or-equal, and return the boolean result (0 or 1). -ReduceArgumentPS(src1[31:0], imm8[7:0]) -{ - IF src1[31:0] == NAN - RETURN (convert src1[31:0] to QNaN) - FI - - m := imm8[7:4] // number of fraction bits after the binary point to be preserved - rc := imm8[1:0] // round control - rc_src := imm8[2] // round ccontrol source - spe := 0 - tmp[31:0] := pow(2, -m)*ROUND(pow(2, m)*src1[31:0], spe, rc_source, rc) - tmp[31:0] := src1[31:0] - tmp[31:0] - RETURN tmp[31:0] -} -FOR j := 0 to 15 - i := j*32 - IF k[j] - dst[i+31:i] := ReduceArgumentPS(src[i+31:i], imm8[7:0]) - ELSE - dst[i+31:i] := src[i+31:i] - FI -ENDFOR -dst[MAX:512] := 0 +RETURN ( a[63:0] >= b[63:0] ) ? 1 : 0 - -
immintrin.h
+ +
emmintrin.h
 - + Floating Point - AVX512DQ - Miscellaneous - - - - Extract the reduced argument of packed single-precision (32-bit) floating-point elements in "a" by the number of bits specified by "imm8", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Flag + SSE2 + Compare + + + + Compare the lower double-precision (64-bit) floating-point element in "a" and "b" for not-equal, and return the boolean result (0 or 1). -ReduceArgumentPS(src1[31:0], imm8[7:0]) -{ - IF src1[31:0] == NAN - RETURN (convert src1[31:0] to QNaN) - FI - - m := imm8[7:4] // number of fraction bits after the binary point to be preserved - rc := imm8[1:0] // round control - rc_src := imm8[2] // round ccontrol source - spe := 0 - tmp[31:0] := pow(2, -m)*ROUND(pow(2, m)*src1[31:0], spe, rc_source, rc) - tmp[31:0] := src1[31:0] - tmp[31:0] - RETURN tmp[31:0] -} -FOR j := 0 to 15 - i := j*32 - IF k[j] - dst[i+31:i] := ReduceArgumentPS(src[i+31:i], imm8[7:0]) - ELSE - dst[i+31:i] := 0 - FI -ENDFOR -dst[MAX:512] := 0 +RETURN ( a[63:0] != b[63:0] ) ? 1 : 0 - -
immintrin.h
+ +
emmintrin.h
 - + Floating Point - AVX512DQ - Miscellaneous - - - - - Extract the reduced argument of packed single-precision (32-bit) floating-point elements in "a" by the number of bits specified by "imm8", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - [round_note] + Flag + SSE2 + Compare + + + + Compare the lower double-precision (64-bit) floating-point element in "a" and "b" for equality, and return the boolean result (0 or 1). This instruction will not signal an exception for QNaNs. -ReduceArgumentPS(src1[31:0], imm8[7:0]) -{ - IF src1[31:0] == NAN - RETURN (convert src1[31:0] to QNaN) - FI - - m := imm8[7:4] // number of fraction bits after the binary point to be preserved - rc := imm8[1:0] // round control - rc_src := imm8[2] // round ccontrol source - spe := 0 - tmp[31:0] := pow(2, -m)*ROUND(pow(2, m)*src1[31:0], spe, rc_source, rc) - tmp[31:0] := src1[31:0] - tmp[31:0] - RETURN tmp[31:0] -} -FOR j := 0 to 15 - i := j*32 - IF k[j] - dst[i+31:i] := ReduceArgumentPS(src[i+31:i], imm8[7:0]) - ELSE - dst[i+31:i] := 0 - FI -ENDFOR -dst[MAX:512] := 0 +RETURN ( a[63:0] == b[63:0] ) ? 1 : 0 - -
immintrin.h
+ +
emmintrin.h
 - + Floating Point - AVX512DQ - Miscellaneous - - - Extract the reduced argument of packed single-precision (32-bit) floating-point elements in "a" by the number of bits specified by "imm8", and store the results in "dst". + Flag + SSE2 + Compare + + + + Compare the lower double-precision (64-bit) floating-point element in "a" and "b" for less-than, and return the boolean result (0 or 1). This instruction will not signal an exception for QNaNs. -ReduceArgumentPS(src1[31:0], imm8[7:0]) -{ - IF src1[31:0] == NAN - RETURN (convert src1[31:0] to QNaN) - FI - - m := imm8[7:4] // number of fraction bits after the binary point to be preserved - rc := imm8[1:0] // round control - rc_src := imm8[2] // round ccontrol source - spe := 0 - tmp[31:0] := pow(2, -m)*ROUND(pow(2, m)*src1[31:0], spe, rc_source, rc) - tmp[31:0] := src1[31:0] - tmp[31:0] - RETURN tmp[31:0] -} -FOR j := 0 to 15 - i := j*32 - dst[i+31:i] := ReduceArgumentPS(src[i+31:i], imm8[7:0]) -ENDFOR -dst[MAX:512] := 0 +RETURN ( a[63:0] < b[63:0] ) ? 1 : 0 - -
immintrin.h
+ +
emmintrin.h
 - + Floating Point - AVX512DQ - Miscellaneous - - - - Extract the reduced argument of packed single-precision (32-bit) floating-point elements in "a" by the number of bits specified by "imm8", and store the results in "dst". - [round_note] + Flag + SSE2 + Compare + + + + Compare the lower double-precision (64-bit) floating-point element in "a" and "b" for less-than-or-equal, and return the boolean result (0 or 1). This instruction will not signal an exception for QNaNs. -ReduceArgumentPS(src1[31:0], imm8[7:0]) -{ - IF src1[31:0] == NAN - RETURN (convert src1[31:0] to QNaN) - FI - - m := imm8[7:4] // number of fraction bits after the binary point to be preserved - rc := imm8[1:0] // round control - rc_src := imm8[2] // round ccontrol source - spe := 0 - tmp[31:0] := pow(2, -m)*ROUND(pow(2, m)*src1[31:0], spe, rc_source, rc) - tmp[31:0] := src1[31:0] - tmp[31:0] - RETURN tmp[31:0] -} -FOR j := 0 to 15 - i := j*32 - dst[i+31:i] := ReduceArgumentPS(src[i+31:i], imm8[7:0]) -ENDFOR -dst[MAX:512] := 0 +RETURN ( a[63:0] <= b[63:0] ) ? 1 : 0 - -
immintrin.h
+ +
emmintrin.h
 - + Floating Point - AVX512VL - AVX512DQ - Miscellaneous - - - - - Extract the reduced argument of packed single-precision (32-bit) floating-point elements in "a" by the number of bits specified by "imm8", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Flag + SSE2 + Compare + + + + Compare the lower double-precision (64-bit) floating-point element in "a" and "b" for greater-than, and return the boolean result (0 or 1). This instruction will not signal an exception for QNaNs. -ReduceArgumentPS(src1[31:0], imm8[7:0]) -{ - IF src1[31:0] == NAN - RETURN (convert src1[31:0] to QNaN) - FI - - m := imm8[7:4] // number of fraction bits after the binary point to be preserved - rc := imm8[1:0] // round control - rc_src := imm8[2] // round ccontrol source - spe := 0 - tmp[31:0] := pow(2, -m)*ROUND(pow(2, m)*src1[31:0], spe, rc_source, rc) - tmp[31:0] := src1[31:0] - tmp[31:0] - RETURN tmp[31:0] -} -FOR j := 0 to 3 - i := j*32 - IF k[j] - dst[i+31:i] := ReduceArgumentPS(src[i+31:i], imm8[7:0]) - ELSE - dst[i+31:i] := src[i+31:i] - FI -ENDFOR -dst[MAX:128] := 0 +RETURN ( a[63:0] > b[63:0] ) ? 1 : 0 - -
immintrin.h
+ +
emmintrin.h
 - + Floating Point - AVX512VL - AVX512DQ - Miscellaneous - - - - Extract the reduced argument of packed single-precision (32-bit) floating-point elements in "a" by the number of bits specified by "imm8", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Flag + SSE2 + Compare + + + + Compare the lower double-precision (64-bit) floating-point element in "a" and "b" for greater-than-or-equal, and return the boolean result (0 or 1). This instruction will not signal an exception for QNaNs. + +RETURN ( a[63:0] >= b[63:0] ) ? 1 : 0 + + +
emmintrin.h
+ + + Floating Point + Flag + SSE2 + Compare + + + + Compare the lower double-precision (64-bit) floating-point element in "a" and "b" for not-equal, and return the boolean result (0 or 1). This instruction will not signal an exception for QNaNs. -ReduceArgumentPS(src1[31:0], imm8[7:0]) -{ - IF src1[31:0] == NAN - RETURN (convert src1[31:0] to QNaN) - FI - - m := imm8[7:4] // number of fraction bits after the binary point to be preserved - rc := imm8[1:0] // round control - rc_src := imm8[2] // round ccontrol source - spe := 0 - tmp[31:0] := pow(2, -m)*ROUND(pow(2, m)*src1[31:0], spe, rc_source, rc) - tmp[31:0] := src1[31:0] - tmp[31:0] - RETURN tmp[31:0] -} -FOR j := 0 to 3 - i := j*32 - IF k[j] - dst[i+31:i] := ReduceArgumentPS(src[i+31:i], imm8[7:0]) - ELSE - dst[i+31:i] := 0 - FI -ENDFOR -dst[MAX:128] := 0 +RETURN ( a[63:0] != b[63:0] ) ? 1 : 0 - -
immintrin.h
+ +
emmintrin.h
 - + Floating Point - AVX512VL - AVX512DQ - Miscellaneous - - - Extract the reduced argument of packed single-precision (32-bit) floating-point elements in "a" by the number of bits specified by "imm8", and store the results in "dst". + SSE2 + Convert + + + Convert packed double-precision (64-bit) floating-point elements in "a" to packed single-precision (32-bit) floating-point elements, and store the results in "dst". -ReduceArgumentPS(src1[31:0], imm8[7:0]) -{ - IF src1[31:0] == NAN - RETURN (convert src1[31:0] to QNaN) - FI - - m := imm8[7:4] // number of fraction bits after the binary point to be preserved - rc := imm8[1:0] // round control - rc_src := imm8[2] // round ccontrol source - spe := 0 - tmp[31:0] := pow(2, -m)*ROUND(pow(2, m)*src1[31:0], spe, rc_source, rc) - tmp[31:0] := src1[31:0] - tmp[31:0] - RETURN tmp[31:0] -} -FOR j := 0 to 3 - i := j*32 - dst[i+31:i] := ReduceArgumentPS(src[i+31:i], imm8[7:0]) +FOR j := 0 to 1 + i := 32*j + k := 64*j + dst[i+31:i] := Convert_FP64_To_FP32(a[k+63:k]) ENDFOR -dst[MAX:128] := 0 +dst[127:64] := 0 - -
immintrin.h
+ +
emmintrin.h
 - + Floating Point - AVX512DQ - Miscellaneous - - - - - - Extract the reduced argument of the lower double-precision (64-bit) floating-point element in "a" by the number of bits specified by "imm8", store the result in the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and copy the upper element from "b" to the upper element of "dst". - -ReduceArgumentPD(src1[63:0], imm8[7:0]) -{ - m := imm8[7:4] // number of fraction bits after the binary point to be preserved - rc := imm8[1:0] // round control - rc_src := imm8[2] // round ccontrol source - spe := 0 - tmp[63:0] := pow(2, -m) * ROUND(pow(2, m) * src1[63:0], spe, rc_source, rc) - tmp[63:0] := src1[63:0] - tmp[63:0] - RETURN tmp[63:0] -} - -IF k[0] - dst[63:0] := ReduceArgumentPD(a[63:0], imm8[7:0]) -ELSE - dst[63:0] := src[63:0] -FI -dst[127:64] := b[127:64] -dst[MAX:128] := 0 + SSE2 + Convert + + + Convert packed single-precision (32-bit) floating-point elements in "a" to packed double-precision (64-bit) floating-point elements, and store the results in "dst". + +FOR j := 0 to 1 + i := 64*j + k := 32*j + dst[i+63:i] := Convert_FP32_To_FP64(a[k+31:k]) +ENDFOR - -
immintrin.h
+ +
emmintrin.h
 - + Floating Point - AVX512DQ - Miscellaneous - - - - - - - Extract the reduced argument of the lower double-precision (64-bit) floating-point element in "a" by the number of bits specified by "imm8", store the result in the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and copy the upper element from "b" to the upper element of "dst". - [round_note] + Integer + SSE2 + Convert + + + Convert packed double-precision (64-bit) floating-point elements in "a" to packed 32-bit integers, and store the results in "dst". -ReduceArgumentPD(src1[63:0], imm8[7:0]) -{ - m := imm8[7:4] // number of fraction bits after the binary point to be preserved - rc := imm8[1:0] // round control - rc_src := imm8[2] // round ccontrol source - spe := 0 - tmp[63:0] := pow(2, -m) * ROUND(pow(2, m) * src1[63:0], spe, rc_source, rc) - tmp[63:0] := src1[63:0] - tmp[63:0] - RETURN tmp[63:0] -} - -IF k[0] - dst[63:0] := ReduceArgumentPD(a[63:0], imm8[7:0]) -ELSE - dst[63:0] := src[63:0] -FI -dst[127:64] := b[127:64] -dst[MAX:128] := 0 +FOR j := 0 to 1 + i := 32*j + k := 64*j + dst[i+31:i] := Convert_FP64_To_Int32(a[k+63:k]) +ENDFOR - -
immintrin.h
+ +
emmintrin.h
 - + Floating Point - AVX512DQ - Miscellaneous - - - - - Extract the reduced argument of the lower double-precision (64-bit) floating-point element in "a" by the number of bits specified by "imm8", store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper element from "b" to the upper element of "dst". - -ReduceArgumentPD(src1[63:0], imm8[7:0]) -{ - m := imm8[7:4] // number of fraction bits after the binary point to be preserved - rc := imm8[1:0] // round control - rc_src := imm8[2] // round ccontrol source - spe := 0 - tmp[63:0] := pow(2, -m) * ROUND(pow(2, m) * src1[63:0], spe, rc_source, rc) - tmp[63:0] := src1[63:0] - tmp[63:0] - RETURN tmp[63:0] -} - -IF k[0] - dst[63:0] := ReduceArgumentPD(a[63:0], imm8[7:0]) -ELSE - dst[63:0] := 0 -FI -dst[127:64] := b[127:64] -dst[MAX:128] := 0 + Integer + SSE2 + Convert + + + Convert the lower double-precision (64-bit) floating-point element in "a" to a 32-bit integer, and store the result in "dst". + +dst[31:0] := Convert_FP64_To_Int32(a[63:0]) - -
immintrin.h
+ +
emmintrin.h
 - + Floating Point - AVX512DQ - Miscellaneous - - - - - - Extract the reduced argument of the lower double-precision (64-bit) floating-point element in "a" by the number of bits specified by "imm8", store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper element from "b" to the upper element of "dst". - [round_note] + Integer + SSE2 + Convert + + + Convert the lower double-precision (64-bit) floating-point element in "a" to a 64-bit integer, and store the result in "dst". -ReduceArgumentPD(src1[63:0], imm8[7:0]) -{ - m := imm8[7:4] // number of fraction bits after the binary point to be preserved - rc := imm8[1:0] // round control - rc_src := imm8[2] // round ccontrol source - spe := 0 - tmp[63:0] := pow(2, -m) * ROUND(pow(2, m) * src1[63:0], spe, rc_source, rc) - tmp[63:0] := src1[63:0] - tmp[63:0] - RETURN tmp[63:0] -} - -IF k[0] - dst[63:0] := ReduceArgumentPD(a[63:0], imm8[7:0]) -ELSE - dst[63:0] := 0 -FI -dst[127:64] := b[127:64] -dst[MAX:128] := 0 +dst[63:0] := Convert_FP64_To_Int64(a[63:0]) - -
immintrin.h
+ +
emmintrin.h
 - + Floating Point - AVX512DQ - Miscellaneous - - - - Extract the reduced argument of the lower double-precision (64-bit) floating-point element in "a" by the number of bits specified by "imm8", store the result in the lower element of "dst", and copy the upper element from "b" to the upper element of "dst". - -ReduceArgumentPD(src1[63:0], imm8[7:0]) -{ - m := imm8[7:4] // number of fraction bits after the binary point to be preserved - rc := imm8[1:0] // round control - rc_src := imm8[2] // round ccontrol source - spe := 0 - tmp[63:0] := pow(2, -m) * ROUND(pow(2, m) * src1[63:0], spe, rc_source, rc) - tmp[63:0] := src1[63:0] - tmp[63:0] - RETURN tmp[63:0] -} - -dst[63:0] := ReduceArgumentPD(a[63:0], imm8[7:0]) -dst[127:64] := b[127:64] -dst[MAX:128] := 0 + Integer + SSE2 + Convert + + + Convert the lower double-precision (64-bit) floating-point element in "a" to a 64-bit integer, and store the result in "dst". + +dst[63:0] := Convert_FP64_To_Int64(a[63:0]) - -
immintrin.h
+ +
emmintrin.h
 - + Floating Point - AVX512DQ - Miscellaneous - - - - - Extract the reduced argument of the lower double-precision (64-bit) floating-point element in "a" by the number of bits specified by "imm8", store the result in the lower element of "dst", and copy the upper element from "b" to the upper element of "dst". - [round_note] + SSE2 + Convert + + + + Convert the lower double-precision (64-bit) floating-point element in "b" to a single-precision (32-bit) floating-point element, store the result in the lower element of "dst", and copy the upper 3 packed elements from "a" to the upper elements of "dst". -ReduceArgumentPD(src1[63:0], imm8[7:0]) -{ - m := imm8[7:4] // number of fraction bits after the binary point to be preserved - rc := imm8[1:0] // round control - rc_src := imm8[2] // round ccontrol source - spe := 0 - tmp[63:0] := pow(2, -m) * ROUND(pow(2, m) * src1[63:0], spe, rc_source, rc) - tmp[63:0] := src1[63:0] - tmp[63:0] - RETURN tmp[63:0] -} - -dst[63:0] := ReduceArgumentPD(a[63:0], imm8[7:0]) -dst[127:64] := b[127:64] +dst[31:0] := Convert_FP64_To_FP32(b[63:0]) +dst[127:32] := a[127:32] dst[MAX:128] := 0 - -
immintrin.h
+ +
emmintrin.h
 - + Floating Point - AVX512DQ - Miscellaneous - - - - - - Extract the reduced argument of the lower single-precision (32-bit) floating-point element in "a" by the number of bits specified by "imm8", store the result in the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and copy the upper 3 packed elements from "b" to the upper elements of "dst". + SSE2 + Convert + + + Copy the lower double-precision (64-bit) floating-point element of "a" to "dst". -ReduceArgumentPS(src1[31:0], imm8[7:0]) -{ - IF src1[31:0] == NAN - RETURN (convert src1[31:0] to QNaN) - FI - - m := imm8[7:4] // number of fraction bits after the binary point to be preserved - rc := imm8[1:0] // round control - rc_src := imm8[2] // round ccontrol source - spe := 0 - tmp[31:0] := pow(2, -m)*ROUND(pow(2, m)*src1[31:0], spe, rc_source, rc) - tmp[31:0] := src1[31:0] - tmp[31:0] - RETURN tmp[31:0] -} - -IF k[0] - dst[31:0] := ReduceArgumentPS(a[31:0], imm8[7:0]) -ELSE - dst[31:0] := src[31:0] -FI -dst[127:64] := b[127:32] -dst[MAX:128] := 0 +dst[63:0] := a[63:0] - -
immintrin.h
+ +
emmintrin.h
 - + Floating Point - AVX512DQ - Miscellaneous - - - - - - - Extract the reduced argument of the lower single-precision (32-bit) floating-point element in "a" by the number of bits specified by "imm8", store the result in the lower element of "dst" using writemask "k" (the element is copied from "src" when mask bit 0 is not set), and copy the upper 3 packed elements from "b" to the upper elements of "dst". - [round_note] + SSE2 + Convert + + + + Convert the lower single-precision (32-bit) floating-point element in "b" to a double-precision (64-bit) floating-point element, store the result in the lower element of "dst", and copy the upper element from "a" to the upper element of "dst". -ReduceArgumentPS(src1[31:0], imm8[7:0]) -{ - IF src1[31:0] == NAN - RETURN (convert src1[31:0] to QNaN) - FI - - m := imm8[7:4] // number of fraction bits after the binary point to be preserved - rc := imm8[1:0] // round control - rc_src := imm8[2] // round ccontrol source - spe := 0 - tmp[31:0] := pow(2, -m)*ROUND(pow(2, m)*src1[31:0], spe, rc_source, rc) - tmp[31:0] := src1[31:0] - tmp[31:0] - RETURN tmp[31:0] -} - -IF k[0] - dst[31:0] := ReduceArgumentPS(a[31:0], imm8[7:0]) -ELSE - dst[31:0] := src[31:0] -FI -dst[127:64] := b[127:32] +dst[63:0] := Convert_FP32_To_FP64(b[31:0]) +dst[127:64] := a[127:64] dst[MAX:128] := 0 - -
immintrin.h
+ +
emmintrin.h
 - + Floating Point - AVX512DQ - Miscellaneous - - - - - Extract the reduced argument of the lower single-precision (32-bit) floating-point element in "a" by the number of bits specified by "imm8", store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper 3 packed elements from "b" to the upper elements of "dst". + Integer + SSE2 + Convert + + + Convert packed double-precision (64-bit) floating-point elements in "a" to packed 32-bit integers with truncation, and store the results in "dst". -ReduceArgumentPS(src1[31:0], imm8[7:0]) -{ - IF src1[31:0] == NAN - RETURN (convert src1[31:0] to QNaN) - FI - - m := imm8[7:4] // number of fraction bits after the binary point to be preserved - rc := imm8[1:0] // round control - rc_src := imm8[2] // round ccontrol source - spe := 0 - tmp[31:0] := pow(2, -m)*ROUND(pow(2, m)*src1[31:0], spe, rc_source, rc) - tmp[31:0] := src1[31:0] - tmp[31:0] - RETURN tmp[31:0] -} - -IF k[0] - dst[31:0] := ReduceArgumentPS(a[31:0], imm8[7:0]) -ELSE - dst[31:0] := 0 -FI -dst[127:64] := b[127:32] -dst[MAX:128] := 0 +FOR j := 0 to 1 + i := 32*j + k := 64*j + dst[i+31:i] := Convert_FP64_To_Int32_Truncate(a[k+63:k]) +ENDFOR - -
immintrin.h
+ +
emmintrin.h
 - + Floating Point - AVX512DQ - Miscellaneous - - - - - - Extract the reduced argument of the lower single-precision (32-bit) floating-point element in "a" by the number of bits specified by "imm8", store the result in the lower element of "dst" using zeromask "k" (the element is zeroed out when mask bit 0 is not set), and copy the upper 3 packed elements from "b" to the upper elements of "dst". - [round_note] + Integer + SSE2 + Convert + + + Convert the lower double-precision (64-bit) floating-point element in "a" to a 32-bit integer with truncation, and store the result in "dst". -ReduceArgumentPS(src1[31:0], imm8[7:0]) -{ - IF src1[31:0] == NAN - RETURN (convert src1[31:0] to QNaN) - FI - - m := imm8[7:4] // number of fraction bits after the binary point to be preserved - rc := imm8[1:0] // round control - rc_src := imm8[2] // round ccontrol source - spe := 0 - tmp[31:0] := pow(2, -m)*ROUND(pow(2, m)*src1[31:0], spe, rc_source, rc) - tmp[31:0] := src1[31:0] - tmp[31:0] - RETURN tmp[31:0] -} - -IF k[0] - dst[31:0] := ReduceArgumentPS(a[31:0], imm8[7:0]) -ELSE - dst[31:0] := 0 -FI -dst[127:64] := b[127:32] -dst[MAX:128] := 0 +dst[31:0] := Convert_FP64_To_Int32_Truncate(a[63:0]) - -
immintrin.h
+ +
emmintrin.h
 - + Floating Point - AVX512DQ - Miscellaneous - - - - Extract the reduced argument of the lower single-precision (32-bit) floating-point element in "a" by the number of bits specified by "imm8", store the result in the lower element of "dst", and copy the upper 3 packed elements from "b" to the upper elements of "dst". + Integer + SSE2 + Convert + + + Convert the lower double-precision (64-bit) floating-point element in "a" to a 64-bit integer with truncation, and store the result in "dst". -ReduceArgumentPS(src1[31:0], imm8[7:0]) -{ - IF src1[31:0] == NAN - RETURN (convert src1[31:0] to QNaN) - FI - - m := imm8[7:4] // number of fraction bits after the binary point to be preserved - rc := imm8[1:0] // round control - rc_src := imm8[2] // round ccontrol source - spe := 0 - tmp[31:0] := pow(2, -m)*ROUND(pow(2, m)*src1[31:0], spe, rc_source, rc) - tmp[31:0] := src1[31:0] - tmp[31:0] - RETURN tmp[31:0] -} - -dst[31:0] := ReduceArgumentPS(a[31:0], imm8[7:0]) -dst[127:64] := b[127:32] -dst[MAX:128] := 0 +dst[63:0] := Convert_FP64_To_Int64_Truncate(a[63:0]) - -
immintrin.h
+ +
emmintrin.h
 - + Floating Point - AVX512DQ - Miscellaneous - - - - - Extract the reduced argument of the lower single-precision (32-bit) floating-point element in "a" by the number of bits specified by "imm8", store the result in the lower element of "dst", and copy the upper 3 packed elements from "b" to the upper elements of "dst". - [round_note] + Integer + SSE2 + Convert + + + Convert the lower double-precision (64-bit) floating-point element in "a" to a 64-bit integer with truncation, and store the result in "dst". -ReduceArgumentPS(src1[31:0], imm8[7:0]) -{ - IF src1[31:0] == NAN - RETURN (convert src1[31:0] to QNaN) - FI - - m := imm8[7:4] // number of fraction bits after the binary point to be preserved - rc := imm8[1:0] // round control - rc_src := imm8[2] // round ccontrol source - spe := 0 - tmp[31:0] := pow(2, -m)*ROUND(pow(2, m)*src1[31:0], spe, rc_source, rc) - tmp[31:0] := src1[31:0] - tmp[31:0] - RETURN tmp[31:0] -} - -dst[31:0] := ReduceArgumentPS(a[31:0], imm8[7:0]) -dst[127:64] := b[127:32] -dst[MAX:128] := 0 +dst[63:0] := Convert_FP64_To_Int64_Truncate(a[63:0]) - -
immintrin.h
+ +
emmintrin.h
 - + Floating Point - AVX512VL - AVX512F - Miscellaneous - - - - - Round packed double-precision (64-bit) floating-point elements in "a" to the number of fraction bits specified by "imm8", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Integer + SSE2 + Convert + + + Convert packed single-precision (32-bit) floating-point elements in "a" to packed 32-bit integers, and store the results in "dst". -RoundTo_IntegerPD(src[63:0], imm8[7:0]){ - IF(imm8[2] == 1) - rounding_direction := MXCSR.RC //Use the rounding mode specified by MXCSR.RC - ELSE - rounding_direction := imm8[1:0] //Use the rounding mode specified by imm8[1:0] - FI - - M := imm8[7:4] // The scaling factor (number of fraction bits to round to) - - CASE(rounding_direction) - 0: tmp[63:0] := round_to_nearest_even_integer(2^M * src[63:0]) - 1: tmp[63:0] := round_to_equal_or_smaller_integer(2^M * src[63:0]) - 2: tmp[63:0] := round_to_equal_or_larger_integer(2^M * src[63:0]) - 3: tmp[63:0] := round_to_nearest_smallest_magnitude_integer(2^M * src[63:0]) - ESAC - - dst[63:0] := 2^-M * tmp[63:0] // scale back down - - IF imm8[3] == 0 //check SPE - IF src[63:0] != dst[63:0] //check if precision has been lost - set_precision() //set #PE - FI - FI - RETURN dst[63:0] -} - FOR j := 0 to 3 - i := j*64 - IF k[j] - dst[i+63:i] := RoundTo_IntegerPD(a[i+63:i], imm8[7:0]) - ELSE - dst[i+63:i] := src[i+63:i] - FI + i := 32*j + dst[i+31:i] := Convert_FP32_To_Int32(a[i+31:i]) ENDFOR -dst[MAX:256] := 0 - -
immintrin.h
+ +
emmintrin.h
 - + Floating Point - AVX512VL - AVX512F - Miscellaneous - - - - Round packed double-precision (64-bit) floating-point elements in "a" to the number of fraction bits specified by "imm8", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + Integer + SSE2 + Convert + + + Convert packed single-precision (32-bit) floating-point elements in "a" to packed 32-bit integers with truncation, and store the results in "dst". -RoundTo_IntegerPD(src[63:0], imm8[7:0]){ - IF(imm8[2] == 1) - rounding_direction := MXCSR.RC //Use the rounding mode specified by MXCSR.RC - ELSE - rounding_direction := imm8[1:0] //Use the rounding mode specified by imm8[1:0] - FI - - M := imm8[7:4] // The scaling factor (number of fraction bits to round to) - - CASE(rounding_direction) - 0: tmp[63:0] := round_to_nearest_even_integer(2^M * src[63:0]) - 1: tmp[63:0] := round_to_equal_or_smaller_integer(2^M * src[63:0]) - 2: tmp[63:0] := round_to_equal_or_larger_integer(2^M * src[63:0]) - 3: tmp[63:0] := round_to_nearest_smallest_magnitude_integer(2^M * src[63:0]) - ESAC - - dst[63:0] := 2^-M * tmp[63:0] // scale back down - - IF imm8[3] == 0 //check SPE - IF src[63:0] != dst[63:0] //check if precision has been lost - set_precision() //set #PE - FI - FI - RETURN dst[63:0] -} - FOR j := 0 to 3 - i := j*64 - IF k[j] - dst[i+63:i] := RoundTo_IntegerPD(a[i+63:i], imm8[7:0]) - ELSE - dst[i+63:i] := 0 - FI + i := 32*j + dst[i+31:i] := Convert_FP32_To_Int32_Truncate(a[i+31:i]) ENDFOR -dst[MAX:256] := 0 - -
immintrin.h
+ +
emmintrin.h
 - + Floating Point - AVX512VL - AVX512F - Miscellaneous - - - Round packed double-precision (64-bit) floating-point elements in "a" to the number of fraction bits specified by "imm8", and store the results in "dst". - + Integer + SSE2 + Convert + + + Convert packed double-precision (64-bit) floating-point elements in "a" to packed 32-bit integers, and store the results in "dst". -RoundTo_IntegerPD(src[63:0], imm8[7:0]){ - IF(imm8[2] == 1) - rounding_direction := MXCSR.RC //Use the rounding mode specified by MXCSR.RC - ELSE - rounding_direction := imm8[1:0] //Use the rounding mode specified by imm8[1:0] - FI - - M := imm8[7:4] // The scaling factor (number of fraction bits to round to) - - CASE(rounding_direction) - 0: tmp[63:0] := round_to_nearest_even_integer(2^M * src[63:0]) - 1: tmp[63:0] := round_to_equal_or_smaller_integer(2^M * src[63:0]) - 2: tmp[63:0] := round_to_equal_or_larger_integer(2^M * src[63:0]) - 3: tmp[63:0] := round_to_nearest_smallest_magnitude_integer(2^M * src[63:0]) - ESAC - - dst[63:0] := 2^-M * tmp[63:0] // scale back down - - IF imm8[3] == 0 //check SPE - IF src[63:0] != dst[63:0] //check if precision has been lost - set_precision() //set #PE - FI - FI - RETURN dst[63:0] -} - -FOR j := 0 to 3 - i := j*64 - dst[i+63:i] := RoundTo_IntegerPD(a[i+63:i], imm8[7:0]) +FOR j := 0 to 1 + i := 32*j + k := 64*j + dst[i+31:i] := Convert_FP64_To_Int32(a[k+63:k]) ENDFOR -dst[MAX:256] := 0 - -
immintrin.h
+ +
emmintrin.h
 - + Floating Point - AVX512VL - AVX512F - Miscellaneous - - - - - Round packed double-precision (64-bit) floating-point elements in "a" to the number of fraction bits specified by "imm8", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + Integer + SSE2 + Convert + + + Convert packed double-precision (64-bit) floating-point elements in "a" to packed 32-bit integers with truncation, and store the results in "dst". -RoundTo_IntegerPD(src[63:0], imm8[7:0]){ - IF(imm8[2] == 1) - rounding_direction := MXCSR.RC //Use the rounding mode specified by MXCSR.RC - ELSE - rounding_direction := imm8[1:0] //Use the rounding mode specified by imm8[1:0] - FI - - M := imm8[7:4] // The scaling factor (number of fraction bits to round to) - - CASE(rounding_direction) - 0: tmp[63:0] := round_to_nearest_even_integer(2^M * src[63:0]) - 1: tmp[63:0] := round_to_equal_or_smaller_integer(2^M * src[63:0]) - 2: tmp[63:0] := round_to_equal_or_larger_integer(2^M * src[63:0]) - 3: tmp[63:0] := round_to_nearest_smallest_magnitude_integer(2^M * src[63:0]) - ESAC - - dst[63:0] := 2^-M * tmp[63:0] // scale back down - - IF imm8[3] == 0 //check SPE - IF src[63:0] != dst[63:0] //check if precision has been lost - set_precision() //set #PE - FI - FI - RETURN dst[63:0] -} - FOR j := 0 to 1 - i := j*64 - IF k[j] - dst[i+63:i] := RoundTo_IntegerPD(a[i+63:i], imm8[7:0]) - ELSE - dst[i+63:i] := src[i+63:i] - FI + i := 32*j + k := 64*j + dst[i+31:i] := Convert_FP64_To_Int32_Truncate(a[k+63:k]) ENDFOR -dst[MAX:128] := 0 - -
immintrin.h
+ +
emmintrin.h
 - + Floating Point - AVX512VL - AVX512F - Miscellaneous - - - - Round packed double-precision (64-bit) floating-point elements in "a" to the number of fraction bits specified by "imm8", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + SSE2 + Set + + + Copy double-precision (64-bit) floating-point element "a" to the lower element of "dst", and zero the upper element. -RoundTo_IntegerPD(src[63:0], imm8[7:0]){ - IF(imm8[2] == 1) - rounding_direction := MXCSR.RC //Use the rounding mode specified by MXCSR.RC - ELSE - rounding_direction := imm8[1:0] //Use the rounding mode specified by imm8[1:0] - FI - - M := imm8[7:4] // The scaling factor (number of fraction bits to round to) - - CASE(rounding_direction) - 0: tmp[63:0] := round_to_nearest_even_integer(2^M * src[63:0]) - 1: tmp[63:0] := round_to_equal_or_smaller_integer(2^M * src[63:0]) - 2: tmp[63:0] := round_to_equal_or_larger_integer(2^M * src[63:0]) - 3: tmp[63:0] := round_to_nearest_smallest_magnitude_integer(2^M * src[63:0]) - ESAC - - dst[63:0] := 2^-M * tmp[63:0] // scale back down - - IF imm8[3] == 0 //check SPE - IF src[63:0] != dst[63:0] //check if precision has been lost - set_precision() //set #PE - FI - FI - RETURN dst[63:0] -} - -FOR j := 0 to 1 - i := j*64 - IF k[j] - dst[i+63:i] := RoundTo_IntegerPD(a[i+63:i], imm8[7:0]) - ELSE - dst[i+63:i] := 0 - FI -ENDFOR -dst[MAX:128] := 0 +dst[63:0] := a[63:0] +dst[127:64] := 0 - -
immintrin.h
+
emmintrin.h
 - + Floating Point - AVX512VL - AVX512F - Miscellaneous - - - Round packed double-precision (64-bit) floating-point elements in "a" to the number of fraction bits specified by "imm8", and store the results in "dst". - + SSE2 + Set + + + Broadcast double-precision (64-bit) floating-point value "a" to all elements of "dst". -RoundTo_IntegerPD(src[63:0], imm8[7:0]){ - IF(imm8[2] == 1) - rounding_direction := MXCSR.RC //Use the rounding mode specified by MXCSR.RC - ELSE - rounding_direction := imm8[1:0] //Use the rounding mode specified by imm8[1:0] - FI - - M := imm8[7:4] // The scaling factor (number of fraction bits to round to) - - CASE(rounding_direction) - 0: tmp[63:0] := round_to_nearest_even_integer(2^M * src[63:0]) - 1: tmp[63:0] := round_to_equal_or_smaller_integer(2^M * src[63:0]) - 2: tmp[63:0] := round_to_equal_or_larger_integer(2^M * src[63:0]) - 3: tmp[63:0] := round_to_nearest_smallest_magnitude_integer(2^M * src[63:0]) - ESAC - - dst[63:0] := 2^-M * tmp[63:0] // scale back down - - IF imm8[3] == 0 //check SPE - IF src[63:0] != dst[63:0] //check if precision has been lost - set_precision() //set #PE - FI - FI - RETURN dst[63:0] -} - FOR j := 0 to 1 i := j*64 - dst[i+63:i] := RoundTo_IntegerPD(a[i+63:i], imm8[7:0]) + dst[i+63:i] := a[63:0] ENDFOR -dst[MAX:128] := 0 - -
immintrin.h
+
emmintrin.h
 - + Floating Point - AVX512VL - AVX512F - Miscellaneous - - - - - Round packed single-precision (32-bit) floating-point elements in "a" to the number of fraction bits specified by "imm8", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + SSE2 + Set + + + Broadcast double-precision (64-bit) floating-point value "a" to all elements of "dst". -RoundTo_IntegerPS(src[31:0], imm8[7:0]){ - IF(imm8[2] == 1) - rounding_direction := MXCSR.RC //Use the rounding mode specified by MXCSR.RC - ELSE - rounding_direction := imm8[1:0] //Use the rounding mode specified by imm8[1:0] - FI - - M := imm8[7:4] // The scaling factor (number of fraction bits to round to) - - CASE(rounding_direction) - 0: tmp[31:0] := round_to_nearest_even_integer(2^M * src[31:0]) - 1: tmp[31:0] := round_to_equal_or_smaller_integer(2^M * src[31:0]) - 2: tmp[31:0] := round_to_equal_or_larger_integer(2^M * src[31:0]) - 3: tmp[31:0] := round_to_nearest_smallest_magnitude_integer(2^M * src[31:0]) - ESAC - - dst[31:0] := 2^-M * tmp[31:0] // scale back down - - IF imm8[3] == 0 //check SPE - IF src[31:0] != dst[31:0] //check if precision has been lost - set_precision() //set #PE - FI - FI - RETURN dst[31:0] -} - -FOR j := 0 to 7 - i := j*32 - IF k[j] - dst[i+31:i] := RoundTo_IntegerPS(a[i+31:i], imm8[7:0]) - ELSE - dst[i+31:i] := src[i+31:i] - FI +FOR j := 0 to 1 + i := j*64 + dst[i+63:i] := a[63:0] ENDFOR -dst[MAX:256] := 0 - -
immintrin.h
+
emmintrin.h
 - + Floating Point - AVX512VL - AVX512F - Miscellaneous - - - - Round packed single-precision (32-bit) floating-point elements in "a" to the number of fraction bits specified by "imm8", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + SSE2 + Set + + + + Set packed double-precision (64-bit) floating-point elements in "dst" with the supplied values. -RoundTo_IntegerPS(src[31:0], imm8[7:0]){ - IF(imm8[2] == 1) - rounding_direction := MXCSR.RC //Use the rounding mode specified by MXCSR.RC - ELSE - rounding_direction := imm8[1:0] //Use the rounding mode specified by imm8[1:0] - FI - - M := imm8[7:4] // The scaling factor (number of fraction bits to round to) - - CASE(rounding_direction) - 0: tmp[31:0] := round_to_nearest_even_integer(2^M * src[31:0]) - 1: tmp[31:0] := round_to_equal_or_smaller_integer(2^M * src[31:0]) - 2: tmp[31:0] := round_to_equal_or_larger_integer(2^M * src[31:0]) - 3: tmp[31:0] := round_to_nearest_smallest_magnitude_integer(2^M * src[31:0]) - ESAC - - dst[31:0] := 2^-M * tmp[31:0] // scale back down - - IF imm8[3] == 0 //check SPE - IF src[31:0] != dst[31:0] //check if precision has been lost - set_precision() //set #PE - FI - FI - RETURN dst[31:0] -} - -FOR j := 0 to 7 - i := j*32 - IF k[j] - dst[i+31:i] := RoundTo_IntegerPS(a[i+31:i], imm8[7:0]) - ELSE - dst[i+31:i] := 0 - FI -ENDFOR -dst[MAX:256] := 0 +dst[63:0] := e0 +dst[127:64] := e1 - -
immintrin.h
+
emmintrin.h
 - + Floating Point - AVX512VL - AVX512F - Miscellaneous - - - Round packed single-precision (32-bit) floating-point elements in "a" to the number of fraction bits specified by "imm8", and store the results in "dst". - + SSE2 + Set + + + + Set packed double-precision (64-bit) floating-point elements in "dst" with the supplied values in reverse order. -RoundTo_IntegerPS(src[31:0], imm8[7:0]){ - IF(imm8[2] == 1) - rounding_direction := MXCSR.RC //Use the rounding mode specified by MXCSR.RC - ELSE - rounding_direction := imm8[1:0] //Use the rounding mode specified by imm8[1:0] - FI - - M := imm8[7:4] // The scaling factor (number of fraction bits to round to) - - CASE(rounding_direction) - 0: tmp[31:0] := round_to_nearest_even_integer(2^M * src[31:0]) - 1: tmp[31:0] := round_to_equal_or_smaller_integer(2^M * src[31:0]) - 2: tmp[31:0] := round_to_equal_or_larger_integer(2^M * src[31:0]) - 3: tmp[31:0] := round_to_nearest_smallest_magnitude_integer(2^M * src[31:0]) - ESAC - - dst[31:0] := 2^-M * tmp[31:0] // scale back down - - IF imm8[3] == 0 //check SPE - IF src[31:0] != dst[31:0] //check if precision has been lost - set_precision() //set #PE - FI - FI - RETURN dst[31:0] -} - -FOR j := 0 to 7 - i := j*32 - dst[i+31:i] := RoundTo_IntegerPS(a[i+31:i], imm8[7:0]) -ENDFOR -dst[MAX:256] := 0 +dst[63:0] := e1 +dst[127:64] := e0 - -
immintrin.h
+
emmintrin.h
 - + Floating Point - AVX512VL - AVX512F - Miscellaneous - - - - - Round packed single-precision (32-bit) floating-point elements in "a" to the number of fraction bits specified by "imm8", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -RoundTo_IntegerPS(src[31:0], imm8[7:0]){ - IF(imm8[2] == 1) - rounding_direction := MXCSR.RC //Use the rounding mode specified by MXCSR.RC - ELSE - rounding_direction := imm8[1:0] //Use the rounding mode specified by imm8[1:0] - FI - - M := imm8[7:4] // The scaling factor (number of fraction bits to round to) - - CASE(rounding_direction) - 0: tmp[31:0] := round_to_nearest_even_integer(2^M * src[31:0]) - 1: tmp[31:0] := round_to_equal_or_smaller_integer(2^M * src[31:0]) - 2: tmp[31:0] := round_to_equal_or_larger_integer(2^M * src[31:0]) - 3: tmp[31:0] := round_to_nearest_smallest_magnitude_integer(2^M * src[31:0]) - ESAC - - dst[31:0] := 2^-M * tmp[31:0] // scale back down - - IF imm8[3] == 0 //check SPE - IF src[31:0] != dst[31:0] //check if precision has been lost - set_precision() //set #PE - FI - FI - RETURN dst[31:0] -} - -FOR j := 0 to 3 - i := j*32 - IF k[j] - dst[i+31:i] := RoundTo_IntegerPS(a[i+31:i], imm8[7:0]) - ELSE - dst[i+31:i] := src[i+31:i] - FI -ENDFOR -dst[MAX:128] := 0 + SSE2 + Set + + + Return vector of type __m128d with all elements set to zero. + +dst[MAX:0] := 0 - -
immintrin.h
+ +
emmintrin.h
 - + Floating Point - AVX512VL - AVX512F - Miscellaneous - - - - Round packed single-precision (32-bit) floating-point elements in "a" to the number of fraction bits specified by "imm8", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + SSE2 + Load + + + Load 128-bits (composed of 2 packed double-precision (64-bit) floating-point elements) from memory into "dst". + "mem_addr" must be aligned on a 16-byte boundary or a general-protection exception may be generated. -RoundTo_IntegerPS(src[31:0], imm8[7:0]){ - IF(imm8[2] == 1) - rounding_direction := MXCSR.RC //Use the rounding mode specified by MXCSR.RC - ELSE - rounding_direction := imm8[1:0] //Use the rounding mode specified by imm8[1:0] - FI - - M := imm8[7:4] // The scaling factor (number of fraction bits to round to) - - CASE(rounding_direction) - 0: tmp[31:0] := round_to_nearest_even_integer(2^M * src[31:0]) - 1: tmp[31:0] := round_to_equal_or_smaller_integer(2^M * src[31:0]) - 2: tmp[31:0] := round_to_equal_or_larger_integer(2^M * src[31:0]) - 3: tmp[31:0] := round_to_nearest_smallest_magnitude_integer(2^M * src[31:0]) - ESAC - - dst[31:0] := 2^-M * tmp[31:0] // scale back down - - IF imm8[3] == 0 //check SPE - IF src[31:0] != dst[31:0] //check if precision has been lost - set_precision() //set #PE - FI - FI - RETURN dst[31:0] -} - -FOR j := 0 to 3 - i := j*32 - IF k[j] - dst[i+31:i] := RoundTo_IntegerPS(a[i+31:i], imm8[7:0]) - ELSE - dst[i+31:i] := 0 - FI -ENDFOR -dst[MAX:128] := 0 +dst[127:0] := MEM[mem_addr+127:mem_addr] - -
immintrin.h
+ +
emmintrin.h
 - + Floating Point - AVX512VL - AVX512F - Miscellaneous - - - Round packed single-precision (32-bit) floating-point elements in "a" to the number of fraction bits specified by "imm8", and store the results in "dst". - + SSE2 + Load + + + Load a double-precision (64-bit) floating-point element from memory into both elements of "dst". -RoundTo_IntegerPS(src[31:0], imm8[7:0]){ - IF(imm8[2] == 1) - rounding_direction := MXCSR.RC //Use the rounding mode specified by MXCSR.RC - ELSE - rounding_direction := imm8[1:0] //Use the rounding mode specified by imm8[1:0] - FI - - M := imm8[7:4] // The scaling factor (number of fraction bits to round to) - - CASE(rounding_direction) - 0: tmp[31:0] := round_to_nearest_even_integer(2^M * src[31:0]) - 1: tmp[31:0] := round_to_equal_or_smaller_integer(2^M * src[31:0]) - 2: tmp[31:0] := round_to_equal_or_larger_integer(2^M * src[31:0]) - 3: tmp[31:0] := round_to_nearest_smallest_magnitude_integer(2^M * src[31:0]) - ESAC - - dst[31:0] := 2^-M * tmp[31:0] // scale back down - - IF imm8[3] == 0 //check SPE - IF src[31:0] != dst[31:0] //check if precision has been lost - set_precision() //set #PE - FI - FI - RETURN dst[31:0] -} - -FOR j := 0 to 3 - i := j*32 - dst[i+31:i] := RoundTo_IntegerPS(a[i+31:i], imm8[7:0]) -ENDFOR -dst[MAX:128] := 0 +dst[63:0] := MEM[mem_addr+63:mem_addr] +dst[127:64] := MEM[mem_addr+63:mem_addr] - -
immintrin.h
+ +
emmintrin.h
 - + Floating Point - AVX512VL - AVX512F - Arithmetic - - - - Compute the approximate reciprocal square root of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). The maximum relative error for this approximation is less than 2^-14. + SSE2 + Load + + + Load a double-precision (64-bit) floating-point element from memory into both elements of "dst". -FOR j := 0 to 3 - i := j*64 - IF k[j] - dst[i+63:i] := APPROXIMATE(1.0 / SQRT(a[i+63:i])) - ELSE - dst[i+63:i] := src[i+63:i] - FI -ENDFOR -dst[MAX:256] := 0 +dst[63:0] := MEM[mem_addr+63:mem_addr] +dst[127:64] := MEM[mem_addr+63:mem_addr] - -
immintrin.h
+ +
emmintrin.h
 - + Floating Point - AVX512VL - AVX512F - Arithmetic - - - Compute the approximate reciprocal square root of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). The maximum relative error for this approximation is less than 2^-14. + SSE2 + Load + + + Load 2 double-precision (64-bit) floating-point elements from memory into "dst" in reverse order. mem_addr must be aligned on a 16-byte boundary or a general-protection exception may be generated. -FOR j := 0 to 3 - i := j*64 - IF k[j] - dst[i+63:i] := APPROXIMATE(1.0 / SQRT(a[i+63:i])) - ELSE - dst[i+63:i] := 0 - FI -ENDFOR -dst[MAX:256] := 0 +dst[63:0] := MEM[mem_addr+127:mem_addr+64] +dst[127:64] := MEM[mem_addr+63:mem_addr] - -
immintrin.h
+ +
emmintrin.h
 - + Floating Point - AVX512VL - AVX512F - Arithmetic - - - - Compute the approximate reciprocal square root of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). The maximum relative error for this approximation is less than 2^-14. + SSE2 + Load + + + Load 128-bits (composed of 2 packed double-precision (64-bit) floating-point elements) from memory into "dst". + "mem_addr" does not need to be aligned on any particular boundary. -FOR j := 0 to 1 - i := j*64 - IF k[j] - dst[i+63:i] := APPROXIMATE(1.0 / SQRT(a[i+63:i])) - ELSE - dst[i+63:i] := src[i+63:i] - FI -ENDFOR -dst[MAX:128] := 0 +dst[127:0] := MEM[mem_addr+127:mem_addr] - -
immintrin.h
+ +
emmintrin.h
 - + Floating Point - AVX512VL - AVX512F - Arithmetic - - - Compute the approximate reciprocal square root of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). The maximum relative error for this approximation is less than 2^-14. + SSE2 + Load + + + Load a double-precision (64-bit) floating-point element from memory into the lower of "dst", and zero the upper element. "mem_addr" does not need to be aligned on any particular boundary. -FOR j := 0 to 1 - i := j*64 - IF k[j] - dst[i+63:i] := APPROXIMATE(1.0 / SQRT(a[i+63:i])) - ELSE - dst[i+63:i] := 0 - FI -ENDFOR -dst[MAX:128] := 0 +dst[63:0] := MEM[mem_addr+63:mem_addr] +dst[127:64] := 0 - -
immintrin.h
+ +
emmintrin.h
 - + Floating Point - AVX512VL - AVX512F - Arithmetic - - - - Compute the approximate reciprocal square root of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). The maximum relative error for this approximation is less than 2^-14. + SSE2 + Load + + + + Load a double-precision (64-bit) floating-point element from memory into the upper element of "dst", and copy the lower element from "a" to "dst". "mem_addr" does not need to be aligned on any particular boundary. -FOR j := 0 to 7 - i := j*32 - IF k[j] - dst[i+31:i] := APPROXIMATE(1.0 / SQRT(a[i+31:i])) - ELSE - dst[i+31:i] := src[i+31:i] - FI -ENDFOR -dst[MAX:256] := 0 +dst[63:0] := a[63:0] +dst[127:64] := MEM[mem_addr+63:mem_addr] - -
immintrin.h
+ +
emmintrin.h
 - + Floating Point - AVX512VL - AVX512F - Arithmetic - - - Compute the approximate reciprocal square root of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). The maximum relative error for this approximation is less than 2^-14. + SSE2 + Load + + + + Load a double-precision (64-bit) floating-point element from memory into the lower element of "dst", and copy the upper element from "a" to "dst". "mem_addr" does not need to be aligned on any particular boundary. -FOR j := 0 to 7 - i := j*32 - IF k[j] - dst[i+31:i] := APPROXIMATE(1.0 / SQRT(a[i+31:i])) - ELSE - dst[i+31:i] := 0 - FI -ENDFOR -dst[MAX:256] := 0 +dst[63:0] := MEM[mem_addr+63:mem_addr] +dst[127:64] := a[127:64] - -
immintrin.h
+ +
emmintrin.h
 - + Floating Point - AVX512VL - AVX512F - Arithmetic - - - - Compute the approximate reciprocal square root of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). The maximum relative error for this approximation is less than 2^-14. + SSE2 + Store + + + + Store 128-bits (composed of 2 packed double-precision (64-bit) floating-point elements) from "a" into memory using a non-temporal memory hint. + "mem_addr" must be aligned on a 16-byte boundary or a general-protection exception may be generated. -FOR j := 0 to 3 - i := j*32 - IF k[j] - dst[i+31:i] := APPROXIMATE(1.0 / SQRT(a[i+31:i])) - ELSE - dst[i+31:i] := src[i+31:i] - FI -ENDFOR -dst[MAX:128] := 0 +MEM[mem_addr+127:mem_addr] := a[127:0] - -
immintrin.h
+ +
emmintrin.h
 - + Floating Point - AVX512VL - AVX512F - Arithmetic - - - Compute the approximate reciprocal square root of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). The maximum relative error for this approximation is less than 2^-14. + SSE2 + Store + + + + Store the lower double-precision (64-bit) floating-point element from "a" into memory. "mem_addr" does not need to be aligned on any particular boundary. -FOR j := 0 to 3 - i := j*32 - IF k[j] - dst[i+31:i] := APPROXIMATE(1.0 / SQRT(a[i+31:i])) - ELSE - dst[i+31:i] := 0 - FI -ENDFOR -dst[MAX:128] := 0 +MEM[mem_addr+63:mem_addr] := a[63:0] - -
immintrin.h
+ +
emmintrin.h
 - + Floating Point - AVX512VL - AVX512F - Miscellaneous - - - - - Scale the packed double-precision (64-bit) floating-point elements in "a" using values from "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + SSE2 + Store + + + + Store the lower double-precision (64-bit) floating-point element from "a" into 2 contiguous elements in memory. "mem_addr" must be aligned on a 16-byte boundary or a general-protection exception may be generated. -SCALE(src1, src2){ - IF (src2 == NaN) - IF (src2 == SNaN) - RETURN QNAN(src2) - FI - ELSE IF (src1 == NaN) - IF (src1 == SNaN) - RETURN QNAN(src1) - FI - IF (src2 != INF) - RETURN QNAN(src1) - FI - ELSE - tmp_src2 := src2 - tmp_src1 := src1 - IF (src2 is denormal AND MXCSR.DAZ) - tmp_src2 := 0 - FI - IF (src1 is denormal AND MXCSR.DAZ) - tmp_src1 := 0 - FI - FI - dst[63:0] := tmp_src1[63:0] * POW(2, FLOOR(tmp_src2[63:0])) - RETURN dst[63:0] -} - -FOR j := 0 to 3 - i := j*64 - IF k[j] - dst[i+63:i] := SCALE(a[i+63:0], b[i+63:i]) - ELSE - dst[i+63:i] := src[i+63:i] - FI -ENDFOR -dst[MAX:256] := 0 +MEM[mem_addr+63:mem_addr] := a[63:0] +MEM[mem_addr+127:mem_addr+64] := a[63:0] - -
immintrin.h
+
emmintrin.h
 - + Floating Point - AVX512VL - AVX512F - Miscellaneous - - - - Scale the packed double-precision (64-bit) floating-point elements in "a" using values from "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + SSE2 + Store + + + + Store the lower double-precision (64-bit) floating-point element from "a" into 2 contiguous elements in memory. "mem_addr" must be aligned on a 16-byte boundary or a general-protection exception may be generated. -SCALE(src1, src2){ - IF (src2 == NaN) - IF (src2 == SNaN) - RETURN QNAN(src2) - FI - ELSE IF (src1 == NaN) - IF (src1 == SNaN) - RETURN QNAN(src1) - FI - IF (src2 != INF) - RETURN QNAN(src1) - FI - ELSE - tmp_src2 := src2 - tmp_src1 := src1 - IF (src2 is denormal AND MXCSR.DAZ) - tmp_src2 := 0 - FI - IF (src1 is denormal AND MXCSR.DAZ) - tmp_src1 := 0 - FI - FI - dst[63:0] := tmp_src1[63:0] * POW(2, FLOOR(tmp_src2[63:0])) - RETURN dst[63:0] -} - -FOR j := 0 to 3 - i := j*64 - IF k[j] - dst[i+63:i] := SCALE(a[i+63:0], b[i+63:i]) - ELSE - dst[i+63:i] := 0 - FI -ENDFOR -dst[MAX:256] := 0 +MEM[mem_addr+63:mem_addr] := a[63:0] +MEM[mem_addr+127:mem_addr+64] := a[63:0] - -
immintrin.h
+
emmintrin.h
 - + Floating Point - AVX512VL - AVX512F - Miscellaneous - - - Scale the packed double-precision (64-bit) floating-point elements in "a" using values from "b", and store the results in "dst". + SSE2 + Store + + + + Store 128-bits (composed of 2 packed double-precision (64-bit) floating-point elements) from "a" into memory. + "mem_addr" must be aligned on a 16-byte boundary or a general-protection exception may be generated. -SCALE(src1, src2){ - IF (src2 == NaN) - IF (src2 == SNaN) - RETURN QNAN(src2) - FI - ELSE IF (src1 == NaN) - IF (src1 == SNaN) - RETURN QNAN(src1) - FI - IF (src2 != INF) - RETURN QNAN(src1) - FI - ELSE - tmp_src2 := src2 - tmp_src1 := src1 - IF (src2 is denormal AND MXCSR.DAZ) - tmp_src2 := 0 - FI - IF (src1 is denormal AND MXCSR.DAZ) - tmp_src1 := 0 - FI - FI - dst[63:0] := tmp_src1[63:0] * POW(2, FLOOR(tmp_src2[63:0])) - RETURN dst[63:0] -} - -FOR j := 0 to 3 - i := j*64 - dst[i+63:i] := SCALE(a[i+63:0], b[i+63:i]) -ENDFOR -dst[MAX:256] := 0 +MEM[mem_addr+127:mem_addr] := a[127:0] - -
immintrin.h
+ +
emmintrin.h
 - + Floating Point - AVX512VL - AVX512F - Miscellaneous - - - - - Scale the packed double-precision (64-bit) floating-point elements in "a" using values from "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + SSE2 + Store + + + + Store 128-bits (composed of 2 packed double-precision (64-bit) floating-point elements) from "a" into memory. + "mem_addr" does not need to be aligned on any particular boundary. -SCALE(src1, src2){ - IF (src2 == NaN) - IF (src2 == SNaN) - RETURN QNAN(src2) - FI - ELSE IF (src1 == NaN) - IF (src1 == SNaN) - RETURN QNAN(src1) - FI - IF (src2 != INF) - RETURN QNAN(src1) - FI - ELSE - tmp_src2 := src2 - tmp_src1 := src1 - IF (src2 is denormal AND MXCSR.DAZ) - tmp_src2 := 0 - FI - IF (src1 is denormal AND MXCSR.DAZ) - tmp_src1 := 0 - FI - FI - dst[63:0] := tmp_src1[63:0] * POW(2, FLOOR(tmp_src2[63:0])) - RETURN dst[63:0] -} - -FOR j := 0 to 1 - i := j*64 - IF k[j] - dst[i+63:i] := SCALE(a[i+63:0], b[i+63:i]) - ELSE - dst[i+63:i] := src[i+63:i] - FI -ENDFOR -dst[MAX:128] := 0 +MEM[mem_addr+127:mem_addr] := a[127:0] - -
immintrin.h
+ +
emmintrin.h
 - + Floating Point - AVX512VL - AVX512F - Miscellaneous - - - - Scale the packed double-precision (64-bit) floating-point elements in "a" using values from "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + SSE2 + Store + + + + Store 2 double-precision (64-bit) floating-point elements from "a" into memory in reverse order. + "mem_addr" must be aligned on a 16-byte boundary or a general-protection exception may be generated. -SCALE(src1, src2){ - IF (src2 == NaN) - IF (src2 == SNaN) - RETURN QNAN(src2) - FI - ELSE IF (src1 == NaN) - IF (src1 == SNaN) - RETURN QNAN(src1) - FI - IF (src2 != INF) - RETURN QNAN(src1) - FI - ELSE - tmp_src2 := src2 - tmp_src1 := src1 - IF (src2 is denormal AND MXCSR.DAZ) - tmp_src2 := 0 - FI - IF (src1 is denormal AND MXCSR.DAZ) - tmp_src1 := 0 - FI - FI - dst[63:0] := tmp_src1[63:0] * POW(2, FLOOR(tmp_src2[63:0])) - RETURN dst[63:0] -} - -FOR j := 0 to 1 - i := j*64 - IF k[j] - dst[i+63:i] := SCALE(a[i+63:0], b[i+63:i]) - ELSE - dst[i+63:i] := 0 - FI -ENDFOR -dst[MAX:128] := 0 +MEM[mem_addr+63:mem_addr] := a[127:64] +MEM[mem_addr+127:mem_addr+64] := a[63:0] - -
immintrin.h
+
emmintrin.h
 - + Floating Point - AVX512VL - AVX512F - Miscellaneous - - - Scale the packed double-precision (64-bit) floating-point elements in "a" using values from "b", and store the results in "dst". + SSE2 + Store + + + + Store the upper double-precision (64-bit) floating-point element from "a" into memory. -SCALE(src1, src2){ - IF (src2 == NaN) - IF (src2 == SNaN) - RETURN QNAN(src2) - FI - ELSE IF (src1 == NaN) - IF (src1 == SNaN) - RETURN QNAN(src1) - FI - IF (src2 != INF) - RETURN QNAN(src1) - FI - ELSE - tmp_src2 := src2 - tmp_src1 := src1 - IF (src2 is denormal AND MXCSR.DAZ) - tmp_src2 := 0 - FI - IF (src1 is denormal AND MXCSR.DAZ) - tmp_src1 := 0 - FI - FI - dst[63:0] := tmp_src1[63:0] * POW(2, FLOOR(tmp_src2[63:0])) - RETURN dst[63:0] -} - -FOR j := 0 to 1 - i := j*64 - dst[i+63:i] := SCALE(a[i+63:0], b[i+63:i]) -ENDFOR -dst[MAX:128] := 0 +MEM[mem_addr+63:mem_addr] := a[127:64] - -
immintrin.h
+ +
emmintrin.h
 - + Floating Point - AVX512VL - AVX512F - Miscellaneous - - - - - Scale the packed single-precision (32-bit) floating-point elements in "a" using values from "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + SSE2 + Store + + + + Store the lower double-precision (64-bit) floating-point element from "a" into memory. -SCALE(src1, src2){ - IF (src2 == NaN) - IF (src2 == SNaN) - RETURN QNAN(src2) - FI - ELSE IF (src1 == NaN) - IF (src1 == SNaN) - RETURN QNAN(src1) - FI - IF (src2 != INF) - RETURN QNAN(src1) - FI - ELSE - tmp_src2 := src2 - tmp_src1 := src1 - IF (src2 is denormal AND MXCSR.DAZ) - tmp_src2 := 0 - FI - IF (src1 is denormal AND MXCSR.DAZ) - tmp_src1 := 0 - FI - FI - dst[31:0] := tmp_src1[31:0] * POW(2, FLOOR(tmp_src2[31:0])) - RETURN dst[31:0] -} - -FOR j := 0 to 7 - i := j*32 - IF k[j] - dst[i+31:i] := SCALE(a[i+31:0], b[i+31:i]) - ELSE - dst[i+31:i] := src[i+31:i] - FI -ENDFOR -dst[MAX:256] := 0 +MEM[mem_addr+63:mem_addr] := a[63:0] - -
immintrin.h
+ +
emmintrin.h
 - + Floating Point - AVX512VL - AVX512F - Miscellaneous - - - - Scale the packed single-precision (32-bit) floating-point elements in "a" using values from "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - -SCALE(src1, src2){ - IF (src2 == NaN) - IF (src2 == SNaN) - RETURN QNAN(src2) - FI - ELSE IF (src1 == NaN) - IF (src1 == SNaN) - RETURN QNAN(src1) - FI - IF (src2 != INF) - RETURN QNAN(src1) - FI - ELSE - tmp_src2 := src2 - tmp_src1 := src1 - IF (src2 is denormal AND MXCSR.DAZ) - tmp_src2 := 0 - FI - IF (src1 is denormal AND MXCSR.DAZ) - tmp_src1 := 0 - FI - FI - dst[31:0] := tmp_src1[31:0] * POW(2, FLOOR(tmp_src2[31:0])) - RETURN dst[31:0] + SSE2 + Swizzle + + + + Unpack and interleave double-precision (64-bit) floating-point elements from the high half of "a" and "b", and store the results in "dst". + +DEFINE INTERLEAVE_HIGH_QWORDS(src1[127:0], src2[127:0]) { + dst[63:0] := src1[127:64] + dst[127:64] := src2[127:64] + RETURN dst[127:0] } - -FOR j := 0 to 7 - i := j*32 - IF k[j] - dst[i+31:i] := SCALE(a[i+31:0], b[i+31:i]) - ELSE - dst[i+31:i] := 0 - FI -ENDFOR -dst[MAX:256] := 0 +dst[127:0] := INTERLEAVE_HIGH_QWORDS(a[127:0], b[127:0]) - -
immintrin.h
+ +
emmintrin.h
 - + Floating Point - AVX512VL - AVX512F - Miscellaneous - - - Scale the packed single-precision (32-bit) floating-point elements in "a" using values from "b", and store the results in "dst". + SSE2 + Swizzle + + + + Unpack and interleave double-precision (64-bit) floating-point elements from the low half of "a" and "b", and store the results in "dst". -SCALE(src1, src2){ - IF (src2 == NaN) - IF (src2 == SNaN) - RETURN QNAN(src2) - FI - ELSE IF (src1 == NaN) - IF (src1 == SNaN) - RETURN QNAN(src1) - FI - IF (src2 != INF) - RETURN QNAN(src1) - FI - ELSE - tmp_src2 := src2 - tmp_src1 := src1 - IF (src2 is denormal AND MXCSR.DAZ) - tmp_src2 := 0 - FI - IF (src1 is denormal AND MXCSR.DAZ) - tmp_src1 := 0 - FI - FI - dst[31:0] := tmp_src1[31:0] * POW(2, FLOOR(tmp_src2[31:0])) - RETURN dst[31:0] +DEFINE INTERLEAVE_QWORDS(src1[127:0], src2[127:0]) { + dst[63:0] := src1[63:0] + dst[127:64] := src2[63:0] + RETURN dst[127:0] } - -FOR j := 0 to 7 - i := j*32 - dst[i+31:i] := SCALE(a[i+31:0], b[i+31:i]) -ENDFOR -dst[MAX:256] := 0 +dst[127:0] := INTERLEAVE_QWORDS(a[127:0], b[127:0]) - -
immintrin.h
+ +
emmintrin.h
 - + Floating Point - AVX512VL - AVX512F + SSE2 Miscellaneous - - - - - Scale the packed single-precision (32-bit) floating-point elements in "a" using values from "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + + Set each bit of mask "dst" based on the most significant bit of the corresponding packed double-precision (64-bit) floating-point element in "a". -SCALE(src1, src2){ - IF (src2 == NaN) - IF (src2 == SNaN) - RETURN QNAN(src2) - FI - ELSE IF (src1 == NaN) - IF (src1 == SNaN) - RETURN QNAN(src1) - FI - IF (src2 != INF) - RETURN QNAN(src1) - FI - ELSE - tmp_src2 := src2 - tmp_src1 := src1 - IF (src2 is denormal AND MXCSR.DAZ) - tmp_src2 := 0 - FI - IF (src1 is denormal AND MXCSR.DAZ) - tmp_src1 := 0 - FI - FI - dst[31:0] := tmp_src1[31:0] * POW(2, FLOOR(tmp_src2[31:0])) - RETURN dst[31:0] -} - -FOR j := 0 to 3 - i := j*32 - IF k[j] - dst[i+31:i] := SCALE(a[i+31:0], b[i+31:i]) +FOR j := 0 to 1 + i := j*64 + IF a[i+63] + dst[j] := 1 ELSE - dst[i+31:i] := src[i+31:i] + dst[j] := 0 FI ENDFOR -dst[MAX:128] := 0 +dst[MAX:2] := 0 - -
immintrin.h
+ +
emmintrin.h
 - + Floating Point - AVX512VL - AVX512F - Miscellaneous - - - - Scale the packed single-precision (32-bit) floating-point elements in "a" using values from "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + SSE2 + Swizzle + + + + + Shuffle double-precision (64-bit) floating-point elements using the control in "imm8", and store the results in "dst". + +dst[63:0] := (imm8[0] == 0) ? a[63:0] : a[127:64] +dst[127:64] := (imm8[1] == 0) ? b[63:0] : b[127:64] + + +
emmintrin.h
+ + + Floating Point + SSE2 + Move + + + + Move the lower double-precision (64-bit) floating-point element from "b" to the lower element of "dst", and copy the upper element from "a" to the upper element of "dst". + +dst[63:0] := b[63:0] +dst[127:64] := a[127:64] + + +
emmintrin.h
+ + + Floating Point + SSE2 + Cast + + + Cast vector of type __m128d to type __m128. This intrinsic is only used for compilation and does not generate any instructions, thus it has zero latency. +
emmintrin.h
+ + + Floating Point + Integer + SSE2 + Cast + + + Cast vector of type __m128d to type __m128i. This intrinsic is only used for compilation and does not generate any instructions, thus it has zero latency. +
emmintrin.h
+ + + Floating Point + SSE2 + Cast + + + Cast vector of type __m128 to type __m128d. This intrinsic is only used for compilation and does not generate any instructions, thus it has zero latency. +
emmintrin.h
+ + + Floating Point + Integer + SSE2 + Cast + + + Cast vector of type __m128 to type __m128i. This intrinsic is only used for compilation and does not generate any instructions, thus it has zero latency. +
emmintrin.h
+ + + Floating Point + SSE2 + Cast + + + Cast vector of type __m128i to type __m128d. This intrinsic is only used for compilation and does not generate any instructions, thus it has zero latency. +
emmintrin.h
+ + + Floating Point + SSE2 + Cast + + + Cast vector of type __m128i to type __m128. This intrinsic is only used for compilation and does not generate any instructions, thus it has zero latency. +
emmintrin.h
+ + + Floating Point + SSE3 + Arithmetic + + + + Alternatively add and subtract packed single-precision (32-bit) floating-point elements in "a" to/from packed elements in "b", and store the results in "dst". -SCALE(src1, src2){ - IF (src2 == NaN) - IF (src2 == SNaN) - RETURN QNAN(src2) - FI - ELSE IF (src1 == NaN) - IF (src1 == SNaN) - RETURN QNAN(src1) - FI - IF (src2 != INF) - RETURN QNAN(src1) - FI - ELSE - tmp_src2 := src2 - tmp_src1 := src1 - IF (src2 is denormal AND MXCSR.DAZ) - tmp_src2 := 0 - FI - IF (src1 is denormal AND MXCSR.DAZ) - tmp_src1 := 0 - FI - FI - dst[31:0] := tmp_src1[31:0] * POW(2, FLOOR(tmp_src2[31:0])) - RETURN dst[31:0] -} - FOR j := 0 to 3 i := j*32 - IF k[j] - dst[i+31:i] := SCALE(a[i+31:0], b[i+31:i]) + IF ((j & 1) == 0) + dst[i+31:i] := a[i+31:i] - b[i+31:i] ELSE - dst[i+31:i] := 0 + dst[i+31:i] := a[i+31:i] + b[i+31:i] FI ENDFOR -dst[MAX:128] := 0 - -
immintrin.h
+ +
pmmintrin.h
 - + Floating Point - AVX512VL - AVX512F - Miscellaneous - - - Scale the packed single-precision (32-bit) floating-point elements in "a" using values from "b", and store the results in "dst". + SSE3 + Arithmetic + + + + Alternatively add and subtract packed double-precision (64-bit) floating-point elements in "a" to/from packed elements in "b", and store the results in "dst". -SCALE(src1, src2){ - IF (src2 == NaN) - IF (src2 == SNaN) - RETURN QNAN(src2) - FI - ELSE IF (src1 == NaN) - IF (src1 == SNaN) - RETURN QNAN(src1) - FI - IF (src2 != INF) - RETURN QNAN(src1) - FI +FOR j := 0 to 1 + i := j*64 + IF ((j & 1) == 0) + dst[i+63:i] := a[i+63:i] - b[i+63:i] ELSE - tmp_src2 := src2 - tmp_src1 := src1 - IF (src2 is denormal AND MXCSR.DAZ) - tmp_src2 := 0 - FI - IF (src1 is denormal AND MXCSR.DAZ) - tmp_src1 := 0 - FI + dst[i+63:i] := a[i+63:i] + b[i+63:i] FI - dst[31:0] := tmp_src1[31:0] * POW(2, FLOOR(tmp_src2[31:0])) - RETURN dst[31:0] -} - -FOR j := 0 to 3 - i := j*32 - dst[i+31:i] := SCALE(a[i+31:0], b[i+31:i]) ENDFOR -dst[MAX:128] := 0 - -
immintrin.h
+ +
pmmintrin.h
 - + Floating Point - AVX512VL - AVX512F - Store - - - - - Scatter double-precision (64-bit) floating-point elements from "a" into memory using 32-bit indices. 64-bit elements are stored at addresses starting at "base_addr" and offset by each 32-bit element in "vindex" (each index is scaled by the factor in "scale"). "scale" should be 1, 2, 4 or 8. - + SSE3 + Arithmetic + + + + Horizontally add adjacent pairs of double-precision (64-bit) floating-point elements in "a" and "b", and pack the results in "dst". -FOR j := 0 to 3 - i := j*64 - l := j*32 - MEM[base_addr + SignExtend(vindex[l+31:l])*scale] := a[i+63:i] -ENDFOR +dst[63:0] := a[127:64] + a[63:0] +dst[127:64] := b[127:64] + b[63:0] - -
immintrin.h
+ +
pmmintrin.h
 - + Floating Point - AVX512VL - AVX512F - Store - - - - - - Scatter double-precision (64-bit) floating-point elements from "a" into memory using 32-bit indices. 64-bit elements are stored at addresses starting at "base_addr" and offset by each 32-bit element in "vindex" (each index is scaled by the factor in "scale") subject to mask "k" (elements are not stored when the corresponding mask bit is not set). "scale" should be 1, 2, 4 or 8. - + SSE3 + Arithmetic + + + + Horizontally add adjacent pairs of single-precision (32-bit) floating-point elements in "a" and "b", and pack the results in "dst". -FOR j := 0 to 3 - i := j*64 - l := j*32 - IF k[j] - MEM[base_addr + SignExtend(vindex[l+31:l])*scale] := a[i+63:i] - k[j] := 0 - FI -ENDFOR -k[MAX:4] := 0 +dst[31:0] := a[63:32] + a[31:0] +dst[63:32] := a[127:96] + a[95:64] +dst[95:64] := b[63:32] + b[31:0] +dst[127:96] := b[127:96] + b[95:64] - -
immintrin.h
+ +
pmmintrin.h
 - + Floating Point - AVX512VL - AVX512F - Store - - - - - Scatter double-precision (64-bit) floating-point elements from "a" into memory using 32-bit indices. 64-bit elements are stored at addresses starting at "base_addr" and offset by each 32-bit element in "vindex" (each index is scaled by the factor in "scale"). "scale" should be 1, 2, 4 or 8. - + SSE3 + Arithmetic + + + + Horizontally subtract adjacent pairs of double-precision (64-bit) floating-point elements in "a" and "b", and pack the results in "dst". -FOR j := 0 to 1 - i := j*64 - l := j*32 - MEM[base_addr + SignExtend(vindex[l+31:l])*scale] := a[i+63:i] -ENDFOR +dst[63:0] := a[63:0] - a[127:64] +dst[127:64] := b[63:0] - b[127:64] - -
immintrin.h
+ +
pmmintrin.h
 - + Floating Point - AVX512VL - AVX512F - Store - - - - - - Scatter double-precision (64-bit) floating-point elements from "a" into memory using 32-bit indices. 64-bit elements are stored at addresses starting at "base_addr" and offset by each 32-bit element in "vindex" (each index is scaled by the factor in "scale") subject to mask "k" (elements are not stored when the corresponding mask bit is not set). "scale" should be 1, 2, 4 or 8. - + SSE3 + Arithmetic + + + + Horizontally add adjacent pairs of single-precision (32-bit) floating-point elements in "a" and "b", and pack the results in "dst". -FOR j := 0 to 1 - i := j*64 - l := j*32 - IF k[j] - MEM[base_addr + SignExtend(vindex[l+31:l])*scale] := a[i+63:i] - k[j] := 0 - FI -ENDFOR -k[MAX:2] := 0 +dst[31:0] := a[31:0] - a[63:32] +dst[63:32] := a[95:64] - a[127:96] +dst[95:64] := b[31:0] - b[63:32] +dst[127:96] := b[95:64] - b[127:96] - -
immintrin.h
+ +
pmmintrin.h
+ + + Integer + SSE3 + Load + + + Load 128-bits of integer data from unaligned memory into "dst". This intrinsic may perform better than "_mm_loadu_si128" when the data crosses a cache line boundary. + +dst[127:0] := MEM[mem_addr+127:mem_addr] + + +
pmmintrin.h
 - + Floating Point - AVX512VL - AVX512F - Store - - - - - Scatter single-precision (32-bit) floating-point elements from "a" into memory using 32-bit indices. 32-bit elements are stored at addresses starting at "base_addr" and offset by each 32-bit element in "vindex" (each index is scaled by the factor in "scale"). "scale" should be 1, 2, 4 or 8. + SSE3 + Move + + + Duplicate the low double-precision (64-bit) floating-point element from "a", and store the results in "dst". -FOR j := 0 to 7 - i := j*32 - MEM[base_addr + SignExtend(vindex[i+31:i])*scale] := a[i+31:i] -ENDFOR +dst[63:0] := a[63:0] +dst[127:64] := a[63:0] - -
immintrin.h
+ +
pmmintrin.h
 - + Floating Point - AVX512VL - AVX512F - Store - - - - - - Scatter single-precision (32-bit) floating-point elements from "a" into memory using 32-bit indices. 32-bit elements are stored at addresses starting at "base_addr" and offset by each 32-bit element in "vindex" (each index is scaled by the factor in "scale") subject to mask "k" (elements are not stored when the corresponding mask bit is not set). "scale" should be 1, 2, 4 or 8. + SSE3 + Load + + + Load a double-precision (64-bit) floating-point element from memory into both elements of "dst". -FOR j := 0 to 7 - i := j*32 - IF k[j] - MEM[base_addr + SignExtend(vindex[i+31:i])*scale] := a[i+31:i] - k[j] := 0 - FI -ENDFOR -k[MAX:8] := 0 +dst[63:0] := MEM[mem_addr+63:mem_addr] +dst[127:64] := MEM[mem_addr+63:mem_addr] - -
immintrin.h
+ +
pmmintrin.h
 - + Floating Point - AVX512VL - AVX512F - Store - - - - - Scatter single-precision (32-bit) floating-point elements from "a" into memory using 32-bit indices. 32-bit elements are stored at addresses starting at "base_addr" and offset by each 32-bit element in "vindex" (each index is scaled by the factor in "scale"). "scale" should be 1, 2, 4 or 8. + SSE3 + Move + + + Duplicate odd-indexed single-precision (32-bit) floating-point elements from "a", and store the results in "dst". -FOR j := 0 to 3 - i := j*32 - MEM[base_addr + SignExtend(vindex[i+31:i])*scale] := a[i+31:i] +dst[31:0] := a[63:32] +dst[63:32] := a[63:32] +dst[95:64] := a[127:96] +dst[127:96] := a[127:96] + + +
pmmintrin.h
+ + + Floating Point + SSE3 + Move + + + Duplicate even-indexed single-precision (32-bit) floating-point elements from "a", and store the results in "dst". + +dst[31:0] := a[31:0] +dst[63:32] := a[31:0] +dst[95:64] := a[95:64] +dst[127:96] := a[95:64] + + +
pmmintrin.h
+ + + Floating Point + SSE4.1 + Swizzle + + + + + Blend packed double-precision (64-bit) floating-point elements from "a" and "b" using control mask "imm8", and store the results in "dst". + +FOR j := 0 to 1 + i := j*64 + IF imm8[j] + dst[i+63:i] := b[i+63:i] + ELSE + dst[i+63:i] := a[i+63:i] + FI ENDFOR - -
immintrin.h
+ +
smmintrin.h
 - + Floating Point - AVX512VL - AVX512F - Store - - - - - - Scatter single-precision (32-bit) floating-point elements from "a" into memory using 32-bit indices. 32-bit elements are stored at addresses starting at "base_addr" and offset by each 32-bit element in "vindex" (each index is scaled by the factor in "scale") subject to mask "k" (elements are not stored when the corresponding mask bit is not set). "scale" should be 1, 2, 4 or 8. + SSE4.1 + Swizzle + + + + + Blend packed single-precision (32-bit) floating-point elements from "a" and "b" using control mask "imm8", and store the results in "dst". FOR j := 0 to 3 i := j*32 - IF k[j] - MEM[base_addr + SignExtend(vindex[i+31:i])*scale] := a[i+31:i] - k[j] := 0 + IF imm8[j] + dst[i+31:i] := b[i+31:i] + ELSE + dst[i+31:i] := a[i+31:i] FI ENDFOR -k[MAX:4] := 0 - -
immintrin.h
+ +
smmintrin.h
 - + Floating Point - AVX512VL - AVX512F - Store - - - - - Scatter double-precision (64-bit) floating-point elements from "a" into memory using 64-bit indices. 64-bit elements are stored at addresses starting at "base_addr" and offset by each 64-bit element in "vindex" (each index is scaled by the factor in "scale"). "scale" should be 1, 2, 4 or 8. - + SSE4.1 + Swizzle + + + + + Blend packed double-precision (64-bit) floating-point elements from "a" and "b" using "mask", and store the results in "dst". -FOR j := 0 to 3 +FOR j := 0 to 1 i := j*64 - MEM[base_addr + SignExtend(vindex[i+63:i])*scale] := a[i+63:i] + IF mask[i+63] + dst[i+63:i] := b[i+63:i] + ELSE + dst[i+63:i] := a[i+63:i] + FI ENDFOR - -
immintrin.h
+ +
smmintrin.h
 - + Floating Point - AVX512VL - AVX512F - Store - - - - - - Scatter double-precision (64-bit) floating-point elements from "a" into memory using 64-bit indices. 64-bit elements are stored at addresses starting at "base_addr" and offset by each 64-bit element in "vindex" (each index is scaled by the factor in "scale") subject to mask "k" (elements are not stored when the corresponding mask bit is not set). "scale" should be 1, 2, 4 or 8. - + SSE4.1 + Swizzle + + + + + Blend packed single-precision (32-bit) floating-point elements from "a" and "b" using "mask", and store the results in "dst". FOR j := 0 to 3 - i := j*64 - IF k[j] - MEM[base_addr + SignExtend(vindex[i+63:i])*scale] := a[i+63:i] - k[j] := 0 + i := j*32 + IF mask[i+31] + dst[i+31:i] := b[i+31:i] + ELSE + dst[i+31:i] := a[i+31:i] FI ENDFOR -k[MAX:4] := 0 - -
immintrin.h
+ +
smmintrin.h
 - - Floating Point - AVX512VL - AVX512F - Store - - - - - Scatter double-precision (64-bit) floating-point elements from "a" into memory using 64-bit indices. 64-bit elements are stored at addresses starting at "base_addr" and offset by each 64-bit element in "vindex" (each index is scaled by the factor in "scale"). "scale" should be 1, 2, 4 or 8. - + + Integer + SSE4.1 + Swizzle + + + + + Blend packed 8-bit integers from "a" and "b" using "mask", and store the results in "dst". -FOR j := 0 to 1 - i := j*64 - MEM[base_addr + SignExtend(vindex[i+63:i])*scale] := a[i+63:i] +FOR j := 0 to 15 + i := j*8 + IF mask[i+7] + dst[i+7:i] := b[i+7:i] + ELSE + dst[i+7:i] := a[i+7:i] + FI ENDFOR - -
immintrin.h
+ +
smmintrin.h
 - - Floating Point - AVX512VL - AVX512F - Store - - - - - - Scatter double-precision (64-bit) floating-point elements from "a" into memory using 64-bit indices. 64-bit elements are stored at addresses starting at "base_addr" and offset by each 64-bit element in "vindex" (each index is scaled by the factor in "scale") subject to mask "k" (elements are not stored when the corresponding mask bit is not set). "scale" should be 1, 2, 4 or 8. - + + Integer + SSE4.1 + Swizzle + + + + + Blend packed 16-bit integers from "a" and "b" using control mask "imm8", and store the results in "dst". -FOR j := 0 to 1 - i := j*64 - IF k[j] - MEM[base_addr + SignExtend(vindex[i+63:i])*scale] := a[i+63:i] - k[j] := 0 +FOR j := 0 to 7 + i := j*16 + IF imm8[j] + dst[i+15:i] := b[i+15:i] + ELSE + dst[i+15:i] := a[i+15:i] FI ENDFOR -k[MAX:2] := 0 - -
immintrin.h
+ +
smmintrin.h
 - + Floating Point - AVX512VL - AVX512F - Store - - - - - Scatter single-precision (32-bit) floating-point elements from "a" into memory using 64-bit indices. 32-bit elements are stored at addresses starting at "base_addr" and offset by each 64-bit element in "vindex" (each index is scaled by the factor in "scale") subject to mask "k" (elements are not stored when the corresponding mask bit is not set). "scale" should be 1, 2, 4 or 8. - + SSE4.1 + Arithmetic + + + + + Conditionally multiply the packed double-precision (64-bit) floating-point elements in "a" and "b" using the high 4 bits in "imm8", sum the four products, and conditionally store the sum in "dst" using the low 4 bits of "imm8". + +DEFINE DP(a[127:0], b[127:0], imm8[7:0]) { + FOR j := 0 to 1 + i := j*64 + IF imm8[(4+j)%8] + temp[i+63:i] := a[i+63:i] * b[i+63:i] + ELSE + temp[i+63:i] := 0.0 + FI + ENDFOR + + sum[63:0] := temp[127:64] + temp[63:0] + + FOR j := 0 to 1 + i := j*64 + IF imm8[j%8] + tmpdst[i+63:i] := sum[63:0] + ELSE + tmpdst[i+63:i] := 0.0 + FI + ENDFOR + RETURN tmpdst[127:0] +} +dst[127:0] := DP(a[127:0], b[127:0], imm8[7:0]) + + +
smmintrin.h
+ + + Floating Point + SSE4.1 + Arithmetic + + + + + Conditionally multiply the packed single-precision (32-bit) floating-point elements in "a" and "b" using the high 4 bits in "imm8", sum the four products, and conditionally store the sum in "dst" using the low 4 bits of "imm8". + +DEFINE DP(a[127:0], b[127:0], imm8[7:0]) { + FOR j := 0 to 3 + i := j*32 + IF imm8[(4+j)%8] + temp[i+31:i] := a[i+31:i] * b[i+31:i] + ELSE + temp[i+31:i] := 0 + FI + ENDFOR + + sum[31:0] := (temp[127:96] + temp[95:64]) + (temp[63:32] + temp[31:0]) + + FOR j := 0 to 3 + i := j*32 + IF imm8[j%8] + tmpdst[i+31:i] := sum[31:0] + ELSE + tmpdst[i+31:i] := 0 + FI + ENDFOR + RETURN tmpdst[127:0] +} +dst[127:0] := DP(a[127:0], b[127:0], imm8[7:0]) + + +
smmintrin.h
+ + + Floating Point + SSE4.1 + Swizzle + + + + Extract a single-precision (32-bit) floating-point element from "a", selected with "imm8", and store the result in "dst". -FOR j := 0 to 3 - i := j*32 - l := j*64 - MEM[base_addr + SignExtend(vindex[l+63:l])*scale] := a[i+31:i] -ENDFOR +dst[31:0] := (a[127:0] >> (imm8[1:0] * 32))[31:0] - -
immintrin.h
+ +
smmintrin.h
 - - Floating Point - AVX512VL - AVX512F - Store - - - - - - Scatter single-precision (32-bit) floating-point elements from "a" into memory using 64-bit indices. 32-bit elements are stored at addresses starting at "base_addr" and offset by each 64-bit element in "vindex" (each index is scaled by the factor in "scale") subject to mask "k" (elements are not stored when the corresponding mask bit is not set). "scale" should be 1, 2, 4 or 8. - + + Integer + SSE4.1 + Swizzle + + + + Extract an 8-bit integer from "a", selected with "imm8", and store the result in the lower element of "dst". -FOR j := 0 to 3 - i := j*32 - l := j*64 - IF k[j] - MEM[base_addr + SignExtend(vindex[l+63:l])*scale] := a[i+31:i] - k[j] := 0 - FI -ENDFOR -k[MAX:4] := 0 +dst[7:0] := (a[127:0] >> (imm8[3:0] * 8))[7:0] +dst[31:8] := 0 - -
immintrin.h
+ +
smmintrin.h
 - - Floating Point - AVX512VL - AVX512F - Store - - - - - Scatter single-precision (32-bit) floating-point elements from "a" into memory using 64-bit indices. 32-bit elements are stored at addresses starting at "base_addr" and offset by each 64-bit element in "vindex" (each index is scaled by the factor in "scale") subject to mask "k" (elements are not stored when the corresponding mask bit is not set). "scale" should be 1, 2, 4 or 8. - + + Integer + SSE4.1 + Swizzle + + + + Extract a 32-bit integer from "a", selected with "imm8", and store the result in "dst". -FOR j := 0 to 1 - i := j*32 - l := j*64 - MEM[base_addr + SignExtend(vindex[l+63:l])*scale] := a[i+31:i] -ENDFOR +dst[31:0] := (a[127:0] >> (imm8[1:0] * 32))[31:0] - -
immintrin.h
+ +
smmintrin.h
 - - Floating Point - AVX512VL - AVX512F - Store - - - - - - Scatter single-precision (32-bit) floating-point elements from "a" into memory using 64-bit indices. 32-bit elements are stored at addresses starting at "base_addr" and offset by each 64-bit element in "vindex" (each index is scaled by the factor in "scale") subject to mask "k" (elements are not stored when the corresponding mask bit is not set). "scale" should be 1, 2, 4 or 8. - + + Integer + SSE4.1 + Swizzle + + + + Extract a 64-bit integer from "a", selected with "imm8", and store the result in "dst". -FOR j := 0 to 1 - i := j*32 - l := j*64 - IF k[j] - MEM[base_addr + SignExtend(vindex[l+63:l])*scale] := a[i+31:i] - k[j] := 0 - FI -ENDFOR -k[MAX:2] := 0 +dst[63:0] := (a[127:0] >> (imm8[0] * 64))[63:0] - -
immintrin.h
+ +
smmintrin.h
 - + Floating Point - AVX512VL - AVX512F - Miscellaneous - - - - - - Shuffle 128-bits (composed of 4 single-precision (32-bit) floating-point elements) selected by "imm8" from "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + SSE4.1 + Swizzle + + + + + Copy "a" to "tmp", then insert a single-precision (32-bit) floating-point element from "b" into "tmp" using the control in "imm8". Store "tmp" to "dst" using the mask in "imm8" (elements are zeroed out when the corresponding bit is set). -SELECT2(src, control){ - CASE(control[0]) - 0: tmp[127:0] := src[127:0] - 1: tmp[127:0] := src[255:128] - ESAC - RETURN tmp[127:0] -} - -tmp_dst[127:0] := SELECT2(a[255:0], imm8[0]) -tmp_dst[255:128] := SELECT2(b[255:0], imm8[1]) - -FOR j := 0 to 7 +tmp2[127:0] := a[127:0] +CASE (imm8[7:6]) OF +0: tmp1[31:0] := b[31:0] +1: tmp1[31:0] := b[63:32] +2: tmp1[31:0] := b[95:64] +3: tmp1[31:0] := b[127:96] +ESAC +CASE (imm8[5:4]) OF +0: tmp2[31:0] := tmp1[31:0] +1: tmp2[63:32] := tmp1[31:0] +2: tmp2[95:64] := tmp1[31:0] +3: tmp2[127:96] := tmp1[31:0] +ESAC +FOR j := 0 to 3 i := j*32 - IF k[j] - dst[i+31:i] := tmp_dst[i+31:i] + IF imm8[j%8] + dst[i+31:i] := 0 ELSE - dst[i+31:i] := src[i+31:i] + dst[i+31:i] := tmp2[i+31:i] FI ENDFOR -dst[MAX:256] := 0 - -
immintrin.h
+ +
smmintrin.h
 - - Floating Point - AVX512VL - AVX512F - Miscellaneous - - - - - Shuffle 128-bits (composed of 4 single-precision (32-bit) floating-point elements) selected by "imm8" from "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Integer + SSE4.1 + Swizzle + + + + + Copy "a" to "dst", and insert the lower 8-bit integer from "i" into "dst" at the location specified by "imm8". -SELECT2(src, control){ - CASE(control[0]) - 0: tmp[127:0] := src[127:0] - 1: tmp[127:0] := src[255:128] - ESAC - RETURN tmp[127:0] -} - -tmp_dst[127:0] := SELECT2(a[255:0], imm8[0]) -tmp_dst[255:128] := SELECT2(b[255:0], imm8[1]) - -FOR j := 0 to 7 - i := j*32 - IF k[j] - dst[i+31:i] := tmp_dst[i+31:i] - ELSE - dst[i+31:i] := 0 - FI -ENDFOR -dst[MAX:256] := 0 +dst[127:0] := a[127:0] +sel := imm8[3:0]*8 +dst[sel+7:sel] := i[7:0] - -
immintrin.h
+ +
smmintrin.h
 - - Floating Point - AVX512VL - AVX512F - Miscellaneous - - - - Shuffle 128-bits (composed of 4 single-precision (32-bit) floating-point elements) selected by "imm8" from "a" and "b", and store the results in "dst". - + + Integer + SSE4.1 + Swizzle + + + + + Copy "a" to "dst", and insert the 32-bit integer "i" into "dst" at the location specified by "imm8". -SELECT2(src, control){ - CASE(control[0]) - 0: tmp[127:0] := src[127:0] - 1: tmp[127:0] := src[255:128] - ESAC - RETURN tmp[127:0] -} - -dst[127:0] := SELECT2(a[255:0], imm8[0]) -dst[255:128] := SELECT2(b[255:0], imm8[1]) -dst[MAX:256] := 0 +dst[127:0] := a[127:0] +sel := imm8[1:0]*32 +dst[sel+31:sel] := i[31:0] - -
immintrin.h
+ +
smmintrin.h
 - - Floating Point - AVX512VL - AVX512F - Miscellaneous - - - - - - Shuffle 128-bits (composed of 2 double-precision (64-bit) floating-point elements) selected by "imm8" from "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Integer + SSE4.1 + Swizzle + + + + + Copy "a" to "dst", and insert the 64-bit integer "i" into "dst" at the location specified by "imm8". -SELECT4(src, control){ - CASE(control[0]) - 0: tmp[127:0] := src[127:0] - 1: tmp[127:0] := src[255:128] - ESAC - RETURN tmp[127:0] -} - -tmp_dst[127:0] := SELECT2(a[255:0], imm8[0]) -tmp_dst[255:128] := SELECT2(b[255:0], imm8[1]) - -FOR j := 0 to 3 - i := j*64 - IF k[j] - dst[i+63:i] := tmp_dst[i+63:i] - ELSE - dst[i+63:i] := src[i+63:i] - FI -ENDFOR -dst[MAX:256] := 0 +dst[127:0] := a[127:0] +sel := imm8[0]*64 +dst[sel+63:sel] := i[63:0] - -
immintrin.h
+ +
smmintrin.h
 - - Floating Point - AVX512VL - AVX512F - Miscellaneous - - - - - Shuffle 128-bits (composed of 2 double-precision (64-bit) floating-point elements) selected by "imm8" from "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Integer + SSE4.1 + Special Math Functions + + + + Compare packed signed 8-bit integers in "a" and "b", and store packed maximum values in "dst". -SELECT4(src, control){ - CASE(control[0]) - 0: tmp[127:0] := src[127:0] - 1: tmp[127:0] := src[255:128] - ESAC - RETURN tmp[127:0] -} - -tmp_dst[127:0] := SELECT2(a[255:0], imm8[0]) -tmp_dst[255:128] := SELECT2(b[255:0], imm8[1]) - -FOR j := 0 to 3 - i := j*64 - IF k[j] - dst[i+63:i] := tmp_dst[i+63:i] - ELSE - dst[i+63:i] := 0 - FI +FOR j := 0 to 15 + i := j*8 + dst[i+7:i] := MAX(a[i+7:i], b[i+7:i]) ENDFOR -dst[MAX:256] := 0 - -
immintrin.h
+ +
smmintrin.h
 - - Floating Point - AVX512VL - AVX512F - Miscellaneous - - - - Shuffle 128-bits (composed of 2 double-precision (64-bit) floating-point elements) selected by "imm8" from "a" and "b", and store the results in "dst". + + Integer + SSE4.1 + Special Math Functions + + + + Compare packed signed 32-bit integers in "a" and "b", and store packed maximum values in "dst". -SELECT4(src, control){ - CASE(control[0]) - 0: tmp[127:0] := src[127:0] - 1: tmp[127:0] := src[255:128] - ESAC - RETURN tmp[127:0] -} - -dst[127:0] := SELECT2(a[255:0], imm8[0]) -dst[255:128] := SELECT2(b[255:0], imm8[1]) -dst[MAX:256] := 0 +FOR j := 0 to 3 + i := j*32 + dst[i+31:i] := MAX(a[i+31:i], b[i+31:i]) +ENDFOR - -
immintrin.h
+ +
smmintrin.h
 - - AVX512VL - AVX512F - Miscellaneous - - - - - - Shuffle 128-bits (composed of 4 32-bit integers) selected by "imm8" from "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Integer + SSE4.1 + Special Math Functions + + + + Compare packed unsigned 32-bit integers in "a" and "b", and store packed maximum values in "dst". -SELECT2(src, control){ - CASE(control[0]) - 0: tmp[127:0] := src[127:0] - 1: tmp[127:0] := src[255:128] - ESAC - RETURN tmp[127:0] -} - -tmp_dst[127:0] := SELECT2(a[255:0], imm8[1:0]) -tmp_dst[255:128] := SELECT2(b[255:0], imm8[3:2]) - -FOR j := 0 to 7 +FOR j := 0 to 3 i := j*32 - IF k[j] - dst[i+31:i] := tmp_dst[i+31:i] - ELSE - dst[i+31:i] := src[i+31:i] - FI + dst[i+31:i] := MAX(a[i+31:i], b[i+31:i]) ENDFOR -dst[MAX:256] := 0 - -
immintrin.h
+ +
smmintrin.h
 - - AVX512VL - AVX512F - Miscellaneous - - - - - Shuffle 128-bits (composed of 4 32-bit integers) selected by "imm8" from "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Integer + SSE4.1 + Special Math Functions + + + + Compare packed unsigned 16-bit integers in "a" and "b", and store packed maximum values in "dst". -SELECT2(src, control){ - CASE(control[0]) - 0: tmp[127:0] := src[127:0] - 1: tmp[127:0] := src[255:128] - ESAC - RETURN tmp[127:0] -} - -tmp_dst[127:0] := SELECT2(a[255:0], imm8[1:0]) -tmp_dst[255:128] := SELECT2(b[255:0], imm8[3:2]) - FOR j := 0 to 7 - i := j*32 - IF k[j] - dst[i+31:i] := tmp_dst[i+31:i] - ELSE - dst[i+31:i] := 0 - FI + i := j*16 + dst[i+15:i] := MAX(a[i+15:i], b[i+15:i]) ENDFOR -dst[MAX:256] := 0 - -
immintrin.h
+ +
smmintrin.h
 - - AVX512VL - AVX512F - Miscellaneous - - - - Shuffle 128-bits (composed of 4 32-bit integers) selected by "imm8" from "a" and "b", and store the results in "dst". + + Integer + SSE4.1 + Special Math Functions + + + + Compare packed signed 8-bit integers in "a" and "b", and store packed minimum values in "dst". -SELECT2(src, control){ - CASE(control[0]) - 0: tmp[127:0] := src[127:0] - 1: tmp[127:0] := src[255:128] - ESAC - RETURN tmp[127:0] -} - -dst[127:0] := SELECT2(a[255:0], imm8[1:0]) -dst[255:128] := SELECT2(b[255:0], imm8[3:2]) -dst[MAX:256] := 0 +FOR j := 0 to 15 + i := j*8 + dst[i+7:i] := MIN(a[i+7:i], b[i+7:i]) +ENDFOR - -
immintrin.h
+ +
smmintrin.h
 - - AVX512VL - AVX512F - Miscellaneous - - - - - - Shuffle 128-bits (composed of 2 64-bit integers) selected by "imm8" from "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Integer + SSE4.1 + Special Math Functions + + + + Compare packed signed 32-bit integers in "a" and "b", and store packed minimum values in "dst". -SELECT2(src, control){ - CASE(control[0]) - 0: tmp[127:0] := src[127:0] - 1: tmp[127:0] := src[255:128] - ESAC - RETURN tmp[127:0] -} - -tmp_dst[127:0] := SELECT2(a[255:0], imm8[1:0]) -tmp_dst[255:128] := SELECT2(b[255:0], imm8[3:2]) - FOR j := 0 to 3 - i := j*64 - IF k[j] - dst[i+63:i] := tmp_dst[i+63:i] - ELSE - dst[i+63:i] := src[i+63:i] - FI + i := j*32 + dst[i+31:i] := MIN(a[i+31:i], b[i+31:i]) ENDFOR -dst[MAX:256] := 0 - -
immintrin.h
+ +
smmintrin.h
 - - AVX512VL - AVX512F - Miscellaneous - - - - - Shuffle 128-bits (composed of 2 64-bit integers) selected by "imm8" from "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Integer + SSE4.1 + Special Math Functions + + + + Compare packed unsigned 32-bit integers in "a" and "b", and store packed minimum values in "dst". -SELECT2(src, control){ - CASE(control[0]) - 0: tmp[127:0] := src[127:0] - 1: tmp[127:0] := src[255:128] - ESAC - RETURN tmp[127:0] -} - -tmp_dst[127:0] := SELECT2(a[255:0], imm8[1:0]) -tmp_dst[255:128] := SELECT2(b[255:0], imm8[3:2]) - FOR j := 0 to 3 - i := j*64 - IF k[j] - dst[i+63:i] := tmp_dst[i+63:i] - ELSE - dst[i+63:i] := 0 - FI + i := j*32 + dst[i+31:i] := MIN(a[i+31:i], b[i+31:i]) ENDFOR -dst[MAX:256] := 0 - -
immintrin.h
+ +
smmintrin.h
 - - AVX512VL - AVX512F - Miscellaneous - - - - Shuffle 128-bits (composed of 2 64-bit integers) selected by "imm8" from "a" and "b", and store the results in "dst". + + Integer + SSE4.1 + Special Math Functions + + + + Compare packed unsigned 16-bit integers in "a" and "b", and store packed minimum values in "dst". -SELECT2(src, control){ - CASE(control[0]) - 0: tmp[127:0] := src[127:0] - 1: tmp[127:0] := src[255:128] - ESAC - RETURN tmp[127:0] -} - -dst[127:0] := SELECT2(a[255:0], imm8[1:0]) -dst[255:128] := SELECT2(b[255:0], imm8[3:2]) -dst[MAX:256] := 0 +FOR j := 0 to 7 + i := j*16 + dst[i+15:i] := MIN(a[i+15:i], b[i+15:i]) +ENDFOR - -
immintrin.h
+ +
smmintrin.h
 - - Floating Point - AVX512VL - AVX512F + + Integer + SSE4.1 + Convert Miscellaneous - - - - - - Shuffle double-precision (64-bit) floating-point elements within 128-bit lanes using the control in "imm8", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + + + Convert packed signed 32-bit integers from "a" and "b" to packed 16-bit integers using unsigned saturation, and store the results in "dst". -tmp_dst[63:0] := (imm8[0] == 0) ? a[63:0] : a[127:64] -tmp_dst[127:64] := (imm8[1] == 0) ? b[63:0] : b[127:64] -tmp_dst[191:128] := (imm8[2] == 0) ? a[191:128] : a[255:192] -tmp_dst[255:192] := (imm8[3] == 0) ? b[191:128] : b[255:192] - -FOR j := 0 to 3 +dst[15:0] := SaturateU16(a[31:0]) +dst[31:16] := SaturateU16(a[63:32]) +dst[47:32] := SaturateU16(a[95:64]) +dst[63:48] := SaturateU16(a[127:96]) +dst[79:64] := SaturateU16(b[31:0]) +dst[95:80] := SaturateU16(b[63:32]) +dst[111:96] := SaturateU16(b[95:64]) +dst[127:112] := SaturateU16(b[127:96]) + + +
smmintrin.h
+ + + Integer + SSE4.1 + Compare + + + + Compare packed 64-bit integers in "a" and "b" for equality, and store the results in "dst". + +FOR j := 0 to 1 i := j*64 - IF k[j] - dst[i+63:i] := tmp_dst[i+63:i] - ELSE - dst[i+63:i] := src[i+63:i] - FI + dst[i+63:i] := ( a[i+63:i] == b[i+63:i] ) ? 0xFFFFFFFFFFFFFFFF : 0 ENDFOR -dst[MAX:256] := 0 - -
immintrin.h
+ +
smmintrin.h
 - - Floating Point - AVX512VL - AVX512F - Miscellaneous - - - - - Shuffle double-precision (64-bit) floating-point elements within 128-bit lanes using the control in "imm8", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Integer + SSE4.1 + Convert + + + Sign extend packed 8-bit integers in "a" to packed 16-bit integers, and store the results in "dst". + +FOR j := 0 to 7 + i := j*8 + l := j*16 + dst[l+15:l] := SignExtend16(a[i+7:i]) +ENDFOR + + +
smmintrin.h
+ + + Integer + SSE4.1 + Convert + + + Sign extend packed 8-bit integers in "a" to packed 32-bit integers, and store the results in "dst". -tmp_dst[63:0] := (imm8[0] == 0) ? a[63:0] : a[127:64] -tmp_dst[127:64] := (imm8[1] == 0) ? b[63:0] : b[127:64] -tmp_dst[191:128] := (imm8[2] == 0) ? a[191:128] : a[255:192] -tmp_dst[255:192] := (imm8[3] == 0) ? b[191:128] : b[255:192] - FOR j := 0 to 3 - i := j*64 - IF k[j] - dst[i+63:i] := tmp_dst[i+63:i] - ELSE - dst[i+63:i] := 0 - FI + i := 32*j + k := 8*j + dst[i+31:i] := SignExtend32(a[k+7:k]) ENDFOR -dst[MAX:256] := 0 - -
immintrin.h
+ +
smmintrin.h
 - - Floating Point - AVX512VL - AVX512F - Miscellaneous - - - - - - Shuffle double-precision (64-bit) floating-point elements using the control in "imm8", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Integer + SSE4.1 + Convert + + + Sign extend packed 8-bit integers in the low 8 bytes of "a" to packed 64-bit integers, and store the results in "dst". -tmp_dst[63:0] := (imm8[0] == 0) ? a[63:0] : a[127:64] -tmp_dst[127:64] := (imm8[1] == 0) ? b[63:0] : b[127:64] - FOR j := 0 to 1 - i := j*64 - IF k[j] - dst[i+63:i] := tmp_dst[i+63:i] - ELSE - dst[i+63:i] := src[i+63:i] - FI + i := 64*j + k := 8*j + dst[i+63:i] := SignExtend64(a[k+7:k]) ENDFOR -dst[MAX:128] := 0 - -
immintrin.h
+ +
smmintrin.h
 - - Floating Point - AVX512VL - AVX512F - Miscellaneous - - - - - Shuffle double-precision (64-bit) floating-point elements using the control in "imm8", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Integer + SSE4.1 + Convert + + + Sign extend packed 16-bit integers in "a" to packed 32-bit integers, and store the results in "dst". -tmp_dst[63:0] := (imm8[0] == 0) ? a[63:0] : a[127:64] -tmp_dst[127:64] := (imm8[1] == 0) ? b[63:0] : b[127:64] - -FOR j := 0 to 1 - i := j*64 - IF k[j] - dst[i+63:i] := tmp_dst[i+63:i] - ELSE - dst[i+63:i] := 0 - FI +FOR j := 0 to 3 + i := 32*j + k := 16*j + dst[i+31:i] := SignExtend32(a[k+15:k]) ENDFOR -dst[MAX:128] := 0 - -
immintrin.h
+ +
smmintrin.h
 - - Floating Point - AVX512VL - AVX512F - Miscellaneous - - - - - - Shuffle single-precision (32-bit) floating-point elements in "a" within 128-bit lanes using the control in "imm8", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Integer + SSE4.1 + Convert + + + Sign extend packed 16-bit integers in "a" to packed 64-bit integers, and store the results in "dst". -SELECT4(src, control){ - CASE(control[1:0]) - 0: tmp[31:0] := src[31:0] - 1: tmp[31:0] := src[63:32] - 2: tmp[31:0] := src[95:64] - 3: tmp[31:0] := src[127:96] - ESAC - RETURN tmp[31:0] -} - -tmp_dst[31:0] := SELECT4(a[127:0], imm8[1:0]) -tmp_dst[63:32] := SELECT4(a[127:0], imm8[3:2]) -tmp_dst[95:64] := SELECT4(b[127:0], imm8[5:4]) -tmp_dst[127:96] := SELECT4(b[127:0], imm8[7:6]) -tmp_dst[159:128] := SELECT4(a[255:128], imm8[1:0]) -tmp_dst[191:160] := SELECT4(a[255:128], imm8[3:2]) -tmp_dst[223:192] := SELECT4(b[255:128], imm8[5:4]) -tmp_dst[255:224] := SELECT4(b[255:128], imm8[7:6]) - -FOR j := 0 to 7 - i := j*32 - IF k[j] - dst[i+31:i] := tmp_dst[i+31:i] - ELSE - dst[i+31:i] := src[i+31:i] - FI +FOR j := 0 to 1 + i := 64*j + k := 16*j + dst[i+63:i] := SignExtend64(a[k+15:k]) ENDFOR -dst[MAX:256] := 0 - -
immintrin.h
+ +
smmintrin.h
 - - Floating Point - AVX512VL - AVX512F - Miscellaneous - - - - - Shuffle single-precision (32-bit) floating-point elements in "a" within 128-bit lanes using the control in "imm8", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Integer + SSE4.1 + Convert + + + Sign extend packed 32-bit integers in "a" to packed 64-bit integers, and store the results in "dst". + +FOR j := 0 to 1 + i := 64*j + k := 32*j + dst[i+63:i] := SignExtend64(a[k+31:k]) +ENDFOR + + +
smmintrin.h
+ + + Integer + SSE4.1 + Convert + + + Zero extend packed unsigned 8-bit integers in "a" to packed 16-bit integers, and store the results in "dst". -SELECT4(src, control){ - CASE(control[1:0]) - 0: tmp[31:0] := src[31:0] - 1: tmp[31:0] := src[63:32] - 2: tmp[31:0] := src[95:64] - 3: tmp[31:0] := src[127:96] - ESAC - RETURN tmp[31:0] -} - -tmp_dst[31:0] := SELECT4(a[127:0], imm8[1:0]) -tmp_dst[63:32] := SELECT4(a[127:0], imm8[3:2]) -tmp_dst[95:64] := SELECT4(b[127:0], imm8[5:4]) -tmp_dst[127:96] := SELECT4(b[127:0], imm8[7:6]) -tmp_dst[159:128] := SELECT4(a[255:128], imm8[1:0]) -tmp_dst[191:160] := SELECT4(a[255:128], imm8[3:2]) -tmp_dst[223:192] := SELECT4(b[255:128], imm8[5:4]) -tmp_dst[255:224] := SELECT4(b[255:128], imm8[7:6]) - FOR j := 0 to 7 - i := j*32 - IF k[j] - dst[i+31:i] := tmp_dst[i+31:i] - ELSE - dst[i+31:i] := 0 - FI + i := j*8 + l := j*16 + dst[l+15:l] := ZeroExtend16(a[i+7:i]) ENDFOR -dst[MAX:256] := 0 - -
immintrin.h
+ +
smmintrin.h
 - - Floating Point - AVX512VL - AVX512F - Miscellaneous - - - - - - Shuffle single-precision (32-bit) floating-point elements in "a" using the control in "imm8", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Integer + SSE4.1 + Convert + + + Zero extend packed unsigned 8-bit integers in "a" to packed 32-bit integers, and store the results in "dst". -SELECT4(src, control){ - CASE(control[1:0]) - 0: tmp[31:0] := src[31:0] - 1: tmp[31:0] := src[63:32] - 2: tmp[31:0] := src[95:64] - 3: tmp[31:0] := src[127:96] - ESAC - RETURN tmp[31:0] -} - -tmp_dst[31:0] := SELECT4(a[127:0], imm8[1:0]) -tmp_dst[63:32] := SELECT4(a[127:0], imm8[3:2]) -tmp_dst[95:64] := SELECT4(b[127:0], imm8[5:4]) -tmp_dst[127:96] := SELECT4(b[127:0], imm8[7:6]) - FOR j := 0 to 3 - i := j*32 - IF k[j] - dst[i+31:i] := tmp_dst[i+31:i] - ELSE - dst[i+31:i] := src[i+31:i] - FI + i := 32*j + k := 8*j + dst[i+31:i] := ZeroExtend32(a[k+7:k]) ENDFOR -dst[MAX:128] := 0 - -
immintrin.h
+ +
smmintrin.h
 - - Floating Point - AVX512VL - AVX512F - Miscellaneous - - - - - Shuffle single-precision (32-bit) floating-point elements in "a" using the control in "imm8", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Integer + SSE4.1 + Convert + + + Zero extend packed unsigned 8-bit integers in the low 8 byte sof "a" to packed 64-bit integers, and store the results in "dst". -SELECT4(src, control){ - CASE(control[1:0]) - 0: tmp[31:0] := src[31:0] - 1: tmp[31:0] := src[63:32] - 2: tmp[31:0] := src[95:64] - 3: tmp[31:0] := src[127:96] - ESAC - RETURN tmp[31:0] -} - -tmp_dst[31:0] := SELECT4(a[127:0], imm8[1:0]) -tmp_dst[63:32] := SELECT4(a[127:0], imm8[3:2]) -tmp_dst[95:64] := SELECT4(b[127:0], imm8[5:4]) -tmp_dst[127:96] := SELECT4(b[127:0], imm8[7:6]) - -FOR j := 0 to 3 - i := j*32 - IF k[j] - dst[i+31:i] := tmp_dst[i+31:i] - ELSE - dst[i+31:i] := 0 - FI +FOR j := 0 to 1 + i := 64*j + k := 8*j + dst[i+63:i] := ZeroExtend64(a[k+7:k]) ENDFOR -dst[MAX:128] := 0 - -
immintrin.h
+ +
smmintrin.h
 - - Floating Point - AVX512VL - AVX512F - Arithmetic - - - - Compute the square root of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Integer + SSE4.1 + Convert + + + Zero extend packed unsigned 16-bit integers in "a" to packed 32-bit integers, and store the results in "dst". FOR j := 0 to 3 - i := j*64 - IF k[j] - dst[i+63:i] := SQRT(a[i+63:i]) - ELSE - dst[i+63:i] := src[i+63:i] - FI + i := 32*j + k := 16*j + dst[i+31:i] := ZeroExtend32(a[k+15:k]) ENDFOR -dst[MAX:256] := 0 - -
immintrin.h
+ +
smmintrin.h
 - - Floating Point - AVX512VL - AVX512F - Arithmetic - - - Compute the square root of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Integer + SSE4.1 + Convert + + + Zero extend packed unsigned 16-bit integers in "a" to packed 64-bit integers, and store the results in "dst". -FOR j := 0 to 3 - i := j*64 - IF k[j] - dst[i+63:i] := SQRT(a[i+63:i]) - ELSE - dst[i+63:i] := 0 - FI +FOR j := 0 to 1 + i := 64*j + k := 16*j + dst[i+63:i] := ZeroExtend64(a[k+15:k]) ENDFOR -dst[MAX:256] := 0 - -
immintrin.h
+ +
smmintrin.h
 - - Floating Point - AVX512VL - AVX512F - Arithmetic - - - - Compute the square root of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Integer + SSE4.1 + Convert + + + Zero extend packed unsigned 32-bit integers in "a" to packed 64-bit integers, and store the results in "dst". FOR j := 0 to 1 - i := j*64 - IF k[j] - dst[i+63:i] := SQRT(a[i+63:i]) - ELSE - dst[i+63:i] := src[i+63:i] - FI + i := 64*j + k := 32*j + dst[i+63:i] := ZeroExtend64(a[k+31:k]) ENDFOR -dst[MAX:128] := 0 - -
immintrin.h
+ +
smmintrin.h
 - - Floating Point - AVX512VL - AVX512F + + Integer + SSE4.1 Arithmetic - - - Compute the square root of packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + + + Multiply the low signed 32-bit integers from each packed 64-bit element in "a" and "b", and store the signed 64-bit results in "dst". FOR j := 0 to 1 i := j*64 - IF k[j] - dst[i+63:i] := SQRT(a[i+63:i]) - ELSE - dst[i+63:i] := 0 - FI + dst[i+63:i] := SignExtend64(a[i+31:i]) * SignExtend64(b[i+31:i]) ENDFOR -dst[MAX:128] := 0 - -
immintrin.h
+ +
smmintrin.h
 - - Floating Point - AVX512VL - AVX512F + + Integer + SSE4.1 Arithmetic - - - - Compute the square root of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + + + Multiply the packed 32-bit integers in "a" and "b", producing intermediate 64-bit integers, and store the low 32 bits of the intermediate integers in "dst". -FOR j := 0 to 7 +FOR j := 0 to 3 i := j*32 - IF k[j] - dst[i+31:i] := SQRT(a[i+31:i]) - ELSE - dst[i+31:i] := src[i+31:i] - FI + tmp[63:0] := a[i+31:i] * b[i+31:i] + dst[i+31:i] := tmp[31:0] ENDFOR -dst[MAX:256] := 0 - -
immintrin.h
+ +
smmintrin.h
 - - Floating Point - AVX512VL - AVX512F - Arithmetic - - - Compute the square root of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Integer + Flag + SSE4.1 + Logical + + + + Compute the bitwise AND of 128 bits (representing integer data) in "a" and "b", and set "ZF" to 1 if the result is zero, otherwise set "ZF" to 0. Compute the bitwise NOT of "a" and then AND with "b", and set "CF" to 1 if the result is zero, otherwise set "CF" to 0. Return the "ZF" value. -FOR j := 0 to 7 - i := j*32 - IF k[j] - dst[i+31:i] := SQRT(a[i+31:i]) - ELSE - dst[i+31:i] := 0 - FI -ENDFOR -dst[MAX:256] := 0 +IF ((a[127:0] AND b[127:0]) == 0) + ZF := 1 +ELSE + ZF := 0 +FI +IF (((NOT a[127:0]) AND b[127:0]) == 0) + CF := 1 +ELSE + CF := 0 +FI +RETURN ZF - -
immintrin.h
+ +
smmintrin.h
 - - Floating Point - AVX512VL - AVX512F - Arithmetic - - - - Compute the square root of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Integer + Flag + SSE4.1 + Logical + + + + Compute the bitwise AND of 128 bits (representing integer data) in "a" and "b", and set "ZF" to 1 if the result is zero, otherwise set "ZF" to 0. Compute the bitwise NOT of "a" and then AND with "b", and set "CF" to 1 if the result is zero, otherwise set "CF" to 0. Return the "CF" value. -FOR j := 0 to 3 - i := j*32 - IF k[j] - dst[i+31:i] := SQRT(a[i+31:i]) - ELSE - dst[i+31:i] := src[i+31:i] - FI -ENDFOR -dst[MAX:128] := 0 +IF ((a[127:0] AND b[127:0]) == 0) + ZF := 1 +ELSE + ZF := 0 +FI +IF (((NOT a[127:0]) AND b[127:0]) == 0) + CF := 1 +ELSE + CF := 0 +FI +RETURN CF - -
immintrin.h
+ +
smmintrin.h
 - - Floating Point - AVX512VL - AVX512F - Arithmetic - - - Compute the square root of packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Integer + Flag + SSE4.1 + Logical + + + + Compute the bitwise AND of 128 bits (representing integer data) in "a" and "b", and set "ZF" to 1 if the result is zero, otherwise set "ZF" to 0. Compute the bitwise NOT of "a" and then AND with "b", and set "CF" to 1 if the result is zero, otherwise set "CF" to 0. Return 1 if both the "ZF" and "CF" values are zero, otherwise return 0. -FOR j := 0 to 3 - i := j*32 - IF k[j] - dst[i+31:i] := SQRT(a[i+31:i]) - ELSE - dst[i+31:i] := 0 - FI -ENDFOR -dst[MAX:128] := 0 +IF ((a[127:0] AND b[127:0]) == 0) + ZF := 1 +ELSE + ZF := 0 +FI +IF (((NOT a[127:0]) AND b[127:0]) == 0) + CF := 1 +ELSE + CF := 0 +FI +IF (ZF == 0 && CF == 0) + dst := 1 +ELSE + dst := 0 +FI - -
immintrin.h
+ +
smmintrin.h
 - - Floating Point - AVX512VL - AVX512F - Arithmetic - - - - - Subtract packed double-precision (64-bit) floating-point elements in "b" from packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Integer + Flag + SSE4.1 + Logical + + + + Compute the bitwise AND of 128 bits (representing integer data) in "a" and "mask", and return 1 if the result is zero, otherwise return 0. -FOR j := 0 to 3 - i := j*64 - IF k[j] - dst[i+63:i] := a[i+63:i] - b[i+63:i] - ELSE - dst[i+63:i] := src[i+63:i] - FI -ENDFOR -dst[MAX:256] := 0 +IF ((a[127:0] AND mask[127:0]) == 0) + ZF := 1 +ELSE + ZF := 0 +FI +dst := ZF - -
immintrin.h
+ +
smmintrin.h
 - - Floating Point - AVX512VL - AVX512F - Arithmetic - - - - Subtract packed double-precision (64-bit) floating-point elements in "b" from packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Integer + Flag + SSE4.1 + Logical + + + + Compute the bitwise AND of 128 bits (representing integer data) in "a" and "mask", and set "ZF" to 1 if the result is zero, otherwise set "ZF" to 0. Compute the bitwise NOT of "a" and then AND with "mask", and set "CF" to 1 if the result is zero, otherwise set "CF" to 0. Return 1 if both the "ZF" and "CF" values are zero, otherwise return 0. -FOR j := 0 to 3 - i := j*64 - IF k[j] - dst[i+63:i] := a[i+63:i] - b[i+63:i] - ELSE - dst[i+63:i] := 0 - FI +IF ((a[127:0] AND mask[127:0]) == 0) + ZF := 1 +ELSE + ZF := 0 +FI +IF (((NOT a[127:0]) AND mask[127:0]) == 0) + CF := 1 +ELSE + CF := 0 +FI +IF (ZF == 0 && CF == 0) + dst := 1 +ELSE + dst := 0 +FI + + +
smmintrin.h
+ + + Integer + Flag + SSE4.1 + Logical + + + Compute the bitwise NOT of "a" and then AND with a 128-bit vector containing all 1's, and return 1 if the result is zero, otherwise return 0. + +FOR j := 0 to 127 + tmp[j] := 1 ENDFOR -dst[MAX:256] := 0 +IF (((NOT a[127:0]) AND tmp[127:0]) == 0) + CF := 1 +ELSE + CF := 0 +FI +dst := CF - -
immintrin.h
+ + +
smmintrin.h
 - + Floating Point - AVX512VL - AVX512F - Arithmetic - - - - - Subtract packed double-precision (64-bit) floating-point elements in "b" from packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + SSE4.1 + Special Math Functions + + + + Round the packed double-precision (64-bit) floating-point elements in "a" using the "rounding" parameter, and store the results as packed double-precision floating-point elements in "dst". + [round_note] FOR j := 0 to 1 i := j*64 - IF k[j] - dst[i+63:i] := a[i+63:i] - b[i+63:i] - ELSE - dst[i+63:i] := src[i+63:i] - FI + dst[i+63:i] := ROUND(a[i+63:i], rounding) ENDFOR -dst[MAX:128] := 0 - -
immintrin.h
+ +
smmintrin.h
 - + Floating Point - AVX512VL - AVX512F - Arithmetic - - - - Subtract packed double-precision (64-bit) floating-point elements in "b" from packed double-precision (64-bit) floating-point elements in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + SSE4.1 + Special Math Functions + + + Round the packed double-precision (64-bit) floating-point elements in "a" down to an integer value, and store the results as packed double-precision floating-point elements in "dst". FOR j := 0 to 1 i := j*64 - IF k[j] - dst[i+63:i] := a[i+63:i] - b[i+63:i] - ELSE - dst[i+63:i] := 0 - FI + dst[i+63:i] := FLOOR(a[i+63:i]) ENDFOR -dst[MAX:128] := 0 - -
immintrin.h
+ +
smmintrin.h
 - + Floating Point - AVX512VL - AVX512F - Arithmetic - - - - - Subtract packed single-precision (32-bit) floating-point elements in "b" from packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + SSE4.1 + Special Math Functions + + + Round the packed double-precision (64-bit) floating-point elements in "a" up to an integer value, and store the results as packed double-precision floating-point elements in "dst". -FOR j := 0 to 7 - i := j*32 - IF k[j] - dst[i+31:i] := a[i+31:i] - b[i+31:i] - ELSE - dst[i+31:i] := src[i+31:i] - FI +FOR j := 0 to 1 + i := j*64 + dst[i+63:i] := CEIL(a[i+63:i]) ENDFOR -dst[MAX:256] := 0 - -
immintrin.h
+ +
smmintrin.h
 - + Floating Point - AVX512VL - AVX512F - Arithmetic - - - - Subtract packed single-precision (32-bit) floating-point elements in "b" from packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + SSE4.1 + Special Math Functions + + + + Round the packed single-precision (32-bit) floating-point elements in "a" using the "rounding" parameter, and store the results as packed single-precision floating-point elements in "dst". + [round_note] -FOR j := 0 to 7 +FOR j := 0 to 3 i := j*32 - IF k[j] - dst[i+31:i] := a[i+31:i] - b[i+31:i] - ELSE - dst[i+31:i] := 0 - FI + dst[i+31:i] := ROUND(a[i+31:i], rounding) ENDFOR -dst[MAX:256] := 0 - -
immintrin.h
+ +
smmintrin.h
 - + Floating Point - AVX512VL - AVX512F - Arithmetic - - - - - Subtract packed single-precision (32-bit) floating-point elements in "b" from packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + SSE4.1 + Special Math Functions + + + Round the packed single-precision (32-bit) floating-point elements in "a" down to an integer value, and store the results as packed single-precision floating-point elements in "dst". FOR j := 0 to 3 i := j*32 - IF k[j] - dst[i+31:i] := a[i+31:i] - b[i+31:i] - ELSE - dst[i+31:i] := src[i+31:i] - FI + dst[i+31:i] := FLOOR(a[i+31:i]) ENDFOR -dst[MAX:128] := 0 - -
immintrin.h
+ +
smmintrin.h
 - + Floating Point - AVX512VL - AVX512F - Arithmetic - - - - Subtract packed single-precision (32-bit) floating-point elements in "b" from packed single-precision (32-bit) floating-point elements in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + SSE4.1 + Special Math Functions + + + Round the packed single-precision (32-bit) floating-point elements in "a" up to an integer value, and store the results as packed single-precision floating-point elements in "dst". FOR j := 0 to 3 i := j*32 - IF k[j] - dst[i+31:i] := a[i+31:i] - b[i+31:i] - ELSE - dst[i+31:i] := 0 - FI + dst[i+31:i] := CEIL(a[i+31:i]) ENDFOR -dst[MAX:128] := 0 - -
immintrin.h
+ +
smmintrin.h
 - + Floating Point - AVX512VL - AVX512F - Miscellaneous - - - - - Unpack and interleave double-precision (64-bit) floating-point elements from the high half of each 128-bit lane in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + SSE4.1 + Special Math Functions + + + + + Round the lower double-precision (64-bit) floating-point element in "b" using the "rounding" parameter, store the result as a double-precision floating-point element in the lower element of "dst", and copy the upper element from "a" to the upper element of "dst". + [round_note] -INTERLEAVE_HIGH_QWORDS(src1[127:0], src2[127:0]){ - dst[63:0] := src1[127:64] - dst[127:64] := src2[127:64] - RETURN dst[127:0] -} - -tmp_dst[127:0] := INTERLEAVE_HIGH_QWORDS(a[127:0], b[127:0]) -tmp_dst[255:128] := INTERLEAVE_HIGH_QWORDS(a[255:128], b[255:128]) - -FOR j := 0 to 3 - i := j*64 - IF k[j] - dst[i+63:i] := tmp_dst[i+63:i] - ELSE - dst[i+63:i] := src[i+63:i] - FI -ENDFOR -dst[MAX:256] := 0 +dst[63:0] := ROUND(b[63:0], rounding) +dst[127:64] := a[127:64] - -
immintrin.h
+ +
smmintrin.h
 - + Floating Point - AVX512VL - AVX512F - Miscellaneous - - - - Unpack and interleave double-precision (64-bit) floating-point elements from the high half of each 128-bit lane in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + SSE4.1 + Special Math Functions + + + + Round the lower double-precision (64-bit) floating-point element in "b" down to an integer value, store the result as a double-precision floating-point element in the lower element of "dst", and copy the upper element from "a" to the upper element of "dst". -INTERLEAVE_HIGH_QWORDS(src1[127:0], src2[127:0]){ - dst[63:0] := src1[127:64] - dst[127:64] := src2[127:64] - RETURN dst[127:0] -} - -tmp_dst[127:0] := INTERLEAVE_HIGH_QWORDS(a[127:0], b[127:0]) -tmp_dst[255:128] := INTERLEAVE_HIGH_QWORDS(a[255:128], b[255:128]) - -FOR j := 0 to 3 - i := j*64 - IF k[j] - dst[i+63:i] := tmp_dst[i+63:i] - ELSE - dst[i+63:i] := 0 - FI -ENDFOR -dst[MAX:256] := 0 +dst[63:0] := FLOOR(b[63:0]) +dst[127:64] := a[127:64] - -
immintrin.h
+ +
smmintrin.h
 - + Floating Point - AVX512VL - AVX512F - Miscellaneous - - - - - Unpack and interleave double-precision (64-bit) floating-point elements from the high half of "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + SSE4.1 + Special Math Functions + + + + Round the lower double-precision (64-bit) floating-point element in "b" up to an integer value, store the result as a double-precision floating-point element in the lower element of "dst", and copy the upper element from "a" to the upper element of "dst". -INTERLEAVE_HIGH_QWORDS(src1[127:0], src2[127:0]){ - dst[63:0] := src1[127:64] - dst[127:64] := src2[127:64] - RETURN dst[127:0] -} - -tmp_dst[127:0] := INTERLEAVE_HIGH_QWORDS(a[127:0], b[127:0]) - -FOR j := 0 to 1 - i := j*64 - IF k[j] - dst[i+63:i] := tmp_dst[i+63:i] - ELSE - dst[i+63:i] := src[i+63:i] - FI -ENDFOR -dst[MAX:128] := 0 +dst[63:0] := CEIL(b[63:0]) +dst[127:64] := a[127:64] + + +
smmintrin.h
+ + + Floating Point + SSE4.1 + Special Math Functions + + + + + Round the lower single-precision (32-bit) floating-point element in "b" using the "rounding" parameter, store the result as a single-precision floating-point element in the lower element of "dst", and copy the upper 3 packed elements from "a" to the upper elements of "dst". + [round_note] + +dst[31:0] := ROUND(b[31:0], rounding) +dst[127:32] := a[127:32] - -
immintrin.h
+ +
smmintrin.h
 - + Floating Point - AVX512VL - AVX512F - Miscellaneous - - - - Unpack and interleave double-precision (64-bit) floating-point elements from the high half of "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + SSE4.1 + Special Math Functions + + + + Round the lower single-precision (32-bit) floating-point element in "b" down to an integer value, store the result as a single-precision floating-point element in the lower element of "dst", and copy the upper 3 packed elements from "a" to the upper elements of "dst". -INTERLEAVE_HIGH_QWORDS(src1[127:0], src2[127:0]){ - dst[63:0] := src1[127:64] - dst[127:64] := src2[127:64] - RETURN dst[127:0] -} - -tmp_dst[127:0] := INTERLEAVE_HIGH_QWORDS(a[127:0], b[127:0]) - -FOR j := 0 to 1 - i := j*64 - IF k[j] - dst[i+63:i] := tmp_dst[i+63:i] - ELSE - dst[i+63:i] := 0 - FI -ENDFOR -dst[MAX:128] := 0 +dst[31:0] := FLOOR(b[31:0]) +dst[127:32] := a[127:32] - -
immintrin.h
+ +
smmintrin.h
 - + Floating Point - AVX512VL - AVX512F - Miscellaneous - - - - - Unpack and interleave single-precision (32-bit) floating-point elements from the high half of each 128-bit lane in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + SSE4.1 + Special Math Functions + + + + Round the lower single-precision (32-bit) floating-point element in "b" up to an integer value, store the result as a single-precision floating-point element in the lower element of "dst", and copy the upper 3 packed elements from "a" to the upper elements of "dst". -INTERLEAVE_HIGH_DWORDS(src1[127:0], src2[127:0]){ - dst[31:0] := src1[95:64] - dst[63:32] := src2[95:64] - dst[95:64] := src1[127:96] - dst[127:96] := src2[127:96] - RETURN dst[127:0] -} - -tmp_dst[127:0] := INTERLEAVE_HIGH_DWORDS(a[127:0], b[127:0]) -tmp_dst[255:128] := INTERLEAVE_HIGH_DWORDS(a[255:128], b[255:128]) - -FOR j := 0 to 7 - i := j*32 - IF k[j] - dst[i+31:i] := tmp_dst[i+31:i] - ELSE - dst[i+31:i] := src[i+31:i] - FI -ENDFOR -dst[MAX:256] := 0 +dst[31:0] := CEIL(b[31:0]) +dst[127:32] := a[127:32] - -
immintrin.h
+ +
smmintrin.h
 - - Floating Point - AVX512VL - AVX512F + + Integer + SSE4.1 Miscellaneous - - - - Unpack and interleave single-precision (32-bit) floating-point elements from the high half of each 128-bit lane in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + + Horizontally compute the minimum amongst the packed unsigned 16-bit integers in "a", store the minimum and index in "dst", and zero the remaining bits in "dst". -INTERLEAVE_HIGH_DWORDS(src1[127:0], src2[127:0]){ - dst[31:0] := src1[95:64] - dst[63:32] := src2[95:64] - dst[95:64] := src1[127:96] - dst[127:96] := src2[127:96] - RETURN dst[127:0] -} - -tmp_dst[127:0] := INTERLEAVE_HIGH_DWORDS(a[127:0], b[127:0]) -tmp_dst[255:128] := INTERLEAVE_HIGH_DWORDS(a[255:128], b[255:128]) - +index[2:0] := 0 +min[15:0] := a[15:0] FOR j := 0 to 7 - i := j*32 - IF k[j] - dst[i+31:i] := tmp_dst[i+31:i] - ELSE - dst[i+31:i] := 0 + i := j*16 + IF a[i+15:i] < min[15:0] + index[2:0] := j + min[15:0] := a[i+15:i] FI ENDFOR -dst[MAX:256] := 0 +dst[15:0] := min[15:0] +dst[18:16] := index[2:0] +dst[127:19] := 0 - -
immintrin.h
+ +
smmintrin.h
 - - Floating Point - AVX512VL - AVX512F + + Integer + SSE4.1 + Arithmetic Miscellaneous - - - - - Unpack and interleave single-precision (32-bit) floating-point elements from the high half of "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + + + + Compute the sum of absolute differences (SADs) of quadruplets of unsigned 8-bit integers in "a" compared to those in "b", and store the 16-bit results in "dst". + Eight SADs are performed using one quadruplet from "b" and eight quadruplets from "a". One quadruplet is selected from "b" starting at on the offset specified in "imm8". Eight quadruplets are formed from sequential 8-bit integers selected from "a" starting at the offset specified in "imm8". -INTERLEAVE_HIGH_DWORDS(src1[127:0], src2[127:0]){ - dst[31:0] := src1[95:64] - dst[63:32] := src2[95:64] - dst[95:64] := src1[127:96] - dst[127:96] := src2[127:96] - RETURN dst[127:0] -} - -tmp_dst[127:0] := INTERLEAVE_HIGH_DWORDS(a[127:0], b[127:0]) - -FOR j := 0 to 3 - i := j*32 - IF k[j] - dst[i+31:i] := tmp_dst[i+31:i] - ELSE - dst[i+31:i] := src[i+31:i] - FI -ENDFOR -dst[MAX:128] := 0 +DEFINE MPSADBW(a[127:0], b[127:0], imm8[2:0]) { + a_offset := imm8[2]*32 + b_offset := imm8[1:0]*32 + FOR j := 0 to 7 + i := j*8 + k := a_offset+i + l := b_offset + tmp[i*2+15:i*2] := ABS(Signed(a[k+7:k] - b[l+7:l])) + ABS(Signed(a[k+15:k+8] - b[l+15:l+8])) + \ + ABS(Signed(a[k+23:k+16] - b[l+23:l+16])) + ABS(Signed(a[k+31:k+24] - b[l+31:l+24])) + ENDFOR + RETURN tmp[127:0] +} +dst[127:0] := MPSADBW(a[127:0], b[127:0], imm8[2:0]) - -
immintrin.h
+ +
smmintrin.h
 - - Floating Point - AVX512VL - AVX512F - Miscellaneous - - - - Unpack and interleave single-precision (32-bit) floating-point elements from the high half of "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Integer + SSE4.1 + Load + + + Load 128-bits of integer data from memory into "dst" using a non-temporal memory hint. + "mem_addr" must be aligned on a 16-byte boundary or a general-protection exception may be generated. -INTERLEAVE_HIGH_DWORDS(src1[127:0], src2[127:0]){ - dst[31:0] := src1[95:64] - dst[63:32] := src2[95:64] - dst[95:64] := src1[127:96] - dst[127:96] := src2[127:96] - RETURN dst[127:0] -} - -tmp_dst[127:0] := INTERLEAVE_HIGH_DWORDS(a[127:0], b[127:0]) - -FOR j := 0 to 3 - i := j*32 - IF k[j] - dst[i+31:i] := tmp_dst[i+31:i] - ELSE - dst[i+31:i] := 0 - FI -ENDFOR -dst[MAX:128] := 0 +dst[127:0] := MEM[mem_addr+127:mem_addr] - -
immintrin.h
+ +
smmintrin.h
 - - Floating Point - AVX512VL - AVX512F - Miscellaneous - - - - - Unpack and interleave double-precision (64-bit) floating-point elements from the low half of each 128-bit lane in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + SSE4.2 + String Compare + + + + + Compare packed strings with implicit lengths in "a" and "b" using the control in "imm8", and store the generated mask in "dst". + [strcmp_note] -INTERLEAVE_QWORDS(src1[127:0], src2[127:0]){ - dst[63:0] := src1[63:0] - dst[127:64] := src2[63:0] - RETURN dst[127:0] -} - -tmp_dst[127:0] := INTERLEAVE_QWORDS(a[127:0], b[127:0]) -tmp_dst[255:128] := INTERLEAVE_QWORDS(a[255:128], b[255:128]) - -FOR j := 0 to 3 - i := j*64 - IF k[j] - dst[i+63:i] := tmp_dst[i+63:i] - ELSE - dst[i+63:i] := src[i+63:i] +size := (imm8[0] ? 16 : 8) // 8 or 16-bit characters +UpperBound := (128 / size) - 1 +BoolRes := 0 +// compare all characters +aInvalid := 0 +bInvalid := 0 +FOR i := 0 to UpperBound + m := i*size + FOR j := 0 to UpperBound + n := j*size + BoolRes.word[i].bit[j] := (a[m+size-1:m] == b[n+size-1:n]) ? 1 : 0 + + // invalidate characters after EOS + IF a[m+size-1:m] == 0 + aInvalid := 1 + FI + IF b[n+size-1:n] == 0 + bInvalid := 1 + FI + + // override comparisons for invalid characters + CASE (imm8[3:2]) OF + 0: // equal any + IF (!aInvalid && bInvalid) + BoolRes.word[i].bit[j] := 0 + ELSE IF (aInvalid && !bInvalid) + BoolRes.word[i].bit[j] := 0 + ELSE IF (aInvalid && bInvalid) + BoolRes.word[i].bit[j] := 0 + FI + 1: // ranges + IF (!aInvalid && bInvalid) + BoolRes.word[i].bit[j] := 0 + ELSE IF (aInvalid && !bInvalid) + BoolRes.word[i].bit[j] := 0 + ELSE IF (aInvalid && bInvalid) + BoolRes.word[i].bit[j] := 0 + FI + 2: // equal each + IF (!aInvalid && bInvalid) + BoolRes.word[i].bit[j] := 0 + ELSE IF (aInvalid && !bInvalid) + BoolRes.word[i].bit[j] := 0 + ELSE IF (aInvalid && bInvalid) + BoolRes.word[i].bit[j] := 1 + FI + 3: // equal ordered + IF (!aInvalid && bInvalid) + BoolRes.word[i].bit[j] := 0 + ELSE IF (aInvalid && !bInvalid) + BoolRes.word[i].bit[j] := 1 + ELSE IF (aInvalid && bInvalid) + BoolRes.word[i].bit[j] := 1 + FI + ESAC + ENDFOR +ENDFOR +// aggregate results +CASE (imm8[3:2]) OF +0: // equal any + IntRes1 := 0 + FOR i := 0 to UpperBound + FOR j := 0 to UpperBound + IntRes1[i] := IntRes1[i] OR BoolRes.word[i].bit[j] + ENDFOR + ENDFOR +1: // ranges + IntRes1 := 0 + FOR i := 0 to UpperBound + FOR j := 0 to UpperBound + IntRes1[i] := IntRes1[i] OR (BoolRes.word[i].bit[j] AND BoolRes.word[i].bit[j+1]) + j += 2 + ENDFOR + ENDFOR +2: // equal each + IntRes1 := 0 + FOR i := 0 to UpperBound + IntRes1[i] := BoolRes.word[i].bit[i] + ENDFOR +3: // equal ordered + IntRes1 := (imm8[0] ? 0xFF : 0xFFFF) + FOR i := 0 to UpperBound + k := i + FOR j := 0 to UpperBound-i + IntRes1[i] := IntRes1[i] AND BoolRes.word[k].bit[j] + k := k+1 + ENDFOR + ENDFOR +ESAC +// optionally negate results +bInvalid := 0 +FOR i := 0 to UpperBound + IF imm8[4] + IF imm8[5] // only negate valid + IF b[n+size-1:n] == 0 + bInvalid := 1 + FI + IF bInvalid // invalid, don't negate + IntRes2[i] := IntRes1[i] + ELSE // valid, negate + IntRes2[i] := -1 XOR IntRes1[i] + FI + ELSE // negate all + IntRes2[i] := -1 XOR IntRes1[i] + FI + ELSE // don't negate + IntRes2[i] := IntRes1[i] FI ENDFOR -dst[MAX:256] := 0 +// output +IF imm8[6] // byte / word mask + FOR i := 0 to UpperBound + j := i*size + IF IntRes2[i] + dst[j+size-1:j] := (imm8[0] ? 0xFF : 0xFFFF) + ELSE + dst[j+size-1:j] := 0 + FI + ENDFOR +ELSE // bit mask + dst[UpperBound:0] := IntRes2[UpperBound:0] + dst[127:UpperBound+1] := 0 +FI - -
immintrin.h
+ +
nmmintrin.h
 - - Floating Point - AVX512VL - AVX512F - Miscellaneous - - - - Unpack and interleave double-precision (64-bit) floating-point elements from the low half of each 128-bit lane in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Flag + SSE4.2 + String Compare + + + + + Compare packed strings with implicit lengths in "a" and "b" using the control in "imm8", and store the generated index in "dst". + [strcmp_note] -INTERLEAVE_QWORDS(src1[127:0], src2[127:0]){ - dst[63:0] := src1[63:0] - dst[127:64] := src2[63:0] - RETURN dst[127:0] -} - -tmp_dst[127:0] := INTERLEAVE_QWORDS(a[127:0], b[127:0]) -tmp_dst[255:128] := INTERLEAVE_QWORDS(a[255:128], b[255:128]) - -FOR j := 0 to 3 - i := j*64 - IF k[j] - dst[i+63:i] := tmp_dst[i+63:i] - ELSE - dst[i+63:i] := 0 +size := (imm8[0] ? 16 : 8) // 8 or 16-bit characters +UpperBound := (128 / size) - 1 +BoolRes := 0 +// compare all characters +aInvalid := 0 +bInvalid := 0 +FOR i := 0 to UpperBound + m := i*size + FOR j := 0 to UpperBound + n := j*size + BoolRes.word[i].bit[j] := (a[m+size-1:m] == b[n+size-1:n]) ? 1 : 0 + + // invalidate characters after EOS + IF a[m+size-1:m] == 0 + aInvalid := 1 + FI + IF b[n+size-1:n] == 0 + bInvalid := 1 + FI + + // override comparisons for invalid characters + CASE (imm8[3:2]) OF + 0: // equal any + IF (!aInvalid && bInvalid) + BoolRes.word[i].bit[j] := 0 + ELSE IF (aInvalid && !bInvalid) + BoolRes.word[i].bit[j] := 0 + ELSE IF (aInvalid && bInvalid) + BoolRes.word[i].bit[j] := 0 + FI + 1: // ranges + IF (!aInvalid && bInvalid) + BoolRes.word[i].bit[j] := 0 + ELSE IF (aInvalid && !bInvalid) + BoolRes.word[i].bit[j] := 0 + ELSE IF (aInvalid && bInvalid) + BoolRes.word[i].bit[j] := 0 + FI + 2: // equal each + IF (!aInvalid && bInvalid) + BoolRes.word[i].bit[j] := 0 + ELSE IF (aInvalid && !bInvalid) + BoolRes.word[i].bit[j] := 0 + ELSE IF (aInvalid && bInvalid) + BoolRes.word[i].bit[j] := 1 + FI + 3: // equal ordered + IF (!aInvalid && bInvalid) + BoolRes.word[i].bit[j] := 0 + ELSE IF (aInvalid && !bInvalid) + BoolRes.word[i].bit[j] := 1 + ELSE IF (aInvalid && bInvalid) + BoolRes.word[i].bit[j] := 1 + FI + ESAC + ENDFOR +ENDFOR +// aggregate results +CASE (imm8[3:2]) OF +0: // equal any + IntRes1 := 0 + FOR i := 0 to UpperBound + FOR j := 0 to UpperBound + IntRes1[i] := IntRes1[i] OR BoolRes.word[i].bit[j] + ENDFOR + ENDFOR +1: // ranges + IntRes1 := 0 + FOR i := 0 to UpperBound + FOR j := 0 to UpperBound + IntRes1[i] := IntRes1[i] OR (BoolRes.word[i].bit[j] AND BoolRes.word[i].bit[j+1]) + j += 2 + ENDFOR + ENDFOR +2: // equal each + IntRes1 := 0 + FOR i := 0 to UpperBound + IntRes1[i] := BoolRes.word[i].bit[i] + ENDFOR +3: // equal ordered + IntRes1 := (imm8[0] ? 0xFF : 0xFFFF) + FOR i := 0 to UpperBound + k := i + FOR j := 0 to UpperBound-i + IntRes1[i] := IntRes1[i] AND BoolRes.word[k].bit[j] + k := k+1 + ENDFOR + ENDFOR +ESAC +// optionally negate results +bInvalid := 0 +FOR i := 0 to UpperBound + IF imm8[4] + IF imm8[5] // only negate valid + IF b[n+size-1:n] == 0 + bInvalid := 1 + FI + IF bInvalid // invalid, don't negate + IntRes2[i] := IntRes1[i] + ELSE // valid, negate + IntRes2[i] := -1 XOR IntRes1[i] + FI + ELSE // negate all + IntRes2[i] := -1 XOR IntRes1[i] + FI + ELSE // don't negate + IntRes2[i] := IntRes1[i] FI ENDFOR -dst[MAX:256] := 0 +// output +IF imm8[6] // most significant bit + tmp := UpperBound + dst := tmp + DO WHILE ((tmp >= 0) AND a[tmp] == 0) + tmp := tmp - 1 + dst := tmp + OD +ELSE // least significant bit + tmp := 0 + dst := tmp + DO WHILE ((tmp <= UpperBound) AND a[tmp] == 0) + tmp := tmp + 1 + dst := tmp + OD +FI - -
immintrin.h
+ +
nmmintrin.h
 - - Floating Point - AVX512VL - AVX512F - Miscellaneous - - - - - Unpack and interleave double-precision (64-bit) floating-point elements from the low half of "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Flag + SSE4.2 + String Compare + + + + + Compare packed strings with implicit lengths in "a" and "b" using the control in "imm8", and returns 1 if any character in "b" was null, and 0 otherwise. + [strcmp_note] -INTERLEAVE_QWORDS(src1[127:0], src2[127:0]){ - dst[63:0] := src1[63:0] - dst[127:64] := src2[63:0] - RETURN dst[127:0] -} - -tmp_dst[127:0] := INTERLEAVE_QWORDS(a[127:0], b[127:0]) - -FOR j := 0 to 1 - i := j*64 - IF k[j] - dst[i+63:i] := tmp_dst[i+63:i] - ELSE - dst[i+63:i] := src[i+63:i] +size := (imm8[0] ? 16 : 8) // 8 or 16-bit characters +UpperBound := (128 / size) - 1 +bInvalid := 0 +FOR j := 0 to UpperBound + n := j*size + IF b[n+size-1:n] == 0 + bInvalid := 1 FI ENDFOR -dst[MAX:128] := 0 +dst := bInvalid - -
immintrin.h
+ +
nmmintrin.h
 - - Floating Point - AVX512VL - AVX512F - Miscellaneous - - - - Unpack and interleave double-precision (64-bit) floating-point elements from the low half of "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Flag + SSE4.2 + String Compare + + + + + Compare packed strings with implicit lengths in "a" and "b" using the control in "imm8", and returns 1 if the resulting mask was non-zero, and 0 otherwise. + [strcmp_note] -INTERLEAVE_QWORDS(src1[127:0], src2[127:0]){ - dst[63:0] := src1[63:0] - dst[127:64] := src2[63:0] - RETURN dst[127:0] -} - -tmp_dst[127:0] := INTERLEAVE_QWORDS(a[127:0], b[127:0]) - -FOR j := 0 to 1 - i := j*64 - IF k[j] - dst[i+63:i] := tmp_dst[i+63:i] - ELSE - dst[i+63:i] := 0 +size := (imm8[0] ? 16 : 8) // 8 or 16-bit characters +UpperBound := (128 / size) - 1 +BoolRes := 0 +// compare all characters +aInvalid := 0 +bInvalid := 0 +FOR i := 0 to UpperBound + m := i*size + FOR j := 0 to UpperBound + n := j*size + BoolRes.word[i].bit[j] := (a[m+size-1:m] == b[n+size-1:n]) ? 1 : 0 + + // invalidate characters after EOS + IF a[m+size-1:m] == 0 + aInvalid := 1 + FI + IF b[n+size-1:n] == 0 + bInvalid := 1 + FI + + // override comparisons for invalid characters + CASE (imm8[3:2]) OF + 0: // equal any + IF (!aInvalid && bInvalid) + BoolRes.word[i].bit[j] := 0 + ELSE IF (aInvalid && !bInvalid) + BoolRes.word[i].bit[j] := 0 + ELSE IF (aInvalid && bInvalid) + BoolRes.word[i].bit[j] := 0 + FI + 1: // ranges + IF (!aInvalid && bInvalid) + BoolRes.word[i].bit[j] := 0 + ELSE IF (aInvalid && !bInvalid) + BoolRes.word[i].bit[j] := 0 + ELSE IF (aInvalid && bInvalid) + BoolRes.word[i].bit[j] := 0 + FI + 2: // equal each + IF (!aInvalid && bInvalid) + BoolRes.word[i].bit[j] := 0 + ELSE IF (aInvalid && !bInvalid) + BoolRes.word[i].bit[j] := 0 + ELSE IF (aInvalid && bInvalid) + BoolRes.word[i].bit[j] := 1 + FI + 3: // equal ordered + IF (!aInvalid && bInvalid) + BoolRes.word[i].bit[j] := 0 + ELSE IF (aInvalid && !bInvalid) + BoolRes.word[i].bit[j] := 1 + ELSE IF (aInvalid && bInvalid) + BoolRes.word[i].bit[j] := 1 + FI + ESAC + ENDFOR +ENDFOR +// aggregate results +CASE (imm8[3:2]) OF +0: // equal any + IntRes1 := 0 + FOR i := 0 to UpperBound + FOR j := 0 to UpperBound + IntRes1[i] := IntRes1[i] OR BoolRes.word[i].bit[j] + ENDFOR + ENDFOR +1: // ranges + IntRes1 := 0 + FOR i := 0 to UpperBound + FOR j := 0 to UpperBound + IntRes1[i] := IntRes1[i] OR (BoolRes.word[i].bit[j] AND BoolRes.word[i].bit[j+1]) + j += 2 + ENDFOR + ENDFOR +2: // equal each + IntRes1 := 0 + FOR i := 0 to UpperBound + IntRes1[i] := BoolRes.word[i].bit[i] + ENDFOR +3: // equal ordered + IntRes1 := (imm8[0] ? 0xFF : 0xFFFF) + FOR i := 0 to UpperBound + k := i + FOR j := 0 to UpperBound-i + IntRes1[i] := IntRes1[i] AND BoolRes.word[k].bit[j] + k := k+1 + ENDFOR + ENDFOR +ESAC +// optionally negate results +bInvalid := 0 +FOR i := 0 to UpperBound + IF imm8[4] + IF imm8[5] // only negate valid + IF b[n+size-1:n] == 0 + bInvalid := 1 + FI + IF bInvalid // invalid, don't negate + IntRes2[i] := IntRes1[i] + ELSE // valid, negate + IntRes2[i] := -1 XOR IntRes1[i] + FI + ELSE // negate all + IntRes2[i] := -1 XOR IntRes1[i] + FI + ELSE // don't negate + IntRes2[i] := IntRes1[i] FI ENDFOR -dst[MAX:128] := 0 +// output +dst := (IntRes2 != 0) - -
immintrin.h
+ +
nmmintrin.h
 - - Floating Point - AVX512VL - AVX512F - Miscellaneous - - - - - Unpack and interleave single-precision (32-bit) floating-point elements from the low half of each 128-bit lane in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Flag + SSE4.2 + String Compare + + + + + Compare packed strings with implicit lengths in "a" and "b" using the control in "imm8", and returns 1 if any character in "a" was null, and 0 otherwise. + [strcmp_note] -INTERLEAVE_DWORDS(src1[127:0], src2[127:0]){ - dst[31:0] := src1[31:0] - dst[63:32] := src2[31:0] - dst[95:64] := src1[63:32] - dst[127:96] := src2[63:32] - RETURN dst[127:0] -} - -tmp_dst[127:0] := INTERLEAVE_DWORDS(a[127:0], b[127:0]) -tmp_dst[255:128] := INTERLEAVE_DWORDS(a[255:128], b[255:128]) - -FOR j := 0 to 7 - i := j*32 - IF k[j] - dst[i+31:i] := tmp_dst[i+31:i] - ELSE - dst[i+31:i] := src[i+31:i] +size := (imm8[0] ? 16 : 8) // 8 or 16-bit characters +UpperBound := (128 / size) - 1 +aInvalid := 0 +FOR i := 0 to UpperBound + m := i*size + IF a[m+size-1:m] == 0 + aInvalid := 1 FI ENDFOR -dst[MAX:256] := 0 +dst := aInvalid - -
immintrin.h
+ +
nmmintrin.h
 - - Floating Point - AVX512VL - AVX512F - Miscellaneous - - - - Unpack and interleave single-precision (32-bit) floating-point elements from the low half of each 128-bit lane in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Flag + SSE4.2 + String Compare + + + + + Compare packed strings with implicit lengths in "a" and "b" using the control in "imm8", and returns bit 0 of the resulting bit mask. + [strcmp_note] -INTERLEAVE_DWORDS(src1[127:0], src2[127:0]){ - dst[31:0] := src1[31:0] - dst[63:32] := src2[31:0] - dst[95:64] := src1[63:32] - dst[127:96] := src2[63:32] - RETURN dst[127:0] -} - -tmp_dst[127:0] := INTERLEAVE_DWORDS(a[127:0], b[127:0]) -tmp_dst[255:128] := INTERLEAVE_DWORDS(a[255:128], b[255:128]) - -FOR j := 0 to 7 - i := j*32 - IF k[j] - dst[i+31:i] := tmp_dst[i+31:i] - ELSE - dst[i+31:i] := 0 - FI +size := (imm8[0] ? 16 : 8) // 8 or 16-bit characters +UpperBound := (128 / size) - 1 +BoolRes := 0 +// compare all characters +aInvalid := 0 +bInvalid := 0 +FOR i := 0 to UpperBound + m := i*size + FOR j := 0 to UpperBound + n := j*size + BoolRes.word[i].bit[j] := (a[m+size-1:m] == b[n+size-1:n]) ? 1 : 0 + + // invalidate characters after EOS + IF a[m+size-1:m] == 0 + aInvalid := 1 + FI + IF b[n+size-1:n] == 0 + bInvalid := 1 + FI + + // override comparisons for invalid characters + CASE (imm8[3:2]) OF + 0: // equal any + IF (!aInvalid && bInvalid) + BoolRes.word[i].bit[j] := 0 + ELSE IF (aInvalid && !bInvalid) + BoolRes.word[i].bit[j] := 0 + ELSE IF (aInvalid && bInvalid) + BoolRes.word[i].bit[j] := 0 + FI + 1: // ranges + IF (!aInvalid && bInvalid) + BoolRes.word[i].bit[j] := 0 + ELSE IF (aInvalid && !bInvalid) + BoolRes.word[i].bit[j] := 0 + ELSE IF (aInvalid && bInvalid) + BoolRes.word[i].bit[j] := 0 + FI + 2: // equal each + IF (!aInvalid && bInvalid) + BoolRes.word[i].bit[j] := 0 + ELSE IF (aInvalid && !bInvalid) + BoolRes.word[i].bit[j] := 0 + ELSE IF (aInvalid && bInvalid) + BoolRes.word[i].bit[j] := 1 + FI + 3: // equal ordered + IF (!aInvalid && bInvalid) + BoolRes.word[i].bit[j] := 0 + ELSE IF (aInvalid && !bInvalid) + BoolRes.word[i].bit[j] := 1 + ELSE IF (aInvalid && bInvalid) + BoolRes.word[i].bit[j] := 1 + FI + ESAC + ENDFOR ENDFOR -dst[MAX:256] := 0 - - -
immintrin.h
- - - Floating Point - AVX512VL - AVX512F - Miscellaneous - - - - - Unpack and interleave single-precision (32-bit) floating-point elements from the low half of "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -INTERLEAVE_DWORDS(src1[127:0], src2[127:0]){ - dst[31:0] := src1[31:0] - dst[63:32] := src2[31:0] - dst[95:64] := src1[63:32] - dst[127:96] := src2[63:32] - RETURN dst[127:0] -} - -tmp_dst[127:0] := INTERLEAVE_DWORDS(a[127:0], b[127:0]) - -FOR j := 0 to 3 - i := j*32 - IF k[j] - dst[i+31:i] := tmp_dst[i+31:i] - ELSE - dst[i+31:i] := src[i+31:i] +// aggregate results +CASE (imm8[3:2]) OF +0: // equal any + IntRes1 := 0 + FOR i := 0 to UpperBound + FOR j := 0 to UpperBound + IntRes1[i] := IntRes1[i] OR BoolRes.word[i].bit[j] + ENDFOR + ENDFOR +1: // ranges + IntRes1 := 0 + FOR i := 0 to UpperBound + FOR j := 0 to UpperBound + IntRes1[i] := IntRes1[i] OR (BoolRes.word[i].bit[j] AND BoolRes.word[i].bit[j+1]) + j += 2 + ENDFOR + ENDFOR +2: // equal each + IntRes1 := 0 + FOR i := 0 to UpperBound + IntRes1[i] := BoolRes.word[i].bit[i] + ENDFOR +3: // equal ordered + IntRes1 := (imm8[0] ? 0xFF : 0xFFFF) + FOR i := 0 to UpperBound + k := i + FOR j := 0 to UpperBound-i + IntRes1[i] := IntRes1[i] AND BoolRes.word[k].bit[j] + k := k+1 + ENDFOR + ENDFOR +ESAC +// optionally negate results +bInvalid := 0 +FOR i := 0 to UpperBound + IF imm8[4] + IF imm8[5] // only negate valid + IF b[n+size-1:n] == 0 + bInvalid := 1 + FI + IF bInvalid // invalid, don't negate + IntRes2[i] := IntRes1[i] + ELSE // valid, negate + IntRes2[i] := -1 XOR IntRes1[i] + FI + ELSE // negate all + IntRes2[i] := -1 XOR IntRes1[i] + FI + ELSE // don't negate + IntRes2[i] := IntRes1[i] FI ENDFOR -dst[MAX:128] := 0 +// output +dst := IntRes2[0] - -
immintrin.h
+ +
nmmintrin.h
 - - Floating Point - AVX512VL - AVX512F - Miscellaneous - - - - Unpack and interleave single-precision (32-bit) floating-point elements from the low half of "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Flag + SSE4.2 + String Compare + + + + + Compare packed strings with implicit lengths in "a" and "b" using the control in "imm8", and returns 1 if "b" did not contain a null character and the resulting mask was zero, and 0 otherwise. + [strcmp_note] -INTERLEAVE_DWORDS(src1[127:0], src2[127:0]){ - dst[31:0] := src1[31:0] - dst[63:32] := src2[31:0] - dst[95:64] := src1[63:32] - dst[127:96] := src2[63:32] - RETURN dst[127:0] -} - -tmp_dst[127:0] := INTERLEAVE_DWORDS(a[127:0], b[127:0]) - -FOR j := 0 to 3 - i := j*32 - IF k[j] - dst[i+31:i] := tmp_dst[i+31:i] - ELSE - dst[i+31:i] := 0 - FI +size := (imm8[0] ? 16 : 8) // 8 or 16-bit characters +UpperBound := (128 / size) - 1 +BoolRes := 0 +// compare all characters +aInvalid := 0 +bInvalid := 0 +FOR i := 0 to UpperBound + m := i*size + FOR j := 0 to UpperBound + n := j*size + BoolRes.word[i].bit[j] := (a[m+size-1:m] == b[n+size-1:n]) ? 1 : 0 + + // invalidate characters after EOS + IF a[m+size-1:m] == 0 + aInvalid := 1 + FI + IF b[n+size-1:n] == 0 + bInvalid := 1 + FI + + // override comparisons for invalid characters + CASE (imm8[3:2]) OF + 0: // equal any + IF (!aInvalid && bInvalid) + BoolRes.word[i].bit[j] := 0 + ELSE IF (aInvalid && !bInvalid) + BoolRes.word[i].bit[j] := 0 + ELSE IF (aInvalid && bInvalid) + BoolRes.word[i].bit[j] := 0 + FI + 1: // ranges + IF (!aInvalid && bInvalid) + BoolRes.word[i].bit[j] := 0 + ELSE IF (aInvalid && !bInvalid) + BoolRes.word[i].bit[j] := 0 + ELSE IF (aInvalid && bInvalid) + BoolRes.word[i].bit[j] := 0 + FI + 2: // equal each + IF (!aInvalid && bInvalid) + BoolRes.word[i].bit[j] := 0 + ELSE IF (aInvalid && !bInvalid) + BoolRes.word[i].bit[j] := 0 + ELSE IF (aInvalid && bInvalid) + BoolRes.word[i].bit[j] := 1 + FI + 3: // equal ordered + IF (!aInvalid && bInvalid) + BoolRes.word[i].bit[j] := 0 + ELSE IF (aInvalid && !bInvalid) + BoolRes.word[i].bit[j] := 1 + ELSE IF (aInvalid && bInvalid) + BoolRes.word[i].bit[j] := 1 + FI + ESAC + ENDFOR ENDFOR -dst[MAX:128] := 0 - - -
immintrin.h
- - - Floating Point - AVX512VL - AVX512DQ - Logical - - - - - Compute the bitwise XOR of packed double-precision (64-bit) floating-point elements in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -FOR j := 0 to 3 - i := j*64 - IF k[j] - dst[i+63:i] := a[i+63:i] XOR b[i+63:i] - ELSE - dst[i+63:i] := src[i+63:i] +// aggregate results +CASE (imm8[3:2]) OF +0: // equal any + IntRes1 := 0 + FOR i := 0 to UpperBound + FOR j := 0 to UpperBound + IntRes1[i] := IntRes1[i] OR BoolRes.word[i].bit[j] + ENDFOR + ENDFOR +1: // ranges + IntRes1 := 0 + FOR i := 0 to UpperBound + FOR j := 0 to UpperBound + IntRes1[i] := IntRes1[i] OR (BoolRes.word[i].bit[j] AND BoolRes.word[i].bit[j+1]) + j += 2 + ENDFOR + ENDFOR +2: // equal each + IntRes1 := 0 + FOR i := 0 to UpperBound + IntRes1[i] := BoolRes.word[i].bit[i] + ENDFOR +3: // equal ordered + IntRes1 := (imm8[0] ? 0xFF : 0xFFFF) + FOR i := 0 to UpperBound + k := i + FOR j := 0 to UpperBound-i + IntRes1[i] := IntRes1[i] AND BoolRes.word[k].bit[j] + k := k+1 + ENDFOR + ENDFOR +ESAC +// optionally negate results +bInvalid := 0 +FOR i := 0 to UpperBound + IF imm8[4] + IF imm8[5] // only negate valid + IF b[n+size-1:n] == 0 + bInvalid := 1 + FI + IF bInvalid // invalid, don't negate + IntRes2[i] := IntRes1[i] + ELSE // valid, negate + IntRes2[i] := -1 XOR IntRes1[i] + FI + ELSE // negate all + IntRes2[i] := -1 XOR IntRes1[i] + FI + ELSE // don't negate + IntRes2[i] := IntRes1[i] FI ENDFOR -dst[MAX:256] := 0 +// output +dst := (IntRes2 == 0) AND bInvalid - -
immintrin.h
+ +
nmmintrin.h
 - - Floating Point - AVX512VL - AVX512DQ - Logical - - - - Compute the bitwise XOR of packed double-precision (64-bit) floating-point elements in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + SSE4.2 + String Compare + + + + + + + Compare packed strings in "a" and "b" with lengths "la" and "lb" using the control in "imm8", and store the generated mask in "dst". + [strcmp_note] -FOR j := 0 to 3 - i := j*64 - IF k[j] - dst[i+63:i] := a[i+63:i] XOR b[i+63:i] - ELSE - dst[i+63:i] := 0 - FI +size := (imm8[0] ? 16 : 8) // 8 or 16-bit characters +UpperBound := (128 / size) - 1 +BoolRes := 0 +// compare all characters +aInvalid := 0 +bInvalid := 0 +FOR i := 0 to UpperBound + m := i*size + FOR j := 0 to UpperBound + n := j*size + BoolRes.word[i].bit[j] := (a[m+size-1:m] == b[n+size-1:n]) ? 1 : 0 + + // invalidate characters after EOS + IF i == la + aInvalid := 1 + FI + IF j == lb + bInvalid := 1 + FI + + // override comparisons for invalid characters + CASE (imm8[3:2]) OF + 0: // equal any + IF (!aInvalid && bInvalid) + BoolRes.word[i].bit[j] := 0 + ELSE IF (aInvalid && !bInvalid) + BoolRes.word[i].bit[j] := 0 + ELSE IF (aInvalid && bInvalid) + BoolRes.word[i].bit[j] := 0 + FI + 1: // ranges + IF (!aInvalid && bInvalid) + BoolRes.word[i].bit[j] := 0 + ELSE IF (aInvalid && !bInvalid) + BoolRes.word[i].bit[j] := 0 + ELSE IF (aInvalid && bInvalid) + BoolRes.word[i].bit[j] := 0 + FI + 2: // equal each + IF (!aInvalid && bInvalid) + BoolRes.word[i].bit[j] := 0 + ELSE IF (aInvalid && !bInvalid) + BoolRes.word[i].bit[j] := 0 + ELSE IF (aInvalid && bInvalid) + BoolRes.word[i].bit[j] := 1 + FI + 3: // equal ordered + IF (!aInvalid && bInvalid) + BoolRes.word[i].bit[j] := 0 + ELSE IF (aInvalid && !bInvalid) + BoolRes.word[i].bit[j] := 1 + ELSE IF (aInvalid && bInvalid) + BoolRes.word[i].bit[j] := 1 + FI + ESAC + ENDFOR ENDFOR -dst[MAX:256] := 0 - - -
immintrin.h
- - - Floating Point - AVX512DQ - Logical - - - - - Compute the bitwise XOR of packed double-precision (64-bit) floating-point elements in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -FOR j := 0 to 7 - i := j*64 - IF k[j] - dst[i+63:i] := a[i+63:i] XOR b[i+63:i] - ELSE - dst[i+63:i] := src[i+63:i] +// aggregate results +CASE (imm8[3:2]) OF +0: // equal any + IntRes1 := 0 + FOR i := 0 to UpperBound + FOR j := 0 to UpperBound + IntRes1[i] := IntRes1[i] OR BoolRes.word[i].bit[j] + ENDFOR + ENDFOR +1: // ranges + IntRes1 := 0 + FOR i := 0 to UpperBound + FOR j := 0 to UpperBound + IntRes1[i] := IntRes1[i] OR (BoolRes.word[i].bit[j] AND BoolRes.word[i].bit[j+1]) + j += 2 + ENDFOR + ENDFOR +2: // equal each + IntRes1 := 0 + FOR i := 0 to UpperBound + IntRes1[i] := BoolRes.word[i].bit[i] + ENDFOR +3: // equal ordered + IntRes1 := (imm8[0] ? 0xFF : 0xFFFF) + FOR i := 0 to UpperBound + k := i + FOR j := 0 to UpperBound-i + IntRes1[i] := IntRes1[i] AND BoolRes.word[k].bit[j] + k := k+1 + ENDFOR + ENDFOR +ESAC +// optionally negate results +FOR i := 0 to UpperBound + IF imm8[4] + IF imm8[5] // only negate valid + IF i >= lb // invalid, don't negate + IntRes2[i] := IntRes1[i] + ELSE // valid, negate + IntRes2[i] := -1 XOR IntRes1[i] + FI + ELSE // negate all + IntRes2[i] := -1 XOR IntRes1[i] + FI + ELSE // don't negate + IntRes2[i] := IntRes1[i] FI ENDFOR -dst[MAX:512] := 0 +// output +IF imm8[6] // byte / word mask + FOR i := 0 to UpperBound + j := i*size + IF IntRes2[i] + dst[j+size-1:j] := (imm8[0] ? 0xFF : 0xFFFF) + ELSE + dst[j+size-1:j] := 0 + FI + ENDFOR +ELSE // bit mask + dst[UpperBound:0] := IntRes2[UpperBound:0] + dst[127:UpperBound+1] := 0 +FI - -
immintrin.h
+ +
nmmintrin.h
 - - Floating Point - AVX512DQ - Logical - - - - Compute the bitwise XOR of packed double-precision (64-bit) floating-point elements in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - + + Flag + SSE4.2 + String Compare + + + + + + + Compare packed strings in "a" and "b" with lengths "la" and "lb" using the control in "imm8", and store the generated index in "dst". + [strcmp_note] -FOR j := 0 to 7 - i := j*64 - IF k[j] - dst[i+63:i] := a[i+63:i] XOR b[i+63:i] - ELSE - dst[i+63:i] := 0 - FI +size := (imm8[0] ? 16 : 8) // 8 or 16-bit characters +UpperBound := (128 / size) - 1 +BoolRes := 0 +// compare all characters +aInvalid := 0 +bInvalid := 0 +FOR i := 0 to UpperBound + m := i*size + FOR j := 0 to UpperBound + n := j*size + BoolRes.word[i].bit[j] := (a[m+size-1:m] == b[n+size-1:n]) ? 1 : 0 + + // invalidate characters after EOS + IF i == la + aInvalid := 1 + FI + IF j == lb + bInvalid := 1 + FI + + // override comparisons for invalid characters + CASE (imm8[3:2]) OF + 0: // equal any + IF (!aInvalid && bInvalid) + BoolRes.word[i].bit[j] := 0 + ELSE IF (aInvalid && !bInvalid) + BoolRes.word[i].bit[j] := 0 + ELSE IF (aInvalid && bInvalid) + BoolRes.word[i].bit[j] := 0 + FI + 1: // ranges + IF (!aInvalid && bInvalid) + BoolRes.word[i].bit[j] := 0 + ELSE IF (aInvalid && !bInvalid) + BoolRes.word[i].bit[j] := 0 + ELSE IF (aInvalid && bInvalid) + BoolRes.word[i].bit[j] := 0 + FI + 2: // equal each + IF (!aInvalid && bInvalid) + BoolRes.word[i].bit[j] := 0 + ELSE IF (aInvalid && !bInvalid) + BoolRes.word[i].bit[j] := 0 + ELSE IF (aInvalid && bInvalid) + BoolRes.word[i].bit[j] := 1 + FI + 3: // equal ordered + IF (!aInvalid && bInvalid) + BoolRes.word[i].bit[j] := 0 + ELSE IF (aInvalid && !bInvalid) + BoolRes.word[i].bit[j] := 1 + ELSE IF (aInvalid && bInvalid) + BoolRes.word[i].bit[j] := 1 + FI + ESAC + ENDFOR ENDFOR -dst[MAX:512] := 0 - - -
immintrin.h
- - - Floating Point - AVX512DQ - Logical - - - Compute the bitwise XOR of packed double-precision (64-bit) floating-point elements in "a" and "b", and store the results in "dst". - - -FOR j := 0 to 7 - i := j*64 - dst[i+63:i] := a[i+63:i] XOR b[i+63:i] +// aggregate results +CASE (imm8[3:2]) OF +0: // equal any + IntRes1 := 0 + FOR i := 0 to UpperBound + FOR j := 0 to UpperBound + IntRes1[i] := IntRes1[i] OR BoolRes.word[i].bit[j] + ENDFOR + ENDFOR +1: // ranges + IntRes1 := 0 + FOR i := 0 to UpperBound + FOR j := 0 to UpperBound + IntRes1[i] := IntRes1[i] OR (BoolRes.word[i].bit[j] AND BoolRes.word[i].bit[j+1]) + j += 2 + ENDFOR + ENDFOR +2: // equal each + IntRes1 := 0 + FOR i := 0 to UpperBound + IntRes1[i] := BoolRes.word[i].bit[i] + ENDFOR +3: // equal ordered + IntRes1 := (imm8[0] ? 0xFF : 0xFFFF) + FOR i := 0 to UpperBound + k := i + FOR j := 0 to UpperBound-i + IntRes1[i] := IntRes1[i] AND BoolRes.word[k].bit[j] + k := k+1 + ENDFOR + ENDFOR +ESAC +// optionally negate results +FOR i := 0 to UpperBound + IF imm8[4] + IF imm8[5] // only negate valid + IF i >= lb // invalid, don't negate + IntRes2[i] := IntRes1[i] + ELSE // valid, negate + IntRes2[i] := -1 XOR IntRes1[i] + FI + ELSE // negate all + IntRes2[i] := -1 XOR IntRes1[i] + FI + ELSE // don't negate + IntRes2[i] := IntRes1[i] + FI ENDFOR -dst[MAX:512] := 0 +// output +IF imm8[6] // most significant bit + tmp := UpperBound + dst := tmp + DO WHILE ((tmp >= 0) AND a[tmp] == 0) + tmp := tmp - 1 + dst := tmp + OD +ELSE // least significant bit + tmp := 0 + dst := tmp + DO WHILE ((tmp <= UpperBound) AND a[tmp] == 0) + tmp := tmp + 1 + dst := tmp + OD +FI - -
immintrin.h
+ +
nmmintrin.h
 - - Floating Point - AVX512VL - AVX512DQ - Logical - - - - - Compute the bitwise XOR of packed double-precision (64-bit) floating-point elements in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Flag + SSE4.2 + String Compare + + + + + + + Compare packed strings in "a" and "b" with lengths "la" and "lb" using the control in "imm8", and returns 1 if any character in "b" was null, and 0 otherwise. + [strcmp_note] -FOR j := 0 to 1 - i := j*64 - IF k[j] - dst[i+63:i] := a[i+63:i] XOR b[i+63:i] - ELSE - dst[i+63:i] := src[i+63:i] - FI -ENDFOR -dst[MAX:128] := 0 +size := (imm8[0] ? 16 : 8) // 8 or 16-bit characters +UpperBound := (128 / size) - 1 +dst := (lb <= UpperBound) - -
immintrin.h
+ +
nmmintrin.h
 - - Floating Point - AVX512VL - AVX512DQ - Logical - - - - Compute the bitwise XOR of packed double-precision (64-bit) floating-point elements in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Flag + SSE4.2 + String Compare + + + + + + + Compare packed strings in "a" and "b" with lengths "la" and "lb" using the control in "imm8", and returns 1 if the resulting mask was non-zero, and 0 otherwise. + [strcmp_note] -FOR j := 0 to 1 - i := j*64 - IF k[j] - dst[i+63:i] := a[i+63:i] XOR b[i+63:i] - ELSE - dst[i+63:i] := 0 +size := (imm8[0] ? 16 : 8) // 8 or 16-bit characters +UpperBound := (128 / size) - 1 +BoolRes := 0 +// compare all characters +aInvalid := 0 +bInvalid := 0 +FOR i := 0 to UpperBound + m := i*size + FOR j := 0 to UpperBound + n := j*size + BoolRes.word[i].bit[j] := (a[m+size-1:m] == b[n+size-1:n]) ? 1 : 0 + + // invalidate characters after EOS + IF i == la + aInvalid := 1 + FI + IF j == lb + bInvalid := 1 + FI + + // override comparisons for invalid characters + CASE (imm8[3:2]) OF + 0: // equal any + IF (!aInvalid && bInvalid) + BoolRes.word[i].bit[j] := 0 + ELSE IF (aInvalid && !bInvalid) + BoolRes.word[i].bit[j] := 0 + ELSE IF (aInvalid && bInvalid) + BoolRes.word[i].bit[j] := 0 + FI + 1: // ranges + IF (!aInvalid && bInvalid) + BoolRes.word[i].bit[j] := 0 + ELSE IF (aInvalid && !bInvalid) + BoolRes.word[i].bit[j] := 0 + ELSE IF (aInvalid && bInvalid) + BoolRes.word[i].bit[j] := 0 + FI + 2: // equal each + IF (!aInvalid && bInvalid) + BoolRes.word[i].bit[j] := 0 + ELSE IF (aInvalid && !bInvalid) + BoolRes.word[i].bit[j] := 0 + ELSE IF (aInvalid && bInvalid) + BoolRes.word[i].bit[j] := 1 + FI + 3: // equal ordered + IF (!aInvalid && bInvalid) + BoolRes.word[i].bit[j] := 0 + ELSE IF (aInvalid && !bInvalid) + BoolRes.word[i].bit[j] := 1 + ELSE IF (aInvalid && bInvalid) + BoolRes.word[i].bit[j] := 1 + FI + ESAC + ENDFOR +ENDFOR +// aggregate results +CASE (imm8[3:2]) OF +0: // equal any + IntRes1 := 0 + FOR i := 0 to UpperBound + FOR j := 0 to UpperBound + IntRes1[i] := IntRes1[i] OR BoolRes.word[i].bit[j] + ENDFOR + ENDFOR +1: // ranges + IntRes1 := 0 + FOR i := 0 to UpperBound + FOR j := 0 to UpperBound + IntRes1[i] := IntRes1[i] OR (BoolRes.word[i].bit[j] AND BoolRes.word[i].bit[j+1]) + j += 2 + ENDFOR + ENDFOR +2: // equal each + IntRes1 := 0 + FOR i := 0 to UpperBound + IntRes1[i] := BoolRes.word[i].bit[i] + ENDFOR +3: // equal ordered + IntRes1 := (imm8[0] ? 0xFF : 0xFFFF) + FOR i := 0 to UpperBound + k := i + FOR j := 0 to UpperBound-i + IntRes1[i] := IntRes1[i] AND BoolRes.word[k].bit[j] + k := k+1 + ENDFOR + ENDFOR +ESAC +// optionally negate results +FOR i := 0 to UpperBound + IF imm8[4] + IF imm8[5] // only negate valid + IF i >= lb // invalid, don't negate + IntRes2[i] := IntRes1[i] + ELSE // valid, negate + IntRes2[i] := -1 XOR IntRes1[i] + FI + ELSE // negate all + IntRes2[i] := -1 XOR IntRes1[i] + FI + ELSE // don't negate + IntRes2[i] := IntRes1[i] FI ENDFOR -dst[MAX:128] := 0 +// output +dst := (IntRes2 != 0) - -
immintrin.h
+ +
nmmintrin.h
 - - Floating Point - AVX512VL - AVX512DQ - Logical - - - - - Compute the bitwise XOR of packed single-precision (32-bit) floating-point elements in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Flag + SSE4.2 + String Compare + + + + + + + Compare packed strings in "a" and "b" with lengths "la" and "lb" using the control in "imm8", and returns 1 if any character in "a" was null, and 0 otherwise. + [strcmp_note] -FOR j := 0 to 7 - i := j*32 - IF k[j] - dst[i+31:i] := a[i+31:i] XOR b[i+31:i] - ELSE - dst[i+31:i] := src[i+31:i] - FI -ENDFOR -dst[MAX:256] := 0 +size := (imm8[0] ? 16 : 8) // 8 or 16-bit characters +UpperBound := (128 / size) - 1 +dst := (la <= UpperBound) - -
immintrin.h
+ +
nmmintrin.h
 - - Floating Point - AVX512VL - AVX512DQ - Logical - - - - Compute the bitwise XOR of packed single-precision (32-bit) floating-point elements in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Flag + SSE4.2 + String Compare + + + + + + + Compare packed strings in "a" and "b" with lengths "la" and "lb" using the control in "imm8", and returns bit 0 of the resulting bit mask. + [strcmp_note] -FOR j := 0 to 7 - i := j*32 - IF k[j] - dst[i+31:i] := a[i+31:i] XOR b[i+31:i] - ELSE - dst[i+31:i] := 0 +size := (imm8[0] ? 16 : 8) // 8 or 16-bit characters +UpperBound := (128 / size) - 1 +BoolRes := 0 +// compare all characters +aInvalid := 0 +bInvalid := 0 +FOR i := 0 to UpperBound + m := i*size + FOR j := 0 to UpperBound + n := j*size + BoolRes.word[i].bit[j] := (a[m+size-1:m] == b[n+size-1:n]) ? 1 : 0 + + // invalidate characters after EOS + IF i == la + aInvalid := 1 + FI + IF j == lb + bInvalid := 1 + FI + + // override comparisons for invalid characters + CASE (imm8[3:2]) OF + 0: // equal any + IF (!aInvalid && bInvalid) + BoolRes.word[i].bit[j] := 0 + ELSE IF (aInvalid && !bInvalid) + BoolRes.word[i].bit[j] := 0 + ELSE IF (aInvalid && bInvalid) + BoolRes.word[i].bit[j] := 0 + FI + 1: // ranges + IF (!aInvalid && bInvalid) + BoolRes.word[i].bit[j] := 0 + ELSE IF (aInvalid && !bInvalid) + BoolRes.word[i].bit[j] := 0 + ELSE IF (aInvalid && bInvalid) + BoolRes.word[i].bit[j] := 0 + FI + 2: // equal each + IF (!aInvalid && bInvalid) + BoolRes.word[i].bit[j] := 0 + ELSE IF (aInvalid && !bInvalid) + BoolRes.word[i].bit[j] := 0 + ELSE IF (aInvalid && bInvalid) + BoolRes.word[i].bit[j] := 1 + FI + 3: // equal ordered + IF (!aInvalid && bInvalid) + BoolRes.word[i].bit[j] := 0 + ELSE IF (aInvalid && !bInvalid) + BoolRes.word[i].bit[j] := 1 + ELSE IF (aInvalid && bInvalid) + BoolRes.word[i].bit[j] := 1 + FI + ESAC + ENDFOR +ENDFOR +// aggregate results +CASE (imm8[3:2]) OF +0: // equal any + IntRes1 := 0 + FOR i := 0 to UpperBound + FOR j := 0 to UpperBound + IntRes1[i] := IntRes1[i] OR BoolRes.word[i].bit[j] + ENDFOR + ENDFOR +1: // ranges + IntRes1 := 0 + FOR i := 0 to UpperBound + FOR j := 0 to UpperBound + IntRes1[i] := IntRes1[i] OR (BoolRes.word[i].bit[j] AND BoolRes.word[i].bit[j+1]) + j += 2 + ENDFOR + ENDFOR +2: // equal each + IntRes1 := 0 + FOR i := 0 to UpperBound + IntRes1[i] := BoolRes.word[i].bit[i] + ENDFOR +3: // equal ordered + IntRes1 := (imm8[0] ? 0xFF : 0xFFFF) + FOR i := 0 to UpperBound + k := i + FOR j := 0 to UpperBound-i + IntRes1[i] := IntRes1[i] AND BoolRes.word[k].bit[j] + k := k+1 + ENDFOR + ENDFOR +ESAC +// optionally negate results +FOR i := 0 to UpperBound + IF imm8[4] + IF imm8[5] // only negate valid + IF i >= lb // invalid, don't negate + IntRes2[i] := IntRes1[i] + ELSE // valid, negate + IntRes2[i] := -1 XOR IntRes1[i] + FI + ELSE // negate all + IntRes2[i] := -1 XOR IntRes1[i] + FI + ELSE // don't negate + IntRes2[i] := IntRes1[i] FI ENDFOR -dst[MAX:256] := 0 +// output +dst := IntRes2[0] - -
immintrin.h
+ +
nmmintrin.h
 - - Floating Point - AVX512DQ - Logical - - - - - Compute the bitwise XOR of packed single-precision (32-bit) floating-point elements in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Flag + SSE4.2 + String Compare + + + + + + + Compare packed strings in "a" and "b" with lengths "la" and "lb" using the control in "imm8", and returns 1 if "b" did not contain a null character and the resulting mask was zero, and 0 otherwise. + [strcmp_note] -FOR j := 0 to 15 - i := j*32 - IF k[j] - dst[i+31:i] := a[i+31:i] XOR b[i+31:i] - ELSE - dst[i+31:i] := src[i+31:i] +size := (imm8[0] ? 16 : 8) // 8 or 16-bit characters +UpperBound := (128 / size) - 1 +BoolRes := 0 +// compare all characters +aInvalid := 0 +bInvalid := 0 +FOR i := 0 to UpperBound + m := i*size + FOR j := 0 to UpperBound + n := j*size + BoolRes.word[i].bit[j] := (a[m+size-1:m] == b[n+size-1:n]) ? 1 : 0 + + // invalidate characters after EOS + IF i == la + aInvalid := 1 + FI + IF j == lb + bInvalid := 1 + FI + + // override comparisons for invalid characters + CASE (imm8[3:2]) OF + 0: // equal any + IF (!aInvalid && bInvalid) + BoolRes.word[i].bit[j] := 0 + ELSE IF (aInvalid && !bInvalid) + BoolRes.word[i].bit[j] := 0 + ELSE IF (aInvalid && bInvalid) + BoolRes.word[i].bit[j] := 0 + FI + 1: // ranges + IF (!aInvalid && bInvalid) + BoolRes.word[i].bit[j] := 0 + ELSE IF (aInvalid && !bInvalid) + BoolRes.word[i].bit[j] := 0 + ELSE IF (aInvalid && bInvalid) + BoolRes.word[i].bit[j] := 0 + FI + 2: // equal each + IF (!aInvalid && bInvalid) + BoolRes.word[i].bit[j] := 0 + ELSE IF (aInvalid && !bInvalid) + BoolRes.word[i].bit[j] := 0 + ELSE IF (aInvalid && bInvalid) + BoolRes.word[i].bit[j] := 1 + FI + 3: // equal ordered + IF (!aInvalid && bInvalid) + BoolRes.word[i].bit[j] := 0 + ELSE IF (aInvalid && !bInvalid) + BoolRes.word[i].bit[j] := 1 + ELSE IF (aInvalid && bInvalid) + BoolRes.word[i].bit[j] := 1 + FI + ESAC + ENDFOR +ENDFOR +// aggregate results +CASE (imm8[3:2]) OF +0: // equal any + IntRes1 := 0 + FOR i := 0 to UpperBound + FOR j := 0 to UpperBound + IntRes1[i] := IntRes1[i] OR BoolRes.word[i].bit[j] + ENDFOR + ENDFOR +1: // ranges + IntRes1 := 0 + FOR i := 0 to UpperBound + FOR j := 0 to UpperBound + IntRes1[i] := IntRes1[i] OR (BoolRes.word[i].bit[j] AND BoolRes.word[i].bit[j+1]) + j += 2 + ENDFOR + ENDFOR +2: // equal each + IntRes1 := 0 + FOR i := 0 to UpperBound + IntRes1[i] := BoolRes.word[i].bit[i] + ENDFOR +3: // equal ordered + IntRes1 := (imm8[0] ? 0xFF : 0xFFFF) + FOR i := 0 to UpperBound + k := i + FOR j := 0 to UpperBound-i + IntRes1[i] := IntRes1[i] AND BoolRes.word[k].bit[j] + k := k+1 + ENDFOR + ENDFOR +ESAC +// optionally negate results +FOR i := 0 to UpperBound + IF imm8[4] + IF imm8[5] // only negate valid + IF i >= lb // invalid, don't negate + IntRes2[i] := IntRes1[i] + ELSE // valid, negate + IntRes2[i] := -1 XOR IntRes1[i] + FI + ELSE // negate all + IntRes2[i] := -1 XOR IntRes1[i] + FI + ELSE // don't negate + IntRes2[i] := IntRes1[i] FI ENDFOR -dst[MAX:512] := 0 +// output +dst := (IntRes2 == 0) AND (lb > UpperBound) - -
immintrin.h
+ +
nmmintrin.h
 - - Floating Point - AVX512DQ - Logical - - - - Compute the bitwise XOR of packed single-precision (32-bit) floating-point elements in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - + + Integer + SSE4.2 + Compare + + + + Compare packed signed 64-bit integers in "a" and "b" for greater-than, and store the results in "dst". -FOR j := 0 to 15 - i := j*32 - IF k[j] - dst[i+31:i] := a[i+31:i] XOR b[i+31:i] - ELSE - dst[i+31:i] := 0 - FI +FOR j := 0 to 1 + i := j*64 + dst[i+63:i] := ( a[i+63:i] > b[i+63:i] ) ? 0xFFFFFFFFFFFFFFFF : 0 ENDFOR -dst[MAX:512] := 0 - -
immintrin.h
+ +
nmmintrin.h
 - - Floating Point - AVX512DQ - Logical - - - Compute the bitwise XOR of packed single-precision (32-bit) floating-point elements in "a" and "b", and store the results in "dst". - - -FOR j := 0 to 15 - i := j*32 - dst[i+31:i] := a[i+31:i] XOR b[i+31:i] -ENDFOR -dst[MAX:512] := 0 + + Integer + SSE4.2 + Cryptography + + + + Starting with the initial value in "crc", accumulates a CRC32 value for unsigned 8-bit integer "v", and stores the result in "dst". + tmp1[7:0] := v[0:7] // bit reflection +tmp2[31:0] := crc[0:31] // bit reflection +tmp3[39:0] := tmp1[7:0] << 32 +tmp4[39:0] := tmp2[31:0] << 8 +tmp5[39:0] := tmp3[39:0] XOR tmp4[39:0] +tmp6[31:0] := MOD2(tmp5[39:0], 0x11EDC6F41) // remainder from polynomial division modulus 2 +dst[31:0] := tmp6[0:31] // bit reflection - -
immintrin.h
+ +
nmmintrin.h
 - - Floating Point - AVX512VL - AVX512DQ - Logical - - - - - Compute the bitwise XOR of packed single-precision (32-bit) floating-point elements in "a" and "b", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -FOR j := 0 to 3 - i := j*32 - IF k[j] - dst[i+31:i] := a[i+31:i] XOR b[i+31:i] - ELSE - dst[i+31:i] := src[i+31:i] - FI -ENDFOR -dst[MAX:128] := 0 + + Integer + SSE4.2 + Cryptography + + + + Starting with the initial value in "crc", accumulates a CRC32 value for unsigned 16-bit integer "v", and stores the result in "dst". + tmp1[15:0] := v[0:15] // bit reflection +tmp2[31:0] := crc[0:31] // bit reflection +tmp3[47:0] := tmp1[15:0] << 32 +tmp4[47:0] := tmp2[31:0] << 16 +tmp5[47:0] := tmp3[47:0] XOR tmp4[47:0] +tmp6[31:0] := MOD2(tmp5[47:0], 0x11EDC6F41) // remainder from polynomial division modulus 2 +dst[31:0] := tmp6[0:31] // bit reflection - -
immintrin.h
+ +
nmmintrin.h
 - - Floating Point - AVX512VL - AVX512DQ - Logical - - - - Compute the bitwise XOR of packed single-precision (32-bit) floating-point elements in "a" and "b", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - -FOR j := 0 to 3 - i := j*32 - IF k[j] - dst[i+31:i] := a[i+31:i] XOR b[i+31:i] - ELSE - dst[i+31:i] := 0 - FI -ENDFOR -dst[MAX:128] := 0 + + Integer + SSE4.2 + Cryptography + + + + Starting with the initial value in "crc", accumulates a CRC32 value for unsigned 32-bit integer "v", and stores the result in "dst". + tmp1[31:0] := v[0:31] // bit reflection +tmp2[31:0] := crc[0:31] // bit reflection +tmp3[63:0] := tmp1[31:0] << 32 +tmp4[63:0] := tmp2[31:0] << 32 +tmp5[63:0] := tmp3[63:0] XOR tmp4[63:0] +tmp6[31:0] := MOD2(tmp5[63:0], 0x11EDC6F41) // remainder from polynomial division modulus 2 +dst[31:0] := tmp6[0:31] // bit reflection - -
immintrin.h
+ +
nmmintrin.h
 - - AVX512IFMA52 - Arithmetic - - - - Multiply packed unsigned 52-bit integers in each 64-bit element of "b" and "c" to form a 104-bit intermediate result. Add the low 52-bit unsigned integer from the intermediate result with the corresponding unsigned 64-bit integer in "a", and store the results in "dst". - -FOR j := 0 to 7 - i := j*64 - tmp[127:0] := ZeroExtend64(b[i+51:i]) * ZeroExtend64(c[i+51:i]) - dst[i+63:i] := a[i+63:i] + ZeroExtend64(tmp[51:0]) -ENDFOR -dst[MAX:512] := 0 + + Integer + SSE4.2 + Cryptography + + + + Starting with the initial value in "crc", accumulates a CRC32 value for unsigned 64-bit integer "v", and stores the result in "dst". + tmp1[63:0] := v[0:63] // bit reflection +tmp2[31:0] := crc[0:31] // bit reflection +tmp3[95:0] := tmp1[31:0] << 32 +tmp4[95:0] := tmp2[63:0] << 64 +tmp5[95:0] := tmp3[95:0] XOR tmp4[95:0] +tmp6[31:0] := MOD2(tmp5[95:0], 0x11EDC6F41) // remainder from polynomial division modulus 2 +dst[31:0] := tmp6[0:31] // bit reflection - -
immintrin.h
+ +
nmmintrin.h
 - - AVX512IFMA52 - Arithmetic - - - - - Multiply packed unsigned 52-bit integers in each 64-bit element of "b" and "c" to form a 104-bit intermediate result. Add the low 52-bit unsigned integer from the intermediate result with the corresponding unsigned 64-bit integer in "a", and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). + + Integer + SSSE3 + Special Math Functions + + + Compute the absolute value of packed signed 8-bit integers in "a", and store the unsigned results in "dst". FOR j := 0 to 7 - i := j*64 - IF k[j] - tmp[127:0] := ZeroExtend64(b[i+51:i]) * ZeroExtend64(c[i+51:i]) - dst[i+63:i] := a[i+63:i] + ZeroExtend64(tmp[51:0]) - ELSE - dst[i+63:i] := a[i+63:i] - FI + i := j*8 + dst[i+7:i] := ABS(Int(a[i+7:i])) ENDFOR -dst[MAX:512] := 0 - -
immintrin.h
+ +
tmmintrin.h
 - - AVX512IFMA52 - Arithmetic - - - - - Multiply packed unsigned 52-bit integers in each 64-bit element of "b" and "c" to form a 104-bit intermediate result. Add the low 52-bit unsigned integer from the intermediate result with the corresponding unsigned 64-bit integer in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Integer + SSSE3 + Special Math Functions + + + Compute the absolute value of packed signed 8-bit integers in "a", and store the unsigned results in "dst". -FOR j := 0 to 7 - i := j*64 - IF k[j] - tmp[127:0] := ZeroExtend64(b[i+51:i]) * ZeroExtend64(c[i+51:i]) - dst[i+63:i] := a[i+63:i] + ZeroExtend64(tmp[51:0]) - ELSE - dst[i+63:i] := 0 - FI +FOR j := 0 to 15 + i := j*8 + dst[i+7:i] := ABS(a[i+7:i]) ENDFOR -dst[MAX:512] := 0 - -
immintrin.h
+ +
tmmintrin.h
 - - AVX512IFMA52 - AVX512VL - Arithmetic - - - - Multiply packed unsigned 52-bit integers in each 64-bit element of "b" and "c" to form a 104-bit intermediate result. Add the low 52-bit unsigned integer from the intermediate result with the corresponding unsigned 64-bit integer in "a", and store the results in "dst". + + Integer + SSSE3 + Special Math Functions + + + Compute the absolute value of packed signed 16-bit integers in "a", and store the unsigned results in "dst". FOR j := 0 to 3 - i := j*64 - tmp[127:0] := ZeroExtend64(b[i+51:i]) * ZeroExtend64(c[i+51:i]) - dst[i+63:i] := a[i+63:i] + ZeroExtend64(tmp[51:0]) + i := j*16 + dst[i+15:i] := ABS(Int(a[i+15:i])) ENDFOR -dst[MAX:256] := 0 - -
immintrin.h
+ +
tmmintrin.h
 - - AVX512IFMA52 - AVX512VL - Arithmetic - - - - - Multiply packed unsigned 52-bit integers in each 64-bit element of "b" and "c" to form a 104-bit intermediate result. Add the low 52-bit unsigned integer from the intermediate result with the corresponding unsigned 64-bit integer in "a", and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). + + Integer + SSSE3 + Special Math Functions + + + Compute the absolute value of packed signed 16-bit integers in "a", and store the unsigned results in "dst". -FOR j := 0 to 3 - i := j*64 - IF k[j] - tmp[127:0] := ZeroExtend64(b[i+51:i]) * ZeroExtend64(c[i+51:i]) - dst[i+63:i] := a[i+63:i] + ZeroExtend64(tmp[51:0]) - ELSE - dst[i+63:i] := a[i+63:i] - FI +FOR j := 0 to 7 + i := j*16 + dst[i+15:i] := ABS(a[i+15:i]) ENDFOR -dst[MAX:256] := 0 - -
immintrin.h
+ +
tmmintrin.h
 - - AVX512IFMA52 - AVX512VL - Arithmetic - - - - - Multiply packed unsigned 52-bit integers in each 64-bit element of "b" and "c" to form a 104-bit intermediate result. Add the low 52-bit unsigned integer from the intermediate result with the corresponding unsigned 64-bit integer in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Integer + SSSE3 + Special Math Functions + + + Compute the absolute value of packed signed 32-bit integers in "a", and store the unsigned results in "dst". -FOR j := 0 to 3 - i := j*64 - IF k[j] - tmp[127:0] := ZeroExtend64(b[i+51:i]) * ZeroExtend64(c[i+51:i]) - dst[i+63:i] := a[i+63:i] + ZeroExtend64(tmp[51:0]) - ELSE - dst[i+63:i] := 0 - FI +FOR j := 0 to 1 + i := j*32 + dst[i+31:i] := ABS(a[i+31:i]) ENDFOR -dst[MAX:256] := 0 - -
immintrin.h
+ +
tmmintrin.h
 - - AVX512IFMA52 - AVX512VL - Arithmetic - - - - Multiply packed unsigned 52-bit integers in each 64-bit element of "b" and "c" to form a 104-bit intermediate result. Add the low 52-bit unsigned integer from the intermediate result with the corresponding unsigned 64-bit integer in "a", and store the results in "dst". + + Integer + SSSE3 + Special Math Functions + + + Compute the absolute value of packed signed 32-bit integers in "a", and store the unsigned results in "dst". -FOR j := 0 to 1 - i := j*64 - tmp[127:0] := ZeroExtend64(b[i+51:i]) * ZeroExtend64(c[i+51:i]) - dst[i+63:i] := a[i+63:i] + ZeroExtend64(tmp[51:0]) +FOR j := 0 to 3 + i := j*32 + dst[i+31:i] := ABS(a[i+31:i]) ENDFOR -dst[MAX:128] := 0 - -
immintrin.h
+ +
tmmintrin.h
 - - AVX512IFMA52 - AVX512VL - Arithmetic - - - - - Multiply packed unsigned 52-bit integers in each 64-bit element of "b" and "c" to form a 104-bit intermediate result. Add the low 52-bit unsigned integer from the intermediate result with the corresponding unsigned 64-bit integer in "a", and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). + + Integer + SSSE3 + Swizzle + + + + Shuffle packed 8-bit integers in "a" according to shuffle control mask in the corresponding 8-bit element of "b", and store the results in "dst". -FOR j := 0 to 1 - i := j*64 - IF k[j] - tmp[127:0] := ZeroExtend64(b[i+51:i]) * ZeroExtend64(c[i+51:i]) - dst[i+63:i] := a[i+63:i] + ZeroExtend64(tmp[51:0]) +FOR j := 0 to 15 + i := j*8 + IF b[i+7] == 1 + dst[i+7:i] := 0 ELSE - dst[i+63:i] := a[i+63:i] + index[3:0] := b[i+3:i] + dst[i+7:i] := a[index*8+7:index*8] FI ENDFOR -dst[MAX:128] := 0 - -
immintrin.h
+ +
tmmintrin.h
 - - AVX512IFMA52 - AVX512VL - Arithmetic - - - - - Multiply packed unsigned 52-bit integers in each 64-bit element of "b" and "c" to form a 104-bit intermediate result. Add the low 52-bit unsigned integer from the intermediate result with the corresponding unsigned 64-bit integer in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Integer + SSSE3 + Swizzle + + + + Shuffle packed 8-bit integers in "a" according to shuffle control mask in the corresponding 8-bit element of "b", and store the results in "dst". -FOR j := 0 to 1 - i := j*64 - IF k[j] - tmp[127:0] := ZeroExtend64(b[i+51:i]) * ZeroExtend64(c[i+51:i]) - dst[i+63:i] := a[i+63:i] + ZeroExtend64(tmp[51:0]) +FOR j := 0 to 7 + i := j*8 + IF b[i+7] == 1 + dst[i+7:i] := 0 ELSE - dst[i+63:i] := 0 + index[2:0] := b[i+2:i] + dst[i+7:i] := a[index*8+7:index*8] FI ENDFOR -dst[MAX:128] := 0 - -
immintrin.h
+ +
tmmintrin.h
 - - AVX512IFMA52 - Arithmetic - - - - Multiply packed unsigned 52-bit integers in each 64-bit element of "b" and "c" to form a 104-bit intermediate result. Add the high 52-bit unsigned integer from the intermediate result with the corresponding unsigned 64-bit integer in "a", and store the results in "dst". + + Integer + SSSE3 + Miscellaneous + + + + + Concatenate 16-byte blocks in "a" and "b" into a 32-byte temporary result, shift the result right by "imm8" bytes, and store the low 16 bytes in "dst". -FOR j := 0 to 7 - i := j*64 - tmp[127:0] := ZeroExtend64(b[i+51:i]) * ZeroExtend64(c[i+51:i]) - dst[i+63:i] := a[i+63:i] + ZeroExtend64(tmp[103:52]) -ENDFOR -dst[MAX:512] := 0 +tmp[255:0] := ((a[127:0] << 128)[255:0] OR b[127:0]) >> (imm8*8) +dst[127:0] := tmp[127:0] - -
immintrin.h
+ +
tmmintrin.h
 - - AVX512IFMA52 + + Integer + SSSE3 + Miscellaneous + + + + + Concatenate 8-byte blocks in "a" and "b" into a 16-byte temporary result, shift the result right by "imm8" bytes, and store the low 16 bytes in "dst". + +tmp[127:0] := ((a[63:0] << 64)[127:0] OR b[63:0]) >> (imm8*8) +dst[63:0] := tmp[63:0] + + +
tmmintrin.h
+ + + Integer + SSSE3 Arithmetic - - - - - Multiply packed unsigned 52-bit integers in each 64-bit element of "b" and "c" to form a 104-bit intermediate result. Add the high 52-bit unsigned integer from the intermediate result with the corresponding unsigned 64-bit integer in "a", and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). + + + + Horizontally add adjacent pairs of 16-bit integers in "a" and "b", and pack the signed 16-bit results in "dst". -FOR j := 0 to 7 - i := j*64 - IF k[j] - tmp[127:0] := ZeroExtend64(b[i+51:i]) * ZeroExtend64(c[i+51:i]) - dst[i+63:i] := a[i+63:i] + ZeroExtend64(tmp[103:52]) - ELSE - dst[i+63:i] := a[i+63:i] - FI -ENDFOR -dst[MAX:512] := 0 +dst[15:0] := a[31:16] + a[15:0] +dst[31:16] := a[63:48] + a[47:32] +dst[47:32] := a[95:80] + a[79:64] +dst[63:48] := a[127:112] + a[111:96] +dst[79:64] := b[31:16] + b[15:0] +dst[95:80] := b[63:48] + b[47:32] +dst[111:96] := b[95:80] + b[79:64] +dst[127:112] := b[127:112] + b[111:96] - -
immintrin.h
+ +
tmmintrin.h
 - - AVX512IFMA52 + + Integer + SSSE3 Arithmetic - - - - - Multiply packed unsigned 52-bit integers in each 64-bit element of "b" and "c" to form a 104-bit intermediate result. Add the high 52-bit unsigned integer from the intermediate result with the corresponding unsigned 64-bit integer in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + + + Horizontally add adjacent pairs of signed 16-bit integers in "a" and "b" using saturation, and pack the signed 16-bit results in "dst". + +dst[15:0] := Saturate16(a[31:16] + a[15:0]) +dst[31:16] := Saturate16(a[63:48] + a[47:32]) +dst[47:32] := Saturate16(a[95:80] + a[79:64]) +dst[63:48] := Saturate16(a[127:112] + a[111:96]) +dst[79:64] := Saturate16(b[31:16] + b[15:0]) +dst[95:80] := Saturate16(b[63:48] + b[47:32]) +dst[111:96] := Saturate16(b[95:80] + b[79:64]) +dst[127:112] := Saturate16(b[127:112] + b[111:96]) + + +
tmmintrin.h
+ + + Integer + SSSE3 + Arithmetic + + + + Horizontally add adjacent pairs of 32-bit integers in "a" and "b", and pack the signed 32-bit results in "dst". -FOR j := 0 to 7 - i := j*64 - IF k[j] - tmp[127:0] := ZeroExtend64(b[i+51:i]) * ZeroExtend64(c[i+51:i]) - dst[i+63:i] := a[i+63:i] + ZeroExtend64(tmp[103:52]) - ELSE - dst[i+63:i] := 0 - FI -ENDFOR -dst[MAX:512] := 0 +dst[31:0] := a[63:32] + a[31:0] +dst[63:32] := a[127:96] + a[95:64] +dst[95:64] := b[63:32] + b[31:0] +dst[127:96] := b[127:96] + b[95:64] - -
immintrin.h
+ +
tmmintrin.h
 - - AVX512IFMA52 - AVX512VL + + Integer + SSSE3 Arithmetic - - - - Multiply packed unsigned 52-bit integers in each 64-bit element of "b" and "c" to form a 104-bit intermediate result. Add the high 52-bit unsigned integer from the intermediate result with the corresponding unsigned 64-bit integer in "a", and store the results in "dst". + + + + Horizontally add adjacent pairs of 16-bit integers in "a" and "b", and pack the signed 16-bit results in "dst". -FOR j := 0 to 3 - i := j*64 - tmp[127:0] := ZeroExtend64(b[i+51:i]) * ZeroExtend64(c[i+51:i]) - dst[i+63:i] := a[i+63:i] + ZeroExtend64(tmp[103:52]) -ENDFOR -dst[MAX:256] := 0 +dst[15:0] := a[31:16] + a[15:0] +dst[31:16] := a[63:48] + a[47:32] +dst[47:32] := b[31:16] + b[15:0] +dst[63:48] := b[63:48] + b[47:32] - -
immintrin.h
+ +
tmmintrin.h
 - - AVX512IFMA52 - AVX512VL + + Integer + SSSE3 Arithmetic - - - - - Multiply packed unsigned 52-bit integers in each 64-bit element of "b" and "c" to form a 104-bit intermediate result. Add the high 52-bit unsigned integer from the intermediate result with the corresponding unsigned 64-bit integer in "a", and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). + + + + Horizontally add adjacent pairs of 32-bit integers in "a" and "b", and pack the signed 32-bit results in "dst". -FOR j := 0 to 3 - i := j*64 - IF k[j] - tmp[127:0] := ZeroExtend64(b[i+51:i]) * ZeroExtend64(c[i+51:i]) - dst[i+63:i] := a[i+63:i] + ZeroExtend64(tmp[103:52]) - ELSE - dst[i+63:i] := a[i+63:i] - FI -ENDFOR -dst[MAX:256] := 0 +dst[31:0] := a[63:32] + a[31:0] +dst[63:32] := b[63:32] + b[31:0] - -
immintrin.h
+ +
tmmintrin.h
 - - AVX512IFMA52 - AVX512VL + + Integer + SSSE3 Arithmetic - - - - - Multiply packed unsigned 52-bit integers in each 64-bit element of "b" and "c" to form a 104-bit intermediate result. Add the high 52-bit unsigned integer from the intermediate result with the corresponding unsigned 64-bit integer in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + + + Horizontally add adjacent pairs of signed 16-bit integers in "a" and "b" using saturation, and pack the signed 16-bit results in "dst". -FOR j := 0 to 3 - i := j*64 - IF k[j] - tmp[127:0] := ZeroExtend64(b[i+51:i]) * ZeroExtend64(c[i+51:i]) - dst[i+63:i] := a[i+63:i] + ZeroExtend64(tmp[103:52]) - ELSE - dst[i+63:i] := 0 - FI -ENDFOR -dst[MAX:256] := 0 +dst[15:0] := Saturate16(a[31:16] + a[15:0]) +dst[31:16] := Saturate16(a[63:48] + a[47:32]) +dst[47:32] := Saturate16(b[31:16] + b[15:0]) +dst[63:48] := Saturate16(b[63:48] + b[47:32]) - -
immintrin.h
+ +
tmmintrin.h
 - - AVX512IFMA52 - AVX512VL + + Integer + SSSE3 Arithmetic - - - - Multiply packed unsigned 52-bit integers in each 64-bit element of "b" and "c" to form a 104-bit intermediate result. Add the high 52-bit unsigned integer from the intermediate result with the corresponding unsigned 64-bit integer in "a", and store the results in "dst". + + + + Horizontally subtract adjacent pairs of 16-bit integers in "a" and "b", and pack the signed 16-bit results in "dst". -FOR j := 0 to 1 - i := j*64 - tmp[127:0] := ZeroExtend64(b[i+51:i]) * ZeroExtend64(c[i+51:i]) - dst[i+63:i] := a[i+63:i] + ZeroExtend64(tmp[103:52]) -ENDFOR -dst[MAX:128] := 0 +dst[15:0] := a[15:0] - a[31:16] +dst[31:16] := a[47:32] - a[63:48] +dst[47:32] := a[79:64] - a[95:80] +dst[63:48] := a[111:96] - a[127:112] +dst[79:64] := b[15:0] - b[31:16] +dst[95:80] := b[47:32] - b[63:48] +dst[111:96] := b[79:64] - b[95:80] +dst[127:112] := b[111:96] - b[127:112] - -
immintrin.h
+ +
tmmintrin.h
 - - AVX512IFMA52 - AVX512VL + + Integer + SSSE3 Arithmetic - - - - - Multiply packed unsigned 52-bit integers in each 64-bit element of "b" and "c" to form a 104-bit intermediate result. Add the high 52-bit unsigned integer from the intermediate result with the corresponding unsigned 64-bit integer in "a", and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). + + + + Horizontally subtract adjacent pairs of signed 16-bit integers in "a" and "b" using saturation, and pack the signed 16-bit results in "dst". + +dst[15:0] := Saturate16(a[15:0] - a[31:16]) +dst[31:16] := Saturate16(a[47:32] - a[63:48]) +dst[47:32] := Saturate16(a[79:64] - a[95:80]) +dst[63:48] := Saturate16(a[111:96] - a[127:112]) +dst[79:64] := Saturate16(b[15:0] - b[31:16]) +dst[95:80] := Saturate16(b[47:32] - b[63:48]) +dst[111:96] := Saturate16(b[79:64] - b[95:80]) +dst[127:112] := Saturate16(b[111:96] - b[127:112]) + + +
tmmintrin.h
+ + + Integer + SSSE3 + Arithmetic + + + + Horizontally subtract adjacent pairs of 32-bit integers in "a" and "b", and pack the signed 32-bit results in "dst". -FOR j := 0 to 1 - i := j*64 - IF k[j] - tmp[127:0] := ZeroExtend64(b[i+51:i]) * ZeroExtend64(c[i+51:i]) - dst[i+63:i] := a[i+63:i] + ZeroExtend64(tmp[103:52]) - ELSE - dst[i+63:i] := a[i+63:i] - FI -ENDFOR -dst[MAX:128] := 0 +dst[31:0] := a[31:0] - a[63:32] +dst[63:32] := a[95:64] - a[127:96] +dst[95:64] := b[31:0] - b[63:32] +dst[127:96] := b[95:64] - b[127:96] - -
immintrin.h
+ +
tmmintrin.h
 - - AVX512IFMA52 - AVX512VL + + Integer + SSSE3 Arithmetic - - - - - Multiply packed unsigned 52-bit integers in each 64-bit element of "b" and "c" to form a 104-bit intermediate result. Add the high 52-bit unsigned integer from the intermediate result with the corresponding unsigned 64-bit integer in "a", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + + + Horizontally subtract adjacent pairs of 16-bit integers in "a" and "b", and pack the signed 16-bit results in "dst". -FOR j := 0 to 1 - i := j*64 - IF k[j] - tmp[127:0] := ZeroExtend64(b[i+51:i]) * ZeroExtend64(c[i+51:i]) - dst[i+63:i] := a[i+63:i] + ZeroExtend64(tmp[103:52]) - ELSE - dst[i+63:i] := 0 - FI -ENDFOR -dst[MAX:128] := 0 +dst[15:0] := a[15:0] - a[31:16] +dst[31:16] := a[47:32] - a[63:48] +dst[47:32] := b[15:0] - b[31:16] +dst[63:48] := b[47:32] - b[63:48] - -
immintrin.h
+ +
tmmintrin.h
 - - AVX512VBMI - Bit Manipulation - - - For each 64-bit element in "b", select 8 unaligned bytes using a byte-granular shift control within the corresponding 64-bit element of "a", and store the 8 assembled bytes to the corresponding 64-bit element of "dst". + + Integer + SSSE3 + Arithmetic + + + + Horizontally subtract adjacent pairs of 32-bit integers in "a" and "b", and pack the signed 32-bit results in "dst". -FOR i := 0 to 7 - q := i * 64 - FOR j := 0 to 7 - tmp8 := 0 - ctrl := a[q+j*8+7:q+j*8] & 63 - FOR l := 0 to 7 - tmp8[l] := b[q+((ctrl+l) & 63)] - ENDFOR - dst[q+j*8+7:q+j*8] := tmp8[7:0] - ENDFOR -ENDFOR -dst[MAX:512] := 0 +dst[31:0] := a[31:0] - a[63:32] +dst[63:32] := b[31:0] - b[63:32] - -
immintrin.h
+ +
tmmintrin.h
 - - AVX512VBMI - Bit Manipulation - - - - - For each 64-bit element in "b", select 8 unaligned bytes using a byte-granular shift control within the corresponding 64-bit element of "a", and store the 8 assembled bytes to the corresponding 64-bit element of "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Integer + SSSE3 + Arithmetic + + + + Horizontally subtract adjacent pairs of signed 16-bit integers in "a" and "b" using saturation, and pack the signed 16-bit results in "dst". -FOR i := 0 to 7 - q := i * 64 - FOR j := 0 to 7 - tmp8 := 0 - ctrl := a[q+j*8+7:q+j*8] & 63 - FOR l := 0 to 7 - tmp8[l] := b[q+((ctrl+l) & 63)] - ENDFOR - IF k[i*8+j] - dst[q+j*8+7:q+j*8] := tmp8[7:0] - ELSE - dst[q+j*8+7:q+j*8] := src[q+j*8+7:q+j*8] - FI - ENDFOR -ENDFOR -dst[MAX:512] := 0 +dst[15:0] := Saturate16(a[15:0] - a[31:16]) +dst[31:16] := Saturate16(a[47:32] - a[63:48]) +dst[47:32] := Saturate16(b[15:0] - b[31:16]) +dst[63:48] := Saturate16(b[47:32] - b[63:48]) - -
immintrin.h
+ +
tmmintrin.h
 - - AVX512VBMI - Bit Manipulation - - - - For each 64-bit element in "b", select 8 unaligned bytes using a byte-granular shift control within the corresponding 64-bit element of "a", and store the 8 assembled bytes to the corresponding 64-bit element of "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Integer + SSSE3 + Arithmetic + + + + Vertically multiply each unsigned 8-bit integer from "a" with the corresponding signed 8-bit integer from "b", producing intermediate signed 16-bit integers. Horizontally add adjacent pairs of intermediate signed 16-bit integers, and pack the saturated results in "dst". -FOR i := 0 to 7 - q := i * 64 - FOR j := 0 to 7 - tmp8 := 0 - ctrl := a[q+j*8+7:q+j*8] & 63 - FOR l := 0 to 7 - tmp8[l] := b[q+((ctrl+l) & 63)] - ENDFOR - IF k[i*8+j] - dst[q+j*8+7:q+j*8] := tmp8[7:0] - ELSE - dst[q+j*8+7:q+j*8] := 0 - FI - ENDFOR +FOR j := 0 to 7 + i := j*16 + dst[i+15:i] := Saturate16( a[i+15:i+8]*b[i+15:i+8] + a[i+7:i]*b[i+7:i] ) ENDFOR -dst[MAX:512] := 0 - -
immintrin.h
+ +
tmmintrin.h
 - - AVX512VBMI - AVX512VL - Bit Manipulation - - - For each 64-bit element in "b", select 8 unaligned bytes using a byte-granular shift control within the corresponding 64-bit element of "a", and store the 8 assembled bytes to the corresponding 64-bit element of "dst". + + Integer + SSSE3 + Arithmetic + + + + Vertically multiply each unsigned 8-bit integer from "a" with the corresponding signed 8-bit integer from "b", producing intermediate signed 16-bit integers. Horizontally add adjacent pairs of intermediate signed 16-bit integers, and pack the saturated results in "dst". -FOR i := 0 to 3 - q := i * 64 - FOR j := 0 to 7 - tmp8 := 0 - ctrl := a[q+j*8+7:q+j*8] & 63 - FOR l := 0 to 7 - tmp8[l] := b[q+((ctrl+l) & 63)] - ENDFOR - dst[q+j*8+7:q+j*8] := tmp8[7:0] - ENDFOR +FOR j := 0 to 3 + i := j*16 + dst[i+15:i] := Saturate16( a[i+15:i+8]*b[i+15:i+8] + a[i+7:i]*b[i+7:i] ) ENDFOR -dst[MAX:256] := 0 - -
immintrin.h
+ +
tmmintrin.h
 - - AVX512VBMI - AVX512VL - Bit Manipulation - - - - - For each 64-bit element in "b", select 8 unaligned bytes using a byte-granular shift control within the corresponding 64-bit element of "a", and store the 8 assembled bytes to the corresponding 64-bit element of "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Integer + SSSE3 + Arithmetic + + + + Multiply packed signed 16-bit integers in "a" and "b", producing intermediate signed 32-bit integers. Truncate each intermediate integer to the 18 most significant bits, round by adding 1, and store bits [16:1] to "dst". -FOR i := 0 to 3 - q := i * 64 - FOR j := 0 to 7 - tmp8 := 0 - ctrl := a[q+j*8+7:q+j*8] & 63 - FOR l := 0 to 7 - tmp8[l] := b[q+((ctrl+l) & 63)] - ENDFOR - IF k[i*8+j] - dst[q+j*8+7:q+j*8] := tmp8[7:0] - ELSE - dst[q+j*8+7:q+j*8] := src[q+j*8+7:q+j*8] - FI - ENDFOR +FOR j := 0 to 7 + i := j*16 + tmp[31:0] := ((SignExtend32(a[i+15:i]) * SignExtend32(b[i+15:i])) >> 14) + 1 + dst[i+15:i] := tmp[16:1] ENDFOR -dst[MAX:256] := 0 - -
immintrin.h
+ +
tmmintrin.h
 - - AVX512VBMI - AVX512VL - Bit Manipulation - - - - For each 64-bit element in "b", select 8 unaligned bytes using a byte-granular shift control within the corresponding 64-bit element of "a", and store the 8 assembled bytes to the corresponding 64-bit element of "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Integer + SSSE3 + Arithmetic + + + + Multiply packed signed 16-bit integers in "a" and "b", producing intermediate signed 32-bit integers. Truncate each intermediate integer to the 18 most significant bits, round by adding 1, and store bits [16:1] to "dst". -FOR i := 0 to 3 - q := i * 64 - FOR j := 0 to 7 - tmp8 := 0 - ctrl := a[q+j*8+7:q+j*8] & 63 - FOR l := 0 to 7 - tmp8[l] := b[q+((ctrl+l) & 63)] - ENDFOR - IF k[i*8+j] - dst[q+j*8+7:q+j*8] := tmp8[7:0] - ELSE - dst[q+j*8+7:q+j*8] := 0 - FI - ENDFOR +FOR j := 0 to 3 + i := j*16 + tmp[31:0] := ((SignExtend32(a[i+15:i]) * SignExtend32(b[i+15:i])) >> 14) + 1 + dst[i+15:i] := tmp[16:1] ENDFOR -dst[MAX:256] := 0 - -
immintrin.h
+ +
tmmintrin.h
 - - AVX512VBMI - AVX512VL - Bit Manipulation - - - For each 64-bit element in "b", select 8 unaligned bytes using a byte-granular shift control within the corresponding 64-bit element of "a", and store the 8 assembled bytes to the corresponding 64-bit element of "dst". + + Integer + SSSE3 + Arithmetic + + + + Negate packed 8-bit integers in "a" when the corresponding signed 8-bit integer in "b" is negative, and store the results in "dst". Element in "dst" are zeroed out when the corresponding element in "b" is zero. -FOR i := 0 to 1 - q := i * 64 - FOR j := 0 to 7 - tmp8 := 0 - ctrl := a[q+j*8+7:q+j*8] & 63 - FOR l := 0 to 7 - tmp8[l] := b[q+((ctrl+l) & 63)] - ENDFOR - dst[q+j*8+7:q+j*8] := tmp8[7:0] - ENDFOR +FOR j := 0 to 15 + i := j*8 + IF b[i+7:i] < 0 + dst[i+7:i] := -(a[i+7:i]) + ELSE IF b[i+7:i] == 0 + dst[i+7:i] := 0 + ELSE + dst[i+7:i] := a[i+7:i] + FI ENDFOR -dst[MAX:128] := 0 - -
immintrin.h
+ +
tmmintrin.h
 - - AVX512VBMI - AVX512VL - Bit Manipulation - - - - - For each 64-bit element in "b", select 8 unaligned bytes using a byte-granular shift control within the corresponding 64-bit element of "a", and store the 8 assembled bytes to the corresponding 64-bit element of "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Integer + SSSE3 + Arithmetic + + + + Negate packed 16-bit integers in "a" when the corresponding signed 16-bit integer in "b" is negative, and store the results in "dst". Element in "dst" are zeroed out when the corresponding element in "b" is zero. -FOR i := 0 to 1 - q := i * 64 - FOR j := 0 to 7 - tmp8 := 0 - ctrl := a[q+j*8+7:q+j*8] & 63 - FOR l := 0 to 7 - tmp8[l] := b[q+((ctrl+l) & 63)] - ENDFOR - IF k[i*8+j] - dst[q+j*8+7:q+j*8] := tmp8[7:0] - ELSE - dst[q+j*8+7:q+j*8] := src[q+j*8+7:q+j*8] - FI - ENDFOR +FOR j := 0 to 7 + i := j*16 + IF b[i+15:i] < 0 + dst[i+15:i] := -(a[i+15:i]) + ELSE IF b[i+15:i] == 0 + dst[i+15:i] := 0 + ELSE + dst[i+15:i] := a[i+15:i] + FI ENDFOR -dst[MAX:128] := 0 - -
immintrin.h
+ +
tmmintrin.h
 - - AVX512VBMI - AVX512VL - Bit Manipulation - - - - For each 64-bit element in "b", select 8 unaligned bytes using a byte-granular shift control within the corresponding 64-bit element of "a", and store the 8 assembled bytes to the corresponding 64-bit element of "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Integer + SSSE3 + Arithmetic + + + + Negate packed 32-bit integers in "a" when the corresponding signed 32-bit integer in "b" is negative, and store the results in "dst". Element in "dst" are zeroed out when the corresponding element in "b" is zero. -FOR i := 0 to 1 - q := i * 64 - FOR j := 0 to 7 - tmp8 := 0 - ctrl := a[q+j*8+7:q+j*8] & 63 - FOR l := 0 to 7 - tmp8[l] := b[q+((ctrl+l) & 63)] - ENDFOR - IF k[i*8+j] - dst[q+j*8+7:q+j*8] := tmp8[7:0] - ELSE - dst[q+j*8+7:q+j*8] := 0 - FI - ENDFOR +FOR j := 0 to 3 + i := j*32 + IF b[i+31:i] < 0 + dst[i+31:i] := -(a[i+31:i]) + ELSE IF b[i+31:i] == 0 + dst[i+31:i] := 0 + ELSE + dst[i+31:i] := a[i+31:i] + FI ENDFOR -dst[MAX:128] := 0 - -
immintrin.h
+ +
tmmintrin.h
 - - AVX512VBMI - Swizzle - - - Shuffle 8-bit integers in "a" across lanes using the corresponding index in "idx", and store the results in "dst". + + Integer + SSSE3 + Arithmetic + + + + Negate packed 8-bit integers in "a" when the corresponding signed 8-bit integer in "b" is negative, and store the results in "dst". Element in "dst" are zeroed out when the corresponding element in "b" is zero. -FOR j := 0 to 63 +FOR j := 0 to 7 i := j*8 - id := idx[i+5:i]*8 - dst[i+7:i] := a[id+7:id] + IF b[i+7:i] < 0 + dst[i+7:i] := -(a[i+7:i]) + ELSE IF b[i+7:i] == 0 + dst[i+7:i] := 0 + ELSE + dst[i+7:i] := a[i+7:i] + FI ENDFOR -dst[MAX:512] := 0 - -
immintrin.h
+ +
tmmintrin.h
 - - AVX512VBMI - Swizzle - - - - - Shuffle 8-bit integers in "a" across lanes using the corresponding index in "idx", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). + + Integer + SSSE3 + Arithmetic + + + + Negate packed 16-bit integers in "a" when the corresponding signed 16-bit integer in "b" is negative, and store the results in "dst". Element in "dst" are zeroed out when the corresponding element in "b" is zero. -FOR j := 0 to 63 - i := j*8 - id := idx[i+5:i]*8 - IF k[j] - dst[i+7:i] := a[id+7:id] +FOR j := 0 to 3 + i := j*16 + IF b[i+15:i] < 0 + dst[i+15:i] := -(a[i+15:i]) + ELSE IF b[i+15:i] == 0 + dst[i+15:i] := 0 ELSE - dst[i+7:i] := src[i+7:i] + dst[i+15:i] := a[i+15:i] FI ENDFOR -dst[MAX:512] := 0 - -
immintrin.h
+ +
tmmintrin.h
 - - AVX512VBMI - Swizzle - - - - Shuffle 8-bit integers in "a" across lanes using the corresponding index in "idx", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). + + Integer + SSSE3 + Arithmetic + + + + Negate packed 32-bit integers in "a" when the corresponding signed 32-bit integer in "b" is negative, and store the results in "dst". Element in "dst" are zeroed out when the corresponding element in "b" is zero. -FOR j := 0 to 63 - i := j*8 - id := idx[i+5:i]*8 - IF k[j] - dst[i+7:i] := a[id+7:id] +FOR j := 0 to 1 + i := j*32 + IF b[i+31:i] < 0 + dst[i+31:i] := -(a[i+31:i]) + ELSE IF b[i+31:i] == 0 + dst[i+31:i] := 0 ELSE - dst[i+7:i] := 0 + dst[i+31:i] := a[i+31:i] FI ENDFOR -dst[MAX:512] := 0 - + +
tmmintrin.h
+ + + TSC + General Support + + + Copy the current 64-bit value of the processor's time-stamp counter into "dst". + dst[63:0] := TimeStampCounter + + +
immintrin.h
+ + + TSXLDTRK + Miscellaneous + + Mark the start of a TSX (HLE/RTM) suspend load address tracking region. If this is used inside a transactional region, subsequent loads are not added to the read set of the transaction. If this is used inside a suspend load address tracking region it will cause transaction abort. If this is used outside of a transactional region it behaves like a NOP. + +
immintrin.h
+ + + TSXLDTRK + Miscellaneous + + Mark the end of a TSX (HLE/RTM) suspend load address tracking region. If this is used inside a suspend load address tracking region it will end the suspend region and all following load addresses will be added to the transaction read set. If this is used inside an active transaction but not in a suspend region it will cause transaction abort. If this is used outside of a transactional region it behaves like a NOP. +
immintrin.h
 - - AVX512VBMI + + Integer AVX512VL - Swizzle - - - Shuffle 8-bit integers in "a" across lanes using the corresponding index in "idx", and store the results in "dst". - -FOR j := 0 to 31 - i := j*8 - id := idx[i+4:i]*8 - dst[i+7:i] := a[id+7:id] + VAES + Cryptography + + + + Perform the last round of an AES encryption flow on data (state) in "a" using the round key in "RoundKey", and store the results in "dst"." + FOR j := 0 to 1 + i := j*128 + a[i+127:i] := ShiftRows(a[i+127:i]) + a[i+127:i] := SubBytes(a[i+127:i]) + dst[i+127:i] := a[i+127:i] XOR RoundKey[i+127:i] ENDFOR dst[MAX:256] := 0 - +
immintrin.h
 - - AVX512VBMI + + Integer AVX512VL - Swizzle - - - - - Shuffle 8-bit integers in "a" across lanes using the corresponding index in "idx", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -FOR j := 0 to 31 - i := j*8 - id := idx[i+4:i]*8 - IF k[j] - dst[i+7:i] := a[id+7:id] - ELSE - dst[i+7:i] := src[i+7:i] - FI + VAES + Cryptography + + + + Perform one round of an AES encryption flow on data (state) in "a" using the round key in "RoundKey", and store the results in "dst"." + FOR j := 0 to 1 + i := j*128 + a[i+127:i] := ShiftRows(a[i+127:i]) + a[i+127:i] := SubBytes(a[i+127:i]) + a[i+127:i] := MixColumns(a[i+127:i]) + dst[i+127:i] := a[i+127:i] XOR RoundKey[i+127:i] ENDFOR dst[MAX:256] := 0 - +
immintrin.h
 - - AVX512VBMI + + Integer AVX512VL - Swizzle - - - - Shuffle 8-bit integers in "a" across lanes using the corresponding index in "idx", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - -FOR j := 0 to 31 - i := j*8 - id := idx[i+4:i]*8 - IF k[j] - dst[i+7:i] := a[id+7:id] - ELSE - dst[i+7:i] := 0 - FI + VAES + Cryptography + + + + Perform the last round of an AES decryption flow on data (state) in "a" using the round key in "RoundKey", and store the results in "dst". + FOR j := 0 to 1 + i := j*128 + a[i+127:i] := InvShiftRows(a[i+127:i]) + a[i+127:i] := InvSubBytes(a[i+127:i]) + dst[i+127:i] := a[i+127:i] XOR RoundKey[i+127:i] ENDFOR dst[MAX:256] := 0 - +
immintrin.h
 - - AVX512VBMI + + Integer AVX512VL - Swizzle - - - Shuffle 8-bit integers in "a" using the corresponding index in "idx", and store the results in "dst". - -FOR j := 0 to 15 - i := j*8 - id := idx[i+3:i]*8 - dst[i+7:i] := a[id+7:id] + VAES + Cryptography + + + + Perform one round of an AES decryption flow on data (state) in "a" using the round key in "RoundKey", and store the results in "dst". + FOR j := 0 to 1 + i := j*128 + a[i+127:i] := InvShiftRows(a[i+127:i]) + a[i+127:i] := InvSubBytes(a[i+127:i]) + a[i+127:i] := InvMixColumns(a[i+127:i]) + dst[i+127:i] := a[i+127:i] XOR RoundKey[i+127:i] ENDFOR -dst[MAX:128] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - - AVX512VBMI - AVX512VL - Swizzle - - - - - Shuffle 8-bit integers in "a" using the corresponding index in "idx", and store the results in "dst" using writemask "k" (elements are copied from "src" when the corresponding mask bit is not set). - -FOR j := 0 to 15 - i := j*8 - id := idx[i+3:i]*8 - IF k[j] - dst[i+7:i] := a[id+7:id] + + Integer + VPCLMULQDQ + Application-Targeted + + + + + Carry-less multiplication of one quadword of + 'b' by one quadword of 'c', stores + the 128-bit result in 'dst'. The immediate 'Imm8' is + used to determine which quadwords of 'b' + and 'c' should be used. + +DEFINE PCLMUL128(X,Y) { + FOR i := 0 to 63 + TMP[i] := X[ 0 ] and Y[ i ] + FOR j := 1 to i + TMP[i] := TMP[i] xor (X[ j ] and Y[ i - j ]) + ENDFOR + DEST[ i ] := TMP[ i ] + ENDFOR + FOR i := 64 to 126 + TMP[i] := 0 + FOR j := i - 63 to 63 + TMP[i] := TMP[i] xor (X[ j ] and Y[ i - j ]) + ENDFOR + DEST[ i ] := TMP[ i ] + ENDFOR + DEST[127] := 0 + RETURN DEST // 128b vector +} +FOR i := 0 to 3 + IF Imm8[0] == 0 + TEMP1 := b.m128[i].qword[0] ELSE - dst[i+7:i] := src[i+7:i] + TEMP1 := b.m128[i].qword[1] FI + IF Imm8[4] == 0 + TEMP2 := c.m128[i].qword[0] + ELSE + TEMP2 := c.m128[i].qword[1] + FI + dst.m128[i] := PCLMUL128(TEMP1, TEMP2) ENDFOR -dst[MAX:128] := 0 +dst[MAX:512] := 0 - +
immintrin.h
 - - AVX512VBMI + + Integer AVX512VL - Swizzle - - - - Shuffle 8-bit integers in "a" using the corresponding index in "idx", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - -FOR j := 0 to 15 - i := j*8 - id := idx[i+3:i]*8 - IF k[j] - dst[i+7:i] := a[id+7:id] + VPCLMULQDQ + Application-Targeted + + + + + Carry-less multiplication of one quadword of + 'b' by one quadword of 'c', stores + the 128-bit result in 'dst'. The immediate 'Imm8' is + used to determine which quadwords of 'b' + and 'c' should be used. + +DEFINE PCLMUL128(X,Y) { + FOR i := 0 to 63 + TMP[i] := X[ 0 ] and Y[ i ] + FOR j := 1 to i + TMP[i] := TMP[i] xor (X[ j ] and Y[ i - j ]) + ENDFOR + DEST[ i ] := TMP[ i ] + ENDFOR + FOR i := 64 to 126 + TMP[i] := 0 + FOR j := i - 63 to 63 + TMP[i] := TMP[i] xor (X[ j ] and Y[ i - j ]) + ENDFOR + DEST[ i ] := TMP[ i ] + ENDFOR + DEST[127] := 0 + RETURN DEST // 128b vector +} +FOR i := 0 to 1 + IF Imm8[0] == 0 + TEMP1 := b.m128[i].qword[0] ELSE - dst[i+7:i] := 0 + TEMP1 := b.m128[i].qword[1] + FI + IF Imm8[4] == 0 + TEMP2 := c.m128[i].qword[0] + ELSE + TEMP2 := c.m128[i].qword[1] FI + dst.m128[i] := PCLMUL128(TEMP1, TEMP2) ENDFOR -dst[MAX:128] := 0 +dst[MAX:256] := 0 - +
immintrin.h
 - - - - - - - AVX512VBMI - Swizzle - - - - Shuffle 8-bit integers in "a" and "b" across lanes using the corresponding selector and index in "idx", and store the results in "dst". - -FOR j := 0 to 63 - i := j*8 - off := 8*idx[i+5:i] - dst[i+7:i] := idx[i+6] ? b[off+7:off] : a[off+7:off] + + Flag + WAITPKG + Miscellaneous + + + + Directs the processor to enter an implementation-dependent optimized state until the TSC reaches or exceeds the value specified in "counter". Bit 0 of "ctrl" selects between a lower power (cleared) or faster wakeup (set) optimized state. Returns the carry flag (CF). + +
immintrin.h
+ + + Flag + WAITPKG + Miscellaneous + + + + Directs the processor to enter an implementation-dependent optimized state while monitoring a range of addresses. The instruction wakes up when the TSC reaches or exceeds the value specified in "counter" (if the monitoring hardware did not trigger beforehand). Bit 0 of "ctrl" selects between a lower power (cleared) or faster wakeup (set) optimized state. Returns the carry flag (CF). + +
immintrin.h
+ + + WAITPKG + Miscellaneous + + + Sets up a linear address range to be + monitored by hardware and activates the + monitor. The address range should be a writeback + memory caching type. The address is + contained in "a". + +
immintrin.h
+ + + WBNOINVD + Miscellaneous + + + Write back and do not flush internal caches. + Initiate writing-back without flushing of external + caches. + +
immintrin.h
+ + + XSAVE + OS-Targeted + + + Copy up to 64-bits from the value of the extended control register (XCR) specified by "a" into "dst". Currently only XFEATURE_ENABLED_MASK XCR is supported. + dst[63:0] := XCR[a] + + +
immintrin.h
+ + + XSAVE + OS-Targeted + + + + Perform a full or partial restore of the enabled processor states using the state information stored in memory at "mem_addr". State is restored based on bits [62:0] in "rs_mask", "XCR0", and "mem_addr.HEADER.XSTATE_BV". "mem_addr" must be aligned on a 64-byte boundary. + st_mask := mem_addr.HEADER.XSTATE_BV[62:0] +FOR i := 0 to 62 + IF (rs_mask[i] AND XCR0[i]) + IF st_mask[i] + CASE (i) OF + 0: ProcessorState[x87_FPU] := mem_addr.FPUSSESave_Area[FPU] + 1: ProcessorState[SSE] := mem_addr.FPUSSESaveArea[SSE] + DEFAULT: ProcessorState[i] := mem_addr.Ext_Save_Area[i] + ESAC + ELSE + // ProcessorExtendedState := Processor Supplied Values + CASE (i) OF + 1: MXCSR := mem_addr.FPUSSESave_Area[SSE] + ESAC + FI + FI + i := i + 1 ENDFOR -dst[MAX:512] := 0 - +
immintrin.h
 - - AVX512VBMI - Swizzle - - - - - Shuffle 8-bit integers in "a" and "b" across lanes using the corresponding selector and index in "idx", and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). - -FOR j := 0 to 63 - i := j*8 - IF k[j] - off := 8*idx[i+5:i] - dst[i+7:i] := idx[i+6] ? b[off+7:off] : a[off+7:off] - ELSE - dst[i+7:i] := a[i+7:i] + + XSAVE + OS-Targeted + + + + Perform a full or partial restore of the enabled processor states using the state information stored in memory at "mem_addr". State is restored based on bits [62:0] in "rs_mask", "XCR0", and "mem_addr.HEADER.XSTATE_BV". "mem_addr" must be aligned on a 64-byte boundary. + st_mask := mem_addr.HEADER.XSTATE_BV[62:0] +FOR i := 0 to 62 + IF (rs_mask[i] AND XCR0[i]) + IF st_mask[i] + CASE (i) OF + 0: ProcessorState[x87_FPU] := mem_addr.FPUSSESave_Area[FPU] + 1: ProcessorState[SSE] := mem_addr.FPUSSESaveArea[SSE] + DEFAULT: ProcessorState[i] := mem_addr.Ext_Save_Area[i] + ESAC + ELSE + // ProcessorExtendedState := Processor Supplied Values + CASE (i) OF + 1: MXCSR := mem_addr.FPUSSESave_Area[SSE] + ESAC + FI FI + i := i + 1 ENDFOR -dst[MAX:512] := 0 - +
immintrin.h
 - - AVX512VBMI - Swizzle - - - - - Shuffle 8-bit integers in "a" and "b" across lanes using the corresponding selector and index in "idx", and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). - -FOR j := 0 to 63 - i := j*8 - IF k[j] - off := 8*idx[i+5:i] - dst[i+7:i] := idx[i+6] ? b[off+7:off] : a[off+7:off] - ELSE - dst[i+7:i] := a[i+7:i] + + XSAVE + OS-Targeted + + + + Perform a full or partial save of the enabled processor states to memory at "mem_addr". State is saved based on bits [62:0] in "save_mask" and "XCR0". "mem_addr" must be aligned on a 64-byte boundary. + mask[62:0] := save_mask[62:0] AND XCR0[62:0] +FOR i := 0 to 62 + IF mask[i] + CASE (i) OF + 0: mem_addr.FPUSSESave_Area[FPU] := ProcessorState[x87_FPU] + 1: mem_addr.FPUSSESaveArea[SSE] := ProcessorState[SSE] + DEFAULT: mem_addr.Ext_Save_Area[i] := ProcessorState[i] + ESAC + mem_addr.HEADER.XSTATE_BV[i] := INIT_FUNCTION[i] FI + i := i + 1 ENDFOR -dst[MAX:512] := 0 - +
immintrin.h
 - - AVX512VBMI - Swizzle - - - - - Shuffle 8-bit integers in "a" and "b" across lanes using the corresponding selector and index in "idx", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - -FOR j := 0 to 63 - i := j*8 - IF k[j] - off := 8*idx[i+5:i] - dst[i+7:i] := idx[i+6] ? b[off+7:off] : a[off+7:off] - ELSE - dst[i+7:i] := 0 + + XSAVE + OS-Targeted + + + + Perform a full or partial save of the enabled processor states to memory at "mem_addr". State is saved based on bits [62:0] in "save_mask" and "XCR0". "mem_addr" must be aligned on a 64-byte boundary. + mask[62:0] := save_mask[62:0] AND XCR0[62:0] +FOR i := 0 to 62 + IF mask[i] + CASE (i) OF + 0: mem_addr.FPUSSESave_Area[FPU] := ProcessorState[x87_FPU] + 1: mem_addr.FPUSSESaveArea[SSE] := ProcessorState[SSE] + DEFAULT: mem_addr.Ext_Save_Area[i] := ProcessorState[i] + ESAC + mem_addr.HEADER.XSTATE_BV[i] := INIT_FUNCTION[i] FI + i := i + 1 ENDFOR -dst[MAX:512] := 0 - - +
immintrin.h
 - - AVX512VBMI - AVX512VL - Swizzle - - - - Shuffle 8-bit integers in "a" and "b" across lanes using the corresponding selector and index in "idx", and store the results in "dst". - -FOR j := 0 to 31 - i := j*8 - off := 8*idx[i+4:i] - dst[i+7:i] := idx[i+6] ? b[off+5:off] : a[off+7:off] + + XSAVE + XSAVEOPT + OS-Targeted + + + + Perform a full or partial save of the enabled processor states to memory at "mem_addr". State is saved based on bits [62:0] in "save_mask" and "XCR0". "mem_addr" must be aligned on a 64-byte boundary. The hardware may optimize the manner in which data is saved. The performance of this instruction will be equal to or better than using the XSAVE instruction. + mask[62:0] := save_mask[62:0] AND XCR0[62:0] +FOR i := 0 to 62 + IF mask[i] + CASE (i) OF + 0: mem_addr.FPUSSESave_Area[FPU] := ProcessorState[x87_FPU] + 1: mem_addr.FPUSSESaveArea[SSE] := ProcessorState[SSE] + 2: mem_addr.EXT_SAVE_Area2[YMM] := ProcessorState[YMM] + DEFAULT: mem_addr.Ext_Save_Area[i] := ProcessorState[i] + ESAC + mem_addr.HEADER.XSTATE_BV[i] := INIT_FUNCTION[i] + FI + i := i + 1 ENDFOR -dst[MAX:256] := 0 - +
immintrin.h
 - - AVX512VBMI - AVX512VL - Swizzle - - - - - Shuffle 8-bit integers in "a" and "b" across lanes using the corresponding selector and index in "idx", and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). - -FOR j := 0 to 31 - i := j*8 - IF k[j] - off := 8*idx[i+4:i] - dst[i+7:i] := idx[i+5] ? b[off+7:off] : a[off+7:off] - ELSE - dst[i+7:i] := a[i+7:i] + + XSAVE + XSAVEOPT + OS-Targeted + + + + Perform a full or partial save of the enabled processor states to memory at "mem_addr". State is saved based on bits [62:0] in "save_mask" and "XCR0". "mem_addr" must be aligned on a 64-byte boundary. The hardware may optimize the manner in which data is saved. The performance of this instruction will be equal to or better than using the XSAVE64 instruction. + mask[62:0] := save_mask[62:0] AND XCR0[62:0] +FOR i := 0 to 62 + IF mask[i] + CASE (i) OF + 0: mem_addr.FPUSSESave_Area[FPU] := ProcessorState[x87_FPU] + 1: mem_addr.FPUSSESaveArea[SSE] := ProcessorState[SSE] + 2: mem_addr.EXT_SAVE_Area2[YMM] := ProcessorState[YMM] + DEFAULT: mem_addr.Ext_Save_Area[i] := ProcessorState[i] + ESAC + mem_addr.HEADER.XSTATE_BV[i] := INIT_FUNCTION[i] FI + i := i + 1 ENDFOR -dst[MAX:256] := 0 - +
immintrin.h
 - - AVX512VBMI - AVX512VL - Swizzle - - - - - Shuffle 8-bit integers in "a" and "b" across lanes using the corresponding selector and index in "idx", and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). + + XSAVE + OS-Targeted + + + + Copy 64-bits from "val" to the extended control register (XCR) specified by "a". Currently only XFEATURE_ENABLED_MASK XCR is supported. -FOR j := 0 to 31 - i := j*8 - IF k[j] - off := 8*idx[i+4:i] - dst[i+7:i] := idx[i+5] ? b[off+7:off] : a[off+7:off] - ELSE - dst[i+7:i] := a[i+7:i] +XCR[a] := val[63:0] + + +
immintrin.h
+ + + XSAVE + XSAVEC + OS-Targeted + + + + Perform a full or partial save of the enabled processor states to memory at "mem_addr"; xsavec differs from xsave in that it uses compaction and that it may use init optimization. State is saved based on bits [62:0] in "save_mask" and "XCR0". "mem_addr" must be aligned on a 64-byte boundary. + mask[62:0] := save_mask[62:0] AND XCR0[62:0] +FOR i := 0 to 62 + IF mask[i] + CASE (i) OF + 0: mem_addr.FPUSSESave_Area[FPU] := ProcessorState[x87_FPU] + 1: mem_addr.FPUSSESaveArea[SSE] := ProcessorState[SSE] + DEFAULT: mem_addr.Ext_Save_Area[i] := ProcessorState[i] + ESAC + mem_addr.HEADER.XSTATE_BV[i] := INIT_FUNCTION[i] FI + i := i + 1 ENDFOR -dst[MAX:256] := 0 - +
immintrin.h
 - - AVX512VBMI - AVX512VL - Swizzle - - - - - Shuffle 8-bit integers in "a" and "b" across lanes using the corresponding selector and index in "idx", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - -FOR j := 0 to 31 - i := j*8 - IF k[j] - off := 8*idx[i+4:i] - dst[i+7:i] := idx[i+5] ? b[off+7:off] : a[off+7:off] - ELSE - dst[i+7:i] := 0 + + XSAVE + XSS + OS-Targeted + + + + Perform a full or partial save of the enabled processor states to memory at "mem_addr"; xsaves differs from xsave in that it can save state components corresponding to bits set in IA32_XSS MSR and that it may use the modified optimization. State is saved based on bits [62:0] in "save_mask" and "XCR0". "mem_addr" must be aligned on a 64-byte boundary. + mask[62:0] := save_mask[62:0] AND XCR0[62:0] +FOR i := 0 to 62 + IF mask[i] + CASE (i) OF + 0: mem_addr.FPUSSESave_Area[FPU] := ProcessorState[x87_FPU] + 1: mem_addr.FPUSSESaveArea[SSE] := ProcessorState[SSE] + DEFAULT: mem_addr.Ext_Save_Area[i] := ProcessorState[i] + ESAC + mem_addr.HEADER.XSTATE_BV[i] := INIT_FUNCTION[i] FI + i := i + 1 ENDFOR -dst[MAX:256] := 0 - - +
immintrin.h
 - - AVX512VBMI - AVX512VL - Swizzle - - - - Shuffle 8-bit integers in "a" and "b" using the corresponding selector and index in "idx", and store the results in "dst". - -FOR j := 0 to 15 - i := j*8 - off := 8*idx[i+3:i] - dst[i+7:i] := idx[i+4] ? b[off+7:off] : a[off+7:off] + + XSAVE + XSAVEC + OS-Targeted + + + + Perform a full or partial save of the enabled processor states to memory at "mem_addr"; xsavec differs from xsave in that it uses compaction and that it may use init optimization. State is saved based on bits [62:0] in "save_mask" and "XCR0". "mem_addr" must be aligned on a 64-byte boundary. + mask[62:0] := save_mask[62:0] AND XCR0[62:0] +FOR i := 0 to 62 + IF mask[i] + CASE (i) OF + 0: mem_addr.FPUSSESave_Area[FPU] := ProcessorState[x87_FPU] + 1: mem_addr.FPUSSESaveArea[SSE] := ProcessorState[SSE] + DEFAULT: mem_addr.Ext_Save_Area[i] := ProcessorState[i] + ESAC + mem_addr.HEADER.XSTATE_BV[i] := INIT_FUNCTION[i] + FI + i := i + 1 ENDFOR -dst[MAX:128] := 0 - +
immintrin.h
 - - AVX512VBMI - AVX512VL - Swizzle - - - - - Shuffle 8-bit integers in "a" and "b" using the corresponding selector and index in "idx", and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). - -FOR j := 0 to 15 - i := j*8 - IF k[j] - off := 8*idx[i+3:i] - dst[i+7:i] := idx[i+4] ? b[off+7:off] : a[off+7:off] - ELSE - dst[i+7:i] := a[i+7:i] + + XSAVE + XSS + OS-Targeted + + + + Perform a full or partial save of the enabled processor states to memory at "mem_addr"; xsaves differs from xsave in that it can save state components corresponding to bits set in IA32_XSS MSR and that it may use the modified optimization. State is saved based on bits [62:0] in "save_mask" and "XCR0". "mem_addr" must be aligned on a 64-byte boundary. + mask[62:0] := save_mask[62:0] AND XCR0[62:0] +FOR i := 0 to 62 + IF mask[i] + CASE (i) OF + 0: mem_addr.FPUSSESave_Area[FPU] := ProcessorState[x87_FPU] + 1: mem_addr.FPUSSESaveArea[SSE] := ProcessorState[SSE] + DEFAULT: mem_addr.Ext_Save_Area[i] := ProcessorState[i] + ESAC + mem_addr.HEADER.XSTATE_BV[i] := INIT_FUNCTION[i] FI + i := i + 1 ENDFOR -dst[MAX:128] := 0 - +
immintrin.h
 - - AVX512VBMI - AVX512VL - Swizzle - - - - - Shuffle 8-bit integers in "a" and "b" using the corresponding selector and index in "idx", and store the results in "dst" using writemask "k" (elements are copied from "a" when the corresponding mask bit is not set). - -FOR j := 0 to 15 - i := j*8 - IF k[j] - off := 8*idx[i+3:i] - dst[i+7:i] := idx[i+4] ? b[off+7:off] : a[off+7:off] - ELSE - dst[i+7:i] := a[i+7:i] + + XSAVE + XSS + OS-Targeted + + + + Perform a full or partial restore of the enabled processor states using the state information stored in memory at "mem_addr". xrstors differs from xrstor in that it can restore state components corresponding to bits set in the IA32_XSS MSR; xrstors cannot restore from an xsave area in which the extended region is in the standard form. State is restored based on bits [62:0] in "rs_mask", "XCR0", and "mem_addr.HEADER.XSTATE_BV". "mem_addr" must be aligned on a 64-byte boundary. + st_mask := mem_addr.HEADER.XSTATE_BV[62:0] +FOR i := 0 to 62 + IF (rs_mask[i] AND XCR0[i]) + IF st_mask[i] + CASE (i) OF + 0: ProcessorState[x87_FPU] := mem_addr.FPUSSESave_Area[FPU] + 1: ProcessorState[SSE] := mem_addr.FPUSSESaveArea[SSE] + DEFAULT: ProcessorState[i] := mem_addr.Ext_Save_Area[i] + ESAC + ELSE + // ProcessorExtendedState := Processor Supplied Values + CASE (i) OF + 1: MXCSR := mem_addr.FPUSSESave_Area[SSE] + ESAC + FI FI + i := i + 1 ENDFOR -dst[MAX:128] := 0 - +
immintrin.h
 - - AVX512VBMI - AVX512VL - Swizzle - - - - - Shuffle 8-bit integers in "a" and "b" using the corresponding selector and index in "idx", and store the results in "dst" using zeromask "k" (elements are zeroed out when the corresponding mask bit is not set). - -FOR j := 0 to 15 - i := j*8 - IF k[j] - off := 8*idx[i+3:i] - dst[i+7:i] := idx[i+4] ? b[off+7:off] : a[off+7:off] - ELSE - dst[i+7:i] := 0 + + XSAVE + XSS + OS-Targeted + + + + Perform a full or partial restore of the enabled processor states using the state information stored in memory at "mem_addr". xrstors differs from xrstor in that it can restore state components corresponding to bits set in the IA32_XSS MSR; xrstors cannot restore from an xsave area in which the extended region is in the standard form. State is restored based on bits [62:0] in "rs_mask", "XCR0", and "mem_addr.HEADER.XSTATE_BV". "mem_addr" must be aligned on a 64-byte boundary. + st_mask := mem_addr.HEADER.XSTATE_BV[62:0] +FOR i := 0 to 62 + IF (rs_mask[i] AND XCR0[i]) + IF st_mask[i] + CASE (i) OF + 0: ProcessorState[x87_FPU] := mem_addr.FPUSSESave_Area[FPU] + 1: ProcessorState[SSE] := mem_addr.FPUSSESaveArea[SSE] + DEFAULT: ProcessorState[i] := mem_addr.Ext_Save_Area[i] + ESAC + ELSE + // ProcessorExtendedState := Processor Supplied Values + CASE (i) OF + 1: MXCSR := mem_addr.FPUSSESave_Area[SSE] + ESAC + FI FI + i := i + 1 ENDFOR -dst[MAX:128] := 0 - - +
immintrin.h
 - + \ No newline at end of file